Journalof BusinessandEconomics

Journal of Business and Economics Volume 5, Number 10, October 2014

Editorial Board Members: Prof. Jeong W. Lee (USA) Prof. Marc Matthias Kuhn (Germany) Prof. Mostafa M. Maksy (USA) Dr. Eugene Kouassi (Abidjan) Prof. Stela Todorova (Bulgaria) Prof. Georg Friedrich Simet (Germany) Dr. Athanasios Mandilas (Greece) Dr. Eugenia Bitsani (Greece) Dr. George M. Korres (Greece) Dr. Somesh K. Mathur (India) Prof. Iltae Kim (Korea) Dr. Masud Chand (USA) Dr. Maria Eugénia Mata (Portugal) Prof. Ulf-Göran Gerdtham (Sweden) Prof. Boban Stojanovic (Serbia) Dr. M. A. Sherif (UK) Dr. Gergana Jostova (USA)

Prof. E. Bruce Hutchinson (USA) Dr. Francesco Vigliarolo (Argentina) Prof. Myro Sanchez Rafael (Spain) Prof. Almira Yusupova (Russia) Prof. Milton Iyoha (Nigeria) Dr. Gergana Jostova (USA) Prof. Juan-Antonio Mondéjar-Jiménez (Spain) Prof. Adam Koronowski (Poland) Dr. Jonathan K. Ohn (USA) Dr. Adiqa Kausar (Pakistan) Prof. Alejandro Prera (USA) Prof. Richard J. Cebula (USA) Dr. Jamal Mattar (Belgium) Dr. Brian W. Sloboda (USA) Prof. Yezdi H. Godiwalla (USA) Prof. Jin Hyo Joseph Yun (Korea) Prof. Christian Seiter (Germany)

Copyright and Permission: Copyright©2014 by Journal of Business and Economics, Academic Star Publishing Company and individual contributors. All rights reserved. Academic Star Publishing Company holds the exclusive copyright of all the contents of this journal. In accordance with the international convention, no part of this journal may be reproduced or transmitted by any media or publishing organs (including various websites) without the written permission of the copyright holder. Otherwise, any conduct would be considered as the violation of the copyright. The contents of this journal are available for any citation. However, all the citations should be clearly indicated with the title of this journal, serial number and the name of the author. Subscription Information: Price: US$800/year (print) Those who want to subscribe to our journal can contact: [email protected]. Peer Review Policy: Journal of Business and Economics (ISSN 2155-7950) is a refereed journal. All research articles in this journal undergo rigorous peer review, based on initial editor screening and anonymous refereeing by at least two anonymous referees. The review process usually takes 4-6 weeks. Papers are accepted for publication subject to no substantive, stylistic editing. The editor reserves the right to make any necessary changes in the papers, or request the author to do so, or reject the paper submitted.

Database Index: Journal of Business and Economics (ISSN 2155-7950) is indexed by SSRN, CrossRef and Ulrich now. Contact Information: Manuscripts can be submitted to: [email protected], [email protected] or [email protected]. Constructions for Authors and Submission Online System are available at our website: http://www.academicstar.us/onlineupload.asp?shaction=show. Address: 228 East 45th Street, Ground Floor, #CN00000267, New York, NY 10024 Tel: 347-566-2153, 347-230-6798 Fax: 646-619-4168, 347-426-1986 E-mail: [email protected], [email protected]

Journal of Business and Economics Volume 5, Number 10, October 2014

Contents Industrial Organization SML Parking and RFID Technology

1709

Elina Ibrayeva, Terrence Sebora Exploring the Forecasting Process in a Fast Moving Consumer Goods Company

1727

Jesus Canduela, Robert Raeside, Ignazio Cabras Logistics Continuing Education: “Berufswertigkeit” and The Duisburg Model

1739

Matthias Klumpp, Stephan Zelewski, Rolf Dobischat, Hella Abidi, Martin Kowalski, Johannes Reidel The Implied Premium and Growth Strategy—Evidence from S&P 500

1754

Sue-Fung Wang, Yi-Cheng Shih, Xiang-Jun Lai Practical Aspects of Pre-evaluation of Eco-efficiency of Environmental Technologies

1773

Włodzimierz A. Sokół

Economic Development, Technological Change, and Growth The Temporal and Spatial Effect of Highways on China’s Economic Growth

1785

Xiugen Mo, Guangqing Chi, Charles Campbell Poverty Assessment in Low-, Medium- and High-Income Arab Countries: Yemen, Egypt and Bahrain

1802

Karima Korayem

Public Economics Matrices of Regional Redistribution of Argentinean Public Revenue and Its Political Appropriation Roque Ruarte Bazán

1817

Mathematical and Quantitative Methods Stochastic Asset Allocation Models

1837

Gavriel Yarmish, Robert Fireworker, Harry Nagel, Joe Thurm Energy Substitution in U.S. Electricity Generation

1845

Osei Yeboah, Saleem Shaik, Afia Fosua Agyekum, Julie Melikpor-Lee A Comperative Regional Analysis for Housing Demand in Turkey

1854

Selahattin Bekmez, Aslı Özpolat Reducing the Administrative Burden in Bulgaria: Single Entry Point for Reporting Fiscal and Statistical Information

1867

Diana Yancheva, Karmen Iskrova

Law and Economics Experiences and Challenges in Public Information Centers in the State of Morelos, Mexico

1872

Roque López Tarángo, Crisóforo Álvarez Violante, Silvia Cartujano Escobar, Selene Viridiana Pérez Ramírez, Paula Ponce Lázaro

Business Administration and Business Economics; Marketing; Accounting Integrating Sustainability Educationin to Business Curriculum: An Analysis of Existing Syllabi 1877 Zabihollah Rezaee, Saeid Homayoun Toward an Integrated Theory of Sustainability

1892

Dilip Mirchandani, Theodoros Peridis Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises

1902

Xin Jin, Song Chen, Jie Wang, Jinghua Xia Managing the Virtual University: A Real Experience

1917

Yannet Liliana Mesa Medina Convergence or Divergence between European HRM and American HRM

1923

Gurhan Uysal The Intervening Effects of Whistleblowing in Reducing the Risk of Asset Misappropriatio

1929

Mohamad Afzhan Khan Bin Mohamad Khalil, Anuar Bin Nawawi, Nurmazilah Dato’ Mahzan The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs

1930

Ricardo Garcia Ramirez, Gonzalo Maldonado Guzman, Maria del Carmen Martinez Serna Financial and Economic Factors Affecting the Entrepreneurial Characteristics of Small and Medium Family Enterprises Roberto Espíritu Olmos, Héctor Priego Huertas, Alejandro Rodríguez Vázquez

1941

Journal of Business and Economics, ISSN 2155-7950, USA October 2014, Volume 5, No. 10, pp. 1709-1726 DOI: 10.15341/jbe(2155-7950)/10.05.2014/001  Academic Star Publishing Company, 2014 http://www.academicstar.us

SML Parking and RFID Technology Elina Ibrayeva, Terrence Sebora (University of Nebraska, Lincoln, Nebraska, 68588, USA)

Abstract: Bruce Keller, general manager of SML Parking in Centerville, is considering integrating RFID technology into his current parking operations. SML is an off-airport parking facility located near Centerville Airfield, and SML currently competes with on-airport parking facilities and three other off-airport parking facilities. This case illustrates that small businesses, like corporations, have limited resources and multiple opportunities. Like many small businesses, SML must make a competitive decision as to whether or not to adopt new technology that impacts how efficiently and effectively customers’ needs are met. RFID technology is emerging in the parking industry. Because SML has no distinctive or relative competitive advantages, adopting RFID technology could help SML leapfrog the competition and become more competitive. The case demonstrates one method by which a small business can conduct research to improve its competitive decision-making. Key words: strategic choice; small business; opportunity research JEL codes: R, L, J, M “It will market itself…it will market itself!” Bruce Keller kept repeating this mantra of the proposed plan for integrating radio frequency identification (RFID) technology into his parking business. With a twinkle in his eye and a boyish grin splayed across his face, it was evident that this was Keller’s pet project. RFID technology would enable Keller to create a transactionless parking system where customers, after a free enrollment in the “FastPark” program, would be able to bypass interactions with parking attendants when entering and exiting a lot. Customers’ personal information would be saved in their individual RFID cards. This would allow SML Parking to automatically bill the customers’ credit cards and email receipts. The program would be similar to EZ pass programs available to individuals who routinely drive on toll roads. Tracey Watkins, an SML employee, was walking past Keller’s office when she heard him. “Excuse me Mr. Keller, did you say something?” “Oh, Tracey! I was just thinking out loud. I’ve been considering this project for years. Sit down and I’ll tell you more about it. I’ve already done some research,” he continued as he deposited a manila folder full of brochures, pamphlets, hand-written notes, and business plans onto his desk. “SML locations in other states have implemented RFID technology into their parking operations. I don’t have any hard numbers, but I visited one of these locations, and liked what I saw. I do not see the benefit of using RFID technology in our operations unless we can provide our customers with something our competitors don’t have—covered parking. This entire project Terrence Sebora, Professor, Department of Management, University of Nebraska Lincoln; research areas/interests: entrepreneurship and strategy. E-mail: [email protected]. Elina Ibrayeva, Professor, Department of Management, University of Nebraska Lincoln; research areas/interests: international management, strategy, organizational behavior. E-mail: [email protected]. 1709

SML Parking and RFID Technology

needs more research, would you like to help?” “Okay!” replied Tracey, somewhat interested. “Can you give me more details?” “Well, I want to start small and convert about 30% of our current lot to use RFID technology, and if that is successful, we can roll out the technology to the entire parking operation.” Keller had plans for his parking facility, but few employees in the company knew he was considering overhauling current operations. Ultimately, Keller’s two-pronged approach would convert a portion of his surface parking lot to RFID technology in conjunction with covered parking. “This Fast Park Software was created by and first proposed to me by Kent Persie, a successful parking entrepreneur. The software isn’t completed just yet, but Persie is committed to finishing it. There are other retail outlets that have completed software packages, but I’ve been working with Persie for a while. Both the completed software and a covered parking structure would require a substantial capital investment, which is why I want you to see if there really is a market for this project.” With that, he handed Tracey the manila folder filled with articles, grabbed his notepad, and left for a meeting without waiting for Tracey to respond.

1. Radio Frequency Identification “I’m not even sure I know what radio frequency identification is!” Tracey thought to herself as she walked out of Keller’s office, “much less how it would work in conjunction with parking. Would our customers even know what it is and how to use it?” Tracey leafed through the pages, and an article “What is RFID?” in something called the “RFID Journal” caught her eye. Radio frequency identification “is a generic term that is used to describe a system that transmits the identity (in the form of a unique serial number) of an object or person wirelessly, using radio waves.” Typical RFID tags, she learned, consist of a microchip that is attached to a radio antenna. Readers are used to retrieve data stored on RFID tags, and the information is then passed in digital form to a computer. RFID microchips can be embedded in plastic cards, and these cards can be read from a few inches to 30 feet away from a reader. After considering the idea for a moment, Tracey realized even her grandparents were accustomed to RFID technology. They had purchased a tollway transponder that allowed them to bypass traditional toll lanes. Readers situated above interstate lanes collect information from the transponder and automatically bill customers’ accounts—creating non-stop toll collection. Tracey continued skimming the articles from the manila envelope and discovered that RFID technology had been used with parking operations for years. TransCore had developed a SecurePass system through which RFID transponders could enable cars to enter a facility without input from a guard. Also, RFID technology was now used in airport parking facilities in larger markets. Back at her desk, Tracey began thinking about the other uses of RFID technology, and began to do a bit more research online. She was surprised to find that RFID permeated business-to-business transactions. One article noted that Wal-Mart began the first phase of its RFID implementation in 2004 when it began tracking pallets and cases of product using RFID technology. In retail settings, RFID tags carry Electronic Product Codes (EPCs). EPCs replace traditional barcodes and allow retailers to track and uniquely identify specific products in the supply chain, much like vehicle identification. Tracey learned that even her neighborhood Walgreens was a pioneer in using RFID in its supply chain management. It seemed RFID was everywhere! Tracey printed the articles and added them to what was beginning to be a tired-looking folder, thinking Keller would be interested to learn about the prevalence of RFID technology

1710


in the rest of the business world. She was starting to see how RFID had impacted the business world and how it could impact the parking industry specifically. In the parking industry, larger markets had already embraced the new technology and were adapting it to meet their specific needs. She continued thinking about how RFID would change SML parking operations as she returned the folder to Keller’s office.

2. SML Parking Overview SML Corporation has operations throughout the globe, primarily as a car rental company. SML of Centerville, Upper Midwestern State, unlike most other SML locations, is comprised of two strategic business units: car rental and parking. The only SML location in Upper Midwestern State with parking operations is located about two miles south of the Centerville Airfield. SML Parking is the oldest off-airport parking facility in the region with approximately 800 stalls of surface level parking. SML offers its customers full service parking, which includes complementary snow removal for more than three inches of snow, jumpstarts, tire inflation, and assistance to customers whose vehicles have been snowed in. Upon arriving at SML, customers take a ticket from a dispenser before passing through a gate and choosing a parking stall. A parking lot attendant then radios a shuttle bus driver, who picks up customers from their vehicles and helps transfer luggage onto the shuttle. Shuttle bus drivers note customers’ parking stalls on slips of paper and give these parking receipts to customers so vehicles can be easily located when customers return. Shuttle drivers drop customers off at the terminal door that corresponds with customers’ airlines, before picking up a load of returning customers and heading back to the SML parking lot. SML is open 24 hours a day and has a minimum of two shuttles operating at a time until the last flight of the night arrives, after which one shuttle operates until the first departure of the next morning. Upon returning to Centerville Airfield, customers either wait curbside until the SML shuttle arrives (this usually takes five to ten minutes) or they can ask SML’s customer service representatives at the airport service desk to page the shuttle to pick them up sooner. Customers are dropped off at their vehicles, after which they stop and pay at the parking attendant booth before heading home.

3. Financial Performance As Tracey continued research on the RFID project, more people around the office heard the news and became interested in the project. However, not everyone wholeheartedly embraced the idea. Wanda Burks, SML’s accountant, was an older woman who had endured the Great Depression and the burst of the Tech Bubble. She had seen tough times before, and the current economy had her concerned. After hearing of Keller’s lofty project plans, Wanda decided to approach him and marched into his office. “Hello Mr. Keller.” “Why, hello Wanda! How are you doing? Have you heard about my RFID idea? The entire office is talking about it!” “That’s actually why I came here. This Fast Park project sounds…interesting…but can SML really afford this project?” Keller was taken aback by her comment. The financial aspect of the project had crossed his mind, and he had always assumed a small loan and SML’s financial resources would be sufficient. “I’m glad you brought this up, Wanda. I’m on my way to meet with Trent Lewis about a broken ticket spitter, 1711


but would you be up for meeting with me Monday to go over some of the details?” Wanda agreed, and she went back to her office to prepare for the meeting. The next Monday, Keller and Wanda sat down and talked business. With its long, stable history in Centerville, SML was sitting on solid financial ground at this point, but Fast Park would require a large capital investment. They began the meeting by first looking over the financial statements of the parking portion of SML over the past three years (see Appendices A, B, and C).

4. The Project Keller revealed to Wanda that he planned to take out a $10,000 loan for the project to cover the cost of the software he was developing with Kent Persie. “Well, what other costs are you accounting for?” Keller proceeded to break down the project into two costs: hardware and software. “The hardware portion of the project is not as simple as the software portion, and I also want to make structural changes with the lot.” “We currently have 800 spaces in our parking lot, and I would like to add 200 more. There’s an unused grassy area and a gravel area that’s used for rental cars, which together equal about 50,000 square feet. I would like to pave these two areas and change the layout of operations,” Keller explained as he showed Wanda how the lot would change (see Appendix D). “Just in case RFID doesn’t go over well here, I don’t want to devote our entire parking operations to the project. Therefore, I’m planning on turning one of our current enter/exit gates into an RFID gate. I want regular parkers to have access to covered parking, so I want to add a separate gate between the RFID lot and regular parking lot. The parking attendant will control this gate. Because we’ll be consolidating our rental parking lot, we’ll add an operations entrance from the rental car lot to the parking lot. This will allow us to use unused spaces in the paid parking lot for rental cars without obstructing the flow surrounding the commercial entrance.” “You may have heard that I am not going to take on this project without covering the RFID portion of the lot. Right now I am looking into a durable, architectural fabric for the covering. It’s relatively cheap in comparison to traditional metal structures. It looks nice and can withstand our tough winters. It’s a similar fabric to what is used to cover Denver International Airport. The canopy comes in two main parts: the metal frame and the fabric. I’m looking to cover 300 stalls.” Keller continued, “We should probably invest in a new fee computer, which will run about $10,000. Recently, Tracey had a meeting with Trent Lewis, president of Forces, a parking hardware company. He sells parking hardware and knows what we would need to get this project done. I always call him when our equipment breaks. According to Trent, I could get a package for $20,000-$25,000. This would include the gates, metal detector, readers, ticket spitter, computer, circuit box, as well as pre-packaged, well-respected software. It would pay for pretty much everything except the lot pavement and the covering. If I chose one of these packages, they would also handle all of the customer billing, which costs about $600 per month to service,” Keller commented as he handed Wanda a table outlining each specific hardware cost of the project (Appendix E). Wanda interrupted Keller with an intriguing question, “So Mr. Keller, you’re going to pay $10,000 to finish the software with Persie, when you can get it included in a full package for $20,000-25,000? Are there a lot of RFID software options available?” “There are a few that specifically deal with parking,” Keller responded. He then handed Wanda a table

1712


outlining software company options along with Tracey’s notes (Appendix F). He continued to summarize Tracey’s meeting. “Apparently, Trent kept raving about netPark, boasting that it gave the most options for the cheapest price. We already have the deal with Kent Persie, though, and if Fast Park is successful here, we could franchise it nationally.” “What did Trent think of that idea?” asked Wanda. “He didn’t like it and he said that we were just reinventing the wheel.”

5. SML’s Competitors As an SML employee, Tracey never had to pay to park when she flew out of Centerville. But the current project made her wonder which company she would choose to go with if she did have to make that decision. On her way home from work, she took a lap around the relatively small airport property to scout out the options. When she got home, she checked out the competitors’ websites and began thinking about the competition in comparison to SML (Appendix G). Smith Airport Parking was a plain parking lot, but it seemed to have a brighter atmosphere than SML. From what she observed on site, Smith’s operations seemed very similar to SML’s operations. Smith’s website was easy to navigate, especially in comparison to SML, as SML’s website primarily focused on its car rental operations and parking information was difficult to find. Tracey was immediately attracted to Robert’s parking. Robert’s had huge, bright signs that made the parking lot noticeable and made it look trustworthy. She had never realized how much more appealing this lot was in comparison to SML. Robert’s website, just like its lot, was also bright and cheery and easy to navigate. Tracey made the loop on Centerville Airport property and discovered there were numerous parking facilities on the airport premises. The parking garage was located right across the street from the terminal doors and the airport surface parking lot was a short walk away from the terminal. The airport also had two other surface lots: North Long Term and South Long Term. Although these lots were on airport property, they were a short shuttle bus ride away from the terminal. The website for airport parking was uninteresting, but overall it was still easier to navigate than SML’s website. John’s parking was the final lot Tracey passed on her way home. It was located in a dingy area of town further away from Airport Drive than the other parking facilities. Customers were required to leave their keys with John’s employees while they were away. Unlike the other competitors, John’s did not have a website, and its shuttle didn’t run continuously—customers had to call to be picked up. She heard this usually required about a 30 minute wait for their customers. After considering the competition, Tracey developed a chart outlining her findings (Appendix H). Tracey was disappointed that SML seemed to lag behind most of its competition in atmosphere and website navigability. She thought about SML and realized that it shouldn’t surprise her that much. The computers she used at work were really old. Also, the ticket dispenser always seemed to be broken. When she met with Trent, he mentioned that the hardware used by SML is no longer produced. Trent admitted he kept some in stock just for Keller. All of the parking options seemed to have the same services available, including snow removal, jump starts, tire inflation, and shuttle services to and from the airport; however, not all parking facilities advertised these services. All of the parking companies were open 24 hours a day, and the majority would pick up and drop off

1713


customers directly at their vehicles. This research made Tracey wonder how SML remained competitive. She thought about its location, and determined that being just off airport property was SML’s primary competitive weapon. She then started thinking about the entire parking industry to see where SML fared nationally.

6. Parking Industry As she started searching online, Tracey thought about her first few trips to the Centerville airport. One summer, she went to a camp in Texas, so her parents dropped her off at the airport. Another time, when her family went to Florida for a month, they took a taxi, because they did not want to leave their car at the airport the entire time. Tracey had also heard about people splurging to take limos to the airport, and many of her friends had taken shuttles to Centerville Airfield from a university in a nearby town when they went home for Christmas. Tracey learned that the parking industry generates over 20 billion dollars of gross revenue annually. Due to increasing airport travel, airport parking was a growing segment within the parking industry, and automated parking had significantly decreased overhead costs for parking companies. One of articles Tracey found claimed that it could be difficult for off-airport parking companies to compete with on-airport parking provided by airports, primarily because of airports’ low cost of capital. The article also noted off-airport parking facilities’ pricing was generally similar to long-term parking on-airport, which Tracey noticed was true for the Centerville area. SML and its competitors were pretty well aligned with the rest of their industry. Nationwide, the majority of off-airport parking facilities were surface lots with enhanced services—including shuttles and “round-the-clock” staffing. Another article mentioned effective marketing and proximity to the airport as important factors when competing as an off-airport parking lot. Tracey searched for “RFID Parking” online and there were nearly a half million hits, indicating a substantial industry-wide trend. She remembered RFID technology was already in use in parking facilities in larger markets, particularly on college campuses and in downtown garages. The technology helped manage space availability. Tracey particularly liked the idea of the database capability that would come along with this technology. SML had no information on its customers, not even who they were, when they parked, or how long they stayed. She knew this type of database could provide endless capabilities for direct marketing.

7. The Technology After thinking about the industry as a whole, Tracey recalled her meeting with Trent Lewis. Trent and Tracey had talked for nearly two hours about the multitude of options Keller could choose from. Trent had owned and managed numerous parking facilities over the past twenty years, and Tracey valued his insight. Trent showed Tracey an example of a major circuit switch box that would essentially control the entire parking facility. It was a metal box attached to a wall, filled with wires and circuit boards. This connected the RFID readers to both the fee and database computers and the gates (Appendix I). Trent spoke about the pros and cons of different parking software programs. Tracey explained to him that Keller and Persie considered SML the flagship user of the yet-to-be-completed software. If the software was successful, Keller and Persie could trademark the Fast Park name and software and then market it throughout the country. Trent viewed the idea skeptically. When Tracey told Trent that Keller would only incorporate RFID technology in conjunction with covered parking, he became even more skeptical. In his experience, parking garages are generally superior to covered parking structures. Because garages supply additional floors of parking 1714


they can generally bring in more revenue than covered structures alone. The difference in cost is about $3,000 per stall for a metal covered parking structure compared to $10,000-15,000 per stall in a garage. Trent explained, “Parking garages are rarely profitable, but rather generally built out of necessity by companies and cities. This is likely especially to be the case for airport parking, as they are unable to charge premium hourly prices for downtown parking. Private garages are normally unprofitable and are left to deteriorate.” Before their meeting adjourned, Trent excitedly spoke about his upcoming trip, and Tracey asked if he would be parking at SML. Trent smiled and told her Robert’s had just finished upgrading its hardware and software through his company, and he would be parking there. Robert’s had installed netPark, the same software Trent recommended for SML. Despite his relationships with both companies, Trent seemed loyal to Robert’s because of its services and updated atmosphere. After reflecting on the meeting, Tracey decided that SML needed primary market data.

8. Market Analysis 8.1 Centerville Travelers’ Survey The next day Tracey went into Keller’s office, and found Keller sifting through the contents of a manila folder. “Oh, Tracey! Hi! How can I help you?” “Hi Mr. Keller! After talking to you about RFID technology the other day, I really think we should do some more research.” “But Tracey! Did I not show you all of the research I have done?” He pointed to another manila folder filled with articles. “Just look at all of this!” “That is a lot, but I still think we need to find out how our current customers and the community feel about this project.” “I guess you have a point.... Okay, I’ll create a survey to hand out to the customers, and you feel free to conduct any surveys that you would like as well.” That weekend, Tracey was flying out of Centerville Airfield—it would be the perfect time to collect random surveys of people at the airport! Surveying current customers would benefit SML, but conducting a random survey would help determine if there was a market in Centerville for RFID technology in parking operations. The night before her trip, Tracey came up with the following questions:  Where was the starting point of your trip?  Is this a business or personal trip?  How did you arrive at the airport?  Where is your car?  Why did you choose this option?  Have you ever considered off-airport parking?  Would you use off-airport parking for $8/$9 per day in Centerville?  Would you be interested in a lot controlled by RFID technology? Even if you have to pay $1/$2 more? The next morning, Tracey arrived at the airport early so she could begin the interviews. Once in flight, Tracey began to compile her results. Only 35.7% of the respondents were from the Centerville area, but of those

1715


from Centerville, 40% were interested in RFID. It was surprising to see that 50% of the patrons of the airport already used off-airport parking and from that number, 71.4% were interested in a program like Fast Park. In the comment portion of her survey, Tracey noticed two interesting trends. Business travelers’ number one priority was convenience. Business travelers traveled often and wish to minimize time travelling to and from the airport. Finally, Tracey noted the older generation was hesitant to accept the idea of RFID technology in parking operations. Tracey continued analyzing the results and summarized them in the graph located in Appendix J. 8.2 Current SML Customer Survey Even at the wedding, RFID continued to consume Tracey’s thoughts. She was actually looking forward to talking to Mr. Keller when she came back. Seconds after walking into the office after her vacation, Tracey’s telephone intercom beeped. It was Keller. “Tracey, can you come into my office? I have something to give you.” Tracey leapt from her chair and wondered what it could be. As she entered Keller’s office, he handed her another folder, filled with 372 SML customer surveys. “While you were gone, I gave our customers surveys. I gave every customer who returned a completed survey a coupon for one free day the next time they parked at SML. Can you analyze these results and tell me what you find?” “Of course!” Tracey replied. Before analyzing the results, she read over the questions:  How many times do you use SML annually?  Are you interested in covered parking?  What is an expected premium rate for covered parking?  Would you be interested in a transactionless “E-Z Pass” System? As she looked over the results, Tracey noted that 30% of customers who used SML once a year or less wanted covered parking, and 58.3% of the customers who used SML 36 or more times a year wanted covered parking. It was surprising to see that even customers who were not interested in covered parking still wrote down their suggested premium. Tracey developed a graph of customers interested in covered parking based on the number of times they parked with SML annually (Appendix K). She then developed a table depicting the results about the premium rate (Appendix L). Although one third of participants did not give a suggestion for a premium rate, many people thought it was sufficient to add $1 extra to the cost. After contemplating the covered parking results, Tracy turned to the main purpose of the surveys: customers’ interest in RFID in the parking operations. Tracey again analyzed the results based on the number of times per year customers parked at SML. She was surprised that customers who parked more times a year seemed less interested in RFID technology. She developed a graph based on these findings, seen in Appendix M.

9. Air Travel and the Expansion of Centerville Airfield Tracey knew the recent recession had negatively impacted the local airport, but she did not know how local traffic flows compared to national ones. Tracey looked at the statistics published by the Centerville Airport Authority and was surprised that the number of passengers decreased from 2008 to 2009 by 3.5% or about 152,000 passengers. Tracey then looked for national statistics and was astounded by what she discovered. According to the Bureau of Transportation Statistics, the number of passengers who flew domestically from

1716


2008-2009 decreased by 20.3%. Tracey remembered she had heard rumors about the possible expansion of Centerville Airfield. She found an article from the Centerville Chamber of Commerce. Tracey read “In October, 2008, the authority finalized its Terminal Development Plan, designed to utilize the current facility in the future through modifications and facility expansion. A comprehensive terminal expansion is years away and will be based on 4.5 million enplanements.” Tracey thought the plan was lofty and very far off. The number of enplanements in 2009, approximately 2.1 million, would have to more than double before expansion of the terminal.

10. Conclusion Tracey organized the mound of manila folders and handed them to Keller. She had added pertinent information to Keller’s initial research, and she hoped he would now be able decide whether or not to integrate RFID technology in SML’s current operations. As she walked back to her desk, she could hear Keller begin sifting through the folder’s various files: RFID technology, company background, financials, and competitors. As she headed home, Tracey passed by Keller’s office, and thought she heard something. She paused, recognizing Keller’s voice, “It will market itself! It will market itself!” She shook her head and smiled. Appendix A Balance sheet Cash Accounts receivable Accrual for bad debts Net accounts receivable Other inventories Prepaid expense Shuttle fleet Accumulated depreciation Net shuttle fleet Fixed assets Accumulated depreciation Net fixed assets Total assets Accounts payable Accrued expenses Payroll taxes payable Total liabilities Additional paid in capital Dividends Current year earnings Retained earnings/ Intercompany transfers Stockholder equity Total liabilities &equity

2009 (000s) $120.0 $122.0 $(10.0) $232.0 $28.0 $46.5 $1,375.0 $980.0 $395.0 $1,237.5 $1,237.5 $0.0 $701.5 $64.5 $57.5 $47.5 $169.5 $126.0 $236.0 $2,130.0 $(1,960.0) $532.0 $701.5

2008 (000s) $84.0 $128.5 $(10.0) $202.5 $13.5 $39.5 $1,300.5 $900.0 $400.5 $1,237.5 $1,237.5 $0.0 $656.0 $69.0 $65.0 $42.0 $176.0 $126.0 $127.0 $2,027.0 $(1,800.0) $480.0 $656.0

2007 (000s) $108.0 $111.5 $ (10.0) $209.5 $19.5 $41.0 $1,300.5 $851.0 $449.5 $1,237.5 $1,237.5 $0.0 $719.5 $102.0 $79.5 $40.0 $221.5 $126.0 $41.0 $2,031.0 $(1,700.0) $498.0 $719.5

1717


Appendix B Income statement Parking revenues Retail Travel industry referrals Monthly Total revenues Expenses Bus depreciation Bus insurance Bus repairs & maintenance Bus fuel Bus licensing & taxes Total bus expenses Travel agent commission Adv & coupon expense Other advertising expense Total advertising & commission Personnel expense Management salary allocation Cashier salaries Courtesy shuttle drivers Payroll taxes and insurance Total personnel & payroll Occupancy and administration Lot repair & maintenance Real estate taxes Rent Computer & supplies Total occupancy & administration Total expenses Net income

2009 (000’s)

2008 (000’s)

2007 (000’s)

$3,050.0 $1,264.0 $126.0 $4,440.0

$3,113.5 $1,251.0 $120.0 $4,484.5

$4,203.0 $1,225.5 $120.0 $5,548.5

$246.0 $99.0 $66.0 $195.0 $22.0 $628.0 $21.0 $29.0 $5.0 $55.0 $0.0 $165.0 $395.5 $696.5 $83.5 $1,340.5 $0.0 $104.0 $57.0 $120.0 $5.5 $286.5 $2,310.0 $2,130.0

$207.0 $115.0 $109.0 $361.5 $19.0 $811.5 $20.0 $28.0 $10.0 $58.0 $0.0 $165.0 $416.0 $665.0 $81.0 $1,327.0 $0.0 $78.5 $57.0 $120.0 $5.5 $261.0 $2,457.5 $2,027.0

$187.0 $126.0 $67.5 $282.5 $19.0 $682.0 $19.5 $28.0 $7.5 $55.0 $0.0 $165.0 $376.5 $693.5 $81.5 $1,316.5 $0.0 $55.0 $57.0 $120.0 $6.5 $238.5 $2,292.0 $3,256.5

Appendix C Cash flow statement Year end net income Cash flows from operations Non cash adjustments Depreciation & amortization Increase (decrease) in accounts receivable Increase (decrease) in payables Net cash flow from operations Other cash flows/disbursements Capital expenditures Net increase (decrease) in cash

1718

2009 (000s) $2,130.0

2008 (000s) $2,027.0

2007 (000s) $2,031.0

$246.0 $6.5 $(6.5) $246.0

$207.0 $(17.0) $(45.5) $144.5

$187.0 $11.5 $26.0 $224.5

$0.0 $2,354.5

$(599.0) $1,580.0

$(97.5) $2,158.0


Appendix D Current Lot

Future Lot

Appendix E

Project Cost

Item Repaving of parking lot: Preparation costs: Gates: Two needed AVI/ RFID reader: Metal detector: Ticket spitter: Fee computer: Calculates fee on exit Circuit box: Holds software that connects gate controls and billing computers Architectural fabric canopy Includes fabric, frame, and installation

Cost $1.50/sq foot $1/sq foot $4,000 each $2000 $800 $8,000 $10,000 $600 $1,677/ stall

Appendix F Other RFID Companies $25,000 + $600/ month netPark Software and Hardware package, including a computer that determines price for parking, RFID reader, panel, and bills customers for you FREE SKNet Available online, but with limited capabilities $12,000 CarNet Very complicated software program

1719

SML Parking and RFID Technology Appendix G Centerville Airfield & Surrounding Parking Facilities

Appendix H

Airport Parking in Centerville

Appendix I

1720

RFID Hardware

SML Parking and RFID Technology Appendix J

Appendix K

Survey of Centerville Travelers

SML Customers’ Interest in Covered Parking & Frequency of Parking

Appendix L What is an Expected Premium Rate for Covered Parking? Premium

None

$1

$2

More than $2

No Response

Number of Responses

52

74

52

32

162

Appendix M

SML Customers’ Interest in RFID Technology & Frequency of Parking

Teaching Note: SML Parking’s Choice Regarding RFID Technology Overview Bruce Keller, general manager of SML Parking in Centerville, is considering integrating RFID technology into his current 1721

SML Parking and RFID Technology parking operations. SML is an off-airport parking facility located near Centerville Airfield, and SML currently competes with on-airport parking facilities and three other off-airport parking facilities. SML is the oldest off-airport parking facility in the Centerville area, though it has no real distinctive competitive advantage. Keller only wants to convert about 30% of his lot to use RFID technology, and he will only do so with the addition of a covered parking structure over the same portion of his uncovered lot. RFID technology has permeated the business world but is just starting to be accepted in the parking industry, though none of SML’s competitors currently have RFID integrated into their operations. Target Audience This case can be used in strategic management and entrepreneurship courses. The scope of this case gives opportunity for students to draw upon and integrate knowledge from multiple areas of business. Teaching Objectives  Provide analysis of internal and external factors that affect how small businesses compete  Evaluate industry critical success factors to determine a firm’s position within the industry  Review financial information to assess project opportunities  Prompt students to consider ethical dilemmas in business negotiations  Help students consider strategy vs. operational effectiveness  Help students recognize and avoid complacency in small businesses Questions The following list of questions is designed to guide students through their analysis of the case: (1) How does SML create value? (2) What are SML’s sustainable competitive advantages? (3) What processes do SML managers use to make the decision on whether RFID technology is a worthy investment? (4)What are the critical success factors of the industry, and how does SML match up with the current critical success factors? (5) What environmental factors could impact the industry and the decision of whether to adopt RFID? (6) If Bruce Keller decides to go forward with the project, should he finish developing the software with Kent Persie or look elsewhere for a completed software package? Is Bruce Keller obligated to finish the project with Kent Persie? (7) Based on the future of the industry, should SML go forward with this project? Answers (1)How does SML create value? This question gives students the opportunity to discuss and analyze ways that firms create value. A resource inventory will help students distinguish between SML’s knowledge, assets, and capabilities. See Exhibit A for SML’s Resource Inventory. RFID, should SML go forward with the project, may be added to SML’s arsenal of assets. Students can also use Porter’s Value chain to record SML’s primary and secondary activities (Exhibit B). (2) What are SML’s sustainable competitive advantages? In order to assess SML’s competitive weapons, students must analyze the key characteristics of the firm’s core competencies: which are hard to copy, durable, superior, or trump-able? (Exhibit C).Considering the competitors outlined in the case, SML has no sustainable relative competitive advantages. While brand name is durable, it can be trumped by other local firms with other well known brand names. While location may be a relative advantage, other competitors are closer to the Centerville Airfield. SML’s lack of distinctive competitive weapons has forced the firm to match or undercut the majority of its competitors’ prices to continue to attract customers. (3) What processes do SML managers use to make the decision on whether RFID technology is a worthy investment? The following are some of the methods Keller, Wanda, and Tracey to seek out the information they needed to make an informed decision regarding RFID technology:  Seeking input from others in organization  Examining financial statements and considering the financing of the project  Researching multiple options of executing the adoption  Researching competitors (website, observation of facilities)  Seeking information on economic and social trends that could impact the project  Finding and reading relevant articles  Online research  Meeting experts  Viewing existing hardware

1722

SML Parking and RFID Technology Potential customer survey Customer survey  Researching relevant statistics  Organizing collected data (4) What are the critical success factors of the industry, and how does SML match up with the current critical success factors? One way to outline SML’s strengths and weaknesses as well as evaluate its opportunities relative to its position among its competitors is by developing a developing a strategy canvas (Exhibit D). Note that at the present, SML is not up to date with the rest of the industry and does not provide a unique service. Introducing RFID could enable SML to take possession of these success factors. (5) What environmental factors could impact the industry and the decision of whether to adopt RFID? This question is intended to push students to recognize the impact of change in the competitive arena, not only for large companies but also for small businesses. Specifically, students should recognize how SML is affected by the changing industry. Environmental factors that impact the industry may be analyzed using a PEST analysis (Exhibit E). Possible discussion points include:  Airport parking success is susceptible to increases and decreases in air travel. Air travel in Centerville decreased by 3.5%, or about 152,000 passengers, from 2008 to 2009, likely because of the recession. However, Centerville was largely insulated from the effects when the statistics are compared nationally.  The Centerville airport may expand, but this would likely be years away as enplanements must nearly double before this expansion takes place.  RFID prominence is significant in this discussion. Students should have opposing viewpoints on whether or not the growing prevalence of RFID is important for SML to consider. The questions raised may involve: o Does this change affect SML? o How soon and to what extent does this affect SML? o Must SML implement RFID now—or should it wait? o Could it be better to not implement RFID considering the local environment in which SML competes? (6) If Bruce Keller decides to go forward with the project, should he finish developing the software with Kent Persie or look elsewhere for a completed software package? Is Bruce Keller obligated to finish the project with Persie? Students should first discuss the ethical implications of finishing the software with Persie or choosing the packaged software. Three possible perspectives are:  Acting for SML, Keller has the responsibility to make the most profit for the company and has the right to choose the software that 1) has already been proven in the market, 2) may turn out to be a lowest cost investment.  Keller should value his relationship with Persie and the expectation Persie has to continue on the project. Therefore, Keller should continue with Persie, even if it could cost more to finish the software, or result in a lower cost product.  More information may be needed to figure out the quality of Persie’s software—will the features it provides compare favorably with the packaged version? What is the possibility that it could prove to be a good investment? How realistic is the franchising option? SNL should investigate further prior to making this decision. The view each student takes is likely based on which personal beliefs each student brings to the case. It is evident here that a manager’s (Keller’s) personal values could have a big impact on the outcome of the project. (7) Based on the future of the industry, should SML go forward with this project? If RFID and/or covered parking is implemented, it may increase customer recognition of SML and eventually cut costs. Students should contemplate the advantages and disadvantages of the project. Exceptional students will identify the other options that Keller has, such as not having covered parking and waiting to implement the project. In assessing attractiveness of the investment the pay-back period of the project should be calculated. See Exhibit F for estimates of initial investment. The revenue from the project may be determined based on the best case scenario. For the best case payback scenario, students should assume that all 300 covered stalls will be used 365 days a year (Exhibit G). Dividing the investment of the project by the revenue of the project, a 1.4 year payback period should be found. It must be emphasized that this is the best case revenue, not the most likely. To give a clear comparison to existing option, the payback period for choosing a packaged software and hardware must be calculated. The ongoing $600 monthly payment is not included for simplicity (Exhibit H). This decreases the payback period to 1.376 years. Because SML has no distinctive or relative competitive advantages, adopting RFID technology would help SML leapfrog the competition and become more competitive—given that Keller and the rest of management make smart decisions regarding the cost and quality of the implementation.  

1723

SML Parking and RFID Technology Conclusion This case illustrates that small businesses, like corporations, have limited resources and multiple opportunities. Being aware of the opportunities requires thought, and creativity, of the managers. Executing these opportunities requires planning and research to figure out the costs and benefits. It is important to note that despite best efforts to collect data required to make a decision, some information required may be ambiguous or unavailable, making estimates and guesses unavoidable. In part due to this uncertainty, a manager’s personal values could significantly influence on the outcome of the project. Adopting RFID technology could give SNL a competitive advantage relative to all other local parking facilities. However, it is how well (or poorly) RFID is adopted—or whether it is adopted at all—that will determine its ultimate effect on the company. Exhibit A Knowledge 40 years experience

Assets Location (proximity to airport) Shuttles Brand Reputation Radios Ticket Dispenser RFID?

Capabilities Customer Service Snow Removal Tire Inflation Parking Space Receipt Ease of Entry/Exit

Exhibit B

Exhibit C Competency

Hard to copy?

Durable?

Superior?

Can be trumped?

Knowledge and experience

Yes

Yes

Possibly

Yes

Customer Service

No

No

No

Yes

Snow Removal

No

No

No

Yes

Tire Inflation

No

No

No

Yes

Location (proximity to airport)

Yes

Yes

Possibly

Yes

Shuttle Frequency

No

No

Yes

Yes

Receipt with space

No

No

Possibly

Yes

Brand Reputation

Yes

Yes

Yes

Possibly

RFID?

No

Yes

Yes

Yes

1724


Exhibit D

Exhibit E  Security Airport expansion Healthcare reform

Political Increasing RFID prevalence Paperless parking transactions

Economic

Pest Analysis

Recession/Boom # Flights/Travel Fuel Prices Shuttles Airplanes Cars

Technological

Trend to travel more Older generation hesitant to accept technology

Social

Exhibit F Investment Hardware Repaving Canopy Software Total:

$29,400 Total of all needed items $125,000 Need to repave 50,000 sq. feet (page 8) $503,100 Need to cover 300 stalls (page 10) $10,000 If fulfilling the agreement with Persie $667,500

1725


Exhibit G Annual Revenue Amount of days Used Covered New Stalls Covered Current Stalls Total:

365 Assuming that the lot will always be filled 200 at $6.00 $1 premium determined from Exhibit 15 100 at $1 SML will already be receiving $5, thus only the premium is used $474,500 Exhibit H

Investment NetPark Repaving Canopy Total:

1726

$25,000 Includes both hardware and software $125,000 Need to repave 50,000 sq. feet (page 8) $503,100 Need to cover 300 stalls (page 10) $653,100


Exploring the Forecasting Process in a Fast Moving Consumer Goods Company Jesus Canduela1, Robert Raeside1, Ignazio Cabras2 (1. Employment Research Institute, Edinburgh Napier University, Edinburgh EH14 1DJ, UK); 2. Newcastle Business School, Northumbria University, Newcastle-upon-Tyne NE1 8ST, UK))

Abstract: The aim of this paper is to explore the forecasting function in a FMCG company, in order to understand actors’ functions and behaviour within the system and the impact of these on forecasting performances. The study intends to expand knowledge related to forecasting performances in complex situations and to address the “theory-practice gap”. The authors develop their study by using the case study of a major UK brewery, by analyzing the company’s documents and by conducting a participatory exploration within the company. Ten key personnel were approached and interviewed with in-depth semi-structured interviews, while four pivotal managers took part in unstructured interviews. Results provide empirical insights about how the forecasting function is perceived in terms of usefulness, the process and also with regard to the environmental elements that may have an impact on its performance. Findings suggest that changes in the marketplace for FMCG companies, and the ever-increasing negotiating power of large customers in particular, have shaped their forecasting strategy. Recommendations are made for FMCG companies to focus less on forecasting at SKU or product levels and to develop more reactive systems, more responsive to customer demand but with reduced focus on building up high stock levels. Key words: fast moving consumer goods; forecasting process; interaction among departments; brewery industry JEL codes: L1, L2, L6

1. Introduction The Fast Moving Consumer Goods (FMCG hereafter) market is characterized by high volume selling often with low profit margins. Baron et al. (1991) observed that FMCG are typically low priced items that are used with a single or limited number of consumptions and are normally sold in large numbers so the cumulative profit can be large. According to Bourlakis and Weightman (2004), the FMCG sector contributed in excess of £125 billion (8% of the GDP) to the UK economy and provided employment for over 3.2 million people (16% of the total UK workforce).

Jesus Canduela, Ph.D., Lecturer in International Business, School of Management, Edinburgh Napier University; research areas/interests: business and employment. E-mail: [email protected]. Robert Raeside, Ph.D., Professor, Chair in Accounting and Statistics, Employment Research Institute, Edinburgh Napier University; research areas/interests: forecasting, demographics and social networks. E-mail: [email protected]. Ignazio Cabras, Ph.D., Reader in Economics, Business and Management, Newcastle Business School, Northumbria University; research areas/interests: regional growth, economics, business and management. E-mail: [email protected]. 1727

Exploring the Forecasting Process in a Fast Moving Consumer Goods Company

FMCG manufacturers tend to produce a relatively small number of goods, but these goods are batched and bundled in many different ways resulting in a large amount of Stock-keeping Units (SKUs), and this aspect adds complexity within the market. In recent times, pressures generated by more competitive markets have increased the need to improve forecast accuracy (Sanders & Manrodt, 2003). Therefore the role of forecasting is perceived as critical for FMCG companies to operate efficiently and to react to changes in demand, whether through customer actions or wider environmental pressures. This process goes beyond the provision of individual forecasts and should encompass a holistic approach in terms of how forecasting is valued, what systems are used to support forecasting, and the economic benefits of forecasting. Wacker and Lummus (2002) commented that “forecasts alone have little or no value, but what is important is how they are used in making managerial decisions” (p. 1014). Due to the large amount of goods (SKUs) produced by each FMCG manufacturer, companies now tend to move to a “just in time supply” or to a “react on time” approach. The objective is to keep the goods in a generic form in order to pack them according to the SKU requirements of the retailer. By doing so, long term predictions become less relevant. This situation shapes the forecasting function and leaves little room for technical improvement. However, this situation requires also more advanced management systems and little research has been done so far in order to understand the management of forecasting (Wacker & Lummus, 2002). FMCG organizations are often suppliers to the penultimate consumer, as they usually supply retailers who sell products on to final consumers. These customers/retailers have increased their negotiating power by demanding low prices, extended shelf life of goods as well as flexible deliveries. The balance of power between demand and supply has changed in favour of these customers/retailers, whose relative importance in the negotiation has been increased by mergers and acquisitions amongst themselves. This is why forecasting has become critical for the effective functioning of manufacturers with regard to satisfying demand, as widely recognized within literature (Sanders, 1995; Van Wezel & Baets, 1995; Tanwari & Betts, 1999; Adebanjo, 2000). Nevertheless, it is difficult for FMCG companies, particularly for those large companies operating in bureaucratic structures, to be flexible and reactive enterprises. Traditionally, each division within a company had to produce its own forecasts and these were then aggregated to give an overall company forecast. In addition, each division set their own planning targets, and these targets tended to be pessimistic in a bid to make targets more competitive at a corporate level. This has resulted in large scale re-organizations within FMCG companies, with traditional production-based forecasts becoming less relevant, and more autonomy and power transferred to sales departments, who could negotiate directly with their powerful retailers. Organizations hence have become more “outward facing”. Little work has been conducted on the analysis of forecasting systems in FMCG organizations, generating some misunderstanding about where forecasting fits in the wider organizational requirements (an exception to this is the case study provided by Adebanjo, 2009). As a consequence, theories and advanced statistical techniques produced by academics are rarely used by practitioners, in this creates situation a “theory-practice gap” (Armstrong & Fildes, 2006; Mahmoud et al., 1992; De Roeck, 1991). However, advanced statistical techniques frequently fail to embrace the complex environment in which forecasts are developed, and do not reach the level of accuracy expected from them. Hence, since little utility appears to be gained from advanced statistical techniques chosen to optimize forecast accuracy (Canduela, 2009), it seems to be better to focus on the management side of the forecasting function, recognizing the need for fast-time information and flexibility in planning. 1728


In this paper, we use a case study based on a major Scottish brewery to explore the impact of changes occurred on the forecasting function within the retail environment. The paper comprises of five sections, including the introduction. Section two provides an overview of the company selected for this case study and the environment in which it operates. Section three describes research methodology and data analysis, Section four discusses findings in the light of the forecasting function and generation of errors. Section five concludes.

2. The Company and Its Environment The British beer industry is rooted in a tradition characterized by a strong “pub culture”, with many characteristics which mark it off as different from the beer markets of other countries. UK off-trade represents around 33% of the country’s beer market, a proportion which differs significantly from those of other countries such as France (61%), Germany (65%) or the USA (70%)1. This situation is slowly changing with the off-trade sector is growing and the on-trade declining. The company selected for this study is located in the Eastern part of Scotland. In early 2000, the company was one of the leading British suppliers with regard to off-trade sales, accounting for 27% of the UK total off-trade beer market (2004). It was probably the last major brewery in the country, with over 300 SKUs and about 150 main customers, employing 200 staff and generating £694 million in turnover and £45 million in profit. Pressures by customers, the volatile nature of the market and aggressive marketing campaigns forced the company to develop new brands and to expand its marketing strategies, with about 150 new SKUs created each year. In addition, according to internal reports provided by the company, recent figures show a 7% increase in the total sells for the sector, with a 2.4% on-trade decrease and a 6.6% off-trade increase. The characteristics and environment of the sales structure at the company are those of a FMCG centred business where complexity, volume and variety are major issues. Business functioning in this environment copes with a number of challenges such as the emergence of more powerful customers, customer consolidation, stiffer competition, promotions, unpredictable orders, changing pricing policies etc. An important challenge in the business environment is being able to deal with and adjust to current trends in consolidation amongst customers. Examples are the acquisitions of TTS stores, operated by Tesco, and of Jaecksen and Bell from Sainsbury’s in 2004: as a result, some of the customers of these buyers companies became larger in size and more powerful in terms of contracting, squeezing the suppliers’ profit margins considerably. Customer consolidation represents another particular concern, given the presence of different price structures for different customers. After consolidation, the new resulting organization expects the new price structure to contain the smallest price among those of the businesses involved in the merger. As a result, the profit margin decreases although the market continues to have steady growth, meaning that the new breed of large customers acquired a much stronger stock associated with the capability to set the final price to be paid for the product at the shop, gives them as retailers more control over consumers’ behaviour and limits the effectiveness of marketing campaigns conducted by the FMCG company. This complex trading condition is exacerbated by the fact that goods produced by FMCG businesses have a relatively short life span and need to be consumed shortly after production. Hence retailers have to purchase products with the longest shelf life possible. With these very stringent requirements, companies must be extremely efficient and accurate with their production and storage times. For instance, if the average can of beer has a life 1

Source: BBPA 2013. 1729


span of 28 weeks, then the retailer would demand it with at least 20 weeks of shelf life remaining, leaving the company with only 8 weeks to produce, package and trade that can. The importance of forecasting is illustrated by the company examined in this case study. Estimations based upon analysis of current stock holding shows that each percentage point improvement in forecast accuracy means a reduction in stock holding needs by 2,300 hectoliters. But these figures are just a proxy: Gormley and MacMillan (2002) showed that the MAPE (Mean Absolute Percentage Error) was around 15% for total beer and 27% for major brands. Mentzer and Cox (1984) and Mentzer and Kahn (1995) reported that typical MAPEs across business sectors were less than 26% with an overall average of 15%. Having said this, the case company had difficulties with assessing forecast errors. These difficulties generated large variations in errors with regard to SKU and compounded by poor record keeping. The latter issue is not surprising, as poor record keeping related to forecasting applies to most companies (Whinkhofer, 1996). Statistical analysis of forecasting conducted by Canduela (2009) and other private consultants employed by the company detected large errors mainly due to the company’s over-forecasting and, more worryingly, related to the company’s under-predicting demand. The company was then far from satisfied with its forecasting performance and recognized that forecasting inefficiencies were directly hindering their competitive strategy. Further investigation of the forecasting system was then sought. Recalling the description made earlier in the paper, the authors considered that any development was unlikely to be achieved by selecting improved methods or with the use of better quality data. Instead, the authors identified an urgent need to explore and examine the soft side of forecasting within the companies, by focusing on the roles, objectives and actions of those taking part in the forecasting system.

3. Research Approach and Preliminary Analysis An observational study was conducted to identify key-figures within the company’s personnel. These figures were approached and interviewed with in-depth semi-structured interviews, with the aim of understanding how the forecasting functions worked and to identify any operational problem associated with the forecasting function. One of the authors worked within the company and reviewed workflow documents, identifying potential key members of staff. Then exploratory unstructured discussions were conducted. A first set of discussions were with the Promotional Analysis Manager. This person was selected due to his background (Statistic and Operational Research), to his interests in research and on the basis of his position which implied dealing with all the company’s departments. The latter aspect ensured sufficient knowledge of the organization’s functioning and made him a valuable support in the selection of interviewees. The exploratory unstructured discussions provided two outcomes: firstly, a general understanding of how forecasts are generated and communicated within the company; and secondly, a comprehensive overview of the company’s structure, to be used as a reference in order to identify all areas of influence within the forecasting function and process. The most suitable individuals in each relevant department were then selected for interview. Other exploratory discussions took place with personnel from Customer Services, Operations Department, Trade Marketing Department and with the Planning and Forecasting Manager. These discussions further improved the authors’ knowledge and understanding of the forecasting system within the company. Forecasting procedures within the company started by using just recorded values from sales from previous periods, with procedures implemented and computed by a statistical package called PROCAST (Cube Software

1730


Limited, 2001). Forecasts were adjusted to the annual target: if the target was not met in the current period, then the following periods of the year would have been amended to meet the annual target (even when this target would have been unfeasible), just to increase pressure on sales. This suggests that forecasts were being used as ‘political’ tools to control sales staff, an aspect frequently observed by Makridakis and Wheelwright (1989) in their study. The forecasts were then presented to managers in monthly business review meetings. The base forecast was reviewed and then communicated to the Marketing, Finance and Operations departments. A top-down review of the base forecast was transmitted to the Marketing and Finance departments on monthly basis. Conversely, operative staff received a bottom-up view and updated this information daily. The scheme is displayed in Figure 1. IPM (Integrated Performance Manager) Business Review Process

Account Activity Account Managers (Agreed/Proposed Events) Brands M arketing Plans

Top-down view

Brands marketing M ONTH LY

BASE FORECAST

Finance

Financial Forecast

M ONTH LY

Bottom-up view

Actual Sales Past/Present/ Future

Company Review

Operations DAILY CHANGES

M onthly Figure 1 Forecasting Process in the Case Company

The exploratory interviews indicated that a consolidated forecasting system was used throughout the company and that it produced the same type of forecasts and accuracy measures (at different levels of aggregation) for all departments. This consolidated forecasting system was part of the Integrated Performance Manager (IPM), a process which includes the generation, circulation and modification of the forecasts throughout the company. The system presented a number of insights and specificities as listed below:  From the production side, the Operations Department was concerned about brewing constraints associated with volumes of liquid produced and required packaging. The department needed a rolling brewing forecast (three weeks ahead) and a rolling packaging forecast (one week ahead) instead of the periodical forecast produced by the forecasting team. An attempt was made to implement the three week rolling forecast throughout the company, but this procedure met strong resistance and a final rejection from the Finance department, which refused to change their practice related to both financial year and period approach.  Both Operations and Planning Departments were concerned with stock issues. The main issues related to identifying the minimum stock requirements, particularly when stock failures affected customer service.  There was a general need for alternative measures with regard to accuracy. Forecast accuracy within periodical reports was calculated by comparing actual sales with Frozen Forecast, set once at the beginning of each period. This forecast could have been altered significantly during the period affecting both brewing and production planning; hence it frequently failed to meet Operations’ requirements. As a result, period forecast accuracy did not reflect the levels of efficacy and efficiency of the Operations department.

1731


The forecasting champion, a figure that Fildes and Hastings (1994) and Mentzer et al. (1997) suggested as essential for successful forecasting in large organizations, did not exist within the case company. The forecasting manager was identified in the company as the champion but his/her role was at a technical level, with its only functions being to control and supervise the forecasting production and its distribution.  Probably the main outcome of this background research was the identification of a wide range of needs among different departments with regard to the forecasting function. The information gathered from preliminary meetings and from the general review of the company enabled the authors to identify and select ten key figures for in-depth interviews. The interviews were planned so that interviewees covered all the important areas of the company which were relevant to the forecasting function. Information about the interviewees is displayed in Table 1. Interviews covered all areas of interests, with interviewees frequently expanding on more than one issue at a time and-most importantly-connecting these issues in a functional manner. A set of specific questions were tailored in order to fit with the role and characteristics of each interviewee. Responses were examined to understand which types of data, estimates or plans are used within departments; how and for what this information was used inside the units, and what was transmitted from each unit to another and the rest of the company. With regard to this, three main issues were explored: (1) the perceived value of the forecast; (2) the level of importance of the forecast and (3) forecasting requirements. Fifteen questions were common to all interviews, while specific points were elaborated upon in discussion. 

Table 1 Position

Demographic Information about the Interviewees.

Business Planning Manager

Department Trade Marketing and Planning Trade Marketing and Planning

Promotional Analysis Manager

Trade Marketing and Planning

Forecasting Manager

Account Director

Trading Wholesale

Business Account Executive

Trading Grocery

Account Director

Trading Grocery

Business Account Manager

Trading Convenience

Demand Planner Demand Plan Manager Financial Planning Manager

Customer Service and Operations Customer Service and Operations Finance

Function

Years in the company

Age

Management of producing the forecast.

6

42

Produces business plans using forecasts.

8

40

4

27

3

31

5

28

15

40

5

30

8

37

Coordinates the team that organizes production.

3

40

Produces and manages the economic business plan referring results to the Central Plan.

1

30

Managed issues affecting the demand and produces analytical support to other departments. Leading a team dealing with the client/account. In charge of the everyday customer relations and informing the Business Account Manager of evolving trading negotiations. Leading the team dealing with the client/account. Assists and reports on major clients to the Account Director. Plans the production of demanded and forecasted beer.

The interviews lasted around one hour and were digitally recorded. All the information collected was then treated as confidential. Answers were analyzed in the light of what was the function or purpose of each department as understood by interviewees. The results are illustrated in Figure 2.

1732


Brewing

Finance

Trade marketing and planning

Case company Sales

Customers

Customer services and operations Figure 2

Forecasting Interaction as Perceived by Department

Views expressed by interviewees about forecasting were classified according to role and department. The information gathered from each interview transcript was then transcribed onto “post-it” notes. Phrases were kept to around 10-12 words, retaining the language of the person providing the account. This exercise generated over 400 post-its which were comprised into groups and, after several iterations, were classified and named according to their characteristics, similarity of form and statements. After groping post-its, an initial model of the forecasting function was developed; its scheme is displayed in Figure 3. PROCESS

FORECAST ENVIRONMENT

•Culture Change •Purpose

•Promotions

•Horizon •Data Sources

•Customers USEFULNESS

•Business Environment

•Production •Perception

•Supply Chain •Organisation

Figure 3

Flow Review

•Bias •Accuracy

Measures Methods Causes

Model of the Forecasting Activity

The main components identified by the forecasting model were named as the “Process” (production forecasting function and related requirements), the “Forecast Environment” (internal and external elements affecting forecasting activity) and the “Usefulness” (perception of the forecast by users, in terms of value and accuracy). These components are not as simple as those portrayed in Figure 3, but are intertwined within a complex system revolving around the volatile business environment. Next, the display of Post-Its was reorganized using the generated framework and, secondarily, according to department and interviewee. This enabled the authors to examine the information sorted by individual, department

1733


and issue, facilitating a cross-case analysis. The number of post-its by individual and issue were counted to compare the responses between the departments, as shown in Table 2. Given that the number of individuals interviewed in each department was uneven, it was important to consider the mean number of responses (per interviewee) for each department. As a result, Customer Services and Operations departments reported the highest number of issues with 46 per interviewee, immediately followed by Trade Marketing, Planning and Trading departments. On the other hand, Finance scoring 24 had the lowest mean per interview. With regard to the forecast process, departments appeared to be more interested in issues related to Process and Environment components rather than Usefulness. Customer Services and Operations, with an average of 22.5 related issues, were the departments with the highest number of mean responses related to Process. The proportions of responses by item were similar for Trade Marketing and Planning, Trading and Customer Service, and Operations departments, but different from the responses reported by the Finance department. The role of Finance within the forecasting function appeared mostly analytical and rather limited: their role was to convert volume forecasts into financial ones. Hence components such as Usefulness and Environment were of little relevance to them. Table 2

Mean Number of Responses by Department (Column Percentages of Statements for Each Element by Department Are Shown in Brackets)

Responses by Department

Trade Marketing and Planning

Trading

Customer Service and Operations

Finance

Process

42 (37%)

63 (44%)

45 (49%)

18 (75%)

Usefulness

29 (26%)

28 (20%)

25 (27%)

3 (12%)

Forecasting Environment

42 (37%)

52 (36%)

22 (24%)

3 (12%)

Total

113 (100%)

43 (100%)

92 (100%)

24 (100%)

4. Analysis of Findings and Causes of Errors Related to Forecasts The framework generated from the initial analysis (Figure 2) was used to investigate relevant issues raised in the interviews. By starting with the component “process”, responses from interviewees clearly indicated that each department had its own needs and motivations, and that departments behaved as independent entities. For instance, the Sales department paid considerable attention to their customers but very little to forecasts provided by the company. Due to their optimistic view in terms of sales and their increasing desire to satisfy customers, they created a situation characterized by inflated forecasts: the company top-down forecast was always about 10% to 15% lower than the sales produced bottom-up forecast. In addition, a general lack of communication, associated with the variety of different needs expressed by departments, created a situation where duplication was a common issue. Interviews provided a number of examples where two different departments conducted the same analysis but tailored to their own needs. The most common result was the generation of far too many reports, with most of them not specific enough to satisfy the needs of any department. This highly judgemental way of forecasting, where figures are continually adjusted and influenced by a variety of people on the basis of different reasons, means that a large amount of information was lost in the process and additional but much less relevant information was artificially created. More problems arise when budget figures produced by the Finance department are taken as targets by other departments, which frequently question their legitimacy by mentioning the incorporation of artificial information.

1734


With regard to “forecast environment”, it could be defined according to three main aspects: (1) the existence of powerful customers, (2) the effect of promotions, especially during Christmas period, and (3) change in business culture. The forecast environment proved to be a very complex environment due to constant adaptation and changes, where processes were discontinuous and disorganized while new efforts were made to make forecasting more structured. Each department “privately” used their own version of the company forecast: this situation generated tensions which created a general lack of trust between departments, with a detrimental impact on forecasting. Interviews indicated that the level of complexity increased even further when powerful customers were considered. Complexity increased because customer accounts differed significantly from each other. A few large customers represented a large proportion of the company’s trading, but each of these powerful customers had very different trading styles. Some customers had a network of depots around the country used to store goods and then to deliver these to different branches. Other customers worked with a just-in-time delivery approach to reduce storing costs: this approach made these customers more demanding in terms of ordering timescales. Due to their different storage and distribution policies, customers’ behaviours caused diversity among departments’ ordering patterns. In particular, departments were receiving information directly from customers, with some providing updated and reliable shop sales data while others did not. Hence the existence of extremely powerful customers with very different ordering styles contributed greatly to the complexity of the environment. For instance, customers would ask for 22-24 weeks shelf life when they were holding only 3 weeks’ worth of stock, expecting to use beer to attract people into shops. Promotions and special events, e.g., major sports events such as the World Cup or festivities such as Christmas, represented complex periods for generating forecasts and were difficult to incorporate into production schedules. However, these events were identified in the interviews as key-aspects within the forecasting process, due to the large proportion of total sales they generated. To cope with difficulties, the company tried to develop and implement a database of promotions and special events called “Events Database” (ED), hoping to incorporate future events in the database and to use it for planning. As the interviews progressed, it was clear that the supply chain side of the business was under continuous stress, mostly due to the size of the portfolio and uncertainty related to the sales process. This tension had resulted in a situation where the supply chain side, including production, of the business became separated from the commercial part. To better adapt to an ever-changing environment, the company implemented a number of structural changes, in order to make a traditional business, as it was, trade more efficiently in a very modern market. The result was a massive business restructuring which mostly affected the Marketing (on the commercial part) and the Supply Chain (on the production part) departments. With regard to Marketing, the objective was to introduce a top-down approach to forecasting by introducing the ED. However, Sales personnel accepted this change with disbelief. This is because the ED tried to take over the task of forecasting events from Sales, using an extended database of historical sales data to produce forecasts. The main task was then to persuade sales staff to do things differently. With regard to the Supply Chain, the main change was to hold more liquid, but less stock of final packaged products. This implied a more agile and effective packaging process, with savings made in terms of storage costs, since storing liquid was easier and cheaper than storing final packaged products. The distribution system changed 1735


too, with 40 existing depots substituted by three larger Regional Distribution Centres, making a reactive system more feasible. Finally, the perception of “usefulness” with regard to forecasts by users appeared to indicate that the forecasting process within the company was a sort of ‘political weapon’. There were frictions among different parts of the business. In particular, Sales and Brewing departments argued over where the forecasting function should be located. Both departments were skeptical about the usefulness of the forecasting function. With regard to forecasting bias, all interviewees appeared to understand the implications derived from overand under-forecasting, and only Sales personnel stated a preference for over-forecasting. In order to control the over-forecasting tendency, the company introduced the Margin Dilution System. This system penalized sale sector budgets when all volume forecasted remained unsold, resulting in the cost of beer forecasted but not sold deducted from their account profits. Interviewees from the Sales department commented that Margin dilution “is good because it shapes the behaviour of the forecaster”, and that they tended to consider over-forecasting better because “that is my nature” (the former quote was by an account director while the latter was made by an account executive). Conversely, staff from the Finance department appeared not to engage much in this system. An interviewee said: “I pass the information to the executive, what they do with it is up to them.” Interviews indicated a lack of robust accuracy measures linked to customer sensitivity with regard to order size and the lead time of brewing. There was also a lack of general understanding about how large errors arise, with no clear or agreed prioritization of the SKU. Thus there was no concept of prioritization in the forecasting system. Understanding what causes forecasting errors is fundamental in order to improve the forecasting function. However, the case company showed a lack of records related to forecast modifications or situations which might generate large errors. For example, in the case of under-forecasted sales there was no recorded attempt to move the customer to another SKU. Again, when volume forecasted for one given period was taken by the customer at the end of that period with sales recorded into the next period, 100% type errors could arise in terms of order timing. As a result, accuracy measures made little or no sense without information regarding what caused these errors arise in the first place. However, while the errors were apparently large—still beer was still being sold and customers were satisfied. The Sales department was also a source of error, with its staff submitting the forecasts to production well before time, with the justification that they were ensuring adequate volume in order to satisfy their customers. Basically they ordered the stock (and so created a forecast) for a period earlier than the one in which sales for that stock were agreed, automatically generating an accuracy error because the stock was not being sold in the period it was forecasted for. To overcome this early order problem, when no stock is available orders get rolled onto the following week, but it is still recorded as a forecasting error. Moreover, the Sales department is reluctant to modify a forecast even when actual sales do not add up to the forecasts and so are accounted as unsold stock. This is because anyone working at Sales never wanted to be in a situation where customer’s demand was not met. An interviewee stated: “after all, although Sales replaces the forecast with their own judgemental forecast and they are very protective and reluctant to alter it (...) they do not seem to be very good at estimating the volume they are going to provide on promotion”. This error can be as high as 50%.

5. Conclusions This paper presented a case study of a traditional brewing company and its attempt to modernize in a very

1736


volatile and agile environment, with a number of restructuring processes being carried out in a short time span. In doing so, the company tried to move to a “react-on-time” mode leaving little scope for the forecasting system. The analysis conducted by the authors identified a number of main issues with regard to the company’s forecasting process. Firstly, changing the business culture in the attempt to modernize and adapt in a very competitive sector had a detrimental impact on forecasting. Secondly, the presence of few powerful customers has led the company to place orders irrespective of marketing efforts, moving the forecasting function from central production into the hands of the Sale department. Although this department had a closer interaction with customers, it tended to provide misleading orders in terms of size and timing, generating a tendency for the FMCG to over produce. This aspect was difficult to manage and control, given that orders were processed at the account level and each customer had different ordering styles. Thirdly, there was a poor opinion and understanding of the forecasting function by both Sales and Brewing departments, which were also working in a “them and us” situation in which each side views the other with suspicion. Fourthly, the lack of robust accuracy measures associated with customer sensitivity, order size and the lead-time of brewing increased the level of difficulty in managing forecasts, especially as each department had its own motivations. Managing forecasts was also not made any easier because of the lack of a clear and agreed prioritization, with too many reports generated and significant information lost during the sales review process. The whole idea behind forecasting is to be proactive, learning how to anticipate the effects of actions or situations. There is no point in forecasting if the approach is reactive. This study demonstrates that, although a company may make a number of efforts in order to be more proactive, the situation within the company itself could remain reactive. Thus the company may simply adopt a “react-on-time” approach, whereas it is important to implement changes and incorporate information as it appears. In the case company presented in this study, there was almost no connection or information flow between departments and Central Planning. This lack of information flow and lack of cooperation among departments raises concerns related to the production of beer. Brewing personnel were given a volume to produce and nothing more in terms of information. This made it difficult to react to changes in demand, especially in a transitional period characterized by frequent organizational changes. Business targets identified by the case company did not produce fixed forecasts in the strict sense of the term. Instead, business targets influenced the sales process in way that targets became forecasts. There was no clear-cut distinction between the use of targets and forecasts to motivate business and for production planning. Forecasts basically started being targets, and then strategic plans started to get ignored. The frequent forecast modification limited the usefulness of quantitative forecasting methods in the case company. Responses from interviews show that there was little faith in initial results provided by the forecasting methods implemented in the company. It is unlikely that improvement in forecast production would have changed these skeptical views. As a result, the idea of “react-on-time” shaped the forecasting function rather than trying to optimize mathematical/statistical forecasting, and the interest slowly shifted from forecast production to the accuracy and appropriateness of reporting measures. It appeared from interviewees’ responses that the accuracy measures used were not appropriate for some departments. This situation reflected the differences in forecast requirements expressed by different departments and depended on how sales staff interacted with customers, affecting the production side of the organization.

1737


Therefore we conclude that it would be better for FMCG companies to focus less on forecasting at SKU or even product level. Instead, FMCG companies should develop more reactive systems which can respond rapidly to customers’ demand without building up high stock levels. These systems need to be monitored and managed. We argue that current time-based accuracy measures, which are derived from errors at the SKU level, are not appropriate measures: FMCG companies should then develop a more holistic approach by assigning forecasts to the correct sales period. References: Adebanjo D. (2009). “Understanding demand management challenges in intermediary food trading: A case study”, Supply Chain Management: An International Journal, Vol. 14, No. 3, pp. 224-233. Adebanjo D., (2000). “Identifying problems in forecasting consumer demand in the fast moving consumer goods sector”, Benchmarking: An International Journal, Vol. 7, No. 3, pp.223-230. Armstrong S. and Fildes R. (2006). “Making progress in forecasting“, International Journal of Forecasting, Vol. 22, Issue 3, pp. 433-441. Baron S., Davies B. and Swindley D. (1991). Macmillan Dictionary of Retailing, Macmillan, Oxford. Bourlakis M. A. and Weightman P. W. H. (2004). Food Supply Chain Management, Blackwell Publishing. British Beer and Pubs Association (2010). Statistical Handbook, London: Brewing Publication Limited Canduela J. (2009). Forecasting in Fast Moving Consumer Goods Organizations: A Case Study, Lambert Academic Publishing. Cube Software Limited (2001). PROCAST, Compton Court, 20-24 Temple End, High Wycombe, Bucks, HP13 5DR. Fildes R. and Hastings R. (1994). “The organization and improvement of market forecasting”, Journal of Operational Research Society, Vol. 45, No. 1, pp. 1-16. Gormley P. and Macmillan G. (2002). Working paper given at the 2002 International Symposium of Forecasting, Dublin. Mahmoud E., DeRoeck R., Brown R. and Rice G. (1992). “Bridging the gap between theory and practice in forecasting“, International Journal of Forecasting, Vol. 8, No. 2, pp. 251-267. Makridakis S. and Wheelwright S. C. (1989). Forecasting Methods for Management, New York: John Wiley. Mentzer J. T., Moon M. A., Kent J. L. and Smith C. D. (1997). “The need for a forecasting champion”, The Journal of Business Forecasting, Vol. 16, No. 3, pp. 3-8. Mentzer J. T. and Cox J. E. (1984). “Familiarity, application, and performance of sales forecasting techniques”, Journal of Forecasting, Vol. 3, pp. 27-36. Mentzer J. T. and Kahn K. B. (1995). “Forecasting technique familiarity, satisfaction, usage, and application”, Journal of Forecasting, Vol.14, pp. 465-476. Sanders N. R. and Manrodt K. B. (2003). “The efficacy of using judgemental versus quantitative forecasting methods in practice”, Omega, Vol. 31, pp. 511-522. Sanders N. R. (1995). “Managing the forecasting function”, Industrial Management & Data Systems, Vol. 95, No. 4, pp. 12-18. Tanwari A. U. and Betts J. (1999). “Impact of forecasting on demand planning”, Production and Inventory Management Journal, Vol. 40, No. 3, pp.31-35. Van Wezel M. C. and Baets W. R. J. (1995). “Predicting market responses with a neural network: The case of fast moving consumer goods”, Marketing Intelligence & Planning, Vol. 13, No. 7, pp. 23-30. Wacker J. G. and Lummus R. R. (2002). “Sales forecasting for strategic resource planning”, International Journal of Operations And Product Management, Vol. 22, No. 9, pp. 1014-1031.

1738


Logistics Continuing Education: “Berufswertigkeit” and The Duisburg Model Matthias Klumpp1, Stephan Zelewski2, Rolf Dobischat3, Hella Abidi1, Martin Kowalski2, Johannes Reidel4 (1. Institute for Logistics and Service Management, FOM University of Applied Sciences, 45141 Essen, Germany; 2. Institute of Production and Industrial Information Management, University of Duisburg-Essen, 45141 Essen, Germany; 3. Institute for Vocational and Further Education, University of Duisburg-Essen, 45141 Essen, Germany; 4. Formerly Centre for Responsibility Research, Institute for Advanced Study in Humanities, 45128 Essen, Germany)

Abstract: The increasingly complex models and processes in logistics require more and more knowledge and competences by personnel in industry and retail companies as well as with logistics service providers. Within the largest European research endeavor in logistics, the “EffizienzCluster LogistikRuhr”, funded by German logistics companies and the German Federal Ministry for Education and Research (BMBF), several individual research projects address the field of logistics knowledge and education management. Three of them form a close cooperation in order to support the innovative continuing education office “DIALOGistik Duisburg” in Europe’s largest inland port of Duisburg, counselling and supporting all personnel from different logistics companies in the area. Based on an analysis part (“Berufswertigkeit” survey) several content and innovation models were developed addressing intermodal transport, sustainability in logistics as well as case-based reasoning in knowledge management. These are bundled to be disseminated in industry and retail by software tools and consulting services. The 2012 empirical study “Berufswertigkeit” in logistics with 1,068 participants and the subsequent development of a qualifications framework derived from EQF structures open a new field for logistics continuing education. The research results presented here will help researchers and practitioners alike in structuring and defining continuing education gaps and efforts in logistics competence fields. Keywords: case-based reasoning; continuing education; corporate social responsibility; human resource management; logistics JEL codes: A2, C8, L9, M5 Matthias Klumpp, Dr., Professor, FOM ild (Institute for Logistics and Service Management); research areas/interests: education in logistics, sustainable logistics, operations research. E-mail: [email protected]. Stephan Zelewski, Dr., Professor, University of Duisburg-Essen, Institute of Production and Industrial Information Management; research areas/interests: logistics, operations research, production management, production theory. E-mail: [email protected]. Rolf Dobischat, Dr., Professor, Department of VET & Further Education, Faculty of Education Sciences, University of Duisburg-Essen; research areas/interests: vocational education and training. E-mail: [email protected]. Hella Abidi, Dipl.-Kffr. (FH), FOM ild (Institute for Logistics and Service Management); research areas/interests: education in logistics, humanitarian supply chain management, e-mobility in the logistics. E-mail: [email protected]. Martin Kowalski, Dipl.-Inf., University of Duisburg-Essen, Institute of Production and Industrial Information Management; research areas/interests: computer science, informatics, logistics. E-mail: [email protected]. Johannes Reidel, Dr., Kulturwissenschaftliches Institut (formerly), research areas/interests: corporate social responsibility, sustainable development, technology assessment. E-mail: [email protected]. 249

Loggistics Continu uing Education n: “Berufswertigkeit” and Th he Duisburg M Model

1. Inttroduction As the world and thhe global traade flows chaange faster th han ever, faccing demandiing requirem ments both byy customers as well as envvironmental and a CSR resttrictions, logiistics is itselff changing faast amidst a technological t l revolution— —some proclaaim the fourthh industrial reevolution—bringing dynaamic and selff-reliable deciision systemss to sub-units of the supplly chain (“intternet of thinngs”). The research question derived ffrom these ch hanges is thee essential queestion if educcation and traaining of logisstics professionals can cattch up with thhese changes and how thiss contributes to the overaall value provvided by loggistics availab bility in a gllobal econom my (Klumpp et al., 2013;; Klumpp, 20012; Roth, 20012). From thhe onset of the t largest lo ogistics reseaarch endeavoor in Europe, the Germann “EffizienzClluster LogisttikRuhr”, wiith more thaan 100 partiicipating insstitutions andd companies led by thee Fraunhofer Institute I (IM ML) Dortmundd, the strateggic role of ed ducation and continuing eeducation in logistics wass emphasized.. Within a straategic researcch alliance thhree major pro ojects addresssed educationn and training g questions inn logistics in a coherent annd role modell concept in order o to supp port the wholee ‘value streaam of logisticcs education”” with new reesearch inputs (Duisburg Model, led by b the Univeersity of Duissburg-Essen):: From the an nalyzing andd conceptualizzing project WiWeLo (Scientific ( F Further Train ning in Loggistics) to thhe corporatee knowledgee management and inform mation sciencce project OrGoLo (Orgaanizational Innovations vvia Good Go overnance inn Logistics Networks) N andd the overarcching topicall project CoR ReLo (Integrrated Corporate Social Responsibility R y Managemennt in Logisticss Networks) addressing a suustainability education e in logistics (see Figure 1).

Figure 1 Logistics Educcation Circle Concept C (Duisb burg Model)

All these three apprroaches are outlined o in orrder to give an a overview over o the uniqque setup and d also resultss from these research proojects—with the mission to change the t way logistics compaanies think, plan p and actt regarding edducation and training of thheir employeees in the futurre. Section 2 provides the key results frrom the largee competence evaluation survey in thhe logistics industry, thee German “B Berufswertiggkeit” survey y with 1,0688 participants (“Analysis”)). Further onn, section 3 discusses co onceptual inssights into thhe innovativee concept off case-based reasoning r in logistics l (“Sttructures”). Section S 4 desccribes the CS SR knowledgee requiremen nts in modernn logistics (“C Content”), whhereas sectionn 5 outlines the DIALOG Gistik conceppt (“Disseminnation”) and section 6 thee conclusion and a further reesearch impacct.

2. Em mpirical An nalysis: “Beerufswertiggkeit” Survey The coompetence evvaluation conncept of “Berrufswertigkeitt” relates to the concept of employab bility and hass

1740

Loggistics Continu uing Education n: “Berufswertigkeit” and Th he Duisburg M Model

been coinedd and used inn the first “B Berufswertigkkeit” survey in Germany in 2007 (Kllumpp, 2007 7; Klumpp & Schaumann,, 2007). The main m objectivve of “Berufsswertigkeit” is an objectivee competencee measuremen nt of a singlee person regarrdless of form mal educationn degrees annd background ds—thereforee typical requuirements off the businesss practice aree used for evvaluation. Thhese criteria for an effecctive compettence measurrement are adapted a from m business praactice. Herebby the differeent educationn degrees cou uld be compaared and the results are practicep andd output-oriennted. This meaasuring conceept considers two importaant elements as a listed here::  First, itt allows for a comparison of activities and compete ences in different industriees in the real--life businesss practicee;  Second d, it enables a comparablee evaluation of o individual persons’ com mpetences annd their value for businesss practicee. requirement criteria thatt The evvaluation insstrument “Beerufswertigkeeitsindex” inccludes 36 qualification q represent thee modern dailly work envirronment and are listed as follows: f 1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12. 13.

Efficienncy Indepenndence and ownn initiativve Flexibillity and adaptabbility Work virtues Stress reesistance Motivattion and ability to lifelongg learning and maintain m to own competence proofile a Coordinnate the work- and lifetimees Creativiity Loyaltyy Risk-takking Charism ma Ability to write and speak in Germann Knowleedge of a foreign languagge

14.. Staff requirem ments and stafff mission planning / staaff developmen nt 15.. Team, staff and a leadership 16.. Improving reesponsible care 17.. Legal knowledge 18.. Ability to apply modern infformationand communnication technollogies (work place) 19.. Communicattion and rhetoric 20.. Assertivenesss 21.. Internationall and interculturral competence 22.. Costumer foccus 23.. Skills in matthematics and sttatistics 24.. Preparation of o cost estimatees and quotations 25.. Planning, im mplementation and documentatioon of orders and d projects 26.. Negotiationss capacity 27.. Analytical prroblem-oriented d work

28. Quality m management (o optimization of processes and produccts or service quality) 29. Concepttual and strategiic implemeentation of indu ustry-specific knowleddge and experience 30. Identificcation with the company c 31. Strategicc orientation, deetermine / control tthe complete co ompany 32. Understaanding of solutiions for complexx technical prob blems 33. Basic knnowledge of bussiness administtration 34. Perception of functionss of managem ment and organ nization 35. Concepttual working in immediate workplace urement and 36. Planningg, control procu logisticss processes

The “B Berufswertigkkeitsindex” is calculated by a summed and unweighhted index off individual ev valuations off the 36 qualiification requuirement criteeria. The valuue range of the t “Berufsw wertigkeitsinddex” (BWI) begins b with 0 and ends at 100 [0: evalluation of alll criteria withh poor and 10 00: evaluatioon of all criteeria with very y good]. Thee following eqquation (Klum mpp et al., 20011, p. 7) reprresents the BW WI calculatioon.

Standardizationn of percent: 25 = ¼ ¼*100

This eqquation includdes a recoding as the achieeved average value is subttracted from tthe value fivee. Hereby thee highest achieved value iss the numericcal figure 4 annd the smalleest possible fiigure is 0. Thhe normalized d index 100% % is calculatedd by multiplication by 25%. Advantagge of this is the t comparabbility of all ppersons and no n prejudicedd importance of one comppetence criteeria over anoother. Disadvvantage is ann exchangeabbility of criteeria which iss 1741

Logistics Continuing Education: “Berufswertigkeit” and The Duisburg Model

sometimes not realistic in business practice (i.e., if specific criteria are necessary for a person). The output-oriented measuring concept “Berufswertigkeit” serves as a basic field-evaluation concept for the development of such an European Qualifications Framework (EQF) for the logistics industry and integrates the required investigation of competences (Klumpp &Schaumann, 2007). At the end of June 2011 a field survey with 1,000 persons within the German BMBF project WiWeLo, part of the “EffizienzCluster LogistikRuhr”, was started. There are three types of survey instruments which could be executed such as written form, telephone and internet. Due to the experiences in the two “Berufswertigkeit” studies of 2007 and 2009 the project consortium decided for a telephone survey. The survey was executed in North Rhine-Westphalia and Hessia. Both states depict a very good representation of the whole country of Germany—as they are both no city states and have no major economic restrictions (e.g., East Germany) but also a representative combination of modern service centers (Frankfurt, Cologne, Düsseldorf), older industrial clusters (Ruhr Area, Rüsselsheim) as well as more rural areas (North Hessia, Westphalia). Since January 2011 the survey was in preparation by development of the survey instrument, a chance-sample of addresses of the logistic industry in both states was drawn. The telephone survey of 1,067 persons begun end of June and took place till September 2012. The interview survey can be found in Klumpp et al. (2013). Altogether 808 persons in North-Rhine Westphalia (NRW) and 259 persons in Hessia with different education levels were asked. Therein existing skills and competences of persons in the logistics industry will be described and the above mentioned draft for an industry qualifications framework logistics will be reworked. Also traditional formal degrees in vocational and academic education will be classified according to evaluated practical competence levels. Finally 1,068 persons from the logistics sector were questioned and these represent 379 female and 689 male respondents. 75.7% of the respondents are from North Rhine-Westphalia and the remaining 24.3% are from Hessia. The respondents represent various professional levels at their current working place. 88.6% are employed as white-collar workers in different levels like branch managers, team leaders and office clerks in their company and 11.4% work as blue-collar workers in warehouses or as truck drivers. The Figure 2 shows the “Berufswertigkeitsindex” which significantly presents that 80% of the competences of 21.8% blue-collar as well as white-collar workers in Hessia are higher evaluated than blue-collar as well as white-collar workers in North-Rhine Westphalia, hereby it is to assume that the German federal state of Hessia indicates a very specialized logistics area by Frankfurt Airport. 25

% of respondents

20 15 NRW 10

Hesse Total

5 0 to 65

to 70

to 75

to 80

to 85

to 90

to 95

to 100

% of Berufswertigkeitsindex

Figure 2 1742

“Berufswertigkeitsindex” for the Logistics Industry (NRW and Hessia)


Furthermore Figure 3 presents the “Berufswertigkeitsindex” of six different fields of logistics business activity in the logistics industry of the 1,068 respondents. It can be noticed that in four fields of logistics business activity a value of more than 75% to 80% namely 26% office clerk, 21% administration level, 19% group leader and 18% managing director is achieved. 40% blue collar-workers indicate a “Berufswertigkeitsindex” of 65% to 70%. 21% academic staff specifies a “Berufswertigkeitsindex” of 80% to 85%.

% of respondents

30 25

Academic [n=197]

20

Managing director [n=231]

15

Blue-collar worker [n=121]

10 5

Administration [n=141]

0

Group leader [n=120] to 65

to 70

to 75

to 80

to 85

to 90

to 95

to 100

Office clerk [n=252]

% of Berufswertigkeitsindex Figure 3 “Berufswertigkeitsindex” Based on Field of Logistics Business Activity

Figure 4 presents the curve of the “Berufswertigkeitsindex” (BWI) across all age groups. The curve that presents the respondents with the age of up to 25 years indicates a high BWI with 75% to 80%. This group estimates themselves with a high BWI due to their specific logistics education and possibly their internet knowledge. In the remaining age groups the curve are flat compared to that. The BWI is equally distributed in the value of 65% to 100%. All graphs of the age groups except the age groups between 46 to 55 years old run in parallel. The remaining age groups achieve the highest BWI with 75% to 80%. The results show that 18% of respondents in the age of 46 to 55 years have a BWI of 70% to 75%. As could have been expected only 2% of the age group till 25 years achieve 95% to 100% and 8% of the age group older than 55 years old has the highest BWI of 95% to 100%.

% of respondents

30 to 25 years old [n=97]

25 20


15


10 5


0 to 65

to70

to 75

to 80

to 85

to 90

% of Berufswertigkeitsindex

to 95

to 100

56 years old und older [n=172]

Figure 4 “Berufswertigkeitsindex” Based on Age of the Respondents

1743


The following figure shows the BWI based on the group of school graduation. The results are as expected the “Berufswertigkeitsindex”. 26% of the respondents with a high school graduation (ISCED level 2—German “Hauptschulabschluss”) specify a BWI of 65%, followed by 21% of the respondents with a high school graduation (ISCED level 3—German “Realschulabschluss”) who declare a BWI of 75% to 80%. The highest BWI between 95% and 100% is quoted by 6% of the respondents that achieved the A-level graduation (German “Abitur”). 30 % of respondents

25 High school graduation (ISCED level 2) [n=129]

20 15

High school graduation (ISCED level 3) [n=364]

10 5

A-level [n=570]

0 to 65 to 70 to 75 to 80 to 85 to 90 to 95 to 100 % of Berufswertigkeitsindex

Figure 5 “Berufswertigkeitsindex” Based on Graduation of the Respondents

In connection with this first empirical evaluation of logistics competences in a field survey the research cluster set out to establish knowledge management tools (e.g., CBR, section 3), knowledge content (e.g., sustainability, section 4) and knowledge and education systems (e.g., DIALOGistik, chapter 5) in order to enhance the competences of logistics employees.

3. Knowledge Management Innovation: Case-based Reasoning in Logistics 3.1 Knowledge Management Research Gap A key requirement of knowledge management, that has been brought forward repeatedly within the scope of business economics as well as in the field of business & information systems research, extends to the reuse of knowledge. Primarily because of two reasons it is recommended to apply the knowledge, which has been acquired in the past to solve problems in business practice, for dealing with new problems as far as possible. Firstly, it would be economically inefficient to re-invest the resources that were used for the initial acquisition of knowledge in case that a reuse of this knowledge for solving new problems is omitted. Secondly, learning curve effects attained by repeated usage of similar knowledge components could not be exploited in such a case of omission. Despite the obvious economic advantages, the demand for a systematic reuse of knowledge is often not implemented in operational practice. The intended reuse of knowledge normally only succeeds with good structured and often quantitative factual knowledge (“know what”) that, for example, can be saved easily and that can be recalled straightforwardly for the purpose of problem-solving (re-)use with the help of conventional data base systems. In the case of poorly structured and normally qualitative knowledge represented mainly in natural language and extending especially to competences in the sense of action-enabling knowledge (“know-how”) and

1744


to “everyday theories” for the pragmatic explanation of processes and systems (“know-why”), the recommended reuse of knowledge is, however, confronted with substantial barriers time and again. These barriers are mainly based on three reasons. Firstly, it is difficult to save and recall qualitative knowledge that is predominantly represented in natural language using conventional database systems (problem of qualitative knowledge). Instead, in the best case it is filed as explicit knowledge in natural language documents (free-formatted “texts”). In the worst case it is even only locked up as implicit knowledge “in the minds” of some professionals (“experts”). Secondly, in the event of a new occurring problem it is very intricate to assess whether it is similar enough to older, already solved problems so that it can be—at least partially—worked out with the knowledge gained from the solving of older problems (problem of sufficient problem similarity). Thirdly, the knowledge, that is generally available in a company on the matter of successfully (or deficiently) treating previous problems, is so extensive that in business practice it is barely possible to realize a systematic reuse of knowledge without computer support. The usage of computers in the field of knowledge management often fails in reality because of the need of being able to process qualitative knowledge that is predominantly represented in natural language (problem of computer support). 3.2 CBR Project Set-up Researchers at the University of Duisburg-Essen have developed an ontology-driven and case-based reasoning tool that can help to solve or at least to alleviate the aforementioned problems of qualitative knowledge, sufficient problem similarity and computer support from the perspective of operational practice. The case-based reasoning technology (Aamodt & Plaza, 1994; Watson, 1997; Avramenko & Kraslawski, 2008), that originates in the research of artificial intelligence, has been selected as an approach to problem solving, because of two reasons. Firstly, it shows a “natural”, direct connection to the both last-mentioned problems. Secondly, it can be ‘enriched’ especially with the help of ontologies (Guarino, 1997; Lin et al., 2011; Zelewski et al., 2012) so that it also shows interesting potential for the first-mentioned problem. Both technologies, the case-based reasoning and the ontologies, have been explored largely independent of one another until now, because of their—at least prima facie—different fields of application. From the point of view of business economics, they have been hardly applied—also within knowledge management—for the solution of practical problems. Hence, the innovative approach lies in the combination of case-based reasoning and ontologies in such a way that they can be used computer-supported for the solution of practical problems. In the context of the joint research project “Organizational Innovations via Good Governance in Logistics Networks” (OrGoLo) it is examined how the management of complex, especially international logistic projects can be supported by such innovative instruments of knowledge management in order to enable logistics companies to achieve sustainable competitive advantages. 3.3 CBR Methodology Case-based reasoning imitates human thinking trying to make a decision based on earlier experiences. The idea of case-based reasoning can be formulated in one sentence: a case-based reasoner solves new problems by reusing solutions that were used to solve similar problems in the past. Problems are generally thematized as cases in the context of case-based reasoning. This specialized terminology is used here for the sake of compatibility with established literature. Each case consists of three characteristic components: the case description (problem description), the result (problem solution) and the evaluation (evaluation of the problem solution). Projects can be seen as a special case of such cases. Therefore, in the following the terms “problems”, “cases” and “projects” can be regarded as synonyms. The knowledge of experience about already conducted logistics projects (old cases) is 1745


stored in the knowledge base (or synonymous case base), i.e., a knowledge base containing the descriptions, results and evaluations of all old cases. The typical case-based reasoning process based on the knowledge stored in the knowledge base is usually divided into four phases of the so called CBR cycle (Aamodt & Plaza, 1994): retrieve, reuse, revise, and retain (see Figure 6). new case presented: only case description

problem

retrieve

aborting due to not achieving minimum level of similarity

knowledge base learned new case: becomes old case

most similar old case

general old cases

(domain) knowledge

retain

new case

reuse CBR cycle

problem solution

validated and evaluated potentially revised new case: case description, case result and case evaluation

revise

solved new case: case description and case result

aborting due to unrealizable requirements

Figure 6

The CBR Cycle (Aamodt & Plaza, 1994; augmented in Kowalski et al., 2012)

The description of a new case is used to search for at least one sufficiently similar and—if there exist several sufficiently similar cases—at least one most similar case in the knowledge base (retrieve). Having found such a sufficiently and most similar case in the knowledge base, it is adopted to the new case (reuse). The quality of the adoption result must be evaluated with respect to user requirements and must therefore potentially be modified (revise). The description, result and evaluation of the new case are combined in order to form a “learned new case” which is stored in the knowledge base (retain). 3.4 Prototyping and Implementation Experiences A prototype CBR tool called “SCM Project Recommender” (Kowalski et al., 2012; Kowalski et al., 2013) was developed using the CBR development and application framework jColibri. This prototype has been customized in a first stage of development in order to collect knowledge of experience on complex, especially international logistics projects as well as to review it in the special knowledge representation form of cases. In the center of the development work were real projects of practice partners which concerned the management of international supply chains (SCM projects). This included, e.g., the suitable packaging for shipping of components of a polar research station and their carriage by sea from Duisburg to the Antarctic and also the multi-modal transport of DIY-products from their factory in the middle of China to their distribution centers of a well-known trade chain in Germany. The CBR tool is planned to be embedded into a “Collaboration Platform” together with the “Supply Chain

1746


Configurator” (Robles et al., 2013), which are developed in the joint research project OrGoLo as two further tools. The collaboration platform is supposed to support especially small and medium-sized companies with the efficient processing of foreign trades. Such foreign trades are often multi-layered and very complex. For example, forwarders are confronted with great challenges concerning import and export prescriptions, especially customs regulations. The resulting burden of responsibility creates substantial requirements of knowledge empowering to act as well as communication among different stakeholders from numerous branches, nations and economic cultures. By developing the collaboration platform a solution for these challenges is presented, which realizes the idea of “good governance” (Lautenschläger & Lautenschläger, 2013) concerning complex logistics projects in international supply chains. The collaboration platform will provide a modern assistance tool for planners, dispatchers and controllers for their intensive cooperation with suppliers and logistics providers. This tool is supposed to preserve for them the power of disposal over their supply chain data at any time, to ensure an efficient use of decentralized available competences for individual supply requirements and to allocate knowledge of experience on already realized logistics projects for continuous organizational learning. The three previously mentioned tools in support of the knowledge management for complex, especially international logistics projects are tested and evaluated in the context of the joint research project OrGoLo on the part of the continuing education office DIALOGistik Duisburg together with partners of the business practice, and developed further in accordance with practical requirements. In general, the DIALOGistik office serves as an innovative institution within the inland port of Duisburg in order to establish a central contact point, especially for small and medium-sized enterprises, for aspects of know-how transfer, qualification and supply chain efficiency. Beyond that it serves for building up a sustainable relation between the University of Duisburg-Essen and local logistics companies, for creating a community of knowledge exchange between practitioners, scientists and other stakeholders, and for enabling knowledge generation and knowledge transfer from science to business and vice versa. In the context of the joint research project OrGoLo, the DIALOGistik office helps to organize and to coordinate the collaboration between scientific and practical partners considering an exchange of experience knowledge, to identify potential business users, willing to participate in a pilot period for the developed software tools, and to promote the developed software tools during and after the initial joint research project OrGoLo.

4. Content Innovation: Sustainability Education in Logistics 4.1 Sustainability Requirements in Logistics Logistics companies, due to their integration into the global value chain, are confronted with the term sustainability, i.e., the challenges of economic, social and ecological responsibility. In the international debate the meaning of this term converges—at least on the company level—with the concept of “Corporate Responsibility” (CR). The growing public awareness for social and environmental standards, the emergent relevance of ethical and eco audits and the necessary preservation of resources and energy efficiency caused by the climate change make sustainability and CR concepts which are geared specifically to the needs of smaller and medium-sized logistics companies and supply chain networks a necessity (Geßner et al., 2013). As the connecting link in global value-creation processes, the logistics sector is confronted with the social, ecological and economic demands of sustainability and CR in a particular way. Firstly, the knowledge gained from climate research on the basis of greenhouse gas (GHG) emissions demonstrates that it is especially the transport sector that suffers from a discrepancy between the actual and the necessary development: On the one hand the

1747


transport of goods and persons causes world-wide just over 13% of GHG emissions, in the EU-27 this percentage totals even 24%. This fact is aggravated by the circumstance that, depending on the individual projection, an increase in the rate of freight transports within the EU by up to a further 80% is expected by 2050. Secondly, “stuck in the middle”, the customer oriented logistics industry has to find solutions to the increasing sustainability and CR requirements that producing companies are facing, while at the same time price is expected to stay low and performance to raise. Thirdly, the logistics sector is primarily made up of small and medium-sized enterprises (SME) and is typically characterized by highly interdependent international network structures, which are structured and organized to greater and lesser degrees, but always highly competitive. Last but not least, the logistics sector faces severe skills shortages; especially against the background of the growing lack of skilled labor in logistics, sustainability and CR strategies, e.g., in human resources, are considered decisive to increase the attractiveness of the sector and to recruit and retain new staff members (Meyer & Schmidt, 2013). 4.2 Project Results Preliminary project results give evidence that SME are quickly overburdened or confused by the many understandings and notions of sustainability and CR and the standards, types of certification and monitoring programs (Meyer et al., 2012). This is an important fact in combination with the dominant economic mindset “If it matters, it’s measured”. Measuring sustainability performance is a heavy task, because the “business case” is hardly to be demonstrated on the basis of the ordinary economic framework of key performance indicators. That’s why up to now sustainability and CR strategies are difficult to implement, especially when companies want to take into account the internalization of social and ecological costs within highly competitive logistics market structures. The crucial lesson to be learned is that an integrated sustainability and CR approach rather relies on a supporting corporate culture. This can be prominently illustrated by means of the standard “ISO 26000 Social Responsibility”. This new standard, launched in 2010 following five years of negotiations between many different stakeholders across the world, provides “only” guidance rather than requirements: It cannot be certified to unlike other ISO standards. In fact, the project result implies that sustainability and CR approaches require structural stimuli that take into account the cultural specifics of logistic companies. The integration of sustainability and CR topics into the organization proves a major challenge for a company’s structure. A fundamental requirement is the commitment to sustainability and CR approaches of the top-level management. Furthermore it will be not sufficient that there is only one sustainability and CR officer. The risk seen in this is that such a person may exhibit a “tunnel vision” that makes him or her less susceptible to the necessities and contradictions of day-to-day business operations. The topic should instead be borne as a topic that touches on every part of the entire organization. This means to set the lowest possible inhibition threshold, e.g., by explaining sustainability and CR as a cross-cutting issue that can be integrated into the daily routines of every employee. Therefore sustainability and CR management should be structurally embedded into the different junction points within the organization. In brief: “Sustainability and CR-education” implies that corporate culture and corporate strategies concur, which means it has to be differently designed and developed for every individual company. The goal of the project is to prove that value-based socio-ecologically responsible supply chain networks enjoy an innovative corporate culture and are marked by lower susceptibility to interference and boast an improved co-ordination potential, higher employee motivation and social reputation. In general this could lead to competitive advantages and new market opportunities on the basis of sustainable management. Most companies already consider service, reputation, cost, security, and safety when choosing logistics service providers. 1748


Sustainability is on the way to be added to the list. To become a champion in the logistics industry it will be clever to become a sustainability champion. The companies that will look at the current economic conditions not as a problem but as an opportunity for their organization will be best prepared for future challenges—time will tell if this assumption can be confirmed.

5. Dissemination and Cluster Approach: DIALOGistik 5.1 Duisburg—Via Structural Change to One of the Most Important Centers of Logistics Logistics is classified as an industry of the future with significant growth potential not only nationally, but also for the regional economy, because with the globally networked division of labor, the logistical performance of a region takes on the role of an important location factor. It is well known that with its large urban metropolises such as Duisburg, the Ruhr Area has been undergoing a major structural change for the past few decades. In the meantime the city of Duisburg has developed into a center for logistics, commerce and services, which with the Rhine-Ruhr port as the largest inland port in Europe (with round about 40 thousand employees) has become a regional logistics hub with a considerable amount of charisma. 5.2 WiWeLo Opens Innovation Corridors by Tailor-Made Qualifications The joint research project “Scientific Further Training in Logistics” (WiWeLo) is a research project sponsored by the Federal Ministry for Education and Research (BMBF) as part of the Excellence Cluster Competition of the Federal Government (High-Tech Strategy) and is a partial-project of the “EffizienzCluster LogistikRuhr” which focuses on the occupational qualification and further education of companies and undertakings. This topic has been chosen, because of inadequate competence on the employees’ side in the logistic industry. The core of the underlying philosophy of the project WiWeLo is the working hypothesis that innovation corridors (technical and organizational, personnel) can only be sustainably opened up by appropriate training measures. However, this implies a scientific analysis of the content dimensions that are to be communicated, likewise a process-related supervision and a testing of the appropriate measures, as well as the evaluation and documentation of the results and effects achieved. Ultimately this involves the development, implementation and evaluation of models of tailor-made, demand-based qualification concepts. Foundations of the conceptual development are scientific analyses of the target corridor of requisite qualification requirements for specific operational target groups on different levels of the job hierarchy. 5.3 DIALOGistik Duisburg between University and Logistics Enterprises Since the scientific project work would be difficult to implement without institutional support, a service organization is being established for knowledge transfer, qualification and logistics efficiency, namely DIALOGistik Duisburg with headquarters in Duisburg. In this organization the project activities of the participating partners are bundled together, so that it acts as a communication platform between science and practice. With the institutionalization of DIALOGistik Duisburg, a company-related network is being implemented which sees itself as a communication platform from the perspective of bundling information, knowledge and product results at the site of the Port of Duisburg and making this available for the transfer to the regional players. From a company-related perspective it will act as a location for the design of further training and education in the context of company personnel policy in order to increase the knowledge base of the employees through work-place-related training and thereby make a contribution to a data-based improvement in the transparency of the regional training and employment market. The innovative approach of DIALOGistik Duisburg

1749


can therefore also be seen as a conception of exemplary, certified training measures in a communication process involving all network partners who are intended to enable permeability between academic and vocational training. The goal is not only to increase the value of vocationally acquired skills through formal recognition processes, but also improving operational and inter-company mobility and opportunities for promotion for employees through extended options within the personnel deployment concepts associated with personnel policy. The consolidation of operational interests with respect to the use of their workforces, as well as individual employee interests in terms of their training and employment aspirations from the point of view of operational and workplace-related career structuring, can be seen as the core tasks of the work of DIALOGistik Duisburg. 5.4 Four Task Areas of DIALOGistik Duisburg The concept of DIALOGistik Duisburg can be broken down into four different task areas:  The first field of activity is a place to exchange experiences between different partners in the logistics industry, and at the same time an interface with other sectors on the peripherals of logistics that are integrated into the logistics value chain (“supply chain”). Through various procedures and tools of community building, special targets are incorporated into a communication process that deals with subject areas which are relevant for practical applications and searches for appropriate solutions to problems.  This exchange process is accompanied by the element of the dialogue between science and practice in order to define new contents and fields of work to create a benefit of all players and feed these in turn into the communication process.  A further range of tasks is an operational and individual consulting and coaching along educationally-relevant issues such as vocational, training and health advice.  This deals with the central aspect of the design of logistical education and training, with the focus of this range of tasks being on the provision of support services for small and medium-sized enterprises, which can in particular offer these businesses cost advantages and synergies within the framework of joint learning networks. 5.5 Instruments, Measures and Institutions On the one hand the aim is to create transparency in the regional education market, while on the other hand it is necessary to solve matching problems between specific company qualification requirements and further training offers outside the companies. In this context an education monitoring, which is currently under construction, will be operated as a permanent regional monitoring and evaluation tool by DIALOGistik Duisburg and will make a valuable contribution. Coupled to the monitoring facility will be an online platform, a logistics wiki, which will provide port-specific knowledge in a compressed and practically-relevant form in order to enable the faster and smoother cross-company transfer of knowledge. The nucleus of the project is the identification of tailor-made education and qualification measures. For this purpose it is necessary to carry out systematizing scientific analyses, which usually implies a multistage procedure for obtaining empirically reliable information. From this it follows that in the future, increasing convergence between operational requirements and the training offered by the regional further training institutions should be initiated by DIALOGistik Duisburg in order to promote the Port of Duisburg site and push forward with the projected aim of a professional school and port academy. At the same time an employee pool could be established for the Port of Duisburg to enable cross-company mobility of the workforces, secure continuous employment and avoid that qualified workers move to other companies. 5.6 Goals, Perspectives and Synergies Up to now the project results show that it has been possible to explore in greater detail the operational qualification requirements of companies. It has also become clear that companies are currently not in a position to 1750


define these requirements by themselves, let alone cover them. For this reason, the external support of the project is required. This also addresses the aspect of linking operational learning processes and human resource strategies, which the project has found to be a key range of tasks for DIALOGistik Duisburg. This is where the joint research project WiWeLo comes in by setting itself the task of developing tailor-made, modular-certified and transferable qualification measures in a close dialogue with company experts for different target groups in order to draw up an overall training concept that is marketable and accepted by the market in the field of “intermodal transport” and inland waterways. The point of reference here is to widen the bridge from vocational qualification to academic education in order to meet the requirement for more permeability in the education system and enable more individualized educational mobility for the purposes of opening up wider professional and career opportunities. Through the project work that it has already begun the joint research project WiWeLo is making an important contribution to the overall direction taken by the “EffizienzCluster LogistikRuhr” by initiating new impulses through qualification concepts and the synergetic release of innovations in research, development and cooperation, which can make a contribution by means of “more intelligent logistics solutions” and support the strategic goal of the “EffizienzCluster LogistikRuhr”. 5.7 The Multimodal Logistics Expert—A Research Model for Scientific Further Education The first trial run of the scientific further education “Expert for Multimodal Logistics” was completed in April 2013. This training was designed in accordance with the demand of Duisburg local logistics enterprises, recorded in expert talks. Participants of the course filled in a questionnaire about the contents of the lessons, methods of teaching, performance of the lecturer etc. At the same time, the lessons were supervised by scientific assistants, monitoring the interaction between lecturers and participants. After the end of all lessons, there were workshops separately for participants and lecturers. The results will be reflected and discussed with managers of logistics enterprises and integrated in a new concept for a sustainable further education “Expert for Multimodal Logistics”.

6. Conclusion and Impact The described integrated approach towards logistics education on a systems-level perspective can be acknowledged as a new and innovative concept which is expected to have major influence especially on a regional level within modern cluster approaches in economic development and support. As increasingly public authorities follow this cluster and branch or discipline approach, the need for such an integrated system view is obvious; and many regions as for example the region of Duisburg—respectively the Ruhr area—are going to look out for such approaches. For Duisburg it is expected that the whole logistics cluster is headed towards a joint “employment pool”, where workers are going to be supported in their education efforts regardless of specific company affiliation—and therefore will be able to transfer more easily and faster between different companies in accordance to the logistics business needs within the whole cluster area. This was specifically recognized in the economic downturn of the 2008/2009 crisis a significantly smaller number of employees than expected lost their jobs permanently within the area of Duisburg. Further on the analysis and standardization efforts within the Duisburg and Ruhr cluster research is directed at establishing an Industry Qualifications Framework (IQF) for logistics modeled on the European Qualifications Framework in order to support especially SME in the logistics industry in their human resource and training management efforts (Klumpp, 2013).

1751


Acknowledgements This research contribution is connected to several projects within the German national research cluster for logistics (“EffizienzCluster LogistikRuhr”, www.effizienzcluster.de), funded by the federal department for research (BMBF), project numbers 01IC10L19D (WiWeLo), 01IC10L20A ff. (OrGoLo), and01IC10L22A ff. (CoReLo). The authors are grateful for this support. References: Aamodt A. and Plaza E. (1994). “Case-based reasoning: Foundational issues, methodological variations, and system approaches”, AI Communications (AICOM), Vol. 7, No. 1, pp. 39-59. Avramenko Y. and Kraslawski A. (2008). Case Based Design–Applications in Process Engineering, Springer, Berlin. Geßner C., Heidbrink L., Kammel V., Kölle A., Kreuels M., Meyer N., Reidel J., Schmidt I. and Schmitz G. (2013). “Integrated corporate social responsibility management in logistics networks (CoReLo)”, in: Clausen U., ten Hompel M., Klumpp M. (Eds.), Efficiency and Logistics, Springer, Berlin Heidelberg, pp. 7-18. Guarino N. (1997). “Understanding, building and using ontologies—A commentary to ‘using explicit ontologies in KBS development’ by van Heijst, Schreiber, and Wielinga”, International Journal of Human-Computer Studies, Vol. 46, No. 2/3, pp. 293-310. Klumpp M. (2013). “Herausforderung Logistikqualifikation: Berufswertigkeitsanalyse und Industrie-Qualifikationsrahmen Logistik”, in: Wolf-Kluthausen H. (Ed.): Jahrbuch Logistik 2013, Free, Korschenbroich, pp. 106-109. Klumpp M. (2012). “Trendimplikationen Logistik 2020: Nachhaltige Bildung für transparente und sozial orientierte Logistikprozesse”, in: Zelewski S., Münchow-Küster A. (Eds.): Logistiktrends in der Dekade 2010-2020–Eine Delphi-Studie, Logos, Berlin, pp. 179-188. Klumpp M. (2007). Begriff und Konzept Berufswertigkeit, Arbeitspapiere der FOM Hochschule für Oekonomie & Management No. 5, 07/2007, Essen. Klumpp M., Abidi H., Krol B., Stender T. andBioly S. (2013). Berufswertigkeit und Logistikqualifikation, Logos, Berlin. Klumpp M., Clausen U. and ten Hompel M. (2013). “Logistics research and the logistics world of 2050”, in: Clausen U., ten Hompel M., Klumpp M. (Eds.), Efficiency in Logistics, Lecture Notes in Logistics, Springer, Berlin Heidelberg, pp. 1-6. Klumpp M., Kriebel K., Beschorner H., Buschfeld D., Dilger B. and Diart M. (2011). Berufswertigkeit konkret, Wissenschaftlicher Abschlussbericht, DHI/FBH Reihe B–Berufsbildung im Handwerk, Issue 68, Eul Verlag, Paderborn. Klumpp M. and Schaumann U. (2007). “Requirements for leadership personnel and the concept Berufswertigkeit”, Kölner Blätter für Wirtschaftspädagogik, Vol. 12, pp. 3-50. Kowalski M., Klüpfel H., Zelewski S. and Bergenrodt D. (2013). “Integration of case-based and ontology-based reasoning for the intelligent reuse of project-related knowledge”, in: Clausen U., ten Hompel M. and Klumpp M. (Eds.), Efficiency and Logistics, Springer, Berlin Heidelberg, pp. 289-299. Kowalski M., Zelewski S. and Bergenrodt D. (2012). “Applying of an ontology-driven case-based reasoning system in logistics”, International Journal of Computers & Technology, Vol. 3, No. 2, pp. 347-350. Lautenschläger H. and Lautenschläger M. (2013). “Good governance in global supply chains from eight perspectives”, in: Clausen U., ten Hompel M. and Klumpp M. (Eds.), Efficiency and Logistics, Springer, Berlin Heidelberg, pp. 19-29. Lin L. F., Zhang W. Y., Öou Y. C., Chu C. Y. and Cai M. (2011). “Developing manufacturing ontologies for knowledge reuse in distributed manufacturing environment”, International Journal of Production Research, Vol. 49, No. 2, pp. 343-359. Meyer N., Reidel J. and Schmidt I. (2012). “Corporate social responsibility—Management in the management sector—Natural conditions and stumbling blocks”, paper presented at the X. International Logistics and Supply Chain Congress, 8-9 November, Istanbul. Meyer N. and Schmidt I. (2013). “Corporate Social Responsibility—Management in small and medium sized logistics service providers”, paper submitted to the International Logistics Science Conference (ILSC), 3-4 September, Dortmund. Roth A. (2012). “Analyse von Anforderungen in der Logistikbranche—Erste Implikationen für die Berufsbildung”, Zeitschrift für Berufs- und Wirtschaftspädagogik, Vol. 108, No. 4, pp. 604-626. Robles M., Wei F. and Noche B. (2013). “Challenges in the planning, organization, execution and control of international supply chains”, in: Clausen U., ten Hompel M. & Klumpp M. (Eds.), Efficiency and Logistics, Springer, Berlin Heidelberg, pp. 245-252.

1752

Logistics Continuing Education: “Berufswertigkeit” and The Duisburg Model Watson I. (1997). Applying Case-Based Reasoning: Techniques for Enterprise Systems, Morgan Kaufmann Publishers, San Francisco, CA. Zelewski S., Bruns A. and Kowalski M. (2012). “Ontologies for guaranteeing the interoperability in e-business: A business economics point of view”, in: Kajan E., Dorloff F. J. and Bedini I. (Eds.), Handbook of Research on E-Business Standards and Protocols: Documents, Data and Advanced Web Technologies, IGI Global, Hershey, PA, pp. 154-184.

1753


The Implied Premium and Growth Strategy—Evidence from S&P 500 Sue-Fung Wang1, Yi-Cheng Shih2, Xiang-Jun Lai1 (1. National Chiao Tung University, Hsinchu, Taiwan; 2. National Taipei University, Taipei, Taiwan)

Abstract: Mispricing means the market price is deviated from intrinsic value. This difference, so called implied premium, might mainly be due to the information asymmetry from firm’s growth strategy. The market tends to have a myth of paying too much premium for firm’s growth. In general, firm pursue growth through diversification or focus strategy. Literatures have seldom discussed the relationship between implied premium and diversification. Therefore, the main purpose in this paper is to examine whether the implied premium is significantly associated with diversification or not. Our results show that diversified firms have higher implied premium. Meanwhile, the degree of diversification of high-tech firms has greater significant relationship than that of non high-tech firms. It implies that investors always pay too much for the growth of high-tech firms. Key words: mispricing; residual income model; diversification JEL codes: D82, G32, L25

1. Introduction Classical finance theory argues that competition among rational investors, who diversify to optimize the statistical properties of their portfolios, will lead to an equilibrium in which prices equal the rationally discounted value of expected cash flows. Even if some investors are irrational, classical theory argues, their demands are offset by arbitrageurs and thus have no significant impact on prices. Eleven years after the influential work of Fama (1970), formulating the efficient market hypothesis (EMH), Shiller (1981) criticized the EMH by providing empirical evidence on the so called overreaction hypothesis. He argued that price volatility is much higher than justified by changes in dividends which lead to periods of strong departures of stock market prices from fundamentally justified values. Among many others, DeBondt and Thaler (1985, 1987) and Chopra et al. (1992) present empirical studies supporting Shiller’s hypothesis of prices overreacting to fundamentals. In contrast, Harris and Ohlsen (1990) and Bernard and Thomas (1989, 1990) find empirical evidence indicating that prices move less than fundamental information would justify. However, some supporters of the EMH do not believe that findings on over- and under-reaction are in contradiction to the EMH at all. According to Fama (1998), overreaction to information is equally likely as under reaction and both can be viewed as chance results. He argues that these “anomalies” disappear with methodical changes and that the

Sue-Fung Wang, Associate Professor, Graduate Institute of Finance, College of Management, National Chiao Tung University; research areas/interests: corporate finance. E-mail: [email protected]. Yi-Cheng Shih, Assistant Professor, Department of Finance and Cooperative Management, College of Business, National Taipei University; research areas/interest: corporate finance and empirical investments. E-mail: [email protected]. Xiang-Jun Lai, Graduate Institute of Finance, College of Management, National Chiao Tung University; research areas/interest: corporate finance. E-mail: [email protected]. 1754

The Implied Premium and Growth Strategy—Evidence from S&P 500

literature may present a biased sample of all studies developed. Since surprising results gain more attention and offer a larger possibility of being published, there may exist a bias towards the publication of such “anomalies”. Anomalies imply market price is deviated from intrinsic value1 of a firm, i.e., there exist mispricing in capital market. Numerous literatures indicate that mispricing is resulted from investor sentiment (Chiang et al., 2011; Stambaugh et al., 2011; Baker & Wurgler, 2006). People from unreasonable expectation of likely returns and so make misguide consumption and investment decisions also easily make bubbles happened (Penman, 2010). Lakonishok et al. (1994) provides evidence that value strategies yield higher returns because these strategies exploit the suboptimal behavioral of the typical investor and not because these strategies are fundamentally riskier. In addition, information asymmetry may also cause mispricing. Information differences across investors (or groups of investors) have been a long-standing concern to price deviation. Uninformed investors require higher cost of capital due to less information obtained than informed investors. Information asymmetry issue can also be discussed among diversified and focused firms. Firm value is created by investing and operating activities (Penman, 20102)3. And the firms’ investing activities are associated with their growth strategies. We can discuss it on focus and diversification perspective by using three measures of diversification as the proxy variables for growth strategies4. Hyland and Diltz (2002) states that the typical firm diversified by making acquisition. Diversifying firms are poorly performing firms in comparison to specialized firms and have lower growth opportunities in their current activities. These diversifiers have accumulated a reserve of liquid assets. They can pay these back directly to shareholders, use the cash to diversify, or invest more in their current activities. The market anticipates that these firms will not return these liquid assets to shareholders and consequently may not be that surprised when firms make a diversifying acquisition. It might even be better for the firm to make such an acquisition that to use these liquid assets to finance investment in poorly performing operations. With this view, management diversifies to assure firm survival and growth when it faces difficulty competing within its industry. Each segment of a firm has its own investing strategy, and it may not consistent with the others. Therefore, it may generate information asymmetry among each segment. Managers frequently cite the desire to mitigate asymmetric information as a motivation for increasing firm focus. An implication of this motivation is that diversified firms are subject to larger asymmetric information problems than are focused firms (Gilson et al., 2000; Habib et al., 1997). The source of the difference in asymmetry could be that diversified firms are less transparent than focused firms. Accounting figures for diversified firms are less transparent relative to those of focused firms. It is possible that asymmetric information problems are more severe for diversified firms. Aggregated cash flows and other diversification-related information problems make it more difficult for analysts (outsiders) to forecast firm cash flows as the transparency hypothesis. The transparency hypothesis predicts that, compared with focused firms, diversified firms should have, all else equal, larger forecast errors, more dispersion among analysts’ forecasts, larger 1

Intrinsic value is also known as true value, fair value, underlying value, or fundamental value. Stephen H. Penman (2010), Financial Statement Analysis and Security Valuation (4th ed.), International: The McGraw-Hill Companies. 3 There are three main activities, operating, investing, and financing activities, of a firm. Operating activities try to maximize the profit of the firm by well operations. Investing activities use the cash raised from financing activities and generated in operations to acquire assets to be employed in operations. Both activities can create additional value for the firm. But the financing activities are investing activities for the claimants not for the firm. It can’t add firm’s value (Penman, 2010). 4 The three measures of diversification are number of segments, Asset-based Herfindahl Index and Sale-based Herfindahl Index. The further information is listed in Section 2.1.2. 2

1755


revaluations around earnings announcements, and smaller earnings response coefficients (ERCs). To the extent that they are less transparent than focused firms, diversified firms will face more difficulty in raising capital, less stock market liquidity, and, therefore, higher costs of capital (Thomas, 2002). Above all, previous studies indicate that mispricing is associated with information asymmetry and investor sentiment, and also found that diversified firms have higher degree of information asymmetry than focused firms. Meanwhile, firm attempts to assure survival and pursue growth through diversification strategy. However, there are few literatures discuss the relation between mispricing and diversification proxy for growth strategy. Therefore, we mainly examine the relationship between implied premium and diversification. The main purpose in this paper is to verify that diversified firms have higher implied premium, resulting from the higher degree of information asymmetry for diversified firms than focus firms. We further investigate this relationship by group. We group our samples by SIC code to examine how the growth strategy affects the implied premium in different groups. We classify the whole sample into two types, one is including financial institutions and another is without them. Furthermore, we classified each group into high-tech and non high-tech firms5. Most of literatures excluded financial institutions (F/I) from their samples because that the financial structure for financial industry is quite different to other industries. However, some literatures indicate that the market value of financial institutions is also significantly associated with diversification. Laeven and Levine (2007) find that there is a diversification discount: The market values of financial conglomerates that engage in multiple activities, e.g., lending and non-lending financial services, are lower than if those financial conglomerates were broken into financial intermediaries that specialize in the individual activities. While difficult to identify a single causal factor, the results are consistent with theories that stress intensified agency problems in financial conglomerates engaged in multiple activities and indicate that economies of scope are not sufficiently large to produce a diversification premium. Thus, in this paper, we particularly examine the relation between implied premium and growth strategy for F/I-included and F/I-excluded samples. From the event of .com Bubble in the period of 1998 to 20006, also called Internet or Technology Bubble, we can observe that investors are always eager to pay more for high growth firms cause that its market value is deviated from fair value. High-tech firms have higher capital expenditure, so they must have greater growth potential and people might anticipate they have better performance in the future than that at the present. In the research of Bessiere and Elkemali (2011), they verify the hypothesis that if analysts exhibit overconfidence, they will overreact before the announcement and underreact after the announcement, and the misreactions (described former) will be greater for high-tech firms compared to low-tech firms is true. This hypothesis indicates that high-tech firms easily have greater misreactions for information. Hence, we divide our sample into high-tech and non high-tech firms to examine whether the implied premium of high-tech firms are really more significant associated with diversification than that of non high-tech firms. The first problem in this paper is the way to determine implied premium. We refer Baginski and Wahlen (2003) but revising the formula slightly that we define implied premium as the difference between intrinsic value 5

Following Brown et al. (2009), the largest three-digit high-tech industries are drugs (SIC 283), office and computing equipment (SIC 357), communications equipment (SIC 366), electronic components (SIC 367), scientific instruments (SIC 382), medical instruments (SIC 384), and software (SIC 737). The SIC code of financial institutions is 6000-6999. 6 Thedot-com bubble (also referred to as the Internet bubble and the Information Technology Bubble) was a speculative bubble covering roughly 1995–2000 (with a climax on March 10, 2000, with the NASDAQ peaking at 5132.52 in intraday trading before closing at 5048.62) during which stock markets in industrialized nations saw their equity value rise rapidly from growth in the Internet sector and related fields (from Wikipedia, the free encyclopedia). 1756


and actual fiscal year-end close price, that is, IP P IV ⁄P , where IPt denotes implied premium at time t, IVt denotes share intrinsic value at time t, and Pt is actual fiscal year-end close price at time t7. When IP is positive, it represents that firm value is overpriced by market. On the opposite, firm value is underpriced by market when IP is negative. For IV, we apply the residual income model8 which is the present value of future residual earning (RE)9 (Ebrahimi & Sarikhani, 2011; Higgins, 2011; Penman, 2010). The second problem is the forecast horizon of RE. According to Richardson and Tinaikar (2004), there exist detective links between historical and forecast data branches, which often produce similar results. Moreover, long-tem analyst earning forecasts into RE have been proven not to improve pricing performance significantly (Lo & Lys, 2001). Some, such as Frankel and Lee (1999), continue to use shorter forecast horizon with one- and two-year ahead analyst earnings forecasts; however, these still suffer from biases in forecasting errors. The limitations encountered by previous studies suggest the validity of using historical EPS over forecast EPS in this paper. The third problem with intrinsic value concerns required return of equity, denoted as r. r must be estimated, and is often viewed exogenous. Yoo et al. (2004) use CAPM-derived ICOE because individual betas predict positive and symmetric association ICOE in the literature. Banginski and Wahlwn (2003) indicate that the accounting-related risk measurements (i.e., the systematic risk and total volatility in a firm’s time series of residual return of equity) are associated with the market’s assessment and pricing of equity risk. Furthermore, their results show that the explanatory power of total volatility is incremental to the Fama and French (1992) factors, market beta, firm size, and the market-to-book ratio. Hence, in our research, we follow Hahn and Lee (2009), using FFrt as our required return of residual income model, and also denoted as r10. Several firm characteristics, including sales growth, financial constraints, growth opportunities, and growth of profitability, are viewed as minor independent variables. These variables are recognized as existing influences on firm value in extensive prior studies. We consider that investors may regard a firm with growth of sales, opportunity, and profitability as a firm with excellent profit performance in the future. And they will be willing to pay more for it, resulting in higher implied premium. The remainder of this paper is organized as follows. Section 2 describe the methodology, including sample selection, research model, and determination of intrinsic value. Section 3 presents and discusses the empirical results. Section 4 provides a summary of our main findings and the conclusion.

2. Methodology 7

In Baginski and Wahlen (2003), the definition of mispricing, IP IV P ⁄P . According to Penman (2010), Firm’s intrinsic value derived from residual income (also known as residual earning) model is consist of three parts, book value, value from short-term forecast, and value from long-term value (continuing value or terminal premium). The detailed information is provided in Section 2.1.1. 9 Residual earning is also known as economic value added (EVA) or residual income (RI). Basic formula of RE in per share is listed below: RE EPS r BPS whereEPSt denotes forecasted EPS at time t-1; r denotes required return of equity; BPSt-1 denotes book value of equity per share at time t. 10 The equation of FFr is listed below: 8

FFr

R

r

β

F

where Rit is stock return of firm i at time t, rft is the risk-free rate at time t, and Fkt denotes one of the Fama an d French four-factor loading (MKT, SMB, HML, and MOM). 1757


2.1 Data Collection and Variable Definition The sample consists of S&P 500 members (COMPUSTAT auto-selection in 2011) over the 1998 to 2007 period. Our sample of firms is the intersection of CRSP, COMPUSTAT, and WRDS, and Research databases meeting the following requirements, applied yearly from 2001 to 2010: (1) From COMPUSTAT, we collect the financial statement and accounting data such as actual fiscal year-end close price (Pit), capital expenditure, book value of total assets, operating cycle, average payment period for accounts payable, debt ratio, and EBITDA. (2) From CRSP11, we estimate individual beta for at least 24-60 month ahead CRSP stock returns. (3) From WRDS12, we obtain segment information to compute the two measurements of diversification. (4) From World Economic Outlook (WEO) published by IMF, we collect the data of GDP growth rate of U.S. The definition of key variables is reported in Table 1. Table 1 Definition of Key Variables This table displays the definition of key variables. The sample period is from 1998 to 2007. Measurement

Variables

Definition

Implied Premium

IP

IP

NoSeg

The number of segments

S_HI

HI

1

SegSalse⁄Sales

Bowen & Wiersema, 2005

A_HI

HI

1

SegAssets⁄Assets

Bowen & Wiersema, 2005

∆Sales

(Salest-Salest-1)/ Salest-1

Pajuste & Benjamin, 2005

Growth Strategy

Sales Growth

P

IV ⁄P

Reference Penman 2010; Baginski & Wahlen 2003 Duchin 2010

Growth Opportunity

CAPEX

Capital expenditure/book value of total assets

Duchin, 2010

Size

LnTA

Profitability Growth

∆ROA

Natural logarithm of total assets ROA = EBITDA/book value of total assets ∆ROA=(ROAt-ROAt-1)/ROAt-1

Gozzi et al., 2008 Hahn & Lee, 2009 Martin & Francis, 2010

2.1.1 Implied Premium Baginski and Wahlen (2003) define the price differentials (PDIFFit) as RFVit minus Pit, where RFVit is intrinsic value computed by residual income model, and Pit is the price per share for firm i as of April 1 of each sample year for which we have analysts’ earnings forecast data. For regression analysis, they take the form as PDIFFit/Pit. But we revise the equation slightly that we define implied premium as the difference between intrinsic value and actual fiscal year-end close price, that is, IP P IV ⁄P , where IPt denotes implied premium at time t, IVt denotes share intrinsic value at time t, and Pt is actual fiscal year-end close price at time t. When IP is positive, it represents that firm value is overpriced by market. On the opposite, firm value is underpriced by market when IP is negative. (1) Intrinsic Value (IV) Felthman and Ohlson (1995) model the relation between a firm value and accounting data concerning operating and financial activities. Book value equals value for financial activities, but they can differ for operating activities. Firm value is assumed to equal the net present value of expected future dividends, and is shown, under clean surplus accounting, to also equal book value of expected future abnormal earnings (which equals accounting 11

COUMSTAT only provides individual beta for recent five years, so we follow the COMPUSTAT procedure for beta estimation of individual firms to compute r. Detail is listed in Section 2.1.1. 12 COUMSTAT only provides segment data for recent five years, so we obtain these data over the 1998 to 2007 period from Historical Segment of COMPUSTAT from WRDS. 1758


earnings minus an interest charge on opening book value. It demonstrates that the conclusions hinge on the extent to which the accounting is conservative as opposed to unbiased. Further, the absence/presence of growth in operating activities is relevant if, and only if, the accounting is conservative. Afterward literatures develop other ways to evaluate firm’s intrinsic value, such as multiples analysis, free cash flow discounted model, dividend valuation model, residual income model, and abnormal earning growth model. The multiples analysis is the easiest but inaccuracy because it considers fewer financial information of the firm, and it doesn’t put future profitability into consideration. In addition, Preinreich (1938), Edwards and Bell (1961), Peasnell (1982), Ohlson (1995), and others show that the dividend valuation model is equivalent to the residual income valuation model. Hence, we follow previous literatures, using the residual income model as the approach to compute the intrinsic value. The computation of intrinsic value from residual income model is composed of Book Value, Short-term Forecast Value, and Continuing Value13. The first portion, Book Value, is known for sure, and so firmly anchors the valuation. The second is based on forecast for two years ahead. These are typically made with some confidence, but with less assurance than the book value component. The value from these forecasts is the sum of the present value of the one- and two-year-head residual earnings. It forecast no growth in residual earnings after two years. The third portion adds value for growth. The long-term growth rate is usually fairly uncertain, so the component of the valuation is the most speculative (Penman, 2010). We estimate the growth rate as GDP growth rate of U.S after two-year-ahead14. The formula of residual income model to compute intrinsic value is shown below: IV

BPS

RE

RE

RE

(1)

Where IVt denotes share intrinsic value for firm i at time t, REt computed by EPS represents residual earnings per share of each firms at time t, and r represents a firm’s required return rate15. g denotes firm’s perpetual growth rate at a constant rate. We calculate intrinsic value by historical EPS (earnings per share), BPS (book value per share), and DPS (dividend per share) collected from COMPUSTAT16. The most important for residual income model is r and g. Because we can’t obtain forecast growth rate for 3-5 year from I/B/E/S on Data stream, we follows Penman (2010) using the GDP growth rate as g17.We obtain GDP growth rate from World Economic Outlook (WEO) of IMF. But this may still suffer some bias. On the other hand, the determination of r is chosen from two types of measurements of required return of equity, reported completely in next section. (2) Determination of Discount Rate (r) We have considered two measurements as our r, CAPM-derived ICOE or Fama-French expected return (FFr). We choose FFr as r finally.

13

Continuing value can be measured under constant growth rate assumption or zero growth assumption. We take the former assumption to fit the reality as likely as possible. 14 Penman (2010), states that GDP growth rate can be used as the perpetual growth rate. We obtain the data of GDP growth rate in U.S. from World Economic Outlook (WEO) published by IMF. 15 r represents the required return of equity, derived from Fama-French four factors, which is listed completely in next section “2.Determination of r”. 16 In general, we compute the continuing value of residual income model by using forecast data of EPS, DPS from I/B/E/S on Datastream, and the forecast BPS which is determined by the following equation, BPSt = BPSt-1+EPSt-DPSt. Richardson and Tinaikar (2004) claim that there exist detective links between historical and forecast data branches, which often produce similar results. It is hard to obtain forecast data (lack of I/B/E/S database) so we use historical data to substitute it. 17 Penman claims that firm’s long-term growth rate must not be larger than national GDP growth rate in general. Thus, Penman use national GDP growth rate as the perpetual growth rate of firms. 1759


We follow the COMPUSTAT procedure for beta estimation of individual firms. The data is only traceable within five years of the present date. For data out of this range, we adapt the formula and steps set in the database using S&P 500 Index returns as market returns (RM), risk-free rate (rf), stock returns of each firms (Ri) in monthly data form, to estimate current individual beta (β ).At least 24-60 previous observations are required to meet the regression requirements. We then substitute individual beta (β ) into CAPM to obtain ICOE for individual firms. The formula of ICOE computation is presented below: CAPMderived ICOE (2) r β RM r Where β is the beta of firm at time t, RMt-rft is the market premium at time t, and rft is the risk-free rate of U.S at time t18. We estimate the Fama-French expected stock return (FFr) by the procedures in (Hahn & Lee, 2009): estimating the Fama and French factor loadings (k) for individual stock i using monthly rolling regressions with a 60-month window every month requires at least 24 monthly return observations in a given window and substituting those betas into the model to obtain expected stock returns. The equation of computing FFr is reported as following: ∑ FFr R r β F (3) Where Rt is stock return of firm at time t, rft is the risk-free rate at time t, and Fkt denotes one of the Fama an d French four-factor loading (MKT, SMB, HML, and MOM)19. Banginski and Wahlwn (2003) indicate that the accounting-related risk measurements (i.e., the systematic risk and total volatility in a firm’s time series of residual return of equity) are associated with the market’s assessment and pricing of equity risk. Furthermore, their results show that the explanatory power of total volatility is incremental to the Fama and French (1992) factors. Scholars also think that FFr can reflect the required return of equity better than CAPM-derived ICOE because it considers more risk factors. Thus, we measure firm’s required return of equity by FFr, denoted as r, of residual income model, and the regression result of using FFr is significant, presented in Section 4. 2.1.2 Diversification Measures (Major Independent Variable) Proxy variables of diversification are extensively discussed in many literatures. Conventional wisdom among finance scholars suggests that corporate diversification, especially conglomerate diversification, destroys shareholder wealth such that the shares of diversified firms sell at a discount. This link between diversification and value destruction is made in virtually every finance text. For example, a leading MBA finance texts put it this way, “diversification, by itself, cannot produce increases in value” (Ross et al., 1999). Furthermore, Brealey and Myers (2000) argue this is because “diversification is easier and cheaper for the stockholder than for the corporation.”Yet, major U.S. corporations remain highly diversified. Montgomery (1994) identifies three main theoretical perspectives that can be used to explain why a firm might choose to diversify: agency theory, the resource based view, and market power. Historically, corporate diversification has been measured using either the business count approach or the strategic approach. Following the business count method, diversification is assessed using Standard Industrial Classification (SIC) codes20 and corporate line-of-business data that are reported to the Securities and Exchange 18

RMt-rftand rft are collected from Fama-French website. Rit is obtained from COUPSTAT. rft and Fama-French four-factor are all collected from Fama-French website. 20 SIC data is comprised of a four-digit scheme that can be used to define increasingly more refined measures of business or industry affiliation. The first two digits of the four-digit code “20” represent the broadest industry grouping. We take Food and Kindred 19

1760


Commission annually. All of these measures share the common feature that they can be objectively calculated from publicly available data. The strategic approach is very subjective and relies less on SIC data and more on the judgment of the researcher (Martin & Sayrak, 2001). The simplest business count measure of corporate diversification is the number of industry groups in which a firm operates. So we defined one of our diversification measurements as number of segment (NoSeg) collected from COMPUSTAT. COMPUSTAT only provides line-of-business information within recent five years, so we obtain these data from WRDS. There’s a problem with simply counting the number of SIC codes for the firm’s different business units to measure diversification is that this measure fails to capture the relative importance or distribution of the firm’s involvement in each industry segment. To solve this problem, Berry (1971) and McVey (1972) suggest that the use of the Herfindahl index, which was originally developed as a measure of industry concentration. The Herfindahl index can be used to capture the relative importance of the firm’s different business segments for a single SIC classification level. There are two types of Herfindahl index, Asset-based and Sale-based. Asset-based Herfindahl index is computed as (SegAsset/Assets)2, and the other, Sale-based Herfindahl index is determined as (SegSales/Sales)2. In our research, we defined the other two diversification measurement as Sale-based and Asset-based Herfindahl Index. Since lower values of the Herfindahl index indicate higher levels of diversification we instead use the inverse measure, HI = 1/(SegSales/Sales)2 for consistency with the other diversification measures used here. This inverse measure equals one for a single business firm and it rises with the level of diversification (Bowen & Wiersema, 2005). 2.1.3 Minor Independent Variables Firm value is created by operating and investing activities. We have mentioned that growth strategy is a kind of way of investing. In this paper, we mainly discuss the relationship between implied premium and growth strategy, which is proxied by diversification, and we consider it as our major independent variable in our regression model. In order to regress our model more completely, we further add some variable affect firm value as our minor independent variables in regression model. For institution, people prefer to invest firms with improvement of sales and operating performance. We use the percentage of change in sales as the proxy for sales growth (Pajuste & Benjamin, 2005). ROA21 is considered as proxy variable for firm performance in a large body of literatures (Hahn & Lee, 2009; Mukherji & Pettus, 2008; Klapper, 2004). We use the change in ROA as the proxy for firm’s improvement of operating performance (Martin & Francis, 2010). ROA can also be used to measure profitability. Firm make investment decisions to pursue growth opportunity, measured by the ratio of capital expenditure (CAPEX) to book value of total assets (Duchin, 2010). Previous studies usually classify state the growth opportunity of investment is realized as future profitability. Thus, profitability and investments are classified as categories of profitability as well (Hahn & Lee, 2009; Tim & Vidhan, 2008). As Penman (2010) says, “Don’t pay too much for the growth.” Growth of profitability has positive contribution to firms value (Hahn & Lee, 2009), and as naturally we expect to observe a positive relation between investments, profitability, and implied premium. Products for an example. After adding a third digit “201”, we narrow the Food and Kindred Products group down to only those firms involved in Meat Products. Finally, adding a fourth digit “2013”, we define the code for firms engaged in Food and Kindred Products—Meat Products—Sausages and other Prepared Meats 21 ROA is the ratio of EBITDA to book value of total assets, and be interpreted as cash-based ROA (Aggarwal & Kyaw, 2006). 1761


We use the natural log if a firm’s assets at the end of the year as the proxy for firm size (Gozzi et al., 2008)22. Firm size is considered a determinant of financial constraints or capital market excess (Timan & Wessels, 1988) that affects decisions of managers and firm value (Lee & Chuang, 2009). It is positively related to firm value (Maury, 2006) because small firms are younger and less well known, and there are therefore more likely to face financing constraints and vulnerable to capital market imperfections arising from information asymmetries and collateral constraints (Gertler & Gilchrist, 1994). Larger size firms have greater degree of information asymmetries, so resulting in higher mispricing (Thomas, 2002). However, in some cases, asset size also serves as proxy for firm risk. Some studies claim that size has positive effect on the risk taking of a firm due to the moral hazard associated with “too-big-to-fail” policy (Boyd et al., 2009), whereas others suggest a negative correlation between firm size and risk. Above all, we expect to find positive correlation between firm size and implied premium. 2.2 Sample Selection Criteria In addition, the sample selection criteria are as following: (1) Criteria 1: missing data (2) Criteria 223: r ≤ 0, r ≤ g and g ≤ 024 (3) Criteria 3: grouping by SIC code We set four criteria for our sample selection. First, we exclude sample with incomplete data. Second, we retain samples satisfied with the assumption of residual income model. At the last, we group our sample by SIC code. The number of sample for each phase is listed in Figure 1. N = 50,000 Criteria 1

N = 3,135 Criteria 2

N = 1,634 Criteria 3

With Financial institutions Without Financial institutions

Whole Sample 1.634 1.518

High-tech firms 367 367

Non high-tech firms 1,267 1,151

Figure 1 Number of Sample for Each Criterion

We have original data of 500 firms with 10 years (5,000 samples). After Criteria 1, we remain 3,135 samples due to trimming off samples with missing data. Then we obey the assumption of residual income model, remaining half approximately samples after Criteria 2. At the last, we group our final sample by SIC code. We can find that high-tech firms and financial institutions account for about 22.5% and 7.1% of whole sample respectively. In the following, we attempt to develop the expected signs of coefficients of those variables reported in Table 1in the following discussion. We provide descriptive statistics of key variables for whole sample in Table 2, by years and groups in Table 3 and Table 4, and we also provide Pearson Correlation Coefficients Test for whole 22

Other proxy variables for firm size also exist, such as natural log of a firm’s total sale or market value of equity. According to Residual Income Model from Penman (2010), the required return of equity (r) has to be positive and more than growth rate (g), and the growth rate need to be positive. The detail of residual income model is listed in Section 2.1.1. 24 The growth rates are negative in 2008 and 2009, so the samples over the 2008 to 2009 period are excluded. 23

1762


sample in Table 5 and for groups in Table 6, all presented in Section 3. 2.3 Robust Regression Model One of the most important statistical tools is a linear regression analysis for many fields. Nearly all regression analysis relies on the method of least squares for estimation of the parameters in the model. A problem that we often encountered in the application of the application of regression is the presence of an outlier or outliers in the data. Outliers can be generated by from a simple operational mistake to including small sample from a different population, and they make serious effects of statistical inference. Even one outlying observation can destroy least squares estimation, resulting in parameter estimates that do not provide useful information for the majority of the data. Robust Regression has been developed as an improvement to least squares estimation in the presence of outliers and to provide us information about what a valid observation is and whether this should be thrown out.25 There are two methods for robust regression, least squares alternatives and parametric alternatives, and we develop the former method by using SAS statistics software. For least squares alternatives method, the simplest methods of estimating parameters in a regression model that are less sensitive to outliers than the least squares estimates, is to use least absolute deviations. Even then, gross outliers can still have a considerable impact on the model, motivating research into even more robust approaches. In 1973, Huber introduced M-estimation for regression. The M in M-estimation stands for “maximum likelihood type”. The method is robust to outliers in the response variable, but turned out not to be resistant to outliers in the explanatory variables (leverage points). In fact, when there are outliers in the explanatory variables, the method has no advantage over least squares. In the 1980s, several alternatives to M-estimation were proposed as attempts to overcome the lack of resistance. See the book by Rousseeuw26 and Leroy for a very practical review. Least trimmed squares (LTS) is a viable alternative and is presently the preferred choice of Rousseeuw and Ryan27. The Theil-Sen estimator has a lower breakdown point than LTS but is statistically efficient and popular. Another proposed solution was S-estimation. This method finds a line that minimizes a robust estimate of the scale (from which the method gets the S in its name) of the residuals. This method is highly resistant to leverage points, and is robust to outliers in the response. However, this method was also found to be inefficient. MM-estimation attempts to retain the robustness and resistance of S-estimation, while gaining the efficiency of M-estimation. The method proceeds by finding a highly robust and resistant S-estimate that minimizes an M-estimate of the scale of the residuals (the first M in the method’s name). The estimated scale is then held constant whilst a close-by M-estimate of the parameters is located (the second M). We conduct robust regression for the following regression equations: Model 1: IP a β NoSeg ∑ β , X ∑ Model 2: IP a β S_HI β , X Model 3: IP

a

β

A_HI

∑

β

,

X

Where IPi denote the implied premium for firm I, Xik denotes minor independent variables as ∆Sales, CAPEX, SIZE, ∆ROA for firm i (definition of each variable is reported in Table 1). Our major independent 25

The description of robust regression analysis is cited from “Robust Regression” written by Lalmohan Bhar. And the following information about robust regression methods is obtained from Wikipedia, the free encyclopedia. 26 Rousseeuw P. J. and Leroy A. M. (2003), Robust Regression and Outlier Detection (1st ed.), John Wiley & Sons Inc. 27 Ryan Thomas P. (2008), Modern Regression Methods(2nd ed.), John Wiley & Sons Inc. 1763


variable, NoSeg, S_HI, and A_HI are three measures of diversification we’ve mentioned in Section 2.1.2.

3. Results We first analyze our effective sample through descriptive statistics. Then we investigate the relationship between implied premium and diversification by robust regression. The results and analysis are presented below. 3.1 Descriptive Statistics of Key Variables28 Table 2 reports the summary statistics of key variables reported as mean and median for whole sample and by groups over the period from 1998 to 2007. In Table 2, we can find that IP are positive whether samples include or exclude financial institutions. It means that firm’s market value is overpriced by market. Furthermore, mean and median of implied premium (IP) for high-tech firms is larger than that for non-high-tech firms, so we infer that the degree of misreaction of high-tech firms is larger than non high-tech firms. It is consistent with Bessiere and Elkemali (2011) that the misreactions will be greater for high-tech firms compared to low-tech firms. Besides, we can also find that financial institutions have positive IP, and it is less than that high-tech firms have, but larger than that non high-tech firms have. There is slightly difference of mean and median by group between F/I-included and F/I-excluded samples for diversification measured by NoSeg, S_HI and A_HI. Non high-tech firms have greater degree of diversification than high-tech firms have, but approximately equal to that financial institutions have. Table 2 Descriptive Statistics of Key Variables This table displays the summary statistics of key variables reported as mean and median for whole sample and by groups over the period from 1998 to 2007. Panel A shows the descriptive statistics, mean and median for sample including financial institutions and for high-tech and non high-tech firms. Otherwise, Panel B reports the descriptive statistics, mean and median for sample excluding financial institutions and for financial institutions and non high-tech firms. The number of sample is presented below, denoted as N. Panel A. Samples include financial institutions Whole sample High-tech firms Non high-tech firms Mean Median Mean Median Mean Median IP(%) 41.02 82.58 57.30 91.29 36.30 79.70 NoSeg 3.21 3.00 2.38 1.00 3.45 3.00 S_HI 6.16 1.52 1.91 1.29 7.40 1.59 A_HI 8.84 1.77 2.73 1.37 10.61 1.86 N 1,634 367 1,267 Panel B. Samples without financial institutions Whole sample Financial Institutions Non high-tech firms Mean Median Mean Median Mean Median IP(%) 40.88 82.49 42.79 84.01 35.64 79.21 NoSeg 3.19 3.00 3.48 3.00 3.45 3.00 S_HI 6.47 1.52 2.14 1.46 7.93 1.60 A_HI 8.48 1.80 13.65 1.56 10.31 1.90 N 1,518 116 1,151

28

We can infer the similar results from descriptive statistics by using risk-free rate as the implied cost of equity of residual income model to compute implied premium, denoted as IPrf. Mean and median of IPrf for high-tech firms is larger than that for non-high-tech firms. Financial institutions have positive IP, and it is less than that high-tech firms have, but larger than that non high-tech firms have. 1764


Table 3 and Table 4 report the summary statistics of key variables reported as mean and median for whole sample by years. We can find that IP for each year are almost positive for each group in Table 3 and Table 4. In Table 3, we find that the degree of IP over the period from 1998 to 2000 is much greater than other years by whether the whole samples include or exclude financial institutions. It means that investors overestimated firm value due to too high anticipation for growth in that period. It also can be verified by .com bubble, which happened in the period of 1995 to 2000. People pay too much for the growth lead bubble happened, resulted in higher implied premium (Penman, 2010). It is consistent with Table 2 that the degree of misreaction of high-tech firms is larger than non high-tech firms by years Non high-tech firms have slightly greater degree of diversified than high-tech firms by years. There is no obvious change of mean and median of diversification measured by NoSeg, S_HI and A_HI for whole samples whether include or exclude financial institutions by each year. We can observe the similar results as in Table 3 and Table 4. In Table 4, the degree of IP over the period from 1998 to 2000 is much greater than other years for high-tech and financial firms (listed in Panel A and D in Table 4). Besides, the degree of IP over the period from 1999 to 2000 is much greater than other years by whether the non high-tech firms include or exclude financial institutions (listed in Panel B and C in Table 4). These are consistent with .com bubble, which happened in the period of 1995 to 2000. There is no obvious change of mean and median of diversification measured by NoSeg, S_HI and A_HI for each gorup by each year. Table 3

Descriptive Statistics of Key Variables for Whole Sample by Years

This table displays the summary statistics of key variables reported as mean and median for whole sample by years over the period from 1998 to 2007. Panel A shows the descriptive statistics, mean and median for sample including financial institutions. Otherwise, Panel B reports the descriptive statistics, mean and median for sample excluding financial institutions. The number of sample is presented below, denoted as N. Panel A. Descriptive statistics of key variables for whole sample with financial institutions IP(%) NoSeg S_HI A_HI Mean Median Mean Median Mean Median Mean Median N 1998 54.29 94.86 2.28 1.00 3.84 1.46 3.63 1.48 129 1999 77.33 98.59 2.92 3.00 2.63 1.57 2.81 1.71 83 2000 75.91 94.22 3.22 3.00 2.14 1.54 3.89 1.66 191 2001 48.59 85.94 3.06 3.00 1.95 1.45 4.27 1.66 172 2002 2.23 63.32 3.15 3.00 2.02 1.42 2.43 1.64 177 2003 12.47 75.72 3.14 3.00 1.87 1.40 4.28 1.70 170 2004 46.87 76.32 3.38 3.00 2.03 1.57 3.55 1.94 176 2005 46.85 83.30 3.33 3.00 1.93 1.47 3.90 1.85 175 2006 16.57 67.86 3.60 3.00 2.18 1.87 5.10 2.04 173 2007 51.60 82.72 3.63 3.50 2.20 1.79 3.58 2.02 188 Panel B. Descriptive statistics of key variables for whole sample without financial institutions IP(%) NoSeg S_HI A_HI Mean Median Mean Median Mean Median Mean Median N 1998 53.45 94.86 2.26 1.00 2.39 1.46 2.87 1.70 123 1999 76.44 99.04 2.86 3.00 2.60 1.44 2.80 1.73 78 2000 74.54 94.07 3.24 3.00 2.15 1.54 3.04 1.65 178 2001 48.16 86.42 3.04 3.00 1.90 1.48 2.48 1.64 165 2002 -1.40 64.65 3.13 3.00 1.97 1.42 2.50 1.72 166 2003 17.33 76.00 3.12 3.00 1.88 1.43 2.59 1.99 158 2004 47.90 76.67 3.35 3.00 2.05 1.64 2.59 1.99 164 2005 41.92 82.58 3.28 3.00 1.91 1.47 3.00 1.92 153 2006 15.87 66.81 3.58 3.00 2.18 1.82 4.28 2.23 161 2007 52.18 82.94 3.65 4.00 2.24 1.85 3.21 2.06 172

1765


Table 4

Descriptive Statistics of Key Variables for High-tech, Non High-tech Firms and Financial Institutions by Years

This table displays the summary statistics of key variables reported as mean and median by years over the period from 1998 to 2007. Panel A, B, C, and D report the descriptive statistics, mean and median for high-tech firms, non high-tech firms with financial institutions, non high-tech firms without financial institutions, and financial firms, respectively. The number of sample is presented below, denoted as N. Panel A. Descriptive statistics of key variables for high-tech firms IP(%) NoSeg S_HI Mean Median Mean Median Mean Median 1998 95.64 97.23 1.87 1.00 2.21 1.46 1999 100.71 100.16 2.17 1.00 3.08 1.37 2000 92.58 97.47 2.48 2.00 1.67 1.48 2001 58.72 91.71 2.36 1.00 1.65 1.00 2002 23.92 81.81 2.00 1.00 1.54 1.00 2003 44.38 78.88 2.24 1.00 1.56 1.29 2004 32.65 88.21 2.90 3.00 1.80 1.73 2005 73.94 88.38 2.65 2.00 1.74 1.15 2006 -27.33 65.42 2.59 2.00 1.84 1.15 2007 47.04 79.88 3.00 3.00 2.04 1.62 Panel B. Descriptive statistics of key variables for non high-tech firms with financial institutions IP(%) NoSeg S_HI Mean Median Mean Median Mean Median 1998 32.14 91.14 2.50 2.00 1.50 1.45 1999 59.43 93.95 3.49 3.00 2.28 1.96 2000 71.50 93.53 3.42 3.00 2.27 1.55 2001 44.44 83.94 3.34 3.00 2.08 1.52 2002 -5.18 -1.76 3.49 3.00 2.16 1.56 2003 2.33 74.83 3.43 3.00 1.97 1.44 2004 49.68 74.38 3.47 3.00 2.08 1.50 2005 42.13 82.58 3.45 3.00 1.97 1.54 2006 24.69 69.41 3.78 4.00 2.24 1.96 2007 52.57 82.85 3.76 4.00 2.24 1.84 Panel C. Descriptive statistics of key variables for non high-tech firms without financial institutions IP(%) NoSeg S_HI Mean Median Mean Median Mean Median 1998 29.11 89.34 2.49 2.00 2.65 1.45 1999 55.63 94.80 3.45 3.00 2.18 1.81 2000 69.31 93.21 3.46 3.00 2.29 1.58 2001 43.56 84.53 3.33 3.00 2.01 1.53 2002 -9.44 56.88 3.49 3.00 2.11 1.56 2003 7.85 75.57 3.43 3.00 2.00 1.47 2004 51.17 74.60 3.44 3.00 2.11 1.60 2005 35.37 81.62 3.41 3.00 1.94 1.47 2006 24.57 68.99 3.78 4.00 2.24 1.90 2007 53.40 83.02 3.81 4.00 2.28 1.91 Panel D. Descriptive statistics of key variables for financial institutions IP(%) NoSeg S_HI Mean Median Mean Median Mean Median 1998 71.49 95.52 2.67 2.50 1.57 1.26 1999 91.35 90.21 3.80 4.00 3.14 2.66 2000 94.66 95.32 3.00 2.00 2.07 1.09 2001 58.80 65.45 3.57 3.00 3.14 1.28 2002 -14.82 37.78 3.45 4.00 2.75 1.33 2003 -51.46 33.24 3.42 3.50 1.74 1.37 2004 32.91 54.95 3.75 3.50 1.74 1.29 2005 81.12 90.83 3.68 4.00 2.09 1.95 2006 25.94 85.89 3.83 3.50 2.24 2.23 2007 45.35 82.45 3.38 3.00 1.82 1.36

1766

Mean 5.19 3.16 2.28 2.03 1.89 2.02 2.50 3.00 2.55 2.60

A_HI Median 1.38 1.52 1.64 1.08 1.00 1.47 1.81 1.42 1.92 2.01

N 45 36 40 50 40 41 29 26 27 33

Mean 1.08 2.54 2.27 5.18 2.58 4.99 3.75 4.05 5.58 3.79

A_HI Median 1.64 1.83 1.55 1.76 1.77 1.77 2.00 1.88 2.09 2.04

N 84 47 151 122 137 129 147 149 146 155

Mean 2.70 2.62 2.96 3.49 2.67 2.67 2.61 3.00 4.63 3.35

A_HI Median 1.65 1.77 1.81 1.74 1.78 1.77 2.02 1.98 2.33 2.16

N 78 42 138 115 126 117 135 127 134 139

Mean 2.12 1.88 8.75 3.04 1.62 7.65 6.58 5.11 6.20 7.56

A_HI Median 1.16 2.13 1.18 1.82 1.41 1.51 1.47 1.63 1.84 1.71

N 6 5 13 7 11 12 12 22 12 16


Before we proceed with our discussion regarding expected signs of coefficients of variables, we must point out the collinearity problems exist in our model. We develop the Pearson Correlation Coefficients Test in Table 5 and Table 6 in the next section. 3.2 Discussion of Collinearity Problems Pearson correlation coefficients test can examine the correlation between each variable. If independent variables have significant correlation with each other, the regression model might suffer serious collinearity problem, violating assumptions of regression. Thus, we take the Pearson test first, and then compute the VIF for each model by groups for further collinearity examination. Table 5 reports the results of Pearson Correlation Coefficients Test for whole samples. In Table 5, we obtain similar results of Pearson Correlation Coefficients Test for whole samples whether include or excluded financial institutions. We observe that all independent variables are significantly related to dependent variable, IP, except for ∆ROA. It is rational that NoSeg is positive related to CAPEX and LnTA. Typical firms diversified by making acquisition (increasing capital expenditure and firm size) to pursue growth (diversification) (Hyland and Diltz, 2002). It can be interpreted by the significant positively correlation between NoSeg and CAPEX (0.1646 with 1% significance level), and NoSeg and LnTA (0.4960 within 1% significance level) in Panel A, for instance. A_HI is also significantly related to LnTA (0.0607 with 5% significance level in Panel A, and 0.0612 within 5% significance level in Panel B). Table 5 Pearson Correlation Coefficients Test for Whole Sample This table displays the Pearson Correlation Coefficients Test of each variable for whole sample. Panel A shows the results for whole sample including financial intuitions. Panel B shows the results for whole sample excluding financial intuitions. IP is dependent variable in our research. Major independent variables are NoSeg, S_HI, and A_HI, and minor independent variables are ∆Sales, CAPEX, LnTA, and ∆ROA. The number of sample is presented below, denoted as N. *, **, and *** indicate significance level at 10%, 5%, and 1%, respectively. Panel A. Whole sample with financial institutions samples (N=1,634) IP NoSeg S_HI A_HI ∆Sales CAPEX LnTA ∆ROA IP 1.0000 NoSeg 0.1069 *** 1.0000 S_HI 0.0041 * -0.0205 1.0000 A_HI 0.0100 * 0.0023 0.0359 1.0000 ∆Sales 0.0772 *** -0.1020 0.0052 0.0084 1.0000 CAPEX 0.0664 *** 0.1646 *** -0.0132 -0.0323 0.0634 ** 1.0000 LnTA 0.1012 *** 0.4960 *** 0.0104 0.0607 ** 0.1354 *** 0.1165 *** 1.0000 ∆ROA 0.0015 -0.0184 -0.0006 -0.0010 0.0038 -0.0079 -0.0120 1.0000 Panel B. Whole sample without financial institutions samples (N=1,518) IP NoSeg S_HI A_HI ∆Sales CAPEX LnTA ∆ROA IP 1.0000 NoSeg 0.1014 *** 1.0000 S_HI 0.0043 * -0.0213 1.0000 A_HI 0.0099 * -0.0076 0.0359 1.0000 ∆Sales 0.0706 *** -0.0884 *** 0.0055 0.0104 1.0000 CAPEX 0.0683 *** 0.1629 *** -0.0153 -0.0316 0.0646 ** 1.0000 LnTA 0.1090 *** 0.5033 *** 0.0126 0.0612 ** 0.1321 *** 0.0576 ** 1.0000 ∆ROA 0.0016 -0.0185 -0.0007 -0.0009 0.0039 -0.0094 -0.0117 1.0000

1767


We take VIF test to illustrate the collinearity problem in Table 6. In statistics, the variance inflation factor (VIF) quantifies the severity of multicollinearity in an ordinary least squares regression analysis. It provides an index that measures how much the variance (the square of the estimate’s standard deviation) of an estimated regression coefficient is increased because of collinearity. In practice, if VIF is larger than 10, collinearity problem exists; less than 10, the problem doesn’t exist. In Table 6, we observe that VIF are almost close to 1 for each model by group. Thus, we can conclude that there is no collinearity problem in our model. Table 6

Variance Inflation Factor (VIF)

This table display the variance inflation factor (VIF) for each variable in each regression model. Panel A reports VIF for whole sample. Panel B and C reports VIF for high-tech and non high-tech firms. IP is dependent variable in our research. Major independent variables are NoSeg, S_HI, and A_HI, and minor independent variables are ∆Sales, CAPEX, LnTA, and ∆ROA. The number of sample is presented below, denoted as N. Panel A. VIF for whole sample NoSeg

with Financial Institutions Samples

without Financial Institutions Samples

1.3488

1.3730

S_HI

1.0003

A_HI

1.0004 1.0047

1.0050

∆Sales

1.0223

1.0211

1.0214

1.0214

1.0212

1.0216

CAPEX

1.0317

1.0164

1.0169

1.0313

1.0070

1.0076

LnTA

1.3410

1.0312

1.0348

1.3548

1.0206

1.0243

∆ROA

1.0005

1.0002

1.0002

1.0005

1.0003

1.0003

N

1,634

1,634

1,634

1,518

1,518

1,518

Panel B. VIF for high-tech firms NoSeg



1.3067

1.3067

S_HI

1.0104

A_HI

1.0104 1.0033

1.0033

∆Sales

1.0712

1.0739

1.0708

1.0712

1.0739

1.0708

CAPEX

1.0093

1.0021

1.0019

1.0093

1.0021

1.0019

LnTA

1.3496

1.0776

1.0718

1.3496

1.0776

1.0718

∆ROA

1.0021

1.0009

1.0008

1.0021

1.0009

1.0008

N

367

367

367

367

367

367

Panel C. VIF for non high-tech firms NoSeg



1.3088

1.3476

S_HI

1.0004

A_HI ∆Sales

1.0006 1.0054

1.0138

1.0132

1.0135

1.0060 1.0155

1.0156

1.0160

CAPEX

1.0687

1.0462

1.0467

1.0702

1.0311

1.0317

LnTA

1.2921

1.0374

1.0412

1.2989

1.0179

1.0221

∆ROA

1.0019

1.0017

1.0017

1.0023

1.0020

1.0020

N

1,267

1,267

1,267

1,151

1,151

1,151

1768


3.3 Robust Regression Results29 Robust Regression is used to eliminate the influence of outliers on regression results. The primary purpose of robust regression analysis is to fit a model which represents the information in the majority of the data. We develop the robust regression by using three kinds of measures of diversification as the major independent variable to examine which measurement has the most significant association with implied premium. We also conduct our model by different groups. In panel A of Table 8, we observe similar results for whole samples whether include or exclude financial institutions. We find that diversification is positively correlated with IP, and NoSeg is the most significant. ∆Sales is positively correlated with IP within 1% significance level. It means people are willing to pay more for firms with sales growth. CAPEX is positively correlated with IP within 1% significance level. It means people are willing to pay more for firms with greater growth opportunities. ∆ROA is positively correlated with IP within 1% significance level. It means that people are willing to pay more for firms with profitability growth (growth of operating performance), but it is has lower significance level, 5%. LnTA is positively correlated with IP within 1% significance level. It is consistent with Thomas (2002), that firms with bigger size have larger degree of information asymmetry. Above all, firms with higher degree of diversification and growth of operating performance and sales, and larger size are easily overpriced by market. The adjusted R-square for whole sample whether includes or exclude financial institutions is located in 13-15%. We further examine the robust regression by high-tech and non high-tech firms. We have mentioned that literatures state that high-tech firms have more growth potential than non high-tech firms. Investors pay high attention on high-tech firms’ performance. Thus, the degree of misreaction of high-tech firms is larger than non high-tech firms. Compared with Panel C, we can find that for high-tech firms, all measures of diversification are much more significant positively correlated with IP than those for non high-tech firms. S_HI is the most significant. As we mentioned at the last paragraph, all minor dependent variables are significant positively correlated with IP. It means high-tech firms with higher degree of diversification and growth of operating performance and sales, and larger size are easily overpriced by market. For non high-tech firms, S_HI is not significant related to IP, but the other measures, NoSeg and A_HI are significant positively correlated with IP within 1% significance level. All minor dependent variables, except for ∆ROA, are significant positively correlated with IP. Non high-tech firms have stable profitability due to fewer investing activities to pursue growth. Therefore, it is rational that ∆ROA is not significant correlated with IP for non high-tech firms because there is no obvious change in ROA for non high-tech firms. The adjusted R-square for high-tech firms is approximately 18 or 19%, much higher than that of non high-tech firms, located in 10-14%.

29

We also use IPrf as the dependent variable in the same regression. We can find that only NoSeg is significant correlated with IPrf, other two measures, S_HI and A_HI are not significant related with IPrf for each group. Besides, adjusted R-square is only 9-10%, lower than the regression we construct in context (13-15%) for whole sample whether includes or exclude financial institutions. For high-tech firms, the adjusted R-square is only 12-15%, lower than the regression we construct in context (18-19%). For Non high-tech firms, the adjusted R-square is only 4-6%, lower than the regression we construct in context (10-15%). Thus, we infer that IP is more effective than IPrf. 1769


Table 7

Robust Regression

This table reports the regression coefficients from the robust regression model over the period from 1998 to 2007. It also reports the associated t-statistics in parentheses. Adjusted R-square for each model is also provided. Three measures of diversification are used as the main independent variable for each of three kinds of models. Panel A, B, and shows the results of robust regression for whole sample, high-tech and non high-tech firms, respectively. IP is dependent variable in our research. Major independent variables are NoSeg, S_HI, and A_HI, and minor independent variables are ∆Sales, CAPEX, LnTA, and ∆ROA. The number of sample is presented below, denoted as N. *, **, and *** indicate significance level at 10%, 5%, and 1%, respectively. Panel A. Robust Regression for whole sample with Financial Institutions Samples without Financial Institutions Samples Dependent variable: IP Intercept 104.40 *** 108.01 *** 108.37 *** 108.59 *** 112.95 *** 113.36 *** (30.46) (31.8) (31.92) (29.98) (31.41) (31.47) NoSeg 1.5346 *** 1.5597 *** (5.47) (5.41) S_HI 0.0020 ** 0.0018 * (2.08) (1.77) A_HI 0.0039 * 0.0042 * (1.7) (1.82) ∆Sales 0.1341 *** 0.1444 *** 0.1431 *** 0.1125 *** 0.1180 *** 0.1167 *** (8.15) (8.48) (8.43) (6.95) (7.03) (6.97) CAPEX 0.2707 *** 0. 3164 *** 0.3222 *** 0.3367 *** 0.4008 *** 0.4079 *** (2.96) (3.42) (3.49) (3.64) (4.26) (4.33) LnTA 2.4095 *** 3.4282 *** 3.4762 *** 2.9364 *** 4.0733 *** 4.1321 *** (5.85) (9.29) (3.69) (6.58) (10.2) (10.33) ∆ROA 0.0003 ** 0.0003 ** 0.0003 ** 0.0003 ** 0.0003 ** 0.0003 ** (2.43) (2.29) (2.3) (2.46) (2.29) (2.28) Adj. R2 0.1448 0.1268 0.1284 0.1545 0.1327 0.1343 N 1,448 1,448 1,448 1,345 1,347 1,347 Panel B. Robust Regression for high-tech firms with Financial Institutions Samples without Financial Institutions Samples Dependent variable: IP Intercept 106.55 *** 108.57 *** 108.52 *** 106.55 *** 108.57 *** 108.52 *** (21.2) (21.48) (21.66) (21.2) (21.48) (21.66) NoSeg 1.0446 ** 1.0446 ** (2.18) (2.18) S_HI 0.0656 *** 0.0656 *** (2.71) (2.71) A_HI 0.0914 * 0.0914 * (1.93) (1.93) ∆Sales 0.0640 *** 0.0771 *** 0.0757 *** 0.0640 *** 0.0771 *** 0.0757 *** (3.77) (4.71) (4.18) (3.77) (4.71) (4.18) CAPEX 0.6217 *** 0.6793 *** 0.6789 *** 0.6217 *** 0.6793 *** 0.6789 *** (3.09) (3.27) (3.30) (3.09) (3.27) (3.30) LnTA 2.1244 2.7928 *** 2.7914 *** 2.1244 2.7928 *** 2.7914 *** (0.00) (4.7) (4.75) (0.00) (4.7) (4.75) ∆ROA 0.0004 *** 0.0004 *** 0.0004 *** 0.0004 *** 0.0004 *** 0.0004 *** (4.08) (3.82) (3.85) (4.08) (3.82) (3.85) Adj. R2 0.1943 0.1831 0.1871 0.1943 0.1831 0.1871 N 321 324 323 321 324 323 Panel C. Robust Regression for non high-tech firms with Financial Institutions Samples without Financial Institutions Samples Dependent variable: IP Intercept 100.80 *** 103.83 *** 104.40 *** 106.51 *** 110.22 *** 110.78 ***

1770


NoSeg

(23.23) 1.5073 *** (4.58)

S_HI

(23.81)

(30.46)

(22.53) 1.48895 *** (4.28)

0.0018 (0.58)

A_HI ∆Sales CAPEX LnTA ∆ROA Adj. R2 N

0.1681 *** (7.18) 0.2567 ** (2.43) 2.2561 *** (4.48) 0.0000 (-0.08) 0.123 1,119

0.1739 *** (7.22) 0.3140 *** (2.93) 3.2244 *** (6.97) 0.0000 (-0.01) 0.1043 1,124

(23.83)

(23.93)

0.0015 (0.49) 1.5346 (5.47) 0.1341 (8.15) 0.2707 (2.96) 2.4095 (5.85) 0.0003 (2.43)

*** *** *** ***

0.1448 1,123

0.1336 (5.6) 0.3664 (3.35) 3.0237 (5.36) 0.0000 (-0.07)

*** *** ***

0.1302 1,020

0.1308 *** (5.47) 0.4439 *** (4.12) 4.0492 *** (8.07) 0.0000 (0.00) 0.1137 1,019

0.0044 (1.84) 0.1290 (5.41) 0.4525 (4.19) 4.1248 (8.20) 0.0000 (0.00)

*** *** *** ***

0.1162 1,019

4. Conclusion Our results show the growth strategy affects the deviation between market value and intrinsic value (also so called implied premium) due to the information asymmetry. Generally, firms take either diversification or focus strategy to create firm value. Prior literatures indicate that mispricing is attributed to asymmetric information or investor sentiment. However, seldom literatures investigated the relationship between implied premium and growth strategy. Therefore, we contribute to the studies by examining this relationship. The evidence shows that diversified firms have higher implied premium, implying that investors overprice diversified firms. Furthermore, we examine the relationship between implied premium and diversification by industry. We find that the degree of diversification of high-tech firms have greater significant association to implied premium than that of non high-tech firms. It implies that market always overreact to the high-tech firms. References: Baginski S. P. and Wahlen J. M. (2003). “Residual income risk, intrinsic values, and share prices”, Accounting Review, pp. 327-351. Baker M. and Wurgler J. (2006). “Investor sentiment and the cross-section of stock returns”, The Journal of Finance, Vol. 61, No. 4, pp. 1645-1680. Berger P. G. and Ofek E. (1995). “Diversification’s effect on firm value”, Journal of Financial Economics, Vol. 37, No. 1, pp. 39-65. Bessiere V. and Elkemali T. (2011). “Uncertainty and financial analysts’ overconfidence: European evidence between high-tech and low-tech firms”, Working Paper, University of Montpellier II. Bowen H. P. and Wiersema M. F. (2005). “Foreign-based competition and corporate diversification strategy”, Strategic Management Journal, Vol. 26, No. 12, pp. 1153-1171. Brown J. R. andFazzari S. M. et al. (2009). “Financing innovation and growth: Cash flow, external equity, and the 1990s R&D bom”, The Journal of Finance, Vol. 64, No. 1, pp. 151-185. Chen F. and Jorgensen B. et al. (2004). “Implied cost of equity capital in earnings-based valuation: International evidence”, Accounting and Business Research, Vol. 34, No. 4, pp. 323-344. Chiang M. C. and Tsai I. et al. (2011). “Fundamental indicators, bubbles in stock returns and investor sentiment”, The Quarterly Review of Economics and Finance, Vol. 51, No. 1, pp. 82-87. Denis D. J. and Sibilkov V. (2010). “Financial constraints, investment, and the value of cash holdings”, Review of Financial Studies, Vol. 23, No. 1, pp. 247-269. Duchin R. (2010). “Cash holdings and corporate diversification”, The Journal of Finance, Vol. 65, No. 3, pp. 955-992. Feltham G. A. and Ohlson J. A. (1995). “Valuation and clean surplus accounting for operating and financial activities”,

1771

The Implied Premium and Growth Strategy—Evidence from S&P 500 Contemporary Accounting Research, Vol. 11, No. 2, pp. 689-731. Francis J. R. and Martin X. (2010). “Acquisition profitability and timely loss recognition”, Journal of Accounting and Economics, Vol. 49, No. 1-2, pp. 161-178. Gozzi J. C. and Levine R. et al. (2008). “Internationalization and the evolution of corporate valuation”, Journal of Financial Economics, Vol. 88, No. 3, pp. 607-632. Hahn J. and Lee H. (2009). “Financial constraints, debt capacity, and the cross-section of stock returns”, The Journal of Finance, Vol. 64, No. 2, pp. 891-921. Hughes J. S. and Liu J. (2007). “Information asymmetry, diversification, and cost of capital”, Accounting Review, Vol. 82, No. 3, p. 705. Hyland D. C. and Diltz J. D. (2002). “Why firms diversify: An empirical examination”, Financial Management, pp. 51-81. Lakonishok J., Shleifer A. and Vishny R. W. (1994). “Contrarian investment, extrapolation, and risk”, Journal of Finance, Vol. 49, pp. 1541-1578. Maury B. and Pajuste A. (2005). “Multiple large shareholders and firm value”, Journal of Banking & Finance, Vol. 29, No. 7, pp. 1813-1834. Ohlson J. A. (1995). “Earnings, book values, and dividends in equity valuation”, Contemporary Accounting Research, Vol. 11, No. 2, pp. 661-687. Rustichini A. (2011). “Decision making and equilibria”, Synthese, pp.1-12. Stambaugh R. F. and Yu J. et al. (2011). “The short of it: Investor sentiment and anomalies”, Journal of Financial Economics, Vol. 104, No. 2, pp. 288-302. Stephen H. Penman (2010). Financial Statement Analysis and Security Valuation (4th ed.), International: The McGraw-Hill Companies. Thomas S. (2002). “Firm diversification and asymmetric information: Evidence from analysts’ forecasts and earnings announcements”, Journal of Financial Economics, Vol. 64, No. 3, pp. 373-396. Yalçın K. C. (2010). “Market rationality: Efficient market hypothesis versus market anomalies”, European Journal of Economic and Political Studies, Vol. 3, No. 2, pp. 23-38.

1772


Practical Aspects of Pre-evaluation of Eco-efficiency of Environmental Technologies Włodzimierz A. Sokół (Central Mining Institute, Plac Gwarkow 1, 40-166 Katowice, Poland)

Abstract: The development of each company operating in the global market and its competitiveness depends on the effectiveness in achieving strategic and operational objectives that are closely related to the use of eco-efficient technologies. In these times of increasing social awareness of protection against industrial pollution, decisions on the choice of technology should subside in the appropriate stage of the company environmental management system. Decisions regarding the selection of eco-efficient technologies follow from the identification of significant environmental aspects of the company’s activities and their impact on the environment. On this basis, long- and short-term objectives and specific projects are determined. These projects are implemented by using environmentaly friendly technologies. There are hundreds technology offers from different countries on the market. For that reason, potential buyers will expect recommendations relating to eco-efficiency technology offers being submitted by suppliers, in order to know which one is better than another without extended and very expensive analysis. For that reason, a simplified pre-evaluation approach was developed for the assessment of the eco-efficiency of a given technology, often based on very limited numerical data. The methodology was tested on more than four hundred offers of environmental technologies submitted by suppliers from seven Baltic Sea Region countries. The results of pre-evaluation of eco-efficiency enhancement depend on the efficient use of resources, energy and reduction of emissions into the air, water and soil are presented on example of searching for clean coal technology offers for heating of buildings. Key words: eco-efficiency; environmental technology; risk management; environmental management JEL codes: L21, O14, Q51

1. Introduction The development of each company operating in the global market and its competitiveness depends on the effectiveness of achieving the strategic and operational objectives of development adopted by its top management. This is closely related to the use of ecologically and economically efficient technologies. These technologies should be socially acceptable, and thus also socially efficient. In these times of increasing social awareness of protection against industrial pollution, decisions about the choice of technology should subside in the appropriate stage of company management, which greatly facilitates the establishment and functioning in an environmental Włodzimierz Antoni Sokół, D.Sc.Eng, Manager of International Projects, Director of the National Contact Point, for Eco-efficient Technologies and Management Systems, Central Mining Institute; research areas/interests: sustainable development, regional environmental management, risk management, environmental technologies, power engineering, energy efficiency, resource and eco-efficiency. E-mail: [email protected]. 1773

Practical Aspects of Pre-evaluation of Eco-efficiency of Environmental Technologies

management system (EMS) that meets the requirements of international standards such as ISO 14001 (ISO 14001, 2004) or better still EMAS (EMAS, 2009). Businesses can take advantage of ready-made tools in this regard, e.g., those available through a network model and e-REMAS software (Sokół, 2011), supporting integrated environmental management at the regional and local level, but also groups of companies and individual enterprises. It is because of the exchangable internal database (Reviewing sheet) and a set of indicators relevant for any company’s activity and structure. An important component of the e-REMAS model (Figure 1) is the management of business risk associated with achievement of the objectives of the organization, because every decision (especially those related to investments) is accompanied by the risk of not achieving the company objectives-ecological, economic and social.

Figure 1 The e-REMAS Model (Sokół, 2013)

In a properly functioning environmental management system, decisions regarding the selection of efficient technologies are adopted as a result of the identification of significant environmental aspects of the company’s activities and their impact on the environment (points 6 and 7 in Figure 1) based on collected and evaluated data (points 2-5 in Figure 1). On the basis of t fact, the environmental policy shall be determined (point 10 in Figure 1); long-and short-term objectives (point 11 in Figure 1) and specific tasks (point 12 in Figure 1). The tasks (projects) are implemented by means of specific technologies, which will enable the company to achieve its goals, including improvement of its environmental performance. They should be innovative and characterized by the highest eco-efficiency of the solutions offered on the market. The next course of action in the EMS is to develop a program of implementation for the tasks (point 13 in Figure 1) and their implementation (point 14 in Figure 1) and monitoring of the environmental performance of the company during implementation of the technology (points 2-5 in Figure 1) to allow possible corrective actions (points 8-9 in Figure 1). Business risk is inherent in environmental management, and this risk must also be managed according to existing schemas (like AIRMIC, 2002; ISO 31000, 2009); however in e-REMAS the AIRMIC approach is applied. The selection process for the implementation of the tasks should be assessed and only those technologies which give

1774


a high probability (confidence) that the objectives (including environmental ones) of the organizations will be achieved shall be selected. Assessment of the probability and impact of a selected technology on the improvement of the environmental performance of the company and its competitiveness provide the basis for risk assessment, and whether the objectives established in the EMS will be achieved, because the technology will ensure success or otherwise, and so there is a risk of failure. Linking of risk management with the environmental management system is presented in Figure 1 and in Sokół (2013), including an example of revitalization of post-industrial site. Conducting assessments of eco-efficiency of the technology provided for implementation leads to the success of the company. There are thousands of technologies from different countries offered on the market. Let us consider only the Polish market for environmental technologies (COM, 2004; UN AGENDA 21, 1992; Lonsdale J. et al., 2011; ETV, 2003). According to the GreenEvo report (GreenEvo, 2010), the Polish market for environmental technologies consists of about 1280 producers and distributors and is under development, searching for new areas for export and innovative solutions for new investments. The main directions of export and import of these technologies in are presented Figure 2 and Figure 3).

Figure 2 Export of Polish environmental technologies. Source (GreenEvo, 2010)

Figure 3 Import of pro-environmental technologies to Poland. (GreenEvo, 2010)

Evaluation of eco-efficiency is mainly available for mass products (Saling et al., 2002; Guinee, 2002; Michelsen et al., 2006; JEMAI, 2004; ISO 14045, 2012). Appropriate methodologies are still being sought for technology assessment. Many evaluation efforts have failed because they lacked important data relating to the life cycle assessment (LCA) of the technology and relied upon simplified assumptions to overcome this lack of sufficient data for an eco-efficiency analysis. Moreover, a full LCA depends on assumed system boundaries and on features of the tools applied, such as ecoindicator 99 (Goedkoop, Effting & Collignon, 2002; ISO 14044, 2006) and others. Meanwhile, potential buyers (very often small, medium and micro companies (in Poland over 1 700 000 are SMEs and 96% are micro-enterprises)), will expect quick recommendations relating to the eco-efficiency of technology offers being submitted by the suppliers, in order to know which one seems to be better than the others, without extended and highly costly analysis. For that reason, a simplified pre-evaluation approach was developed for assessment of the eco-efficiency of a given technology, often based on limited numerical data from the beginning in the framework of the international project SPIN, presented in Kaunus (Sokół, 2011). This was then tested as a task of the EFFECT project (EFFECT 2012) on examples of over 400 technology offers from Baltic Sea Region countries and LONGLIFE-INVEST (LONGLIFE INVEST 2012) for a new investment: an energy efficient dormitory building for Klaipeda University in Lithuania. Examples of implementation of the methodology for sustainable revitalization of post-industrial sites are presented in Sokół (2013). The methodological details and its 1775


practical aspects are discussed more extensively in this paper and illustrated by means of the example of searching for clean coal technology offers as option for heating systems for buildings.

2. Pre-evaluation of Eco-Efficiency of Environmental Technologies Pre-evaluation of the probability that the technology offers are suitable is a part of the project evaluation. Depending on the importance of the technology for the project, coupled with product knowledge which is provided with the technology offer, the technologies are categorized into different “bins” because of the probability of achieving the efficiency declared by the supplier. The probability that the project is able to achieve the expected objectives after implementation of the selected technology (in the framework of a given investment project) influences the allocation of a technology to a specific bin. Putting a technology offer in a selected bin is valid, assuming the supplier describes the technology in sufficient detail. If the supplier submits numeric data on the technology, with a confirmed efficiency in comparison to the situation before implementation of the technology, or to reference solution, then the eco-efficiency of the technology eco-efficiency is pre-evaluated. Inspired by JEMAI (2004), a simple approach to the enhancement of eco-efficiency of an improved technology developed is illustrated as follows (Sokół, 2013):

 E[%]  100  ( Enew / Eold  1)  100  3 / 

I ij 

Enew 

1 ni

Ii 

Where:

1 nj

nj

I k 1

2 ijk

ni

I j 1

2 ij

,

3

I i 1

2 i

  1 

(1)

i =1-3, j=1-ni

(2)

, i =1-3, j=1-ni , k=1-nj

Vnew 1   1/ I new I new / Vnew 

3

I i 1

2 i

and Eold 

(3)

1 3

(4)

In Equations (1) to (4) Iij , Iijk and Ii are the relative environmental impact of a new technology (improved):

I  I I ijk   ijk , new  /  ijk , old  Vnew   Vold

  

(5)

and: I1 – Efficiency of resources, i.e., the degree of reduction in raw material consumption, e.g., water, fossil fuels etc., I2 – Efficient use of energy that includes the degree of the reduction in the consumption of primary energy, I3 – Emissions released, i.e., the degree of reduction of emissions into air, water and soil and the content of environmentally harmful substances etc., ni – quantity of the parameters Iij in the framework of the environmental impacts Ii, nj – quantity of the parameters Iijk in the framework of the environmental impacts Iij, Enew, Eold – the eco-efficiency of the new and old technologies, respectively, Vnew , Vold – technology value of the new and old solutions, respectively, Iij,new, Iij,old – environmental impact of the new and old solutions, respectively. In Equation (1), it is assumed that for an old technology (before implementation of improvements), each 1776


component of the environmental impact is represented by Ii =1. If the new technology does not show improvement in a component, the value remains equal to 1. Equations (2) and (3) (i.e., the average sum of squares) is used when a few improvements Iij were implemented relating to one environmental impact Ii, or a few improvements Iijk were implemented relating to one environmental impact Iij; however Equation (3) is an option for grouping specific impacts under the term Iij for example those relating to waste, emissions into the air, materials etc., if relevant. Pre-assessment of the eco-efficiency enhancement (ΔE) assumes that the system boundaries apply to the technology only and that external changes are negligible. The accuracy of this evaluation depends on the numerical data submitted by the technology supplier. This approach can motivate a potential buyer to ask for more detailed data to allow assessment of eco-efficiency for the full life cycle.

3. Results of Testing the Methodology The methodology was tested on 415 selected offers for environmental technologies submitted by suppliers from seven Baltic Sea Region (BSR) countries: Denmark, Estonia, Finland, Germany, Lithuania, Poland and Sweden, being considered for the SPIN and ACT CLEAN databases (www.actclean.gig.eu). The suppliers of these technology highlights have marked 102 offers as energy efficient, 22 as renewable energy resource technologies, and 16 as remediation offers. Results of a pre-evaluation of the eco-efficiency ΔE [%] of selected energy-efficient technology offers from BSR are presented in Sokół (2013); however, with giving an extended description of site remediation technology. For that reason, this description is developed in this paper for four selected clean coal technology options identified while searching for technology offers for heating systems for dormitory building in Klaipeda (LONGLIFE INVEST 2012) to increase energy efficiency and to reduce CO2 emission. The data of the technology offers for clean use of coal in buildings for heating are presented in Tables 1-4 (Czaplicka, Sciazko, 2004). In this example, the offers contained a lot of data, because they were collected for research purposes, however not coal but wood pellets were finally selected for heating a building in Klaipeda. In general, technology offerings include only a single datum or only a description without any numerical figures. In the case, the data relate to the following: a boiler with a culm chamber (a), a boiler with a fixed grate (b), a boiler with a moving grate (c) and a boiler retort furnace (d). The basic parameters, relevant units and values are given in columns 1-3. The technology parameters are ranked in four groups: technology value V1, resource efficiency I1, energy efficiency I2 and efficiency in reduction of emissions I3. Columns 5 and 6 present the relative values, i.e., Iijk/V1 and their units. The energy efficiency section I2 includes external consumption of electricity I21 for use within the system for driving fans, feeders and powering the control panel; thermal losses I22 are also included because the process efficiency η < 1. In this case, Equation (2) is used to calculate I2 as average sum of squares of I21 and I21. Efficiency in the emission reduction section I3 include: the release of ash I31 and emissions into the air I32 that include five emissions I321 - I325 namely SO2, CO2, NOx, CO and dust, respectively. In this case, Equation (3) is used to calculate the impact I31 as the average sum of squares of I31 and I32; however impact I32 is calculated using Equation (3) as the average sum of squares of I321 - I325. This approach is a useful option because it permits assessment of the separate impacts of ash release and emissions into the air; however, application of Equation (2) only for all six parameters is acceptable, and in that case leads to the same conclusions in relation to the efficiency of the pre-evaluated technologies.

1777

Practical Aspects of Pre-evaluation of Eco-efficiency of Environmental Technologies Table 1 Technology T1-Boiler with Culm Chamber Parameter Unit A 1 2 3 Technology value-V1 Energy production kW 12 1. Resource efficiency-I1 Fuel consumption kg/h 2.1 2. Energy efficiency–I2 Energy consumption kW 0.12 Energy losses kW 2.8380 3. Efficiency in emission reduction–I3 3.1. Wastes–I31 Ash release kg/h 0.497 3.2. Emissions to air–I32 Emission-SO2 kg/h 0.0125 1- solid fuel, 2 – air, 3 – outlet gasses, 4 – ash, 5 – heat Emission-CO2 kg/h 3.3 exchangers, 6 – control panel, 7 – furnace Emission-NOx kg/h 0.0153 Emission-CO kg/h 0.1134 Emission-Dust kg/h 0.0014

B 4

Unit 5

A/ V11 6

V11

kW/kW

1

I11

kg/kWh

0.175

I21 I22

kW/kW kW/kW

0.01000 0.23650

I31

kg/kWh

0.04142

I321 I332 I343 I354 I365

kg/kWh kg/kWh kg/kWh kg/kWh kg/kWh

0.00104 0.27500 0.00128 0.00945 0.00012

B 4

Unit 5

A/ V11 6

V11

kW/kW

1

I11

kg/kWh

0.16000

I21 I22

kW/kW kW/kW

0.00750 0.28013

I31

kg/kWh

0.02550

I321 I322 I323 I324 I325


0.00162 0.62000 0.00053 0.00872 0.00142

Table 2 Technology T2-Boiler with Fixed Grate Parameter Unit A 1 2 3 Technology value -V1 Energy production kW 20 1. Resource efficiency -I1 Fuel consumption kg/h 3.2 2. Energy efficiency -I1 Energy consumption kW 0.15 Energy losses kW 5.6027 3. Efficiency in emission reduction –I3 3.1. Wastes –I31 Ash release kg/h 0.51 3.2. Emissions to air –I32 Emission - SO2 kg/h 0.0323 Emission - CO2 kg/h 12.4 1- solid fuel, 2 – air, 3 – outlet gasses, 4 – ash, 5 – heat Emission - NO kg/h 0.0106 x exchangers, 6 – control panel, 7 – fixed grate Emission - CO kg/h 0.1743 Emission - Dust kg/h 0.0283

In Equations (1), (2) and (3), Iij, Iijk and Ii respectively refer to the relative environmental impact of a new technology (improved) or to a reference solution, and not absolute values. This means that we have to define which technology has to be considered as the reference for pre-evaluation of the eco-efficiency enhancement ΔE of other technologies. Identifying the reference solution is essential because the eco-efficiency enhancement ΔE is not calculated by taking direct the values from column 6 of Tables 1 to 6 to the Equations (2) or (3), but by using Equation (5). If technology T1 (Boiler with culm chamber) is selected as the reference and the aim is the evaluation of the 1778


eco-efficiency enhancement ΔE for technology T2 (Boiler with fixed grate), then applying Equation (5), for example for I11, we obtain:  I (T 2)  0.160   I11 (T 2)   11  0.914  I11 (T1)  0.175

(6)

Table 3 Technology T3-Boiler with Moving Grate Parameter Unit A 1 2 3 Technology value –V1 Energy production kW 35 1. Resource efficiency - I1 Fuel consumption kg/h 5.6 2. Energy efficiency-I2 Energy consumption kW 0.35 Energy losses kW 9.8047 3. Efficiency in emission reduction - I3 3.1. Wastes - I31 Ash release kg/h 0.73 3.2. Emissions to air-I32 Emission - SO2 kg/h 0.0549 Emission - CO2 kg/h 22.4 Emission - NOx kg/h 0.0511 1- solid fuel, 2 – air, 3 – outlet gasses, 4 – ash, 5 – heat Emission - CO kg/h 0.0270 exchangers, 6 – control panel, 7 – feeder Emission - Dust kg/h 0.0062

B 4

Unit 5

A/ V11 6

V11

kW/kW

1

I11

kg/kWh

0.16000

I21 I22

kW/kW kW/kW

0.01000 0.28013

I31

kg/kWh

0.02086

I321 I332 I343 I354 I365


0.00157 0.64000 0.00146 0.00077 0.00018

B 4

Unit 5

A/ V11 6

V11

kW/kW

1

I11

kg/kWh

0.14800

I21 I22

kW/kW kW/kW

0.00840 0.18412

I31

kg/kWh

0.02000

I321 I332 I343 I354 I365


0.00223 0.49200 0.00119 0.00068 0.00021

Table 4 Technology T4–Boiler with Retort Furnace Parameter Unit A 1 2 3 Technology value - V1 Energy production kW 25 1. Resource efficiency – I1 Fuel consumption kg/h 3.7 2. Energy efficiency – I2 Energy consumption kW 0.21 Energy losses kW 4.6031 3. Efficiency in emission reduction – I3 3.1. Wastes – I31 Ash release kg/h 0.5 3.2 Emissions to air – I32 Emission - SO2 kg/h 0.0557 Emission - CO2 kg/h 12.3 1- solid fuel, 2 – air, 3 – outlet gasses, 4 – ash, 5 – heat Emission - NO kg/h 0.0297 x exchangers, 6 – control panel, 7 –feeder, 8 – retort Emission - CO kg/h 0.0169 furnace Emission - Dust kg/h 0.0053

This means that technology T2 reduced the impact I1 (i.e., consumption of fuel) by ΔI1 = 8.6%. This result is presented in Figure 4 as well as results for I2 and I3 and the eco-efficiency enhancement ΔE. Figures 5, 6 and 7 present the same results if technologies T2, T3 and T4 are taken as reference technologies for the others. 1779


Reference technology - T1 Reduced impacts ΔI & Ecoefficiency enhancement ΔE [%]

50 9

15

9

1

19

8

0 -8

-10 -50

-3

-10

-51

-100 -150 -200 -229 T2 - fixed grate

T3-moving grate

ΔI1 - Resources

9

9

15

ΔI2 - Energy

1

-10

19

-229

-8

-10

-51

-3

8

-250

ΔI3 - Emissions ΔE [%]

T4-retort furance

Technology No

Figure 4 Eco-efficiency Enhancement ΔE of Technologies T2, T3 and T4 if the Reference Technology Is T1 Reference technology - T2 Reduced impacts ΔI & Ecoefficiency enhancement ΔE [%]

15 7

10 5

8 5

0

0 -5 -10 -15

-2 -9

-9

-9

-12 -15

-20

-18

-25 -30 -31 T1-culm chamber

T3-moving grate

ΔI1 - Resources

-9

0

7

ΔI2 - Energy

-12

-18

8

ΔI3 - Emissions

-31

-9

-2

ΔE [%]

-15

-9

5

-35

T4-retort furance

Technology No

Figure 5 Eco-efficiency Enhancement ΔE of Technologies T1, T3 and T4 if the Reference Technology Is T2 Reference technology - T3 Reduced impacts ΔI & Ecoefficiency enhancement ΔE [%]

50

7

0

0 -50

12

7

25 0

11

-9 -53

-56

-100 -150 -200 -250

-243 T1-culm chamber

-271 T2 - fixed grate

ΔI1 - Resources

-9

0

7

ΔI2 - Energy

7

12

25

-243

-271

0

-53

-56

11

-300


T4-retort furance

Technology No

Figure 6 Eco-efficiency Enhancement ΔE of Technologies T1, T2 and T4 if the Reference Technology Is T3

1780


Reference technology - T4 Reduced Impacts ΔI & Ecoefficiency enhancement [%]

0 -50

-18

-8

-24 -59

-100

-4

-8

-25 -59

-37

-15

-150 -200 -250 -300 -350

-286

-288

T1-culm chamber

T2 - fixed grate

ΔI1 - Resources

-18

-8

-8

ΔI2 - Energy

-24

-25

-37

-286

-288

-4

-59

-59

-15


T3-moving grate

Technology No

Figure 7 Eco-efficiency Enhancement ΔE of Technologies T1, T2 and T3 if the Reference Technology Is T4

The positive values of Ii in Figures 4, 6 and 7 indicate a reduction of environmental impact Ii (and conversely for negative values). Positive values of ΔE indicate that the eco-efficiency has increased compared with the reference technology; negative values indicate a decrease. Analysis of Figures 4 to 7 leads to the conclusion that technology T4 (boiler with retort furnace) is the most ecoeffective in comparison with the reference technology (T1) and the other technologies (T2 and T3). For benchmarking of technology offers it is best to use a different technology as a reference instead of one of these four. It has to be less eco-efficient. It should be the least effective in relation to the other options. For that reason in Table 5 reference technology T0 having the worst performance of all four technologies is proposed, i.e., the worst parameters of Tables 1, 2, 3 and 4 are chosen. Table 5 Reference Technology T0 Parameter Technology value-V1 Energy production 1. Resource efficiency-I1 Fuel consumption 2. Energy efficiency–I2 Energy consumption Energy losses 3. Efficiency in emission reduction–I3 3.1. Wastes–I31 Ash release 3.2. Emissions to air–I32 Emission-SO2 Emission-CO2 Emission-NOx Emission-CO Emission-Dust

No

Unit

Max Value

V11

kW/kW

1

I11

kg/kWh

0.17500

I21 I22

kW/kW kW/kW

0.01000 0.28013

I31

kg/kWh

0.04142

I321 I332 I343 I354 I365


0.00223 0.64000 0.00146 0.00945 0.00142

The resulting eco-efficiency enhancement ΔE of technologies T1, T2, T3 and T4 if the reference technology is T0 is presented in Figure 8. Eco-efficiency enhancement ΔE of technologies T1, T2, T3 and T4 depends on which 1781


Reduced Impacts ΔI & Ecoefficiency Enhancement ΔE [%]

technology is selected as the reference, as shown in Figure 9. Reference technology - T0 30

25 25

23

25 20

17

15 7

10 5

12

10

15

14

9

6

11

9

0

0

27

0 T1

T2

T3

T4

ΔI1 - Resources

0

9

9

15

ΔI2 - Energy

7

12

0

25

ΔI3 - Emissions

10

17

23

25

ΔE [%]

6

14

11

27

Technology No

Figure 8 Eco-efficiency Enhancement ΔE of Technologies T1, T2, T3 and T4 if the Reference Technology Is T0 40 20

6

14 11

27

0

8

0 -3

-20 -40 -60 -80

5

0

0

11

0

-9

-15

-51

-15 -53 -56

-59 -59

Reference - T0

Reference -T1

Reference - T2

Reference - T3

DE1 for T1-culm chamber

6

0

-15

-53

Reference -T4 -59

DE2 for T2-fixed grate

14

-51

0

-56

-59

DE3 for T3-moved grate

11

-3

-9

0

-15

DE4 for T4-retort furnance

27

8

5

11

0

Figure 9 The Eco-efficiency Enhancement ΔE of Technology Options Depends on which Technology Is Used as the Reference

The use of reference technology T0 is better at presenting the environmental impacts and eco-efficiency enhancement with respect to each technology. The conclusion is the same: that technology T4 (Boiler with retort furnace) is the most ecoeffective in comparison with the reference technology and the other technologies (T1, T2, T3 and T4). The question is how to proceed if the technology offer includes as technology value V1 more than one component, i.e., V11, V12, etc? In this case, the best approach is to calculate the eco-efficiency enhancement separately for each component of V1, i.e., ΔE11, ΔE12, etc. However, the average sum of squares can be used for evaluation of the combined ΔE, but in practice a potential buyer likes to know the eco-efficiency related to each technology value. The next question relates to the economic and social efficiency of the technology. The methodology is general (see Equations (1) to (5)) and different technology values and different impacts can be considered, but assessment of the technology’s economic and social efficiency during relevant steps of company environmental management system is recommended (see Figure 1), taking into account existing and potential risks. In this case, not only the 1782


technology offer is evaluated, but the whole project (the task, or tasks). How this is done for the case of sustainable revitalization of post industrial sites can be seen in Sokół (2013). Eco-efficiency evaluations always consider the whole life cycle (ISO 14045:2012; JEMAI, 2004; Goedkoop, Effting & Collignon, 2002; etc.). Eco-efficiency pre-evaluation is a relative (not absolute) assessment, assuming changes in the technology system, which is why the boundaries of the system mainly include the use of technology. In the cases considered for pre-evaluation of technology offers for clean use of coal for heating in buildings, this assumption is reasonable because changes may involve negligible fuel preparation and transportation of the same from afar. Good practice would be to assess the risks posed by the real emission changes throughout the life cycle of the technologies analyzed. If the supplier provides data on the remaining life cycle stages of the technology (such as the extraction of raw materials and fuels, transportation and disposal) then these impacts will be evaluated. In practice, technology offers include only fragmented data. In the case of the more than 400 technology offers reviewed within the framework of the EFFECT project (EFFECT, 2012), less than 56% included any numeric data and none related the full life cycle. The data contained in offers frequently describe selected technical characteristics of the technology and are not compared with earlier solutions, competitors’ solutions or reference technologies.

4. Conclusions The methodology presented is an easy and useful tool for initial assessment of the eco-efficiency of environmental technologies offered by their respective suppliers. Moreover, the methodology facilitates assessments of how the solutions reducing consumption of resources and energy and emissions into air, water and soil will improve the environmental performance of the purchasing company after the application of the technology in relation to the system previously used or a competitor’s system. The technology’s impact on economic and social efficiency has to be evaluated during the relevant steps of the company environmental management system, taking into account existing and potential risks and rather separately (not by aggregation with eco-efficiency of environmental impacts) to better see all influents. Eco-efficiency pre-evaluation means a relative (not absolute) assessment performed assuming that the boundaries of the system mainly include the use of technology, however, it is worth assessing the risk of that simplification. It does not preclude a full LCA and eco-efficiency analysis for the selected solution within the environmental management system, if the supplier provides sufficient data. References: AIRMIC (2002). “Risk management standards”, ALARM, IRM: 2002. COM (2004). “Final: Communication from the commission to the council and the European Parliament”, Stimulating Technologies for Sustainable Development: An Environmental Technologies Action Plan for the European Union, Brussels, 28.1.2004. Czaplicka K. and Sciazko M. (2004). “Model of prognosis of environmental and economic extraction and using clean coal. Part 2. Eco-efficiency of clean combustion of coal. Central Mining Institute. Katowice, 2004 (in Polish) EMAS (2009). Regulation (EC) No 1221/2009 of The European Parliament and of The Council of 25 November 2009 on the voluntary participation by organizations in a Community eco-management and audit scheme (EMAS), repealing Regulation (EC) No 761/2001 and Commission Decisions 2001/681/EC and 2006/193/EC. EFFECT (2012). “Dialogue platform on energy and resource efficiency in the Baltic Sea Region”, available online at: http://www.cbss.org/environment-and-sustainability/baltic-21-lighthouse-projects/. Environmental Technology Verification (ETV) Program (2003). “Case studies: Demonstrating program outcomes”, EPA/600/R-06/001.

1783

Practical Aspects of Pre-evaluation of Eco-efficiency of Environmental Technologies Goedkoop M., Effting S. and Collignon M. (2002). The Eco-indicator 99: A Damage Oriented Method for Life Cycle Impact Assessment—Manual for Designers (2nd ed.). GreenEvo (2010). “Rynek polskich technologii środowiskowych (Polish market of environmental technologies)”, GreenEvo, September 2010 Guinee J. B. (2002). “Handbook on life cycle assessment: Operational guide to the ISO standards”, Int. J LCA, Vol. 7, No. 5. ISO 31000 (2009). “Risk management—Principles and guidelines”. ISO 14001 (2004). “Environmental management systems—Requirements with guidance for use”. ISO 14044 (2006). “Environmental management—Life cycle assessment—Requirements and guidelines”. ISO 14045 (2012). “Environmental management—Eco-efficiency assessment of product systems—Principles, requirements and guidelines”. Japan Environmental Management Association for Industry (JEMAI) (2004). Eco-Efficiency and Factor Handbook for Products: Japan Environment Public Release Draft, January 2004. LONGLIFE-INVEST (2012). “The implementation of the planned Lithuanian Long life pilot project as a dormitory for Klaipeda University”, accessed 30 March 2014, available online at: http://longlife-invest.eu/jml/. Lonsdale J. et al. (2011). “Detailed assessment of the market potential, and demand for, an EU ETV scheme”, EPEC, June 2011. Michelsen O., Magerholm Fet A. M. and Dahlsrud A. (2006). “Eco-efficiency in extended supply chains: A case study of furniture production”, Journal of Environmental Management, Vol. 79, pp. 290-297. Saling P., Kicherer A., Dittrich-Kramer B. and Wittlinger R. et al. (2002). “Eco-efficiency analysis by BASF: The method”, Int. J LCA, Vol. 7, No. 4, pp. 203-218. Sokół W. A. (2011). “Biogas in regional environmental management system: Eco-energetics-biogas and syngas”, in: Gdanska Wyższa Szkoła Administracji, Technologies, Legal Framework, Policy and Economics in Baltic Sea Region, Gdańsk, pp. 40-41. Sokół W. A. (2011). “Pre-evaluation of eco-efficiency of environmental technologies”, in: International Conference “Sustainable Consumption and Production 2011”, Kaunus, 30 September 2011. Sokół W. A. (2013). “A methodology for pre-evaluation of ecoefficiency of environmental technologies for sustainable revitalization of post-industrial sites”, in: Contaminated Soils, Sediments, Water & Energy, Vol. 19, Proceedings of the 29th Annual International Conference on Soils, Sediments, Water & Energy, Amherst, October 21-24 2013. pp. 132-155, accessed 30 March 2014, available online at: http://www.aehsfoundation.org/ecc-proceedings.aspx. UN AGENDA 21 (1992). United Nations Conference on Environment & Development Rio de Janerio, Brazil, 3 to 14 June 1992.

1784


The Temporal and Spatial Effect of Highways on China’s Economic Growth* Xiugen Mo, Guangqing Chi, Charles Campbell (Mississippi State University, Mississippi State, MS 39762, USA)

Abstract: Infrastructure investment is currently enjoying great popularity among government as a strategy to handle the recent worldwide economic recession. However, the relevant literatures have yet to come to a consistent agreement on how infrastructures, especially that of transportations, impact economic growth. This paper investigates the effects of highway infrastructure on economic growth using spatial regression models estimated from the county data of Guangxi Zhuang Autonomous Region (GXZAR) in China from 1993-2007. The results indicate that relationships between initial highway infrastructure and economic growth can be positive, negative, or negligible. The spatial effects are also inconsistent. Impacts are stronger in near short-term but diminishing over times. We conclude that the highways are a necessary, but not a sufficient condition for economic growth. The highway impact depends on other factors as well. Improving infrastructure alone could not create a continued economic growth. Key words: growth; spatial; infrastructures; roads; China JEL codes: O18, R42, R12, R53

1. Introduction Highway construction is a common-seen policy to stimulate economic development. The newly fiscal-stimulus package announced by Chinese Government in November 2008 includes plans to invest a huge amount of money on transportation infrastructures such as highways and railways. One may cast no doubt on the effect of policies crafted to improve the transportation system because of the common belief that improved transport infrastructures would always result in further development of the region as is the case with most urban areas. However, a clear answer to how the highways impact economic development has not been provided by existing literatures. The role of highway infrastructures in economic development is usually inquired along with the study of the relationships between economic development and infrastructures. Solow (1957) provides a theoretical model to *

This paper was completed at Department of Finance and Economics, School of Business, Mississippi State University in May, 2009. Xiugen Mo, Ph.D., Department of Finance and Economics, Mississippi State University; research areas/interests: policy and program analysis, labor market, sustainable development, inequality and poverty. E-mail: [email protected]. Guangqing Chi, Assistant Professor, Department of Sociological Science, Mississippi State University; research areas/interests: transportation and demography, natural amenities and rural development, and population estimation and forecasting. E-mail: [email protected]. Charles Campbell, Professor of Economics Emeritus, Department of Finance and Economics, Mississippi State University; research areas/interests: economic impact analysis, community development, ecological economics and sustainable development. E-mail: [email protected]. 1785

The Temporal and Spatial Effect of Highways on China’s Economic Growth

investigate economic development. The model predicts that economies of undeveloped regions grow faster. Barro and Sala-I-Martin (1991) gave empirical evidences for the Solow model. They found that the economies of low income states grow faster than the high income states in the United States of America; they conclude that there exists a convergence among states. Barro and Sala-I-Martin estimate that the speed of convergence is about 2 percent. However, Fujita and Hu (2001) argue that the convergence only happens for regions among the coastal provinces while a divergence exists between coastal provinces and inland provinces in China from 1985 to 1994. The economic convergence is conditional. At macro level across countries, Barro (1991) found evidence supporting convergence from 1960 to 1985. Barro reported that GDP growth rate relates to initial GDP and government expenditure negatively and significantly. However, the relationship between public investment and the growth rate in GDP was found to be insignificant. Barro’s result thus implies that investment on transportation infrastructures may not impact economic growth. Nonetheless, evidence supporting the proposition that the development of transport infrastructures play a role in economic development also exists. Demurger (2001) reported that infrastructures produce an impact on economic growth. Bougheas, Demetriades and Mamuneas (2000) also found that infrastructure (transportation and communication) has a nonlinear relationship with economic growth. Furthermore, Berechman, Ozmen and Ozbay (2006) studied the contribution of highways to economic activities in the difference geographical scales in USA. They found that the contribution of highways declines and the spillover increases when the geographical scale decreases. The effect of highways on economic development can also be examined in terms of economic factors such as input costs and population. Cohen and Paul (2007) estimated the highway impact on the shadow values and input costs and found SAR(R) is more efficient method of estimation compared to SUR. Chi, Voss and Delle (2006) also reported that the construction of highways has an impact on population. A review on the debates in literatures was carried out by Button (1998); this review led to a conclusion that the findings from previous research have not provided a conclusive answer as to the role of transport infrastructure on endogenous growth. This paper investigates how initial highway condition impacts the economic growth in the later periods. The relationship between the economic growth and the highway is investigated based on the Solow model, assuming the existence of spatial dependent. Spatial models are used to test the hypothesis that the highway densities have an impact on the regional economic growth and the impact varies over different periods. The other hypothesis can be tested by the models is spillover effect of highway infrastructures. The dataset for parameter estimation includes data of 89 counties of Guangxi Zhuang Autonomous Region in China from 1993 to 2007. The next section of the paper provides the exploratory spatial data analysis (ESDA). The theoretical model and empirical model will be given in section III. In section IV, the estimated results are discussed. The conclusion is made in section V.

2. Data and Spatial Analysis 2.1 Dataset The dataset is collected from the statistic year-books of Guangxi Zhuang Autonomous Region in 1993-2007. The data of 1995 are not included for the reason of non-availability. The dependent variable is the average GDP growth rates over thirteen periods, which are 1993-1994, 1993-1996…, and 1996-1997, 1996-1998…, 1996-2007. The independent variables are the initial conditions of highway densities and other control variables in 1993 or 1996. The results of the exploratory spatial data analysis (ESDA) of the key variables are provided next.

1786


Table 1

Data Description

Variables

Unit

Minimum

Maximum

Growth Rate 96-97

Percent

Mean 11.3391

10.7899

-35.5300

62.5700

Growth Rate 96-98

Percent

2.6278

10.0473

-38.1600

28.1300

Growth Rate 96-99

Percent

3.5343

7.5835

-23.3500

21.7300

Growth Rate 96-00

Percent

3.8189

5.9537

-16.9000

17.7600

Growth Rate 96-01

Percent

4.2772

5.0863

-12.6900

16.3000

Growth Rate 96-02

Percent

4.5630

4.3767

-9.1200

13.9200

Growth Rate 96-03

Percent

4.4724

4.1998

-7.2900

15.6000

Growth Rate 96-04

Percent

5.4800

3.6566

-4.4500

17.3800

Growth Rate 96-05

Percent

6.7091

3.2482

-2.1600

17.3500

Growth Rate 96-06

Percent

7.5530

3.0932

-1.0200

16.4600

Growth Rate 96-07

Percent

8.1091

2.8910

0.9100

17.3400

2

Standard Deviation

Highway Density (HWD) in 1996

KM/KM

0.1768

0.0723

0.0373

0.4608

Population Density in 1996

Person/KM2

225.2813

212.4247

39.9000

1323.5000

Fix Asset Investemtnt in 1996

Yuan per Capita

716.3206

786.5654

55.5200

4155.3800

Middle-Primary School Student Ratio in 1996

Ratio

0.3859

0.4612

0.1201

4.5446

Health Service in 1996

Beds/1000 Persons

17.8721

15.3173

1.5600

82.3600

2.2 The Spatial Distributions of GDP Growth Understanding the spatially distributed pattern of GDP growth rates is great helpful for us to build an appropriate model. Before we go into the major steps to analyze the cluster and the spatial dependency, we first investigate the structure of the spatial weight matrix. A weight matrix has to be chosen before the ESDA. However, no theories are available to specify the structure of the weight matrix. For the purpose of this paper, the economic growth may be influenced by all neighbor regions because they are connected by highways. We consider the queen matrix and the nearest k matrix. The Moran’s I of dependent variable for different periods, using different weight matrixes, are graphed in Figure 1. A pattern can be seen in the Figure 1 is that the magnitudes of Moran’s Is are declined as the farther neighbor counties are included in the weight matrix. The Moran’s Is computed from nearest-3 matrix are higher than that from nearest-5, which are in turn larger than that from nearest-7. Similarly, the Moran’s Is from queen-1 matrix are larger than that from queen-2. Also in Figure 1, the variation of Moran’s Is over periods are similar regardless of the differences of the weight structures. Interestingly, the Moran’s Is of nearest-3 are almost the same as queen-1. In this paper, the spatial model will be based the queen-1 neighbor matrixes. A very important implication of Moran’s I is to analyze the spatial dependence. Figure 1 indicates that the spatial dependences of the GDP growth rates increase quickly in short term, decline slightly in mid-term, and then climb up in long term. It is reasonable for the spatial dependence of economic growth in long term causing by the interaction among regions. However, the decline in the mid-term is puzzling. This may be explained by that economic activities are immediately stimulated by an improvement of infrastructure and then calm down by rational sense. Next we will investigate the distribution pattern of GDP growth. The spatial distribution of the average GDP growth rate over 1993-2006 is shown in Figure 2. The higher-growth-rate counties are concentrated in the northeast and west areas. They also distribute along with the railways or the expressways. On the other hand, the lower-growth-rate counties are those in the southeast and the

1787


mid-north areas. This can be also seen in Figure 3 by LISA. There is a cluster of high-high growth rate in the west, a cluster low-low growth rate in the southeast. 0.4000 0.3500 0.3000 0.2500

NN3

0.2000

NN5

0.1500

NN7 Queen1

0.1000

Queen2 0.0500 0.0000 -0.0500

1

2

3

4

5

6

7

8

9

10

11

12

Figure 1 The Moran’s I of GDP Growth Rate over Different Periods

Figure 2

Spatial Distribution of the GDP Growth Rate of Guangxi in 1993-2007

2.3 Highway Density By comparing Figure 4 and 5, we can see that lower growth rates over 1993-2006 are overlap the higher highway densities in 1993 in the southeast areas.

1788


And a cluster of high-high highway density is in the southeast. The highway densities seem related to GDP growth negatively.

Figure 3 LISA of the GDP Growth Rate of Guangxi in 1993-2006

Figure 4

Spatial Distribution of the Highway Density in Guangxi in 1993

1789


Figure 5 LISA of the Highway Density of Guangxi in 1993-2006

3. Economic Model 3.1 Theoretical Model Economic factors can be divided by their liquidity into two categories—the liquid and the fixed factors. The liquid factors, such as labor and money, move to the regions or economic sectors of higher economic returns. The fixed factors, such as infrastructures and natural resources, are bound to a region. For a specific region, the economic development is easier to seen variation with the flow of liquid factors. However, the fixed factors are usually monotonic. Therefore, the relationship of infrastructure like highways to economic growth is harder to be identified. Among the infrastructures, highways are the one that increases monotonically. Unless in an extreme case such as earthquake, the physical amount of highways in a region is never declined. Thus, in certain regions, the highway densities may either remain unchanged or grow. Furthermore, once a highway is constructed, it exists for a long time. Highways may be upgraded or improved so that they continue to serve economic development. Finally, highways are public goods. Though a fee may be required to use certain segments of highways, most of highways are open to public. These features allow highways produce a different impact on economic development. The impact of highways on economic development can be both temporal and spatial. The time impact of highways is derived from that they serve the economic activities in all times and the impacts usually are through other economic factors. The investment of highway construction is an economic activity, which influences economic development in the same way as other investment such as increase the employment and income. However, after constructed, the physical highways are not an investment but a public good that serves for years. It becomes a necessary environment for economic activities. Only if the improvement of highway condition is

1790


realized, the other factors are mobilized into the region and take the advance of the improved highway services. The impact of the highways depends on the performance of the liquid factors after constructed. In spatial dimension, the impacts of public infrastructures are not confined only in the local regions. The highways facilitate the economic activity such as trade and mobility of economic factors such as technology, labor, and capital by reducing transportation costs. The economic dependency among regions increases. Meanwhile, the economic opportunities created by highways also affect the neighbor regions. The temporal and spatial lag impacts of highways have to be considered in economic model. Let H represents the quantity of highways in a region or the neighbors. Y is the output. The production function is given in the following form. /

, (1) Where, β is parameter. t is the time period. s is a constant and the highest effect of highways is obtained at the period s . The term (s-s/t-t) captures the variation of highway effect over times. When t is small, the whole term is negative and the impact of highway is small. When t = s, the term is 1 and the highway has full impact. f . is the production function of other non-fixed asset factors K and L and is supposed not vary over time period. Therefore, ,

(2)

The function can be rearranged as ⁄

/

(3)

Taking log and divided by t of both sides, derive ⁄

⁄ ⁄ / 1 (4) ⁄ ⁄ The left hand side is the average growth rate. On the right hand side, let 1 which is the coefficient of highway. It can be positive or negative and varies along with the time. The null hypothesis for test is H 0 : a=0 If a = 0 is rejected, economic growth is affected by highway construction. Testing the significance of the coefficient for the highway density in the region or the neighbor region equals to test the highway effect and spillover effect. The sign of the coefficient indicates the direction of impact. A positive sign means that economy benefits from the highways construction. In contrast, a negative sign indicates that the economic activities are drained away. 3.2 Empirical Model The spatial dependence of dependent variable causes bias if empirical model is estimated by OLS. The Moran’s Is of GDP growth rates, shown in section II, imply that the existence of spatial dependency is very possible. The major source of spatial dependence is the interaction of neighbor economic activities which is promoted by the improvement of transportation and information infrastructure. This paper focuses on the effect of highways. The spatial effect of highway density is presented by the spatial lag of highway density. However, the highway may be not the only source of the spillover. A spatial lag of the dependent variable is also added into the equation. (5) Where, y is the average growth rate of GDP over a period of time. H is the highway density. wy and wH are the weighted growth rate and highway density. i represent the regions and j is the neighbor regions (i≠j). Hj is the

1791


average highway density of neighbor regions. e is the error term. a and b are the parameters. X is other control variables included the initial GDP, investment, human capital, and other control variables such as other means of transportations.

4. Estimations and Discussion The estimated results of empirical models are shown in Table 2. We investigates the impact of Highways based on the initial endowment in 1996 (96 models) because the continuity of the available data. The results are also compared to that based on the initial endowment in 1993 (93 models), 1997 (97 models), and 1999 (99 models). The dependent variables for different models are the average GDP growth rate over different periods. Comparing the results of different models allows us to understand how highways impact the economic growth rate over periods. Furthermore, the models provide us evidences to investigate the spatial effect. Next we will discuss the specification of the spatial models and then the direct effect and spillover effect of highways. At the end, the results of control variables are provided. 4.1 Spatial Model Specification The results from OLS in table 2 provide evidences to specify the spatial model. The first step is to determine the existence of spatial dependence. The Moran’s I, LM of the error term, and the LM of lag term are statistics to diagnose for spatial dependence. The Moran’s Is of the error term in Table 2 indicate the spatial dependences are various but significant over periods. Hypothesizes of independence are rejected for all periods except the last. The Moran’s Is increase in short term and then reduce as the period become longer. Beside the shortest and longest periods, the LM-lags and LM-errors are significant. This means that both the lag and error terms are the sources for the spatial dependence for the GDP growth rate and then the OLS models are bias. The bias of spatial dependent model should be corrected by an appropriate model. Table 2 Independent Variables CONSTANT Initial GDP Highway Density (HWD) Weighted HWD Population Density Fix Asset Investemtnt Education Health

1792

Coeff. Std Coeff. Std Coeff. Std Coeff. Std Coeff. Std Coeff. Std Coeff. Std Coeff. Std

The Estimates of OLS Model

Dependent Variable: Average Annual GDP Growth Rate 1996-1997 1996-1998 1996-1999 1996-2000 1996-2001 11.7870 * 5.0106 6.1002 7.1852 ** 6.8313 ** 6.0902 5.3416 3.9096 3.0978 2.6009 -0.0157 *** -0.0212 *** -0.0175 *** -0.0123 *** -0.0110 *** 0.0056 0.0049 0.0036 0.0029 0.0024 42.5835 * 30.9083 20.7945 13.7407 14.0917 21.5955 18.9409 13.8633 10.9845 9.2226 -13.2785 -0.5178 -1.2004 -9.0495 -7.2537 37.1607 32.5928 23.8554 18.9017 15.8699 -0.0088 -0.0111 -0.0095 * -0.0074 * -0.0071 * 0.0088 0.0077 0.0056 0.0045 0.0037 0.01 ** 0.01 *** 0.01 *** 0.01 *** 0.01 *** 0.0034 0.0030 0.0022 0.0017 0.0014 1.1302 0.5535 0.2790 -0.0470 -0.0595 2.6265 2.3036 1.6861 1.3360 1.1217 0.0551 0.1612 0.1712 * 0.1148 0.1057 * 0.1501 0.1316 0.0963 0.0763 0.0641 (Table 2 to be continued)

The Temporal and Spatial Effect of Highways on China’s Economic Growth (Table 2 continued) Railway Dummy

Coeff. Std

R-squared Likelihood AIC BIC Breusch-Pagan Moran's I LM(lag) Robust LM(lag) LM(error) Robust LM(error) LM(SARMA) Independent Variables CONSTANT

Coeff. Std Initial GDP Coeff. Std Highway Density (HWD) Coeff. Std Weighted HWD Coeff. Std Population Density Coeff. Std Fix Asset Investemtnt Coeff. Std Education Coeff. Std Health Coeff. Std Railway Dummy Coeff. Std R-squared Likelihood AIC BIC Breusch-Pagan Moran’s I LM(lag) Robust LM(lag) LM(error) Robust LM(error) LM(SARMA)

3.2997 0.1847 0.4035 -0.1858 2.4451 2.1446 1.5697 1.2437 0.1512 0.1716 0.2919 0.2787 -330.1850 -318.5110 -290.7370 -270.0210 678.3700 655.0230 599.4730 558.0420 700.7670 677.4210 621.8710 580.4400 18.1617 *** 12.9880 15.1911 * 13.7416 * 0.0851 * 0.2902 *** 0.3283 *** 0.2787 *** 1.0201 20.9321 *** 27.5131 *** 19.7716 *** 0.1908 4.4385 *** 6.2997 *** 4.5974 *** 1.43 16.68 *** 21.35 *** 15.39 *** 0.60 0.19 0.14 0.21 1.62 21.12 *** 27.65 *** 19.99 *** Dependent Variable: Average Annual GDP Growth Rate 1996-2002 1996-2003 1996-2004 1996-2005 1996-2006 6.0876 *** 7.0913 *** 8.5304 *** 9.0787 *** 9.9248 *** 2.1960 2.2114 1.9341 1.7499 1.6584 -0.0097 *** -0.0095 *** -0.0080 *** -0.0068 *** -0.0063 *** 0.0020 0.0020 0.0018 0.0016 0.0015 14.3941 * 9.6929 6.9404 4.2370 4.7910 7.7870 7.8413 6.8582 6.2052 5.8805 -0.0867 -7.5717 -9.6310 -6.7610 -7.6396 13.3996 13.4931 11.8013 10.6776 10.1188 -0.0053 * -0.0001 0.0005 -0.0002 -0.0014 0.0032 0.0032 0.0028 0.0025 0.0024 0.01 *** 0.00 ** 0.00 ** 0.00 ** 0.00 ** 0.0012 0.0012 0.0011 0.0010 0.0009 -1.7424 * -0.6596 -0.5281 0.0117 -0.2735 0.9471 0.9537 0.8341 0.7547 0.7152 0.0838 0.1379 ** 0.1104 ** 0.0984 ** 0.0997 ** 0.0541 0.0545 0.0477 0.0431 0.0409 -0.1115 0.1553 -0.1025 0.4737 0.5816 0.8817 0.8878 0.7765 0.7026 0.6658 0.3292 0.2614 0.2546 0.2267 0.2342 -239.4020 -240.0210 -228.0980 -219.1930 -214.4090 496.8050 498.0420 474.1960 456.3860 446.8180 519.2020 520.4400 496.5930 478.7830 469.2160 10.2321 12.0656 7.2647 4.3348 4.2706 0.2745 *** 0.2671 *** 0.2211 *** 0.1622 *** 0.1236 ** 18.9451 *** 16.7905 *** 12.4101 *** 5.9845 ** 4.4225 ** 4.0803 ** 2.6689 * 2.9361 * 0.7697 1.7692 14.93 *** 14.13 *** 9.69 *** 5.21 ** 3.03 * 0.06 0.01 0.21 0.00 0.37 19.01 *** 16.80 *** 12.62 *** 5.98 * 4.79 *

0.2286 1.0442 0.3033 -254.4610 526.9220 549.3200 9.2192 0.2463 16.5228 4.8215 12.02 0.32 16.84 1996-2007 10.6745 1.5740 -0.0054 0.0014 2.9035 5.5813 -7.8330 9.6041 -0.0019 0.0023 0.00 0.0009 0.0199 0.6788 0.0694 0.0388 0.4876 0.6319 0.2102 -209.7630 437.5250 459.9230 7.4124 0.0712 1.6724 1.0645 1.00 0.40 2.07

*** *** *** *** ***

*** ***

**

*

Note: * presents significant at 10 percent level; ** presents significant at 5 percent level; *** presents significant at 1 percent level.

1793


Table 3 Independent Variables CONSTANT Coeff. Std Initial GDP Coeff. Std Highway Density Coeff. (HWD) Std Weighted HWD Coeff. Std Population Density Coeff. Std Fix Asset Investment Coeff. Std Education Coeff. Std Health Coeff. Std Railway Dummy Coeff. Std Weighted Growth Coeff. Rate Std LAMBDA Coeff. Std R-squared Likelihood AIC BIC Breusch-Pagan Test Residual Moran’s I Likelihood Ratio Test Independent Variables CONSTANT Coeff. Std Initial GDP Coeff. Std Highway Density Coeff. (HWD) Std Weighted HWD Coeff. Std Population Density Coeff. Std

1794

The Estimates of SARMA Model

Dependent Variable: Average Annual GDP Growth Rate 1996-1997 1996-1998 1996-1999 1996-2000 14.5312 * 5.8883 * 5.0837 ** 4.1020 ** 8.2673 3.1195 2.5240 1.9991 -0.0178 *** -0.0108 *** -0.0113 *** -0.0082 *** 0.0061 0.0031 0.0025 0.0019

1996-2001 3.2064 * 1.7755 -0.0072 *** 0.0017

1996-2002 3.5020 ** 1.6314 -0.0075 *** 0.0015

42.0375 **

15.7332 *

14.4854 **

19.2827 10.7960 40.1225 -0.0127 0.0093 0.0083 *** 0.0032 -0.0990 2.3340 0.1423 0.1413 3.3463 2.5566 -0.4779 ***

25.6056

19.6073 *

16.1705 11.2705 -45.6009 * -32.5793 * 24.9083 18.1501 0.0052 0.0039 0.0051 0.0042 0.0060 *** 0.0061 *** 0.0020 0.0016 1.5975 0.8993 1.6391 1.2556 0.0037 0.0470 0.0909 0.0711 -1.1861 -0.8631 1.2692 1.0505 0.9607 ***

0.8861 ***

15.9681 *

9.2325 8.1028 -28.2996 * -22.6434 * 14.5454 12.7743 0.0033 0.0021 0.0033 0.0030 0.0049 *** 0.0039 *** 0.0013 0.0011 0.6947 0.6278 1.0068 0.8848 0.0209 0.0331 0.0567 0.0499 -1.0459 -0.6532 0.8272 0.7291 0.9316 ***

0.8842 ***

6.4501 -16.8360 10.8149 0.0019 0.0026 0.0043 *** 0.0010 -1.1313 0.7602 0.0283 0.0430 -0.7570 0.6627 0.8478 ***

0.1792 0.0820 0.0912 0.0979 0.1042 0.1159 0.5361 *** -0.7029 *** -0.3960 ** -0.4737 *** -0.4757 *** -0.2174 *** 0.1113 0.1548 0.1648 0.1635 0.1635 0.1645 0.2306 0.5494 0.5553 0.5317 0.5064 0.5264 -328.94 -299.90 -271.38 -252.72 -241.07 -224.33 677.88 619.80 562.76 525.44 502.14 468.66 702.76 644.69 587.64 550.32 527.03 493.54 14.1694 7.5779 9.0559 5.0045 1.3786 0.9522 0.0055 -0.0423 -0.0213 -0.0274 -0.0273 -0.0092 1.5124 7.2580 *** 1.9090 3.1901 * 2.6957 0.7275 Dependent Variable: Average Annual GDP Growth Rate 1996-2003 1996-2004 1996-2005 1996-2006 1996-2007 2.0396 1.8342 1.7227 2.0981 2.2647 1.5462 1.3952 1.5592 1.5412 1.6817 -0.0062 -0.0047 *** -0.0040 *** -0.0034 *** -0.0029 *** 0.0014 0.0012 0.0012 0.0011 0.0011 12.5201 * 6.5734 -14.6724 10.3593 0.0040 0.0022

10.0902 *

6.5000

6.2097

5.9275 -13.4611 9.0863 0.0037 ** 0.0019

5.7288 -8.4287 8.8334 0.0024 0.0018

5.4320 -9.9807 8.3245 0.0019 0.0017

5.0824 5.3983 -8.0850 8.1291 0.0009 0.0016 (Table 3 to be continued)

The Temporal and Spatial Effect of Highways on China’s Economic Growth (Table 3 continued) Fix Asset Investment Coeff. Std Education Coeff. Std Health Coeff. Std Railway Dummy Coeff. Std Weighted Growth Coeff. Rate Std LAMBDA Coeff. Std R-squared Likelihood AIC BIC Breusch-Pagan Test Residual Moran’s I Likelihood Ratio Test

0.0019 0.0009 -0.4112 0.7299 0.0705 0.0410 -0.4774 0.5998 0.9433 0.1037 -0.4266 0.1644 0.5127 -223.07 466.14 491.03 5.6869 -0.0162 2.4763

0.0015 ** 0.0008 -0.4114 0.6294 0.0441 0.0354 -0.5321 0.4953 0.9738 ***

0.0011 0.0008 0.0862 0.6188 0.0467 0.0346 -0.0283 0.4924 0.9068 ***

0.0008 0.0007 -0.0244 0.5825 0.0458 0.0322 0.0777 0.4568 0.8899

0.1016 0.1228 0.1203 -0.5738 *** -0.5217 *** -0.5761 0.1606 0.1623 0.1605 0.5130 0.4115 0.4291 -211.98 -209.37 -204.18 443.96 438.74 428.36 468.84 463.63 453.24 4.7661 5.9182 8.1990 -0.0242 -0.0300 -0.0364 5.0198 ** 3.1120 * 5.0629

0.0010 0.0007 0.1785 0.5641 0.0253 0.0308 0.0538 0.4323 *** ***

**

0.8727 0.1309 -0.6567 0.1572 0.3815 -202.59 425.17 450.06 7.5054 -0.0462 6.2745

*** ***

**

Note: * presents significant at 10 percent level; ** presents significant at 5 percent level; *** presents significant at 1 percent level.

The candidate models include the spatial lag, the spatial error and SARMA models. Shown in Tables 2, 3 and Appendix A and B, we find that SARMA model is the appropriate model to investigate the highway impact. The OLS model shows the LM-lags are always larger than the LM-errors except the early period of 1996-1997. This implies the spatial lag models are preferred to the spatial error models. However, the tables in appendix A and B indicate the both spatial lag and spatial error do not completely solve the problem of spatial dependence. Although the Moran’s Indexes of the residuals in these two models are all small and insignificant, the likelihood ratio tests are all significant, showing that the certain degree of spatial dependence still exists. As shown in Table 3, the SARMA models give higher the log-likelihoods and smaller value of AICs and BICs. The likelihood ratio tests show that the SARMA model yield a better results in solving the problem of spatial dependence though not completely. Therefore, this paper investigates the impact of highways based on the estimated results of SARMA model. 4.2 The Highway Effect The interesting findings are the positive and diminishing effect highway density on the economic growth. As shown in Table 3, all coefficients of highways density are positive and diminishing. Most of coefficients are statistically significant except the periods of two years and longer than eight years. The coefficient of the two-year period is very close to significant criterion of ten percent. With these results, we are able to investigate the highways density impact on the economic growth both in scale and time. Magnitude of impact is large in the near term and diminishes until dies off. For near next year, the contribution of highways for an average highway density county (0.1768 kilometer per square kilometer) is unbelievable high. However, it declines down to about 91 percent GDP growth rate in the first two year period and then reduces gradually to about 5 percent GDP growth in the period of 1996-2004. The contribution continues to diminish and the impact becomes statistically insignificant in longer periods. These results clearly indicate the 1795


pattern of highway impact on GDP growth. Across the space, the higher highway density a region has today, the higher GDP growth rate will be in following short periods. The highways do not have long time impact. As discussed above, highways are a necessary factor for economic growth. They provide an improved environment for economic activities. Whether not the improved environment improved by highways becomes a real economic growth depends on some other factors. Better transportation condition may draw other investments, the effect of transportation infrastructure depends the investments. Based on this consensus, the results above might hold for every case. Table 4 shows the coefficients of initial highway density in 93 models, 97 models and 99 models. The results do not support the findings of 96 models. In 93 models, the coefficients are negative and statistically significant after 2003. All coefficients in 97 models are negative and statistically insignificant. In 99 models, the coefficient of the first period is negative and statistically significant. The coefficients of other periods are all statistically insignificant but positive or negative. It is difficult to make an inference for these conflict results. However, two patterns seem observable. The magnitudes of the coefficients are diminishing. The other is the significant coefficients seem continue for a period of times. We suspect this is the consequence of average. That is, the potential of highways only result in economic growth in certain year. Then the significant relationship among the highways and growth rate is spurious. Table 4

Coefficients of Initial Highway Density in the SARMA Models for Different Periods

Growth rate 93-94

Std. Error 13.0899 18.0784

Growth rate 93-96 Growth rate 93-97

-3.2143 13.8005 -6.6700 10.4031


-1.9624 -2.0605

8.9936 7.8964

Growth Rate 97-98 Growth Rate 97-99

-20.2799 21.6423 -16.2479 12.8933


-3.8911 -2.1240

7.2454 6.5503


-13.8937 -5.4404

9.5139 8.1986


-11.3851 -2.7307

Growth rate 93-02 Growth rate 93-03 Growth rate 93-04 Growth rate 93-05 Growth rate 93-06 Growth rate 93-07

-2.9126 -9.6869 -10.2848 -10.6642 -10.6477 -10.8078

5.8652 5.0385 4.6742 4.2857 4.3003 4.3132

Growth Rate 97-02 Growth Rate 97-03 Growth Rate 97-04 Growth Rate 97-05 Growth Rate 97-06 Growth Rate 97-07

-2.3217 4.1708 2.4351 0.4617 -0.3417 -0.8858

6.7720 6.4776 5.5987 5.2145 4.9897 4.8037

Growth Rate 99-02 Growth Rate 99-03 Growth Rate 99-04 Growth Rate 99-05 Growth Rate 99-06 Growth Rate 99-07

3.7628 3.1202 -0.7736 0.0683 -1.0081 -2.1555

Independent variables HWD 93

* ** ** ** **

Independent variables

HWD 97 Std. Error Independent variables HWD 99

Std. Error

5.9535 * 4.7827 5.4291 6.2100 5.2773 5.0727 4.9035 4.7078

Note: * presents significant at 10 percent level; ** presents significant at 5 percent level; *** presents significant at 1 percent level。

To verify the real impact, the initial highway densities are regressed to the growth rates by years. Results indicate the highway density in 1993 and 1996 only relate significantly to the GDP growth rate of 2003 and of 1997 respectively. This result confirms that highways are a necessary condition for economic growth but not a sufficient condition. Highways construction itself produces a stronger impact on near term economic. This impact may be significant or insignificant. The impact diminishes in times. 4.3 The Spillover of Highway No strong evidences support the existence of spillover effect. The results of 96 models in Table 2 show the average highway density of neighbor counties have negative and statistically significant coefficients for some periods. However, only the coefficient of the first period of weighted highway density is positive and significant in 94 models in table 5. All coefficients in other models are insignificant and not consistent signs are found. 1796


Table 5 Independent Variables Growth rate 93-94 Growth rate 93-96 Growth rate 93-97 Growth rate 93-98 Growth rate 93-99 Growth rate 93-00 Growth rate 93-01 Growth rate 93-02 Growth rate 93-03 Growth rate 93-04 Growth rate 93-05 Growth rate 93-06 Growth rate 93-07

Coefficients of Initial Highway Density in the SARMA Models for Different Periods

WHWD 93 Std. Error 47.3554 31.9421 24.3043 18.5120 13.5871 9.7394 6.5597 7.9081 9.8902 5.3324 3.2164 1.6134 2.4209

28.8569 25.7701 19.7247 16.0213 13.8308 12.9776 11.6164 10.6284 9.0309 8.4663 7.6848 7.7786 7.7957

Independent Variables

WHWD 97 Std. Error

Growth rate 97-98 Growth rate 97-99 Growth rate 97-00 Growth rate 97-01 Growth rate 97-02 Growth rate 97-03 Growth rate 97-04 Growth rate 97-05 Growth rate 97-06 Growth rate 97-07

-12.6812 -6.7444 -2.3241 -5.1898 1.6979 -2.9739 -2.0349 -0.8530 -4.9266 -1.8443


WHWD 99

Std. Error

*

30.4993 19.4544 14.4464 12.0782 10.0012 9.5813 8.2755 7.6231 7.2615 7.0291

Growth rate 99-00 Growth rate 99-01 Growth rate 99-02 Growth rate 99-03 Growth rate 99-04 Growth rate 99-05 Growth rate 99-06 Growth rate 99-07

14.0873 2.5424 12.5917 2.0755 4.6669 2.0209 1.4076 4.5761

10.1959 9.2574 11.5549 10.9970 8.9583 8.4841 7.9185 7.4392

Note: * presents significant at 10 percent level; ** presents significant at 5 percent level; *** presents significant at 1 percent level

4.4 The Effect of Other Variables The results also allow us to investigate effect of other variables. First, as the prediction of Solow model, the initial economic level (GDP) is found to have negative impact on economic growth in long periods. The fix asset investment is also found to require time to have significant impact on economic growth. The coefficients are positive and significant in short term and insignificant in long term. Their magnitudes decrease in times. Both proxy variables for education, measured by the ratio of middle school students to primary school students, and for Health service have positive and statistically insignificant coefficients. The effect of population density is negative and insignificant in both short and long periods. The effect of railway is positive but insignificant. The coefficients of the dependent lag are positive except the first period, and significant in both long periods. The economic growths in neighbor counties produce positive spillover. Their magnitudes vary in the same pattern as the Moran’s I of the dependent variables, implying the term captures the spatial dependence.

5. Conclusion Spatial dependences of economic growth are found in county-level data of Guangxi Zhuang Autonomous Region in China. The OLS estimator is bias in the estimation of highway impact. SARMA models are used to correct the bias. The empirical results show that the effects of the highways on economic growth are inconsistent. The signs and the significances of the coefficients vary with times and the initial conditions. That is, the impact of initial highway densities can be positive or negative and can be significant or insignificant. No strong evidences indicate the existence of highway spillover effect. However, the magnitudes of the highway impacts and spillover effects are all large in short term and diminishing in time. These results are determined by the property of highways as a public good. Highways are a public infrastructure that increases monotonically while economic growth can fluctuate. The impact of highways can be two folds. The construction of highways is a part of investment so it can produce a sort term impact on economic growth. The sort term impacts are direct and stronger. Then direct impact will die out in times. On the other hand, the service provided by the highways is a necessary condition but not a sufficient condition for economic growth. Therefore, the impact of highways is obtained through the performances of other 1797


factors. The spatial effects from highways in neighbor regions could not be directly and are long term. Similar to the highways inside the region, the spillover effect of neighbor highways are derived from the performances of other economic factors. The signs of the impact cannot be deterministic and depend on the movement direction of other economic factors. References: Barro Robert J. (1991). “Economic growth in a cross section of counties”, The Quarterly Journal of Economics, pp. 407-443. Barro Robert J. and Xavier Sala-I-Martin (1991). “Convergence across states and regions”, Brookings Papers on Economic Activity 1, pp. 107-182. Berechman Joseph, Dilruba Ozmen and Kaan Ozbay (2006). “Empirical analysis of transportation investment and economic development at state, county and municipality levels”, Transportation, Vol. 33, pp. 537-551. Bougheas Spiros, Panicos O. Demetriades and Theofanis P. Mamuneas (2000). “Infrastructure, specialization, and economic growth”, Canadian Journal of Economics, Vol. 33, No. 2, pp. 506-522. Button Kenneth (1998). “Infrastructure investment, endogenous growth and economic convergence”, Annual Regional Science, Vol. 32, pp. 145-162. Chi Guangqing, Paul R. Voss and Steven C. Delle (2006). “Rethinking highway effects on population changes”, Public Works Management & Policy, Vol. 11, No. 1, pp. 18-32. Cohen Jeffrey P. and Catherine Morrison Paul (2007). “The impacts of transportation infrastructure on property values: A higher-order spatial econometrics approach”, Journal of Regional Science, Vol. 47, No. 3, pp. 457-478. Demurger Sylvie (2001). “Infrastructure development and economic growth: An explanation for regional disparities in China?”, Journal of Comparative Economics, Vol. 29, pp. 95-117. Fujita Masahisa and Dapeng Hu (2001). “Regional disparity in China 1985-1994: The effects of globalization and economic liberalization”, Annual Regional Science, Vol. 35, pp. 3-37. Solow Robert M. (1957). “A contribution to the theory of economic growth”, Quarterly Journal of Economics, Vol. 70, No. 1, pp. 65-94. Appendix A The Estimates of Spatial Lag Model Independent Variables CONSTANT

1996-1997 Coeff. Std

Initial GDP

Coeff. Std

Highway Density (HWD) Weighted HWD Population Density

Education Health

5.8664 -0.0157 *** 0.0054

5.3016 4.5378 -0.0166 ***

1996-1999

1996-2000

1996-2001

4.5537

4.9769 *

4.7444 **

3.1733

2.6623

-0.0129 *** -0.0095 ***

2.3083 -0.0089 ***

0.0043

0.0030

0.0025

0.0021

1996-2002 4.4511 ** 1.8982 -0.0078 *** 0.0018

Coeff.

43.3157 **

29.3142 *

17.6161

12.7354

13.1513 *

Std

20.3336

13.0414 **

11.1239 -11.752 0 19.1687

9.1793 -13.272 5 15.8022

7.8812

6.5598

-11.2405

-7.5334

Std

35.2358

16.0452 -18.487 8 27.7087

13.5743

11.3627

Coeff.

-0.0078

-0.0033

-0.0023

-0.0027

-0.0031

-0.0022

0.0083

0.0066

0.0046

0.0038

0.0033

0.0027

Coeff.

Std Fix Asset Investment

10.5756 *

1996-1998

-16.0472

Coeff.

0.0079 **

0.0096

0.0072 *** 0.0061 ***

0.0054 ***

0.0048 ***

Std

0.0032

0.0025

0.0018

0.0015

0.0013

0.0010

Coeff.

1.0396

0.0945

-0.0997

-0.2870

-0.0355

Std

2.4730

1.9511

1.3522

1.1166

0.9586

0.7971

Coeff.

0.0557

0.0795

0.0956

0.0603

0.0597

0.0462

Std

0.1419

0.1125

0.0783

0.0645

0.0555

0.0461

-1.5741 **

(Appendix A to be continued)

1798

The Temporal and Spatial Effect of Highways on China’s Economic Growth (Appendix A continued) Railway Dummy

Coeff. Std

Weighted Growth Rate Coeff. Std R-squared

3.0714

-0.4242

-0.7966 1.0397

-0.2446

-0.4395

2.3167

1.8178

1.2612

0.8942

0.7423

0.1331

0.4545

0.5467 *** 0.4990 ***

0.4545 ***

0.4822 ***

0.1436

0.1135

0.1000

0.1112

0.1072

0.1629

Likelihood

-0.8677

-329.73

0.1075

0.3990

0.4933

0.4396

-310.63

-279.11

-261.44

0.4340

0.4716

-247.37

-231.24

AIC

679.47

641.25

578.22

542.88

514.74

482.48

BIC

704.36

666.14

603.11

567.76

539.62

507.36

Breusch-Pagan

15.0642 *

5.2562

2.9448

1.6697

1.1800

1.3702

Residual Moran's I

0.0161

0.0020

0.0205

0.0125

0.0061

0.0220

Likelihood Ratio Test

0.9012

Independent Variables CONSTANT Initial GDP

15.7718 ***

1996-2003 Coeff.

4.8372

Std

2.0006

Coeff.

-0.0076

** ***

23.2547 *** 17.1652 ***

14.1873 ***

1996-2004

1996-2005

5.8318 ***

6.5981 ***

7.5670

8.9699 ***

1.8918

1.9097

1.9472

2.0199

-0.0066 ***

-0.0060 ***

1996-2006

16.3279 ***

-0.0056

1996-2007

-0.0051 ***

Std

0.0018

0.0016

0.0015

0.0015

0.0014

Coeff.

9.6804

7.3471

4.9997

5.2842

3.3745

Std

6.6978

6.0090

5.6487

5.4119

5.2309

-11.1410

-11.2780

-7.3515

-7.9480

-7.4559

11.5253

10.3402

9.7241

9.3200

9.0181

Coeff.

0.0016

0.0015

0.0005

-0.0007

-0.0015

Std

0.0027

0.0024

0.0023

0.0022

0.0021

Fix Asset Investment

Coeff.

0.0024

0.0022 **

0.0017 *

0.0016

0.0019 **

Education

Coeff. Std

0.8145

Health

Coeff.

0.1047

Std

Highway Density (HWD) Weighted HWD

Coeff. Std

Population Density

Std

Railway Dummy

Coeff. Std

Weighted Growth Rate Coeff. Std R-squared Likelihood

**

0.0011

0.0009

0.0009

0.0009

0.0008

-0.4621

-0.3622

0.1412

-0.1628

0.0964

0.7308

0.6870

0.6582

0.6360

0.0829 **

0.0838 **

0.0884

0.0630 *

0.0472

0.0423

0.0397

0.0381

0.0367

-0.1002

-0.2468

0.3702

0.4497

0.4180

0.7587

0.6805

0.6414

0.6145

0.5937

0.4589

**

0.4097 ***

0.3092 **

0.2661

0.1740

0.1114

***

0.1178

0.1275

0.1313

0.1392

0.4007

0.3634

0.2871

0.2785

0.2287

-232.92

-222.79

-216.51

-212.44

-208.99

AIC

485.84

465.58

453.03

444.89

437.99

BIC

510.73

490.47

477.91

469.78

462.87

Breusch-Pagan Residual Moran’s I Likelihood Ratio Test

8.9333 0.0251 14.1999

***

5.0156

4.1907

3.7088

6.4619

0.0119

0.0221

0.0027

-0.0031

5.3598 **

3.9294 **

10.6112 ***

1.5378

1799


Appendix B Independent Variables CONSTANT

Coeff. Std T-value P-value Initial GDP Coeff. Std T-value P-value Highway Density (HWD) Coeff. Std T-value P-value Weighted HWD Coeff. Std T-value P-value Population Density Coeff. Std T-value P-value Fix Asset Investment Coeff. Std T-value P-value Education Coeff. Std T-value P-value Health Coeff. Std T-value P-value Railway Dummy Coeff. Std T-value P-value LAMBDA Coeff. Std T-value P-value R-squared Likelihood AIC BIC Breusch-Pagan Residual Moran’s I Likelihood Ratio Test Independent Variables CONSTANT Coeff. Std T-value P-value

1996-1997 11.5200 * 6.3095 1.8258 0.0679 -0.0174 *** 0.0056 -3.1162 0.0018 43.4417 ** 19.7885 2.1953 0.0281 -9.5756 35.9540 -0.2663 0.7900 -0.0083 0.0087 -0.9639 0.3351 0.0084 *** 0.0032 2.5983 0.0094 0.9844 2.4529 0.4013 0.6882 0.0814 0.1430 0.5690 0.5693 3.3105 2.4293 1.3628 0.1730 0.1838 0.1474 1.2466 0.2125 0.1705 -329.48 676.96 699.36 14.6886 * -0.0028 1.4098 1996-2003 7.6181 * 2.7406 2.7797 0.0054

The Estimates of Spatial Error Model 1996-1998 8.4372 6.6359 1.2715 0.2036 -0.0194 *** 0.0048 -4.0726 0.0000 27.2831 * 15.5054 1.7596 0.0785 -18.4871 32.5660 -0.5677 0.5703 -0.0020 0.0074 -0.2687 0.7882 0.0101 *** 0.0026 3.8927 0.0001 -0.5531 1.8690 -0.2960 0.7673 0.1115 0.1142 0.9765 0.3288 -1.4041 2.0698 -0.6784 0.4975 0.5485 *** 0.1096 5.0048 0.0000 0.4210 -310.11 638.22 660.61 3.4710 -0.0014 16.8063 *** 1996-2004 8.7502 *** 2.3147 3.7803 0.0002

1996-1999 7.0115 5.1404 1.3640 0.1726 -0.0153 *** 0.0034 -4.5533 0.0000 16.4674 10.9753 1.5004 0.1335 -6.5133 24.0995 -0.2703 0.7870 -0.0018 0.0052 -0.3536 0.7237 0.0076 *** 0.0018 4.2138 0.0000 -0.6206 1.2841 -0.4833 0.6289 0.1250 0.0795 1.5726 0.1158 -0.5990 1.4552 -0.4116 0.6806 0.6429 *** 0.0955 6.7346 0.0000 0.5124 -278.93 575.87 598.26 1.1061 0.0148 23.6073 *** 1996-2005 9.1405 *** 1.9848 4.6053 0.0000

1996-2000 1996-2001 1996-2002 7.7305 ** 7.2286 ** 6.2358 ** 3.9316 3.2143 2.7587 1.9662 2.2489 2.2604 0.0493 0.0245 0.0238 -0.0110 *** -0.0101 *** -0.0094 *** 0.0028 0.0024 0.0020 -3.9908 -4.2826 -4.7372 0.0001 0.0000 0.0000 10.9182 11.3997 13.2945 ** 8.9935 7.7061 6.4115 1.2140 1.4793 2.0735 0.2247 0.1391 0.0381 -11.1801 -9.3973 -2.1523 19.0916 15.9763 13.5024 -0.5856 -0.5882 -0.1594 0.5581 0.5564 0.8734 -0.0031 -0.0030 -0.0024 0.0043 0.0037 0.0031 -0.7187 -0.8281 -0.7853 0.4723 0.4076 0.4323 0.0065 *** 0.0057 *** 0.0051 *** 0.0015 0.0013 0.0011 4.3402 4.4599 4.7575 0.0000 0.0000 0.0000 -0.7694 -0.1518 -1.3829 * 1.0774 0.9350 0.7717 -0.7141 -0.1624 -1.7922 0.4752 0.8710 0.0731 0.0851 0.0747 0.0692 0.0660 0.0569 0.0472 1.2880 1.3132 1.4671 0.1977 0.1891 0.1424 -1.1301 -0.3546 -0.4898 1.2003 1.0277 0.8559 -0.9416 -0.3451 -0.5722 0.3464 0.7300 0.5672 0.5714 *** 0.5202 *** 0.5543 *** 0.1064 0.1134 0.1088 5.3725 4.5862 5.0958 0.0000 0.0000 0.0000 0.4499 0.4370 0.4793 -261.58 -247.90 -231.50 541.17 513.79 481.01 563.56 536.19 503.40 0.3349 1.1524 0.0116 0.0098 0.0202 16.8761 *** 13.1286 *** 15.7986 *** 1996-2006 1996-2007 10.0006 *** 10.6288 *** 1.8150 1.6239 5.5099 6.5453 0.0000 0.0000

(Appendix B to be continued) 1800

The Temporal and Spatial Effect of Highways on China’s Economic Growth (Appendix B continued) Initial GDP

Coeff. -0.0088 Std 0.0020 T-value -4.4020 P-value 0.0000 Highway Density (HWD) Coeff. 8.9306 Std 6.5016 T-value 1.3736 P-value 0.1696 Weighted HWD Coeff. -12.3896 Std 13.5511 T-value -0.9143 P-value 0.3606 Population Density Coeff. 0.0017 Std 0.0031 T-value 0.5486 P-value 0.5833 Fix Asset Investment Coeff. 0.0026 Std 0.0011 T-value 2.3621 P-value 0.0182 Education Coeff. -0.2069 Std 0.7868 T-value -0.2630 P-value 0.7925 Health Coeff. 0.1182 Std 0.0480 T-value 2.4639 P-value 0.0137 Railway Dummy Coeff. -0.1435 Std 0.8675 T-value -0.1655 P-value 0.8686 LAMBDA Coeff. 0.5318 Std 0.1119 T-value 4.7544 P-value 0.0000 R-squared 0.4142 Likelihood -232.77 AIC 483.54 BIC 505.94 Breusch-Pagan 6.3767 Residual Moran’s I 0.0207 Likelihood Ratio Test 14.5037

***

**

**

***

***

-0.0073 0.0018 -4.1027 0.0000 6.5396 5.8341 1.1209 0.2623 -12.8815 11.7929 -1.0923 0.2747 0.0014 0.0028 0.5136 0.6076 0.0023 0.0010 2.3441 0.0191 -0.1739 0.7154 -0.2432 0.8079 0.0927 0.0432 2.1440 0.0320 -0.3011 0.7739 -0.3890 0.6973 0.4634 0.1206 3.8422 0.0001 0.3676 -223.03 464.06 486.45 3.6987 0.0171 10.1405

***

**

**

***

***

-0.0066 0.0016 -4.0280 0.0001 4.7274 5.4504 0.8673 0.3858 -8.9063 10.5541 -0.8439 0.3987 0.0005 0.0025 0.2071 0.8360 0.0018 0.0009 1.9611 0.0499 0.3207 0.6757 0.4746 0.6351 0.0909 0.0403 2.2584 0.0239 0.4548 0.7087 0.6417 0.5211 0.3603 0.1321 2.7282 0.0064 0.2944 -216.42 450.84 473.24 3.7877 0.0187 5.5478

***

**

**

***

**

-0.0060 0.0015 -3.8878 0.0001 4.8370 5.2635 0.9190 0.3581 -9.1969 9.9242 -0.9267 0.3541 -0.0007 0.0024 -0.3115 0.7554 0.0017 0.0009 1.9205 0.0548 -0.0448 0.6541 -0.0685 0.9454 0.0961 0.0386 2.4879 0.0128 0.4619 0.6710 0.6883 0.4913 0.2905 0.1388 2.0935 0.0363 0.2760 -212.73 443.47 465.86 3.4728 0.0129 3.3506

***

*

**

**

*

-0.0054 *** 0.0014 -3.7531 0.0002 3.0953 5.1293 0.6035 0.5462 -8.0111 9.2890 -0.8624 0.3885 -0.0017 0.0022 -0.7698 0.4414 0.0020 0.0008 2.3791 0.0174 0.1516 0.6354 0.2386 0.8114 0.0688 * 0.0370 1.8582 0.0631 0.4415 0.6272 0.7039 0.4815 0.1737 0.1481 1.1725 0.2410 0.2251 -209.20 436.40 458.80 6.6370 0.0091 1.1221

1801


Poverty Assessment in Low-, Medium- and High-Income Arab Countries: Yemen, Egypt and Bahrain Karima Korayem  (Al-Azhar University, Cairo, Egypt)

Abstract: Bahrain, Egypt and Yemen differ significantly in their population size, per capita income, and output production per sectors. Three questions are raised in this paper: First, do those structural differences affect the type of poverty prevailing and its relative spread among the population? Second, what are the main causes of poverty in the three countries and how different they are, given the structural dissimilarity among them? Third, what are the poverty reduction policies applied in the three countries, and how do they differ among themselves? The paper addresses the first question by estimating and assessing the extent and type of poverty prevailing in the three countries, differentiating between absolute and relative poverty. For the second question, the direct and indirect causes of poverty on the macro-level are pointed out theoretically, and the application of those causes to the three countries are examined to find out whether, being rich or poor, has its impact on those causes in the country concerned. For the third question, the paper assesses the poverty eradication policies applied in each of the three countries. Key words: Arab countries; poverty; causes; policy measures JEL codes: O200

1. Introduction The Arab countries differ among themselves in many aspects, economically and socially. Important differences are the sectors’ structure of the gross domestic product (GDP), per capita income and the population size. Will those differences affect the type of poverty, its main causes and the policy measures applied to combat it in the three countries? The study will start by identifying the type of poverty prevailing in the three countries, differentiating between absolute and relative poverty. The households and individuals who are living in absolute poverty are those whose income is not enough to buy their basic needs of food, clothes, shelter, etc; those are living at the poverty line or below. On the other hand, those who are living in relative poverty are the households and individuals whose income is enough to buy the basic needs, but is lower significantly than the average income per capita on the country level. To discuss the main causes of poverty in the three countries, we shall, first, identify those causes on the theoretical level, then examine which of them exist in each of the three countries, given their economic differences. 

Karima Korayem, Ph.D. in Economics, Professor, Al-Azhar University; research areas/interests: poverty & income distribution. E-mail: [email protected] & [email protected]. 1802

Poverty Assessment in Low-, Medium- and High-Income Arab Countries: Yemen, Egypt and Bahrain

Finally, policy measures applied to combat poverty in the three countries will be assessed. Accordingly, the study will consist of seven parts, including the introduction. Part two will discuss the main economic and social characteristics of the three countries: population, income per capita, sectors’ contribution to GDP. Part three will assess the type of poverty prevailing in the three countries, absolute and relative poverty. Part four will estimate poverty and income distribution in the three countries. Part five will present the macro causes of poverty theoretically and examining their existence in the three countries. Part six will assess poverty eradication policy measures applied in those countries. Finally, part seven has the concluding remarks.

2. Main Economic Features of the Three Countries: Yemen, Egypt, Bahrain One may classify the Arab countries into two groups. One group is the oil producing countries which are characterized with relatively large extraction sector’s share in GDP and high income per capita. The second group includes the other Arab countries which are characterized with diversified commodity sectors (agriculture, manufacturing, etc.) and a lower income per capita as compared to the first group. Bahrain falls in the first group, while Yemen and Egypt fall in the second group. Looking at the population size and income per capita in 2010, Bahrain has population of 1.3 million and income per capita of $17464. Egypt has population of 78.5 million and Income per capita of $2783 and Yemen has population of 23.2 million and income per capita of $1265 (Table 1). Thus, income per capita in Bahrain is about 6 times the income per capita in Egypt and 14 times the income per capita in Yemen, while income per capita in Egypt is about twice income per capita in Yemen. Table 1

The Economic Features of Yemen, Egypt & Bahrain 2009 28125

Yemen 2010 29298

Egypt 2009 2010 188489 218393

2009 19586

Bahrain 2010 22945

Gross domestic product (GDP) at market prices ($ millions)(1) Rate of growth of GDP at domestic 3.9 8.0 4.7 5.1 3.1 4.5 currency (constant prices)(2) Population (in thousands)(3) 22492 23154 76822 78462* 1215 1314* Rate of growth of population(2009-2010) % -2.9 -2.1 -8.1 GDP per capita ($)(4) 1250 1265 2454 2783 16120 174654 Rate of growth of GDP per capita at constant prices (%)(5) 1.0 5.1 2.6 3.0 -5.0 -3.6 Percentage of all commodity sectors to GDP (%)(6) 45.4 45.7 38.7 49.1 44.3 47.1 43.8 41.3 29.1 28.0 53.0 51.2 Percentage of the extraction industry to all commodity sectors (%) (7) 41.4 43.2 59.1 60.0 33.8 37.1 Percentage of agriculture and manufacturing to all commodity sectors (%) (7) Note: * Primary data. (1) Arab Leagues & Others (2010& 2011); Appendix 2/2. (2) Same Source above (2011); Table 1. (3) Same Source above; Appendix 8/2; (4) Same Source above; Table 2. (5) Calculated from the above as (= rate of growth of GDP at domestic currency (at constant prices)—rate of population growth), assuming that the population growth rate in 2009 is the same as in 2010. (6)Commodity sectors consist of agriculture, fishing & forestry, extraction industries, manufacturing, construction, electricity, gas & water. (7) Calculated from the same source below (Report 2011), Appendix 2/3 & 2/4. Source: Arab League & Others (2010, 2011).

1803


Looking at the commodities sectors’ share1 in GDP, one finds that it is quite close in the three countries; in 2010, it ranges between a minimum of 45.7% of GDP in Yemen, and a maximum of 49.1% in Egypt, with Bahrain falling in between with 47.1% (Table 1). However, the share of the extraction sectors, including oil, in total commodities’ sectors’ differ significantly among the three countries; it ranges between a minimum of 28% in Egypt and a maximum of 51.2% in Bahrain, with Yemen falling in between with 41.3% (Table 1). But, despite the oil high price and the relatively high share of the extraction sectors in total commodity sectors’ share in Yemen as compared to Egypt, the income per capita is lower in the former as compared to the latter. Comparing the share of the extraction sectors in GDP in the three countries, it is found that in 2010, this share was 24.1% in Bahrain, 18.9% in Yemen, and 13.7% in Egypt2. Consequently, the Bahraini economy has the highest income per capita, but the lowest diversification in economic activities; next comes Egypt after quite a distance in income per capita; and lastly in order, comes Yemen. Given those differences between Bahrain and the two other countries, two questions will be raised in this respect. First, does Bahrain suffer from the same type of poverty prevailing in Egypt and Yemen? Second, is the spread of poverty among the population the same in the three countries?

3. Type of Poverty in Yemen, Egypt, Bahrain Theoretically, there are two types of poverty: absolute and relative poverty. People living in absolute poverty are those who live at the poverty line or below, which means that their income is just sufficient, or less, to cover their necessary living requirements of food, shelter, education, etc. of basic needs. There are three methods of estimating the poverty line. The first method is estimating the expenditure on necessary food which provide the individual with the required calories that make him live a healthy life and the expenditure on necessary non-food items, like shelter, clothes, education, and health. Food items used in estimating the poverty line consists of those items with the lowest price available, which reflect the consumption patterns of the low-income people in the country concerned. This is called the national poverty line. The problem with this poverty line is that the consumption pattern of the low-income people and the prices of the necessary commodities and services differ from one country to another and, also, differ in the same country over time. Estimating the food expenditure at the poverty line, which represents the most important component of basic needs consumption of the poor, depends on the food items chosen and the prices applied, which can be politically manipulated to get a lower poverty line and, hence, a lower poverty level which does not reflect the reality of the size and condition of the poor. Consequently, the national poverty line does not reflect accurately the poverty levels between countries, or in the same country over time. The second method is the international poverty line as defined by the World Bank. It sets the poverty line as equal to $2 or less daily3. According to it, those who are living in poverty are those individuals who live at $2 a day or less. The disadvantage of this poverty line estimate is its dependence on the exchange rate, with the Purchasing Power Parity (PPP) applied. But its advantage as compared to the national poverty line is that it is easier to apply in estimating comparative poverty levels in different countries. The third method is estimating the level of poverty subjectively. This method is based on assuming that the lowest three or four deciles are those who live in poverty, without applying any objective criteria to substantiate 1

The commodity sectors consist of agriculture, fishing and forestry, the extraction industry and manufacturing, construction, electricity and gas, and water. 2 These ratios have been calculated from Table 1. 3 This international poverty line estimate was set, first, at $1 daily, then raised to $2. 1804


the identification of the 30% or 40% of the population as the poor in the society. The second type of poverty, relative poverty, is defined as those who are living above the absolute poverty line and, consequently, have their basic needs satisfied. But, on the other hand, the individual’s income (expenditure) is significantly below the average income (e.g., by half or by 1/3) of the country he is living in. Yemen and Egypt suffer from absolute poverty. National and international poverty lines have been estimated for each of the two countries at different points of time. On the other hand, Bahrain suffers from relative poverty. A field survey carried out on the Bahraini receiving social insurance from the government, who represent the Bahraini poor in the society, found that all of them owned consumer durables, like fridge, stove, washing machine and about 90% of them owe a car; moreover, some of them have two cars4. These consumer durable goods cannot be owned by those who live in absolute poverty who just opt to get their basic needs of food, shelter, etc. The result of this survey shows that Bahrain has relative poverty, and not absolute poverty as in Yemen and Egypt. To avoid the negative aspects of the absolute poverty measures pointed out above, and to make objective comparison between the three countries despite the difference in the type of poverty they have, a suggested methodology will be used, based on the decile distribution estimates in the three countries, to estimate the relative share of the low-income (expenditure) group of people including the poor. According to this methodology, an objective indicator based on economic theory will be applied to differentiate between low-, medium-, and high-income people and an index is derived to estimate the income distribution in each of the three countries.

4. Absolute Poverty, Relative Poverty, and Income Distribution in the Three Countries 4.1 The Methodology Applied5 The low-, middle- and high-income (expenditure) households group will be estimated by applying the decile distribution of households’ income (expenditure). The state of income distribution is measured by applying the income inequality index (III). The development of this index, methodologically, is based on the equal-distribution concept of income. Income is equally distributed among the population, if a given percentage of the population receives an equal percentage of national income; i.e., five percent of the population receives five percent of national income, ten percent of the population receives ten percent of national income, etc. Accordingly, income is unequally distributed if a given percentage of the population receives a smaller percentage share of national income, while other equal percentage of the population receives a greater percentage share of national income. The former group is the population who fall in the lower-income intervals, while the latter group is those who belong to the upper-income intervals. In the decile income-distribution, one may differentiate between three groups: the household deciles whose shares of national income are less than 10 percent for each decile; the household deciles whose shares of national income are around 10 percent of national income for each decile; and those household deciles whose relative shares are greater than 10 percent of national income for each decile. Hence, in the decile income-distribution, the 10 percent share of national income is the equal income-distribution share (EIDS), while in the quintile income-distribution pattern, the EIDS will refer to 20 percent of national income, etc. The EIDS will be used in identifying the three household groups: the lower-income households, the middle-income households, and the upper-income households; the first group includes the poor and the last group includes the rich. 4 5

See Korayem (2012). For the development of the methodology, and its application on Egypt, see Korayem (2002). 1805


When the income distribution data are not available, the expenditure data in the Households Expenditure Surveys can be used to identify the low-income households group in the society, since household’s expenditure and income are closely related. In this case, the equal expenditure-distribution share (EEDS), instead of the EIDS, will be used to differentiate between the three household groups: the lower-expenditure (lower-income) households group include all deciles the expenditure share of each is below the EEDS; the middle-expenditure (middle-income) households group, which consists of all deciles the expenditure share of each is around the EEDS; and the upper-expenditure (upper-income) households group, which encompasses all deciles the expenditure share of each of them is higher than the EEDS. According to this methodology, income distribution is measured by the Income Inequality Index (III), the value of which falls between zero (in case of perfect income equality), and one (for the extreme case of income inequality). The income inequality index (III) is presented as follows: III =

∑N

X –RSi

/

(1)

X

Where X = fixed population interval = equal income-distribution share (EIDS); RSi = relative income share of the ith population interval; N = number of the population intervals; i.e., N = 5 for the quintile distribution of the population, N = 10 for the decile distribution, etc. The Meaning of the III: The numerator: ∑N X– RSi /2 represents the relative share of national income (or expenditure) that is unequally distributed6. The denominator: (100-X ) represents the extreme case of inequality in income distribution, when all the national income is received by one population interval (X), i.e., by one quintile, or one decile, etc. In the extreme case of equality in income distribution, III = 0, since in this case RSi = X for all i, and hence the numerator in equation (1) is equal to zero. In the extreme case of income inequality, the general solution of Equation (1) will be: III =

∑N

Xi–0

 X–100 –X

/

1

(2)

Where N = 5 for the quintile distribution of the population; N = 10 for the decile distribution, etc.7 Thus, the value of the income inequality index falls between 0 and 1, i.e., 0< III< 1. 4.2 Estimating the Three Income (Expenditure) Groups in Yemen, Egypt, and Bahrain The estimation of the low-, medium-, and high-income (expenditure) groups in the three countries has been made by using the Households Budget Survey in Yemen (2006), Egypt (2009), and Bahrain (2002) as shown in Table 2. Applying the methodology explained above and assuming a range of 2% around the EIDS (EEDS) for the 6

The numerator is divided by 2 because we are taking the sum of the absolute values (i.e., disregarding the + & - signs) of the differences between the decile expenditure shares and the EIDS. This entails double counting. To rectify for that, the total of the summation of all those differences should be divided by 2. 7 Applying the general solution (Equation 2) to the quintile and decile distribution of the population, we get the following: (a) For the quintile distribution: X = 20, N = 5 III =

∑

| X

| |X X

|/

= [(4X + 80) / 2] / 80 = 80/80 = 1

(b) For the decile distribution: X = 10, N = 10 III = [(9X + 90)/2]/90 = 90/90 = 1 1806


middle income (expenditure) group, the three groups are defined as follows, using the households’ expenditure data: the low-income (expenditure) group are those households whose expenditure for each decile is less than 8% of total expenditure on the country level; the medium-income (expenditure) group are the households whose expenditure for each decille ranges between 8%-12% of total expenditure; the high-income (expenditure) group are the households whose expenditure for each decile is higher than 12% of total expenditure. Thus, according to Table 2, the low-income (expenditure) households represent 50% of the population in Yemen and Bahrain, having 26.2% of total expenditure in the former country and 24.2% in the latter; while this group represents 40% of the population in Egypt and receives 22.2% of total expenditure. The middle-income (expenditure) group represents 30% of the population in Yemen and Bahrain, having 29.4% of total expenditure in the former country and 30.3% in the latter; in Egypt, the middle-income (expenditure) group represents 40% of households and have 39.3% of total expenditure. The high-income (expenditure) group represents 20% of households in the three countries and its relative share in total expenditure is 44.5% in Yemen, 38.4% in Egypt and 45.9% in Bahrain. Table 2

The Decile Distribution of Households’ Expenditure in Yemen, Egypt & Bahrain(1)

Yemen Egypt Bahrain (2006) (2009) (2002) First Decile 3.1 3.4 2.6 Second Decile 4.4 5.3 4.1 Third Decile 5.3 6.3 4.8 Fourth Decile 6.2 7.2 5.8 Fifth Decile 7.2 8.0 6.9 Sixth Decile 8.3 9.0 8.6 Seventh Decile 9.6 10.3 9.9 Eighth Decile 11.5 11.8 11.8 Ninth Decile 14.5 13.8 16.1 Tenth Decile 30.0 25.1 29.8 Income Distribution (III)(2) 0.288 0.232 0.306 (1) The country order in the Table is according to GDP per capita (see Table 1). (2) Calculated using Equation (1) in the text. Source: for Yemen, Korayem (2011); for Bahrain, Korayem (2012); for Egypt, calculated from: CAPMAS (2009), Table (10.1). Households’ Decile Distribution

To estimate the equality in income distribution in the three countries, two indicators are used: The first indicator is the ratio of the expenditure of the highest household’s decile to the lowest decile. The second indicator is estimating the Income Inequality Index (III) in the three countries, applying Equation 1. Table 2 is used in estimating the two indicators. For the first indicator, the ratio of the households’ expenditure of the highest decile to the lowest decile is 9.7% in Yemen, 7.4% in Egypt and 11.5% in Bahrain. For the second indicator, the III is equal to 0.288 in Yemen, 0.232 in Egypt and 0.306 in Bahrain. These two indicators show that the income distribution is the worse in Bahrain and the best in Egypt, with Yemen falling in the middle. One may conclude that for the two indicators the income distribution is the worse in the rich country, Bahrain, as compared to the two relatively poor countries, Yemen and Egypt.

5. Macro Causes of Poverty in the Three Countries Theoretically, poverty is associated with low income per capita. However, this is just one of the causes of

1807


poverty. There are other factors which are responsible of poverty in absolute and relative terms. Thus, we shall point out, first, the main causes of poverty on the theoretical level in general, then assess the relative effectiveness of those causes in each of the three countries. 5.1 Macro Causes of Poverty on the Theoretical Level8 One may classify the causes of poverty into direct and indirect causes (Figure 1). The first are those factors that have direct impact on the average income generated on the national level and, hence, on poverty. The second are the factors that underlie and affect the direct causes. One may point out four direct causes: low annual rate of growth of gross national product (GNP) per capita, low labour productivity, high support burden ratio9, and unequal income distribution. The assignment of a set of indirect causes to a certain direct cause is somewhat arbitrary, since some of the indirect causes affect more than one direct cause. For example, a high population growth rate is identified as one of the indirect causes responsible for the low growth rate of GNP per capita, while it is responsible also for another direct cause, the high support burden ratio. Also, there is a correlation between some indirect causes (like unemployment and inappropriate macroeconomic policies; inadequate access to assets and credits and the unequal distribution of assets), and some of the direct causes (like low labour productivity and the rate of growth of GNP per capita). For the sake of clarity Figure 1 presents the links between each of the direct causes and a set of specified indirect causes chosen as the ones more directly responsible for the identified direct cause. The first direct cause of poverty is the low annual rate of growth of GNP (GDP) per capita. A high GNP rate of growth per capita implies a rise in production, increased generation of income and, hence, less poverty. Consequently, taking the rate of growth of population into consideration, the low rate of growth of GNP per capita is one of the direct causes of poverty. Factors underlying this direct cause are: high population growth rate, inappropriate macroeconomic policies, and the external factors that affect the availability of resources at the country level. Main external factors are: deterioration in the terms of trade, debt burden, wars, and inadequate regional and international co-operation. Given a certain rate of growth of GNP, the higher the population growth rate, the lower will be the annual rate of growth of GNP per capita. Inappropriate policies will affect the annual rate of growth of GNP per capita through their impact on the availability of resources and their inefficient utilization in production. For example, tight fiscal and monetary policies that reduce investment will decrease production, employment creation and income generation and, hence, GNP per capita. The external factors affect, also, the rate of growth of GNP per capita, and hence poverty, through their impact on the availability of resources in the country concerned. The deteriorating terms of trade and the debt burden reduce the foreign exchange available to the country, which is needed for the import of final goods (e.g., food) and intermediate factors of production (like capital and intermediate goods), which will reduce production and GNP rate of growth. The outbreak of war in a country, not only destroys physical capital which will have to be replaced later, but also directs resources toward the purchase of military equipments and away from investment to raise production and income. Moreover, the negative economic impact of war extends to many countries directly or indirectly involved in the war. Another external factor is inadequate regional and international co-operation; it affects the amount of funds available for economic development and for poverty alleviation.

8

Korayem, 1998. Support burden ratio is defined as the number of persons supported by one worker’s income on average. This is estimated as equal to total population/employed labour force on the country level.

9

1808


1. A. High Population Growth Rate

1. B. Inappropriate Macroeconomic Policies

1. Low Annual Rate of Growth of GNP (GDP) per Capita

4. B. Inadequate Net Transfers to the Poor

4. A. Unequal Distribution of Assets

1. C. External Factors: -Deteriorating Terms of Trade -Debt Burden -Wars -Inadequate Regional and International Cooperation

2. A. Inadequate Access to Education Services 4. Unequal Income Distribution

Poverty

2. Low Labour Productivity

2. C. Inadequate Access to Assets and Credits

3. High Support Burden Ratio

3. C. High Unemployment Rate

Figure 1

3. B. Low Female Participation in the Labour Force

2. B. Inadequate Access to Health Services

3. A. Low Labour Force Participation Rate

Economic Causes of Poverty on the Macro Level

The second direct cause of poverty is low labour productivity. Neo-classical economic theory stipulates a proportional relation between labour marginal productivity and wages. However, there are institutional and structural factors in the contemporary economies that result in wage setting deviating from this rule 10 . Nevertheless, productivity cannot be disregarded as one of the important determinants of labour’s income in any country. Higher labour productivity means higher production and higher income generation. Labour productivity is affected by three factors: access to education, access to health services, and access to assets and credits. Inadequate access to education and health services reduces the work capabilities of labour and reduces its productivity. Inadequate access to assets and credits reduces availability of assets that can co-operate with labour in production and raise its productivity. The third direct cause of poverty is the support burden ratio which indicates how many persons are supported by each worker on average. A high support burden ratio in a country (equals to 3 or 4) means that the income earned by one worker is used to support three or four persons on average which implies, given a certain wage level, a lower income per capita as compared to the countries where this ratio is lower. There is a direct relation between the value of the support burden ratio in a country and the level of poverty in that country, taking labour productivity as given. The support burden ratio is affected by three indirect causes of poverty: the labour force

10

For example, the bargaining power of labour unions is effective in raising wages periodically in the developed countries. Also, it is argued that wages in the government sector in some Arab countries are higher than labour productivity (e.g., the case of Egypt). 1809


participation rate, the female participation in the labour force, and unemployment. The lower the labour force participation rate, the higher will be the support burden ratio. Given the dependency ratio11, the lower the female participation rate in the labour force, the higher is the support burden ratio. The third underlying cause of the support burden ratio is unemployment; the higher the unemployment rate, the smaller is the number of the employed labour force and, hence, the higher is the support burden ratio. The fourth direct cause of poverty is the unequal distribution of income which is affected by two indirect causes. One is the unequal distribution of physical and financial assets among the population, which is aggravated by the inadequate access of the poor to assets and credits. The second indirect cause is the inadequate net transfers to the poor to compensate for the skewed income distribution. Net transfers to the poor are the outcome of the indirect taxes which the poor have to pay on their consumption expenditures, and the subsidies and transfers they receive (in kind or in cash) from the government and the civil society (e.g., the NGO’s). 5.2 Causes of Poverty in the Three Arab Countries Examining the existence of the direct and indirect causes of poverty in the three countries, one finds that these causes differ significantly among them. Some of these causes apply strongly to one country or more while others do not apply, given the differences in the type and extent of poverty in them. The first direct cause of poverty, the annual rate of growth of income per capita, differ considerably in the three countries. In 2010, it was the highest in Yemen (5.1%), with the lowest income per capita ($1265), 3.1% in Egypt with income per capita $2783 and was negative in Bahrain (-3.6%) with the highest income per capita among the three countries ($174654) (see Table 1). Looking at the indirect causes affecting the rate of growth of income per capita (Figure 1), one finds that in Yemen and Egypt where absolute poverty prevails, the population rate of growth which affects income per capita and its rate of growth negatively, is 2.9% in the former and 2.1% in the latter as compared to a significantly high population rate of growth of 8.1% in Bahrain (Table 1). This shows that a high population growth rate and a low rate of growth of GDP per capita, even a negative one like in Bahrain, can be tolerated for some period of time without causing absolute poverty, if income per capita is high enough. On the other hand, if income per capita is low and absolute poverty prevails, like in Egypt and Yemen, having high GDP growth rate and low population growth rate should be put as important national targets, otherwise income per capita will fall and poverty increases. The second indirect cause, the inappropriate macro-economic policies applied on the national level, affect the rate of growth of income per capita through its negative impact on prices, cost of production, employment creation, etc. A good example of that is the Economic Reform and Structural Adjustment Policies (ERSAP) applied in Egypt in 1991, and in Yemen in 1995. Those policies have led to an increase in prices and cost of production in Egypt by the devaluation of the Egyptian pound and raising interest rates. In Yemen, those policies have led to the removal of food subsidy, which has resulted in increasing prices of food commodities that have the largest share in the budget of the poor. However, the positive impact of ERSAP was the reduction in the international debt for Egypt and Yemen. But despite that, the final impact on the poor in the two countries was negative (Korayem, 1995/96; Korayem, 2011). For the external factors, which are among the indirect causes affecting income per capita and its rate of growth,

11

The dependency ratio is the summation of the very young (less than age 15) and very old (greater than age 65) divided by total population. 1810


one finds that the net barter terms of trade estimated by the World Bank fell in Bahrain from 124.7 to 114 during 2007-2010, while increased in Egypt from 131.3 to 139.3, and in Yemen from 149.4 to 151.2 in the same period12, implying that this factor did not have a negative impact on the two countries with absolute poverty, while its fall in Bahrain did not, supposedly, hurt the poor who live in relative poverty, having income above the poverty line. The other external factor that has negative impact on the country is the international debt. This factor does not exist in Bahrain, while it is an important factor in Yemen and Egypt. In the period 2007-2010, the ratio of international debt to GDP fell from 27.5% to 21.0% in Yemen, and from 25.7% to 16.0% in Egypt. Accordingly, the ratio of annual international debt service to export of goods and services fell from 3.4% to 2.8% in Yemen and from 8.9% to 6.0% in Egypt during the same period, 2007-2010 (League of Arab States & others, the Economic Report, 2008 (p.167) and 2011 (p. 166). The other international factor which had a large negative impact on the Yemen economy is the break out of the civil war in 1994, which has cost the country $11.13 billion, amounting to 3 years of the country GDP during that time (Korayem, Poverty in Yemen, 2011). Finally, the last international factor which was effective in the three countries is the reduction in the international and regional cooperation. Table 3

Indicators of the Causes of Poverty in Yemen, Egypt & Bahrain (2009, 2010)

Healtlh & Education Indicators

Yemen 2009 2010

2009

Egypt 2010

Bahrain 2009 2010

1. Health Indicators: A- Life Expectancy at Birth 62.0 -70.0 -76.0 -B- Children Mortality Rate under age 5(for every 1000 children alive) 90.2 -21.0 -12.1 -C- Ratio of Health Expenditure to Total Public Expenditure 5.8 -7.3(1) -9.8(1) -2. Education Indicators: A- Enrollment Rate in Primary Education (%)(2) 85.4(1) -104.7(3) -105.3(1) -(2) (4) 4) B- Enrollment Rate in High School (%) 47.4 -86.2( -96.8(1) -1) (3) C- Illiteracy Rate (15 years & over) (%) 39.1( -33.6 -9.2(1) -(5) D- Education Expenditure to Total Public Expenditure (%) 17.7 -11.9 -11.7(5) -E- Labour Force as a percentage In total Population (%) 34.0 -33.0 -32.7 -F- Ratio of Female Labour Force(15 years & over)to total Labour Force (%) 17.3 -28.0 -32.5 -G- Unemployment Rate (%) 14.6 --8.9 -3.8(6) (7) H- Average Labour Productivity ($) 3678 -7435 -49297 -Note: (1) Year 2008. (2) Enrollment rate is defined as those enrolled in a certain education level to total population of the age which are supposed to enroll in the education level under consideration. (3) Year 2007. (4) Year 2005. (5) During the period2006-2007. (6) Bahraini population only. (7) Estimated as follows: Average labour productivity = GDP at current prices ($) (Table 1)/(Total population (Table 1) x percentage of labour force to population (Table 3) Source: Arab League & Others (2011), Annex 2/10, 2/11, 2/13, 2/15, 2/16, 2/17 & 2/18.

Looking at the second direct cause of poverty, the average labour productivity, it is much higher in Bahrain with relative poverty as compared to Yemen and Egypt which have absolute poverty. In 2009, the annual average labour productivity was $49297 in Bahrain, as compared to $7435 in Egypt and $3678 in Yemen (Table 3). This is 12

The net barter terms of trade is estimated as the percentage of the export under consideration; see: http//data. worldbank.org/indicator.

price index to the import price index in the country

1811


attributed mainly to the capital intensive technique applied in the oil production activity in Bahrain as compared to Egypt and Yemen with more diversified economic activities, given the relative availability of capital and relative scarcity of labour in the former country as compared to the two latter countries. For the indirect factors affecting the average labour productivity, Bahrain, as compared to Egypt and Yemen, is more superior in the education and health services as reflected by the education and health indicators (Table 3), which have positive impact on labour productivity. For health indicators, life expectation at birth in 2009 is 76 years in Bahrain as compared to 70 years in Egypt and 62 years in Yemen. Also, for child deaths before 5 years of age, the rate is 12.1 child per thousand alive child in Bahrain, as compared to 21 child in Egypt and 90.2 child in Yemen. Health expenditure differ, also, in the three countries; it represents 9.8% of total public expenditure in Bahrain with the highest income per capita, and fell to 7.3% in Egypt and to 5.8% in Yemen with the lowest income per capita. For education indicators, enrollment in the primary education is close in Bahrain and Egypt, despite the large difference in income per capita between the two countries; it is 105.3% in Bahrain, 104.7% in Egypt, as compared to 85.4% in Yemen. In the high school enrollment, the three countries differ significantly, with a lower level in Yemen, 47.7, as compared to 86.2 in Egypt and 96.8 in Bahrain (Table 3). However, the illiteracy rate for those 15 years and older is closer in Egypt (33.6%) and Yemen (39.1%) and significantly lower in Bahrain (9.1%). On the other hand, the ratio of education expenditure to total public expenditures is the highest in Yemen with the lowest income per capita reaching 17.7%, as compared to 11.9% in Egypt and 11.7% in Bahrain (Table 3). For the third direct cause of poverty, the support burden ratio, it is 3 in Egypt and 4 in Yemen and Bahrain13. This means that every worker in Yemen and Bahrain supports on average 4 persons, while in Egypt, every worker supports on average 3 persons. In the developed countries, the support burden ratio is 2, indicating that every worker supports on average 2 persons only, despite the high income earned in those countries. It is worth noting that the support burden ratio in Bahrain and Yemen is among the highest in the Arab countries, in which this ratio ranges between 3 and 4 with the exception of Somalia (Korayem, 1998). Among the factors affecting the support burden ratio negatively and, hence, affecting poverty indirectly is the ratio of labour force to population which is almost the same in the three countries. In 2009, the labour force share in total population was 32.7% in Bahrain with relative poverty, as compared to 33% in Egypt and 34% in Yemen, both with absolute poverty. The second factor affecting the support burden ratio is the relative share of women in the labour force (for 15 years and over); it is the lowest in Yemen (17.3%) as compared to 28% in Egypt and 32.5% in Bahrain (Table 3). The unemployment rate, which is the third factor affecting the support burden ratio, is the highest in Yemen (14.6%), as compared to 8.9% in Egypt and 3.8% in Bahrain with respect to Bahraini nationals only (Table 3). For the unemployment rate of all the residence of Bahrain, nationals and others, it was 11.6% in 2002 (20.7% for females and 8.6% for males) (UNDP, 2004). The fourth direct cause of poverty is unequal income distribution. With the estimation of poverty level and income distribution in Yemen, Egypt and Bahrain, it has been found that there is a direct relation between poverty and income per capita, but opposite relation may exist between poverty and income distribution. As has been shown above, income distribution is the worse in Bahrain with high income per capita and relative poverty, as compared to Yemen and Egypt with significantly lower income per capita and absolute poverty. This means that the low income per capita is one of the causes of poverty, but it does not have a clear definite relation with income distribution. The country may have relatively low or high income per capita and accompanied by relatively equal 13

For Egypt (Korayem, Pro- Poor Plicies, 2002); for Yemen (Korayem, 2011); and for Bahrain (Korayem, 2012).

1812


or unequal income distribution. Given the level of income per capita (low or high), the income distribution status affect the extent of poverty prevailing in the society. With low income per capita, for example, the relatively equal income distribution means that the extent of poverty prevailing in the society would be less as compared to the case if national income is more unequally distributed. The same applies to countries with high income per capita. Thus, in the three Arab countries we are studying, one expects that relative poverty would be less in Bahrain, if its national income have been more equally distributed. The same applies to Yemen and Egypt with relatively low income per capita; absolute poverty would be less, if national income would have been more equally distributed than its present status. Thus, the tax system applied, direct and indirect taxes, is supposed to be included among the poverty alleviation policies and, hence, should be assessed carefully in the three countries, which is beyond the scope of this study. One of the indirect factors affecting income distribution is the unequal distribution of assets. Although no data are available on this factor in the three countries, it is expected that the unequal distribution of assets is more prevailing in the relatively rich country, Bahrain as compared to Egypt and Yemen. The second indirect factor is the insufficient net transfers from abroad, which applies to Egypt and Yemen where some of their labour work in the Gulf countries. This factor does not apply to Bahrain, since it is among the labour receiving countries.

6. Poverty Eradication Policy Measures in the Three Countries The three countries differ in the policy measures applied to combat poverty. However, some of these policies, like giving Social Assistance (SA) to the poor by the government and the Non-Governmental Organizations (NGO’s) are applied in the three countries. Yemen, which is the poorest among the three countries, is the least interested regarding applying poverty eradication measures. Yemen has eliminated the subsidy and raised the energy prices in 1995 according to the Structural Adjustment Policies suggested by the International Monetary Fund (IMF) and the World Bank (WB). The policy measures applied in Yemen during 2000-2005 has led to an increase in the number of the poor and the deterioration in their standard of living due to the elimination of food subsidy on the consumers’ goods, the deterioration in the exchange rate of the Yemeni currency and, hence, the increase in the general price level. Moreover, the social expenditure has been reduced to about 7% of the GDP (Korayem, 2011). In Egypt, where there is absolute poverty as in Yemen but at a lower degree, the government gives monthly Social Assistance to the poor who do not have income, as well as a one-time subsidy in specific occasions which the poor may suffer from, like becoming a widow, or losing the job, etc. Also, among the most important poverty eradication policy measures applied in Egypt is the commodity subsidy policy. There are two types of commodity subsidy (Korayem, 2013). One type is the distribution of necessary food commodities (rice, sugar, oil and tea) at subsidized prices through ration cards, which are supposedly distributed to the low-income people in Egypt14. The second type is the subsidized bread (Baladi Bread) which is sold at almost 1/5 of its market price and distributed in certain outlets on the basis of first come first serve. In addition to the food subsidy, there is also energy subsidy covering electricity, gas and other type of fuels which benefit the private car owners, private transportation cars of commodities and individuals and, also, factories which are heavy users of energy. The purpose of the energy subsidy to those factories is to reduce the cost of production of the commodities produced by them to enable them 14

Unfortunately, part of these ration cards are owned by well-off people, which represent a waste that the government is trying to find a way to reduce it. 1813


to sell their product at low prices. Energy subsidy opts, also, to reduce the transportation cost of the commodities and individuals, aiming to reduce the cost of living of the low-income people by reducing market prices of the commodities and private transportation tariffs for the public. But, unfortunately, the energy subsidy is not achieving its purpose efficiently in lowering the prices of heavily used energy commodities and private transportation cost for the public to benefit the low-income people. Accordingly, waste is high. The burden of the energy subsidy is considerably high on the budget; it represents about 80% of total subsidy in the budget. Thus, the government is trying to find out alternative ways to reduce energy subsidy without affecting the cost of living of the low-income people in Egypt (Korayem, 2013). In Bahrain, which is the richest among the three countries and has relative poverty only, one finds that subsidy policies applied there are more intensive and comprehensive as compared to Yemen and Egypt. The Constitution states that the government is responsible for the welfare of the Bahraini people. Accordingly, the government subsidizes bread, rice, oil, sugar and meat to all the population. Also, education is free for the Bahraini nationals15. The Bahraini government subsidizes, also, housing for the relatively low-income nationals, either by assisting them in rent payments, or renewing their old houses, or building houses for them and enabling them to get easy loans from banks. The government subsidizes, also, electricity and water for the low-income people. The Ministry of Social Development gives monthly assistance to the needy households and individuals. Moreover, in December 2001, the King gave the low-income Bahraini nationals, under the name of “El-Makrama El-Malakeya”16, 30% of the shares of a commercial Mall called Seif-Mall to those people receiving assistance from the government to enable them, as owners of those shares, to get their profits and to get capital gains when the shares’ price increases. At the same time, those people continue to receive financial assistance from the government (Korayem, 2012).

7. Concluding Remarks Yemen, Egypt and Bahrain differ among themselves in different aspects. Bahrain has the highest income per capita among the three countries, but the lowest diversification in economic activities; next comes Egypt after quite a distance in income per capita; and lastly in order, comes Yemen. The three countries differ, also, with respect to the type of poverty prevailing and the policy measures applied to combat it. Yemen and Egypt suffer from absolute poverty, while Bahrain suffers from relative poverty. Estimating poverty, it has been found that the low-income (expenditure) households represent 50% of the population in Yemen and Bahrain and 40% of the population in Egypt; the middle-income (expenditure) group represents 30% of the population in Yemen and Bahrain and 40% in Egypt; and the high-income (expenditure) group represents 20% of households in the three countries. The Income Inequality Index (III), which ranged between zero (for full income distribution equality), and 1 (for maximum inequality) is 0.288 in Yemen, 0.232 in Egypt and 0.306 in Bahrain, indicating that the income distribution is the worse in the relatively rich country, Bahrain, and the best in Egypt, with Yemen falling in the middle. Investigating the direct and indirect causes of poverty in the three countries, it has been found that they are significantly different. Also, the policy measures applied to combat poverty differ in the three countries with very few exceptions.

15 Free education in public schools has been also adopted in Egypt after the 1952 revolution. However, over time the quality of education in these schools is deteriorating. 16 This means in the Arabic language “Kingdom Generosity”.

1814

Poverty Assessment in Low-, Medium- and High-Income Arab Countries: Yemen, Egypt and Bahrain References: Abdel-Khalek Gouda (1987). “Estimation of the poverty line of the Bahraini families from the households expenditure survey 1984/84”, UNESCWA, July (in Arabic). Alhagry Sherifa and Gaafar Alsayegh (2003). “Assessment of the cash social assistance program as a mean to combat poverty and attaining social security in the Kingdom of Bahrain”, Center of Bahrain for Studies and Research, Kingdom of Bahrain, May (in Arabic). Central Agency for Public Mobilization and Statistics (CAPMAS) (2009). “Household budget and expenditure survey 2008/2009”, August (in Arabic). El-Laithy Heba (1995). Literature Review on Poverty in the Arab Region, United Nations, Department for Development Support and Management Services (UNDDSMS), December. Elzayani Efnan R. et al. (2006). “The financial recommendation draft of the program, strategy, and budget of the economic empowerment of the Bahraini women”, the Supreme Council for Women, Kingdom of Bahrain, July (in Arabic). Gunaid Abdalla, Naguib Abdel-Gelil Mohamed and Mohamed Sallam Aly (2002). “‘Qat and Health’ in Ministry of Planning and Ministry of Agriculture”, Proceedings of a National Conference on Qat, April 6-7 (in Arabic). International Labour Organization (ILO) (2007). Equality at Work: Tackling the Challenges, Geneva. Korayem Karima (1995/96). “Structural adjustment, stabilization policies and the poor in Egypt”, Cairo Papers in Social Sciences, the American University in Cairo (AUC), Vol. 18, Monograph 4, Winter. Korayem Karima (1996). “Macro-economic policies and the poor in the Arab Region”, United Nations Development Program (UNDP), Preventing and Eradicating Poverty, Report of the Experts’ Meeting on Poverty Alleviation in the Arab States, February. Korayem Karima (1998). “Causes of poverty in the Arab countries: An economic perspective”, in: Karima Korayem & Maria Petmesidou (Eds.), Poverty and Social Exclusion in the Mediterranean Area, Comparative Research Programme on Poverty (CROP), Bergen, Norway. Korayem Karima (2000). “The impact of food subsidy policy on low-income people and the poor in Egypt”, Cairo Paper in Social Sciences, the American University in Cairo, Vol. 23, No. 1, Spring. Korayem Karima (2002). “Pro-poor policies in Egypt: Identification and assessment”, International Journal of Political Economy, Vol. 32, No. 2, Summer. Korayem Karima (2010). “Food subsidy and social assistance programme in Egypt: Assessment and policy options”, a Report prepared for the Ministry of Social Solidarity and World Food Programme in collaboration with Cairo Demographic Center, October 14. Korayem Karima (2011). “Poverty in Yemen: Evolution, causes and poverty reduction policy”, the Middle East Economic Association (MEEA), Economic Development in the MENA Countries: Contemporary Issues, Nova Science Publishers Inc, NY. Korayem Karima (2012). “Poverty of Bahraini females: Identification and assessment”, Arab Economic Journal, the Arab Society for Economic Research, Winter-Spring, pp. 57-58. Korayem Karima (2013). “Food subsidy and the social assistance program in Egypt: Targeting and efficiency assessment”, Topics in Middle Eastern and North African Economies, Vol. 15, No. 1, May. League of Arab States, Arab Fund for Economic Development, Arab Monetary Fund & Arab Organization for Arab Oil Exporting Country the Arab Economic Report for 2010 & 2011 (in Arabic). Ministry of Social Development (2006). “Path toward achieving the millennium development goals in the Kingdom of Bahrain: Draft report”, Kingdom of Bahrain (in Arabic). Morgan James (1962). “The anatomy of income distribution”, Review of Economics and Statistics, Vol. 44, No. 3, August. Mussaiker Abdel-Rahman A. (2003). “Food basket and poverty line in the Bahraini society”, Kingdom of Bahrain, February (in Arabic). Republic of Yemen (2002). “Poverty reduction strategy paper (PRSP) 2003-2005”, Sanaa, May 31. Sub-Regional Resource Facility for Arab States (UNSURF-AS) and the United Nations Development Programme (UNDP) (no date) Poverty, Growth, Employment, and Income Distribution in Yemen, 1998-2006. Supreme Council for Women (2007). “The national plan for the strategy implementation to empower the Bahraini Women”, Kingdom of Bahrain (in Arabic). United Nations Development Program (UNDP) (2004). “Living conditions in the Kingdom of Bahrain at the outset of the third millennium”, Kingdom of Bahrain, Manama, October (in Arabic). United Nations Development Program (UNDP) (no date) “Performance and challenges of human development in the State of

1815

Poverty Assessment in Low-, Medium- and High-Income Arab Countries: Yemen, Egypt and Bahrain Bahrain”, Human Development Report for the State of Bahrain (in Arabic). United Nations Population Fund (UNFPA) (2006). Capacity Building Workshop for Women from Conflict and Post-Conflict Situations, Tunisia, Hammamet, November. United Nations (2003). “Millennium development goals: Kingdom of Bahrain”, Manama, October. World Bank (2001). “Yemen country assistance evaluation”, January 29. World Bank (2002b). Republic of Yemen: Poverty Update (Vols. I and II), December 11. World Bank (2004). World Development Report; Making Services Work for the Poor People. World Bank (2007). World Development Indicators, USA. Available online at: http://data.worldbank.org/indicator/TT.PRI.MRCH.XD.WD. Available online at: http://data.worldbank.org/country/egypt-arab-republic. Available online at: http://go.worldbank org/ASIA.

1816


Matrices of Regional Redistribution of Argentinean Public Revenue and Its Political Appropriation Roque Ruarte Bazán (Universidad Nacional de Córdoba; Universidad Nacional de Chilecito, Argentina)

Abstract: The aim of this paper is to show the matrices that allow to explicitly read cross fiscal transfers that occur among the 24 provinces as well as among the 24 national districts, which derive of the application of the present system of collection and distribution of public funds administered by the central government. In the analysis, the collections made by the provinces were incorporated. Here, these matrices are being related to the partial appropriation of those funds by certain regional political elite in the current democratic period. Key words: fiscal federalism; regions; political elites; Argentina JEL codes: H71, H73, H77

1. Introduction The initial objective of the paper is to explicitly read matrices that enable cross fiscal transfers that occur among the 24 provinces as well as among the 24 national district counterparts, which result in the current application of public fund distribution system operated by the central government and incorporate collection analysis conducted by the provinces. Along with that, we entered the question of ownership of part of these funds to support regional political elite in the current democratic period. In this opportunity we will discuss particularly the financial relationships among the provinces that arise from the distribution of secondary and related sharing, confronting them with the principles of federalism. But the crucial thing is to determine that jurisdictional residents of certain provinces are contributing to the maintenance of provincial treasuries of others in a much higher proportion than the inhabitants of the former. This would be, in itself, an act of domestic geopolitics that determines the institutional delay observable in Argentina and that would transcend even the non-compliance with federal principles, It is possible to access official data related to these provincial interrelationships consistently, but today it is not feasible to obtain official information that allows us to estimate regional fiscal balances relating to the central government and social security subsystem, as we did in previous works. However, we will estimate those referred to the National Administration together, albeit suboptimal. To display the cross-transfer game in particular that of the residents of each of the provinces and the national districts, we developed two matrices. Now we prefer to do it with the updated data available, which are those of 2011. Roque Ruarte Bazán, Ph.D., Professor, Universidad Nacional de Córdoba; Universidad Nacional de Chilecito; research areas/interests: regional fiscal balances, fiscal federalism, interdisciplinary studies. E-mail: [email protected]. 1817

Matrices of Regional Redistribution of Argentinean Public Revenue and Its Political Appropriation

In line with what we have argued in a recent article, we read that the resulting net fiscal flows, now exposed in matrix mode, are not spontaneous but obey a strategy to appropriate a share of central government revenues, developed by a regional elite, established mainly in the north and south of the central strip. This strategy also takes into account the income and expenditure of the provinces themselves located in this region, with actions to preserve or increase the former and to decrease the latter. In what follows, we first give a conceptual frame of the federal jurisdictions and principles, exclusive provincial taxes and charges or those shared by national and provincial jurisdictions (“concurrent” taxes); secondly, we address the construction and analysis of the above-mentioned matrices, methodology and information used, based on official data, in 2011, including provincial revenues in the first array. Below we provide an economic and political interpretation of the recent history of a strategy of appropriation of public revenues and ultimately the overall conclusions.

2. Jurisdictions, Federal Principles and Taxation Jurisdiction means the authority to govern and enforce laws. Federalism: State system where a central power coexists with regional powers. These are often referred to as federated states, to distinguish them from national states; each level has its own jurisdiction to exercise its functions assigned by the Constitution. The people, the land and the government are the state elements in their mutual relationship, as Hermann Heller (p. 256) conceptualized. So, it is up to the provinces that their governments exercise their authority over residents within its geographic boundaries; taxpayers in tax matters. The distribution of powers between the two types of jurisdiction is established in the Constitution. In USA what it is not explicitly granted to the central government is in the States of the Union, similar to the case of Argentina, with respect to the provinces, while the opposite occurs in Canada. There is no constitutional treatment in terms of transfer of sovereignty from one federated state to another, at any constitution. Constitutional provisions have their basis in the federal principles developed in a doctrine by several authors, from the creation of federal states to the recent past, under whose rationale should be interpreted. Héctor López Bofill (pp. 245-246) has stated such principles according to the text of the 1994 Argentinean Constitution, such as: federative unit, federal (national) sovereignty, federal (national) constitutional supremacy, provincial autonomy, competence reserves, equality among provinces, indestructibility of states, provincial territorial integrity, participation of the provinces. Less numerous is the list of other authors, but this would come to reflect that some principles are extended to contain others. So Rojo Salgado, Argimiro (pp. 56-62), citing G. Héraud, condenses these principles in only four: autonomy, participation, cooperation and subsidiarity. Consistent with the purpose of this study, we selected the following principles (1) Autonomy: ability to manage resources and issue legal standards independently, so that each state takes care of its own affairs. (2) Competence reserves: they presuppose the existence of exclusive areas of competence, for which it must establish safeguards that prevent an entity from intruding on the powers of another (3) Equality of the provinces: it prevents exceptions or privileges that exclude some of the provinces in favor of others. These are complemented by (4) subsidiarity: it means replacement and supplementarity, refers to the ideas of help and support, from the top level toward the lower, without canceling it. We may add, with regard to the competence reserve, the distinction between exclusive and concurrent

1818


competences. López Bofill (p. 251), citing C. M. Bidegain, mentioned directives to troubleshoot demarcation of competences: “(a) If the issue is national by area and dimension, competence will be national; (b) The affairs to be distributed among the national government and the provinces are not a closed list, but open and growing; (c) The national competence may arise due to the inability of the provinces to address problems that necessarily have to be solved.” Now it is appropriate to connect the previous doctrinal issues with the approach taken by the Public Finance in the issue of fiscal federalism. As such, the federal government model would represent a solution to the proposition of optimal allocation of functions that require the benefit principle, according to which: every service should be determined and paid within the confines of the jurisdiction in which the benefits occur. This requires the use of national taxes in the financing of national services and funding for local services is made through taxes whose burden is borne within the jurisdiction where such services are provided, as Musgrave and Musgrave (pp. 557-591) have taught. Therefore, these authors propose the criteria for determining the functions of (1) allocation and (2) distribution, in a federal system. As to the first, for national goods, the theory of fiscal federalism requires a central provision, while goods whose benefits are limited to regional level, should be provided by lower-level jurisdictions. And the role of getting equitable distribution of income among individuals must be exercised by the central (national) government. Similarly, the national government should take care of the issue of equity among provinces if their residents differ in their per capita income and, therefore, in their fiscal capacity. This equity would be achieved through policies of subsidies. It should be borne in mind, for the purposes of this paper, what the fiscal doctrine proposes: the help must go from the national to the sub-national jurisdictions, not among the latter. In Argentina the Constitution (Sections 4º and 75º) provides concurrent tax powers among the national government and the provinces in indirect taxes. These taxes may be levied by the provincial and national governments, only in emergency situations, for a limited time and concurrently. The most important part of the Argentine tax system is the subsystem called “regime of coparticipation” or simply “coparticipation” now governed by Act No. 23,548 of 1988. This, like its predecessors, has been wrongly called “Agreement Act”, since there is no prior agreement, but operates as a contract of adhesion, imposed by the national government. The revenue sharing system in force in Argentina, or “coparticipation” is a mechanism by which the provinces delegated to the Nation the collection of concurrent taxes and the subsequent distribution of funds among the provincial treasuries. It forms a collection “coparticipable mass” together. Primary distribution is called the division of revenue among the national jurisdiction and the provinces as a whole. Redistribution among provinces is called secondary distribution1. This redistribution is done taking into account the social and economic situation of each province, according to preset ratios. Currently the original mechanism of the Agreement Act has spread to other shared taxes. 1

Percentual distribution: Buenos Aires: 21.3, Ciudad Autónoma de Buenos Aires: 1.9, Catamarca: 2.6, Córdoba: 4.7, Corrientes: 1.8, Chaco: 8.4, Chubut: 3.6, Entre Ríos: 4.7, Formosa 3.4, Jujuy: 2.8, La Pampa 1.9, La Rioja: 2.0, Mendoza: 4.1, Misiones 3.3, Neuquén: 1.8, Río Negro: 2.5, Salta: 3.8, San Juan: 3.2, San Luis: 2.2, Santa Cruz: 1.7, Santa Fe: 8.7, Santiago del Estero 3.9; Tucumán: 4.6, Tierra del Fuego: 1.3. 1819


The constitutionality of the original “coparticipation” has been discussed. Opponents to subsystem argued that the provinces transferred inalienable powers to the Nation. Conversely, it was argued that provinces have delegated only the authority to collect and administer taxes. The latter, for us, is not what happens. Some provinces are giving up some of their own measurable powers by the transfers of funds to the treasures of other provinces, as we show below. It has been said that these discussions have been settled after the constitutional reform of 1994, due to the insertion of the coparticipation in the new text. For us it is not thus, because it has opened a legal conflict between constitutional norms: the “Agreement Act” scheme confronts to the existence of a federal system of government established in Section 1 of the Argentinean Constitution2, as we explain above and justify below indicating monetary amounts at stake.

3. The Regional Fiscal Balances (RFBs) and Matrices 3.1 Concept The Regional Fiscal Balances scheme (RFBs) is an instrument of economic information that shows an estimate of net fiscal flows among regions occurring through mechanisms operating a central government. These flows are calculated as the difference between inputs and outputs of fiscal funds for a specified period, in each of the parts of the territory. This allows us to distinguish two regions: a donor region and a recipient region of funds transferred from one to another, in the sense that its residents are those who give or receive, depending on the sign of the balances. Mathematical symbols: Inflowi = Ii; Outflowi = Oi; i = 1 to 24. RFBi = Ii – Oi; i = 1 to 24. If: RFBi < 0, dadori; RFBi > 0, recipienti. Transferred from the donor region: TDR = ∑nRFBi, n = 1 to N Received by the receiving region: RRR = ∑mRFBi, m = 1 to M; N + M = 24 TDR = RRR ± deficit/surplus of central government. As in our previous papers on regional fiscal balances (Ruarte Bazán, 2011, 2012), we tried to calculate and display the flows separated into three types: (1) Provincial (2) Central Government (3) Social Security. By adding them we can get one general. For reasons explained later, in this study we had to present (2) and (3) together, which we call the National Administration, as is usual in Argentina. Therefore, we have developed two kinds of flows calculated for 24 partitions in each, corresponding to the 23 provinces and the Autonomous City of Buenos Aires (CABA) in the first case and 24 national districts identically called, in the second kind. That is, hereinafter we call District to each geographical partition used to try National Administration’s fluxes. The collection of all involved taxes is the responsibility of the Federal Administration of Public Revenue (AFIP, acronym in Spanish). In addition, AFIP collects pension contributions. Redistribution toward the provincial treasuries is done by other central government agencies, while social benefits are basically distributed by the National Social Security Administration (ANSES). Although the pension system should be autonomous, according to the Argentinean Constitution, it is actually manipulated by the central government, both the collection of its resources (through AFIP) and the payment of benefits. Its operation has distorted the Argentinean taxation. Today 2

Section 1—The Argentina Nation adopts for its government the federal, republican, representative form, according to what this Constitution establishes. 1820


its resources come in a 55% of tax collection but, in turn, a part of its funds are used to meet the expenses of government and the needs of central treasury. In the fiscal flow of each province, the input is what it receives from the central government according to Act No. 23548—secondary distribution—and similar acts. Individual output is what is collected by the AFIP in its geography, estimated by distributing the provincial portion of the primary distribution, as explained in the methodology. In the fiscal flow of the National Administration, each output of the district is estimated similarly with the central government primary distribution more contributions; each entry is the amount spent by the Nation in districts. In place of RFBs, we can use Rates of Fiscal Geographical Return (RFGRs), which we define as the ratio of what is returned in the form of remittances from the central government regarding the proceeds from each province or district. Therefore, the role of donor or recipient of each partition depends on whether the value is greater or less than 1. Although the distinction between the two is arithmetic (quotient, instead of subtraction), RFGRs exempt us from inflation adjustments. It should be noted that in terms of fiscal policy, a central government, within its jurisdiction, uses alternately both devolution or redistribution criteria to design or evaluate net flows. According to the devolution criterion, funds should be allocated to each region according to the amount paid by taxpayers who reside there. Redistribution criterion is pursuing a regional redistribution of income and/or of the provincial fiscal capacity, by transferring funds to correct personal or regional economic asymmetries, equitably. In the case of provincial funds raised by AFIP and intended for secondary distribution among provinces, the redistributive criterion should not apply because it is not the responsibility or provincial function to correct fiscal imbalances from another province or region asymmetries, but the national government, as conceptualized before. However, for the determination of coparticipable remittances to the provinces, the redistributive criterion is used, which is a departure from the federal principles and the Theory of Fiscal Federalism. 3.2 Matrices One objective of this paper is to obtain further conclusions than usual, by reformulating the RFB scheme known. For this we thought to insert a matrix detailing what each receiver unit receives from each donor and, simultaneously, alter the sequence location of columns of the original scheme. So the matrix of the provinces was developed in order to make explicit the financial relationship of each of the donor states with each receiver. When developing the national case, we show the transfers of public funds among districts. In the provincial case also we incorporate provincial tax revenues and other transfers from the national government. Denote matrices new schemes outlined.

4. Methodology To calculate the net flows, you can use different output/ input criteria. These approaches are usually classified as: levying/payment; load/benefit; load/spending; income/spending (Barberán Ortí, 2005). We choose load for outflows. In estimating provincial and security flows there is no difference between benefit and spending criteria because transfers go directly to the owner or to the recipient in each case. We opt for criteria of expense or benefit to calculate the central government inflows which will influence the result, especially in the role of the City of Buenos Aires, where most national employees reside.

1821


4.1 Provincial RFBs We show outflows in real terms directly when possible or, if not, we estimate them. In the latter case, the amount obtained for each tribute/province, in turn depend on the criteria we choose to allocate geographic origin of primary levying, here in its provincial portion. Besides, these estimates depend on the availability of information. As for small taxpayers, we learned of the geographical origin of each collected tax through accurate information published by the AFIP. Instead, to distribute receipts from large taxpayers (these with tax residence assigned by law only in the City of Buenos Aires), it is necessary to use coefficients, as just discussed. We use the same criteria justified in Ruarte Bazán et al. (2011). Regarding remittances of secondary distribution, we take the 2011 official information of Boletín Fiscal de Argentina (2012) where it shows those concerning Act No. 23,548 (1988) and its subsequent amendments and extensions. The central government returns to the provinces that capture the same amount, ergo no sense to talk of redistributing surplus or deficit. As for the type of record used, we chose cash basis exclusively. 4.2 National Administration RFBs As for the geographical allocation of the AFIP revenue collection, we follow the same methodology explained in previous RFBs, here the part belonging to the central government but now we must include exclusive national taxes, as tariffs, besides the pension system. The geographic location information relating to the execution of expenditure of the central government and benefit of social security, is not accessible from the 2011 fiscal year, what before was explicited by the General Accounting Office. Thus we have seen the need to use an approximation: the obtained proportions for each district on the total country budget. In addition there is the limitation that central expenditure with the social security benefits come together. For this reason, we can only design a joint national RFB. We decided not to distribute surplus or deficit to balance inputs and outputs as other authors do, because we would obtain a fictitious equality between revenues and expenditures of the national administration, which is difficult to justify. We prefer using the spending approach to calculate inflows, due to criticism in Argentina when it intends to distribute salaries of public employees resident in the City of Buenos Aires among all districts as required by the use of the criterion of benefit. 4.3 Composition of Public Revenue in Argentina (no municipalities) In recent decades the composition of this income has undergone structural changes. In the 90s it was subtracted revenue at the central and provincial governments by concurrent tax diversion (about 15%) to the social security system. In the 90s and 2000s, other drawdowns with different purpose and destination (Provinces, social security, Central Government) were set for levying each concurrent important tax. The resultant residue is distributed according to effective percentages of primary distribution: 40% central government, 60% Provinces (including CABA). The transfers toward the provinces—some of which originated in new special taxes shared—are about 25% of its original primary distribution portion, both are automatic. In addition, the central government makes discretionary transfers. So the Argentinian tax system is very complex and is known as a “fiscal maze”. 1822


The following approximate percentages on average during 2000s and italics for 2011 are shown. By collecting agencies: DGRs (Provinces and CABA): 16%, 15%. AFIP: 84%, 85%. Principal taxes, duties and contributions collected by AFIP: Concurrents: Value-added (29%, 21%), income - profits (21%, 15%), others (20%, 14%). Exclusive: Duties on exports and tariffs and others (13%, 19%). Social security: Employer and employee contributions (17%, 31%). Principal taxes and duties collected by DGRs: Gross receipts (70%, 76%), seals (8%, 9%), real estate (10%, 7%), others (12%, 8%) By destination of proceeds: Provinces: 40%, Central Government: 35%. Social Security: 25%. Or, Provinces: 40%, 36%. National Administration: 60%, 64%.

  



5. Results

Recipient region

Donor region

Table 1

Total

Province CABA Buenos Aires Santa Fe Córdoba Subtotal Chubut Neuquén San Luis Tierra del Fuego Santa Cruz La Pampa Mendoza Río Negro La Rioja Misiones Catamarca San Juan Jujuy Entre Ríos Salta Tucumán Corrientes Formosa Santiago del Estero Chaco Subtotal

Provincial RFBs. 2010 ($ million Argentine)

Acronym CAB BAS SFE CBA CHU NEU SLU TFU SCR LAP MZA RNG LRJ MIS CAT SJU JUJ ERI SAL TUC CRR FOR SES CHA

Ii 1.963 19.947 8.968 8.821 39.699 1.678 1.825 2.274 1.266 1.654 1.875 4.180 2.536 2.050 3.432 2.681 3.316 2.870 4.865 3.932 4.759 3.821 3.610 4.133 4.980 61.737 101.435

Oi 34.181 25.284 11.176 9.798 80.439 1.329 1.256 1.484 349 677 816 2.815 1.069 475 1.186 360 911 413 2.178 1.184 1.774 797 346 610 967 20.996 101.435

RFBi -32.218 -5.337 -2.208 -977 -40.740 349 569 791 918 977 1.059 1.365 1.467 1.575 2.246 2.321 2.405 2.457 2.687 2.748 2.985 3.024 3.264 3.523 4.012 40.741 0

RFGRi 0.057 0.789 0.802 0.900 0.494 1.263 1.453 1.532 3.628 2.443 2.298 1.485 2.372 4.316 2.894 7.447 3.640 6.949 2.234 3.321 2.683 4.794 10.434 6.775 5.150 2.940

Usually the RFBs scheme is tabulated for each partition in the sequence of the arithmetic operation, first the data: minuend and subtrahend, then the result or subtraction: inflowi (Ii), outflowi (Oi), balancei (RFBi) = Ii - Oi. By example, we show below Provincial RFB 2010 in Table 1, extracted from Ruarte Bazán 2011, incorporated 1823


Rates of Fiscal Geographical Returni RFGRi = Ii/Oi, in the last column. By locating the provinces in ascending order of their RFB values, the two regions are bounded: top donor and bottom recipient, and thus can subtotal by region. We include the meaning of the acronyms used to identify geographical units in the second column. Beside Table 1 these provinces are presented on a map and colored by type of region.

Map 1

Argentina’s Regions: Provincial RFBs 2010—Donors: light blue. Receivers: pink.

5.1 Provincial Matrix The purpose of this new scheme is to interpret the RFBs not merely as residues but as what is removed or added to the amount paid by local taxpayers depending on givers or receivers and as indicators of individual relationships between givers and receivers. We obtain this different and useful perspective if, first, we change the location of the first 3 numerical columns of the previous scheme: in the first column we put values picked up by AFIP from each geographical unit (Oi), in the second column we locate amounts RFBi transferred (-) or received (+) and the third column we put the funds going to the treasuries Ii = Oi ± RFBi. Furthermore, we modify the recipient region by inserting a matrix in the strict sense, which allows us to identify what each province receives from each of the donor provinces. This fact partly determines the title of this article and the name of the next Table 2. We also extend the schema by adding new columns showing: the collections of the General Directorate of Provincial Revenue (DGRs); two columns for percentages indicators, then the subtotal of provincial revenues, other column of index percentage and finally a column which records other transfers from the central government (%). The design of the table leads us to locate the Ciudad Autónoma de Buenos Aires (CAB), Buenos Aires (BAS), 1824


Santa Fe (SFE) and Córdoba (CBA) at the top, forming the donor region occupying a central zone of the country. To its north, south and west is the receiving region that includes the other 20 provinces. Here and in other comments below it will be appreciated that CAB has a singular behavior, by its historical and institutional situation3. A brief economic characterization of the two regions can be found in the Annex. Now it suffices to say that it resides in the first 62% of the population and concentrated 73% of the GDP. The proceeds for the payment of the provincial part of the primary distribution is displayed in column (1) which indicates that the donor region provides 79% ($106 022 mill.) of the total ($134 132 mill.) and thus its contributions quadruples the receiving region’s funds ($28,111 mil.). In the second column we see that the funds deducted and transferred from the donor region total $53,775 million, for whose formation the CAB residents contribute 76% ($40,792 million.). It is clear that all recipient provinces received input beyond what is collected locally by the AFIP ($53.774 million against $28.111 million), i.e., from outside they get twice the amount contributed by residents. It also highlights the fact that the contribution of CAB residents exceed local input, in most of these jurisdictions (13 out of 20): TFU, RNG, LRJ, SJU, MIS, CAT, JUJ (467%), SAL, CRR, FOR (438%), TUC, SES (402%), CHA (300%). In others: SCR, LAP and ERI, CAB’s contributions resemble the local contribution. Funds from other givers are generally lower than local revenue, except with respect to BAS, CAT (95%), JUJ (93%), FOR (88%) and LRJ (55%). However, for some recipient provinces the input from top provinces together is less than the local contributions, see column (4) (2)/(1): Chubut (CHU): 30%, Neuquén (NEU): 44%, Mendoza (MZA) 54% and San Luis (SLU) 69%. While some are at the opposite extreme: Catamarca (CAT): 630%, Jujuy (JUJ): 617%, Formosa (FOR): 578%, Santiago del Estero (SES): 531%. The reading of column (3) highlights the fact that although on the donors lie the highest tax effort as seen, it reaches them less ($52,247 thousand) to that obtained by the recipient ($81,885 thousand): only 39% of the total raised ($134,132 million). The percentages in column (4), confirm this fact in detail: the first region’s residents together give 51%, while the receivers increase their perception 191%. Among these the particular situation of CAB (94%) is re-emphasized. Among the receivers there are disparities, from Chubut (CHU) with 30% to CAT and JUJ exceeding 600%. If provincial levyings (5) placed in scene versus the local contribution to the AFIP, we read column (6) that the location of NEU, CHU and MZA remains below 100%; while the opposite happens for FOR (1,133%), La Rioja (LRJ) followed by (953%) and reiterated SES (694%), with CAT and JUJ (646%). If relatively we measure the provincial revenue compared to total provincial revenue (7), we could visualize the degree of fiscal autonomy of each subnational jurisdiction and/or the fiscal effort that can be done to their residents. We do this in (8) and naturally we found the special case of CAB (89%), highlighting the place of BAS (57%). In the middle are placed: NEU (45%), CBA (37%) and SFE (36%) and CHU (36%). By contrast, there are 3

The Ciudad Autónoma de Buenos Aires was established in the 1994 Constitution and took its first government in 1996. Its jurisdiction is exercised within the limits of a city and has provincial and municipal roles. It collects revenues of provincial and municipal nature and has expenses of a province (not completely) and of a municipality. It was not included in the secondary distribution of the Act 23548 (1988) until the 2000s and only with an initial 1.4%. Still it is argued if it can be considered a province, but derives its resources in a high percentage of its residents and mostly provides services proper to a province. Thus, it is justified for its singular fiscal behavior. 1825


provincial treasures that are financed by only one digit of their local revenues, in the case of FOR (7%) and LRJ (8%), others with figures close to 10% as SES (11%), CAT (12%), JUJ (12%), and CHA (14%). This situation does not change essentially, in percentage terms, if you take into account the current transfers, usually drawn directly from the central government to the provinces, as you can see the last column of Table 2.

Geographic location

(2)/(1) %

(6) (2)/(5) %

(8)

(7) Subtotal of provincial revenues

(5)/(7) %

(9)

43.392

40.792

2.600

94

21.624

189

24.224

89

88

BAS

34.153

8.137

26.016

24

35.193

23

61.209

57

50

SFE

15.350

3.445

11.905

22

6.606

52

18.511

36

33

CBA

13.126

1.400

11.725

11

7.029

20

18.754

37

35

53.775

52.247

51

70.452

76

122.699

57

52

(3) (2) Allocated Subtotal value from provincial non-local treasuries residents according to (RFBs) secondary distribution

(5) Collected Tax by provincial (2)/(1) DGRs (excludes royalties and % other non-tax revenue)

Geographic location

(1) Contributions to provincial Primary treasuries of non local distribution residents from: provincial and similar, collected by CAB BAS SFE CBA AFIP from local residents

(4)

(6)

(7)

(8)

(2)/(5)

Subtotal of provincial revenues

(5)/(7)

%

%

(9) (5) / (7) + Current Transfers (%)

RECIPIENT REGION

(5) Collected Tax by provincial DGRs (excludes royalties and other non-tax revenue)

(4)

CAB

Subtotal 106.022

Total

Provincial Matrix. 2011 ($ million Argentine) (5) / (7) + Current Transfers (%)

Donor region

Table 2

(1) (2) (3) Primary Decrease directed Allocated value distribution to supply provincial provincial and non-local treasuries similar, collected provincial according to by AFIP from treasuries secondary local residents (RFBs) distribution

CHU

1.697

390

78

33

13

514

2.210

30

1.262

41

3.472

36

35

NEU

1.675

561

112

47

19

740

2.415

44

2.004

37

4.418

45

42

SCR

956

918

183

78

32

1.211

2.166

127

938

129

3.104

30

22

SLU

1.784

933

186

79

32

1.229

3.013

69

867

142

3.881

22

22

TFU

431

935

186

79

32

1.232

1.663

286

572

215

2.236

26

25

LAP

1.103

1.038 207

88

36

1.369

2.472

124

729

188

3.202

23

22

MZA

3.612

1.469 293

124

50

1.936

5.548

54

2.912

66

8.460

34

32

RNG

1.324

1.546 308

131

53

2.038

3.362

154

1.104

185

4.466

25

24

LRJ

578

1.619 323

137

56

2.134

2.712

369

224

953

2.936

8

6

SJU

1.721

2.035 406

172

70

2.682

4.403

156

880

305

5.284

17

16

MIS

1.554

2.277 454

192

78

3.002

4.556

193

1.308

230

5.864

22

19

CAT

487

2.326 464

196

80

3.067

3.554

630

475

646

4.028

12

11

JUJ

531

2.485 496

210

85

3.276

3.807

617

507

646

4.314

12

10

ERI

2.990

2.630 525

222

90

3.467

6.457

116

1.955

177

8.413

23

21

SAL

1.494

2.829 564

239

97

3.730

5.224

250

1.397

267

6.621

21

19

CRR

1.068

3.041 607

257

104

4.009

5.078

375

797

503

5.875

14

13

FOR

707

3.102 619

262

106

4.089

4.796

578

361

1133

5.156

7

7

TUC

2.193

3.135 625

265

108

4.132

6.325

188

2.300

180

8.626

27

24

SES

870

3.503 699

296

120

4.618

5.488

531

665

694

6.153

11

10

CHA

1.337

4.019 802

339

138

5.299

6.635

396

1.120

473

7.756

14

13

81.885

191

22.378

240

104.264

21

20

Sub 28.111 total 134.132

40792 8.137 3.445 1.400 53.774

134.132

226.963

Compilation, according to official figures. AFIP, Dirección Nacional de Coordinación Fiscal con las Provincias (DNCFP) and other sources.

5.2 National Administration Matrix. Similarly to the above matrix, we construct Table 3 which central government and pension system flows.

1826


RECIPIENT REGION

National Administration Matrix 2011 ($ million Argentine)

22.206

BAS

110.633

18.962

17

35,7

91.671

CBA

35.409

13.278

37

25,0

22.130

CHU

7.240

572

8

1,1

6.668

NEU

5.900

108

2

0,2

5.792

Total

201.543

53.074

26

100,0

148.468

Geographic location SFE

(3) Difference which is applied in expenses of Central Government and payment of previsional benefits, local.

SLU

(1) (RFBs) Primary distribution and Contributions to expenses from non-local residents from: exclusive taxes of Central Government, previsional Subtotal contributions, collected SFE BAS CBA CHU NEU by AFIP from local $ % residents. 4.983 75 70 49 2 0 196 4

48

5.227

MZA

12.819

156

147

103

4

1

411

3

99

13.329

TFU

2.543

373

350

245

10

2

980

39

237

3.759

Geographic location

Donor region

Table 3

(1) (2) Primary distribution and Decrease directed to cover expenditures in other districts. exclusive taxes of Central (RFBs) Government, previsional contributions, collected (2) / (1) $ % by AFIP from local % residents. 42.361 20.155 48 38,0

(3) Sum which is applied in expenses of Central Government and Deficit payment of previsional benefits, local.

SJU

5.850

507

477

334

14

3

1335

23

322

7.507

CAT

3.581

573

540

377

16

3

1510

42

365

5.456

ERI

8.210

599

564

395

17

3

1578

19

381

10.170

SCR

4.129

607

571

399

17

3

1597

39

385

6.112

LAP

3.014

833

783

549

23

4

2192

73

529

5.735

RNG

5.164

872

821

574

25

5

2297

44

554

8.015

MIS

4.830

892

839

587

25

5

2348

49

565

7.744

SAL

5.975

1058

996

697

30

6

2787

47

673

9.435

LRJ

2.206

1090

1025

718

31

6

2870

130

692

5.768

JUJ

2.823

1197

1126

789

34

6

3152

112

761

6.737

TUC

7.939

1241

1167

817

35

6

3266

41

789

11.994

FOR

1.660

1292

1216

851

37

7

3404

205

821

5.885

CHA

3.771

1490

1401

981

43

8

3923

104

947

8.641

CRR

3.168

1517

1428

999

43

8

3995

126

964

8.126

SES

2.918

1625

1530

1071

46

9

4281

147

1033

8.232

CAB

120.208

4159

3913

2740

118

22

10952

9

2643

133.803

20.155

18.962

13.278

572

108

53074

26

12808

271.674

356

256

Sub Totals 205.792

Deficit Total

407.335

4.733

4.256

3.254

12808 420.142

Compilation, according to official figures. AFIP, Oficina Nacional de Presupuesto (ONP) and other specific sources

This array allows us to see what portion of the funds provided by residents of the national giver districts are transferred towards receivers: $53,074 to $201,543 (million): 26%. This deviation is very relevant to SFE (48%) and CBA (37%) and significant for BAS (17%). The relative weight of the total of the contribution is primarily supported by SFE (38.0%), BAS (35.7%) and CBA (25.0), totaling 98.7% of the transfer. Recipients receive a bonus of 26% (53.074/205.792) on what they contribute and 32% including undistributed deficit (65.882/205.7929). Formosa (FOR) excels which a triple of what it contributes (205%), and

1827


SES (147%), LRJ (130%), CRR (126%), JUJ (112%) and CHA (104%), more than double these percentages without taking deficit into account, which would rise by a third these percentages. CAB special situation excels because, on the one hand, it is located among those that have one of the lowest percentages of additional perception (9%) and on the other, it has a high relative share of around 20% (10.952/53.074). 5.3 Federal Solidarity Fund (FFS) and the Royalties Table 4

Provinces

Provincial Distribution of Duty Collections of Exports of Soybean and Derivates of Oil and Mining Royalties 2011 ($ million Argentine) Regional Fiscal Balance of Federal Solidarity Fund Duties by geographical Provinces origin of soy 30% for Remittances Balances exports this Fund to provinces

Santa Fe

13543

4063

626

3437

SFE

Córdoba

5367

1610

622

988

CBA

Sub total

18910

5673

1248

4425

Tierra del Fuego

0

7029

(5) (4) (1) / (2) Residues (1) – (2)

%

0 0

0

0

86

-86

301

572

-271

53

90

27

132

-105

LPA

204

729

-525

28

Chubut

0

0

111

-111

CHU

2059

1262

797

163

Santa Cruz

0

0

111

-111

SCZ

983

938

45

105

Neuquén

0

0

122

-122

NEU

2341

2004

337

117

San Luis

60

18

160

-142

SLU

0

0

0

145

-145

LRJ

0

224

-224

467

140

289

-150

SES

2

665

-663

0

0

173

-173

CAB

0

0

0

177

-177

RNG

590

1104

-514

53

Entre Ríos

533

160

342

-182

ERI

311

1955

-1644

16

Salta

263

79

268

-190

SAL

276

1397

-1121

20

Catamarca

0

0

193

-193

204

475

-271

32

Jujuy

0

0

199

-199

JUJ

3

507

-504

1

Misiones

0

0

231

-231

MIS

47

1308

-1261

4

San Juan

0

0

237

-237

198

880

-682

22

Formosa

0

0

255

-255

FOR

21

361

-340

4

Corrientes

0

0

260

-260

CRR

75

797

-722

9

243

73

349

-276

CHA

0

0

0

292

-292

MZA

969

La Pampa

La Rioja Santiago del Estero Ciudad de Buenos Aires Río Negro

Chaco Mendoza Tucumán

TFU

Collection of provincial taxes (1) (2) Collection of (3) Provinces that royalties Other charge provinces royalties 0 6606

CAT

SJU

867

21624

1120 2912

0

0

0 -1943

110

33

333

-301

TUC

0

2300

Buenos Aires

3493

1048

1538

-490

BAS

0

35193

0

Sub total

5260

1578

6003

-4428

74739

9894

Sub total

8584

18090

33

0 47

Total 24170 7251 7251 -3 Total 8584 92829 9 Compilation, according to official figures. Instituto Nacional de Estadísticas y Censos (INdeC), DNCFP, Subsecretaria de Relaciones con las Provincias. Ministerio de Economía de la Nación and other sources.

These two tax issues could not be incorporated in the previous matrices but they are interesting for the purpose of this paper. Both have to do with the exploitation of natural resources. In 2009 the so-called Federal Solidarity Fund was created which comprised 30% of the amounts received by way of export duty on soybean and derivates and aims to strengthen funding for infrastructure in the provinces 1828


which is distributed under the same percentages of secondary distribution of Act No. 23,548. Before its existence the redistribution of these export duties was under the jurisdiction of the central government and thus the inherent fiscal flows in its entirety would fit into the national matrix and so continues in 70% but it is possible now to make a special regional fiscal balance for the 30% including the provinces, as in Table 4. Moreover, royalties are important to provincial taxation as seen in the columns attached to the same Table. It can be interpreted that this Table complements and completes the Provincial Matrix. Royalties represent, as a whole, about half of the revenue from taxes in the provinces that charge them and for some more: CHU, SCZ, NEU, RNG, TFU and they are important for MZA, CAT and SJU. This should increase the fiscal autonomy of those provinces to a greater degree than the one shown in the Provincial Matrix. As it is read, both are redistribution mechanisms that show, together, the same recipient provinces of previous matrices. In one case the sums collected are shared among all provinces, not in the other.

6. The Regionalized Appropriation of Public Revenues Table 5 Rates of Fiscal Geographical Return Criterion of Load/Spending Donor region: shaded light blue; Intermediate: green; Recipient: pink 2006-2010

2005-2009

2011

General

General1

Provincial

Central Government

Social Security

General

CAB

0,06

SFE

0.39

TFU

0.17

SFE

BAS

0.78

CBA

0.55

NEU

0.25

CBA

SFE

0.89

CHU

0.66

CHU

0.28

CHU

CBA

1

BAS

0.68

SCR

0.33

NEU

0.77

CHU

1.16

NEU

0.69

SLU

0.43

BAS

0.81

BAS

0.78

CHU

0.99

NEU

1.39

SLU

0.74

SFE

0.49

CAB

0.9

CAB

0.83

NEU

1.08

MZA

1.58

CAT

0.84

CBA

0.56

SLU

0.97

SLU

0.99

MZA

1.15

SLU

1.77

MZA

0.94

MZA

0.58

MZA

1.02

MZA

1.03

SLU

1.22

LAP

2.37

ERI

1.06

RNG

0.63

LAP

1.4

LAP

1.34

ERI

1.48

SCR

2.37

TFU

1.15

LAP

0.64

RNG

1.43

RNG

1.37

SJU

1.57

RNG

2.51

LAP

1.18

MIS

0.66

SCR

1.45

SCR

1.39

SCR

1.63

ERI

2.6

RNG

1.24

ERI

0.74

ERI

1.46

ERI

1.51

RNG

1.75

MIS

2.98

SJU

1.51

SJU

0.8

TFU

1.62

MIS

1.66

TUC

1.81

TUC

3.04

CAB

1.53

SAL

0.89

MIS

1.72

TFU

1.81

TFU

1.82

TFU

3.82

SAL

1.53

BAS

0.96

TUC

1.86

TUC

1.87

MIS

1.93

SAL

4.02

MIS

1.54

TUC

0.98

SAL

2.1

SAL

2.14

SAL

1.96

0.59

SFE

0.63

SFE

0.7

CHU

0.65

CBA

0.70

0.71

CBA

0.72

BAS

0.81

NEU

0.76

CAB

0.83

0.59

SJU

4.39

TUC

1.66

CAB

1.04

SJU

2.16

SJU

2.25

LAP

1.99

LRJ

4.87

SCR

1.68

JUJ

1.04

LRJ

2.75

CHA

2.79

CAT

2.21

CHA

5.31

CRR

2.24

LRJ

1.07

CAT

2.86

LRJ

2.81

CHA

2.99

CRR

5.47

CHA

2.45

CRR

1.08

CRR

2.9

CRR

2.84

LRJ

3.05

CAT

7.35

LRJ

2.53

CAT

1.27

CHA

2.99

CAT

3.08

CRR

3.12

JUJ

7.46

JUJ

2.61

CHA

1.28

JUJ

3.5

JUJ

3.59

JUJ

3.14

SES

8.37

SES

3.55

SES

1.44

SES

4.35

SES

4.23

SES

3.62

FOR 12.37 FOR 5.68 FOR 1.58 FOR 6.52 FOR 6.17 FOR 4.51 1 Note: Calculated using values of Provincial and National Administration Matrices. Based on data from AFIP, Contabilidad General de la Nación (CGN), DNCFP and other. 1829


In previous publications (Ruarte Bazán R. et al., 2011, 2012) we have discussed the mechanisms of appropriation of Argentinean public revenues and for this we use the tool we proposed and defined above as Rates of Fiscal Geographical Return (RFGRs), among other social and political indicators. Its use facilitates historical analysis because it allows us to frame periods longer than a year, to avoid the correction for inflation that would imply the use of Regional Fiscal Balances (RFBs), as said before. The next Table 5 shows the 2006-2010 period separated in three types of RFGR and the General RFGR as sums of the 3 partial RFGRs, and the 2005-2009 period for General RFGR used in the following econometric demonstrations which highlights that the donor region now includes San Luis, a place without major natural resources. Also a final column for 2011 General RFGR, calculated from the previous matrices is added. 6.1 The RFGRs as a Measure of Regional Appropriation of Fiscal Flows Operated by the Central Government We can visualize the RFGR as a measure of regional appropriation of these fiscal flows. This table, and also the matrices, show that the Central Government makes a distribution of own resources in their districts, with similar priorities to the secondary redistribution between provinces. In Ruarte Bazán (2012, p. 2) we present the ordinal correlation (Spearman) between provincial and national RFGR as  = 0.86 for 2005-2009 and  = 0.82 for 2010 and cardinal correlation R2 is 0.73 and 0.71, respectively, and higher for previous years. That is, there exists a pattern of behavior that would respond to the same strategy and would not distinguish whether it is national or provincial jurisdiction. Then, the regional belonging would be decisive in the redistribution of public revenues collected centrally. From this point of view, it would become somewhat abstract discussion of whether that redistribution favors the nation or the provinces in general, imploring the federal principles. In fact, the combination of provincial and national districts forms two resident´s regions, being the transfer of funds from one to another very pronounced. In any tax system, whether federal or unitary, it is inevitable the occurrence of an asymmetry in the geographical distribution of government revenue and therefore it is appropriate that there is a geographical redistribution to correct inherent regional economic imbalances. But in Argentina, this redistribution is very disproportionate, unjustified and permanent, which alters the federal principles without getting redistributive effects on income, as we have shown in Ruarte Bazán et al. (2011). 6.2 The TRG Is a Better Indicator of the Political and Socio-cultural Behavior In this last article we proposed to construct a regional index of alignment of politicians by origin district, considering the behavior of legislators in the National Congress, provincial governors and President of the Nation. This index, with variability between 0 and 1, allowed us to demarcate two political regions, according to whether each of 24 geographical partitions is above or below a critical value (0.66), for the period 2005-2009. Then we correlate this politic regional index with other known regional, political and economic indicators: political overrepresentation, the “pampean”4 character, metropolitan attribute and also with RFGR. From the comparison it resulted that the latter ones arose as the best indicator of regional, political alignment. Here too, the belonging to a region explains the alignment better than party membership. Indeed, for the period it was found that 3-4 governors of the recipient region were aligned but affiliated to different parties than the Peronist party in the central government while a Peronist governor of the donor region was not aligned with the elite. 4

Pampa is an Aboriginal word used in Argentina to denote an area of fertile plains conducive to agricultural development, traditionally cattle and grain and now soybeans also. Here there are the largest number of rural and urban population and greater industrial production.

1830


In order to deepen the analysis, we were interested in investigating the regionalized behavior of politicians not only in their political alignment but also to society from which they come and as functional performance. The intention was to understand in part how the common sociocultural factors associated with the analyzed regions play and the tax reasons. This is why we built another index that combined the transparency of political activities, the level of patronage or political commercialism and management quality. The idea was to see how these variables are related to the same index used as political and fiscal behavior reflected in the RFGR. We found that the RFGR is a good descriptor of political behavior expressed in the combined ratio and the same intensity as the political alignment. From what it is inferred or deduced that, regionally, the political, cultural and fiscal behaviors are conjugated in the same social behavior. 6.3 Regionalization in Provincial Budgets A similar strategy of regional appropriation can be seen when we extend our approach to the composition of resources and expenditures of provincial budgets. The donor provinces are forced to increase proportion of own collections, according the Provincial Matrix, and thus they must raise the tax burden on their residents and their socio-economic indicators are expected to get worse. On the other hand, in the recipient region, the mining and oil provinces consolidated their income by royalties recognized in the Constitution of 1994, while North and Midwest provinces were allowed to transfer their deficit pension funds, toward the national level. 6.4 Concept of Elite and Public Choice Theory The concept of elite emerged in the field of political science in late nineteenth and early twentieth centuries and was first proposed by the Italian Gaetano Mosca to refer to an organized minority that deals with the political leadership in a society with some degree of development. It is possible to talk about elites, depending on their origin and the social dimension. The members of the elites are articulated by interest and are homogenized by membership to certain groups (Imaz José L., 1973). The region of origin may be among these factors of agglutination. Among many examples, let’s take Brazil, where it is common to refer to “paulista” (Sao Paulo) and “carioca” (Rio de Janeiro) elites as dominant, according to the times. Although there is a history on the regionalization of the elites in Argentina, we were interested in the formation of that occurred since the advent of democracy in 1983, especially in the current era. In the construction of political index mentioned above, we were able to visualize that some Argentine political leaders tend to have permanent regionalized alliances beyond their party identification and the network of fiscal interests that it represents is manifested systemically in Congress and in the behavior of presidents and governors. So only in the receiving region bounded by the RFGR and Conurbation of Buenos Aires City5, especially south and west, it could be said that a political elite has emerged, in the sense that those who integrate it have an internalized behavior based on the agreement or covenant to capture a disproportionate share of government 5

The Buenos Aires Conurbation, however, belongs geographically to Buenos Aires (BAS), province or district. In Ruarte Bazán et al. (2010, pp. 18-19), we decided to simulate the Buenos Aires district split into two territories: the Courbanation and another called New Buenos Aires, because it was not possible to calculate directly the fiscal flows since the information concerning transfers and national expenditure in towns and cities, although existent, is inaccessible, all in function that there is strong evidence (preference in geographical execution of national public works, municipal budgets, political alignment of municipal mayors, etc) that fiscal and politician’s behavior in both territories is different. Thus, this simulation allowed us to reinforce the conclusion that the net fiscal flow between regions, measured by RFGR, is the best indicator of political behavior by origin. 1831


revenue. However, despite the huge funds recepted, we find that they have not improved the levels of poverty and the per capita GDP of the population of the region where the elite is located6. Here we find that the theory of public choice explains the observed behavior by politicians in question. This is because this theory holds that politicians seek to maximize the public budget according to their own interests, giving them priority over social welfare. As it is known, this theory links the economy with politics, with application of rational choice model to political phenomena. It is a general theory about how private interests act in the public activity. It denies the assumption of benevolent or honest bureaucrat, rejecting that those who act selfishly in private are disinterested when they assume the public office. 6.5 Significant Political Events in the Establishment and Consolidation of the Preeminent Regional Elite Most of these political and legal facts have impact on the budgets and result directly or indirectly in a benefit to the district or region governed by politicians of the acting elite. (1) 1986. Interim Agreement between the northern governors and the national government which derived in the Act 23548 that meant the loss and transfer of 6 1/2 points percentage in secondary coparticipation by the Province of Buenos Aires toward the other provinces. This was a quart of its revenues by this origin. (2) 1988. Justicialist Party primaries. Two pairs were formed for president and vice president. One formed by a politician from the North (Carlos Menem) and another politician from the Conurbanation of Buenos Aires (Eduardo Duhalde). The other pair, originating in Buenos Aires (Antonio Cafiero) and Córdoba (José de la Sota). The triumph of the first candidates, though by a narrow margin, 54% to 46%, is representative of the emerging elite that we have proposed. The second pair won in Cordoba and City of Buenos Aires with good results in Santa Fe and Buenos Aires, while Menem-Duhalde imposed by high-margin in the North and Patagonia, except a few districts. By the political events of the ensuing decades, it could be interpreted that said electoral victory consolidated the elite. (3) 1992. Creation of Historic Repair Fund of Greater Buenos Aires. Section 40 Act 24073, introduced with the promise of President C. Menem to keep it for the entire period of E. Duhalde in the governance of Buenos Aires province. (4) April, 1992. Transfer of deficient Retirement Systems of provincial employees to the ANSES. 1992 Fiscal Agreement. Second Clause. Numeral 6. It benefited particularly the provinces of the West and the North. (5) 1994. Granting constitutional status to the Coparticipation regime (Act 23548) with clauses that prevent modification. Section 75, subsection 2 and transitional clause. (6) 1994. To enshrine the original domain of natural resources to the provinces. Constitution, Section 124, in fine, which mainly benefits the Andean provinces. (7) From 2001 to 2003. Political crisis in favor of the emerging regional elite. Because it meant the displacement of two resident presidents of the Central Region: Fernando de la Rua (City of Buenos Aires) who was elected by vote and a designated President by Congress, Adolfo Rodriguez Saa (San Luis), replaced by a President from the region of the new elite Eduardo Duhalde (south western Conurbation) also designated, who 6

In Ruarte Bazán et al. (2010), using an econometric model that takes as dependent variable the Percentage Households below the poverty line and as independent variable the Rates of Fiscal Geographical Return in each district, using Gross National Product per capita as a control variable, based on the initial 141 observations between 2002 to 2007, it was found that those regions which are net recipients of funds tend to have higher levels of poverty. 1832


promotes as his successor and later elected Nestor Kirchner (Santa Cruz), of the same acting elite region. (8) July. 2006. Enactment of the Act of hydrocarbons, No. 26.197. Called “short Act”, which transfers to the oil provinces the permits and oil concessions granted in each district. (9) 2009. Creation of the Federal Solidarity Fund, Decree 206, which allocates 30% of export duties on soybeans and soy products, to be distributed according to the percentages established by Act 23,548. This fund has an unbalanced distribution against soybean producing regions, as seen in Table 4. (10) 2012. Expropriation of YPF. Through Act No. 26,741, which secures an equity stake and share and the directory between the national government and only those provinces in whose territories oil is extracted, in circumstances that the activities of the company are spread throughout all the country in production and marketing of fuels.

7. Conclusions 7.1 From the Provincial Matrix  In general, we can say that there is a disproportionate transfer of fiscal funds toward the treasures of recipient provinces located in the North, South and West of Argentina, from residents of the other donor provinces located in a Central strip.  This transfer exceeds the levying of Federal Administration of Public Revenues-AFIP-of local residents (191%) of recipient provinces altogether and the levying that performs the Provincial Directorates of Income-DGRS-(240%) of same provinces. This latter collection is only a 20% of total provincial revenues of all origin.  In contrast, we can discern provinces geographically located at the center and the East (CAB, SFE, BAS and CBA), where the transfer to other provinces represents the 51% of the funds collected by AFIP. Their DGRs collect 52% Provincial revenues by all concepts.  It is a fact that the receiver provincial treasuries get most of their resources beyond its borders which mainly come from the CAB’s residents. The contribution of these residents exceeds the AFIP collection by secondary coparticipation in 13 provinces and if it includes DGRs it exceeds them in seven provinces: LRJ, CAT, JUJ, CRR, FOR, SES and CHA. These would be absolutely unsustainable provinces without input from the CAB residents.  The City of Buenos Aires as a legal person has a unique historic, geographic and institutional situation. It is the more autonomous jurisdiction in the sense that it gets its resources to 88% of own levying. At the opposite extreme are the provincial treasures of La Rioja and Formosa which have 6% and 7% resources of that origin, respectively.  The Province of Buenos Aires holds the second index of autonomy if it is measured by its own levying: 50% of total revenues.  When we incorporate information about Provincial distribution of duty collections of exports of soybean and derivates and of oil and mining royalties, the tendency that favors generally the same provinces is confirmed.  All this allows us to infer that the Argentine fiscal system does not respect the federal principles of autonomy, the competence reserves, equality of the provinces and subsidiarity. Nor does the theory of fiscal federalism apply, as to the criteria for allocation and distribution for both national and provincial level.  Argentina is not a confederal state which requires a greater degree of autonomy and the possibility of state segregation.

1833


Neither is it a unitary state, because the provinces retain their own tax levying power, have the original domain of natural resources and the provision of certain services such as primary education; there is a judicial system of separate functions between nation and provinces.  Ergo, it is sui generis.  Disparate behavior between the recipient provinces is observed, but in some of them the degree of financial dependence is extremely high. It could be said that these are failed provincial governments: by themselves they would not be able to provide the most basic government services. 7.2 From National Administration Matrix  The fact that residents of the Center East: SFE, BAS and CBA (no CAB) are the donors is repeated, in this case from districts. Some of these strongly contribute to the maintenance of receptors located mainly in the North, West and South: Its contribution is around 26%, standing out SFE with 48%, followed by CBA with 37% and 17% BAS. Other giver districts in the South: CHU and NEU contribute with lower percentages.  While recipients receive a bonus of 26% of what they contribute or 32% if the undistributed deficit is added. The CAB district excels since it receives a low percentage (9%) of what it contributes while it is a high percentage of the total transferred (21%) and, on the contrary, here the AFIP collects ($ A 120208 mill.), the highest percentage: 30% of the national share of the primary distribution and of the social security system ($ A 407335 mill.). It is followed by the BAS (27%) district.  There are six districts located in the North where the AFIP levying is less than the external supply: LRJ, JUJ, FOR, CHA, SES and CRR. 7.3 From the Appropriation of Public Revenue  The territorial behavior of public finances, as measured by Rates of Tax Geographical Return (RFGR), operates as a radiography of good resolution of what regionalized political behavior is in Argentina. These rates are also indicative of regional patterns of cultural behavior.  The appropriation of a portion of the revenue from the central and provincial governments, by the main political leaders of the region formed by the North, South, West and the Greater Buenos Aires, is decisive for the formation of the Argentine dominant political elite which resides in that region.  The installation process of this leadership recognizes historical reasons in the deformation of the fiscal federalism.  The Theory of Public Choice helps to explain this politician behavior.  From the 1980s to the present, this leadership develops a strategy that consolidates its development with important political and legal facts.  For a correct diagnosis of the current situation of public finances and politics in Argentina, it is necessary to understand that the exercise of public power does not pass the federal Nation - provinces axis, but a relationship of hegemony of the elite that has occupied the national government and the governments of the provinces in the recipient region, against the leaders of the donor region.  The variable regional belonging of politicians, used less frequently than their party affiliation and ideology to explain political phenomena, may be of greater importance than the latter in accordance with what we discuss here. This would have to be reevaluated by both political discipline and Argentina’s political leadership, in order to design the investigations and political goals, respectively.  Because the elite is a system that gathers interests, it will not be altered by any leak, transmutation or disappearance of some of its members. The hierarchy does not matter much, even of the President of the Nation, 

1834


who would act in its exercise, as a primus inter pares.  If there is room for the recirculation of elites, as posed by Wilfredo Pareto, it means that there may be renovations within the same elite or a replacement of current elite may occur. In Argentina, it seems that politicians in the Center Region have not matured on the connection of their interests but recent events suggest that it is not ruled out that they will begin to have a change of perspective according to both their interests and the country in general.  A new political elite that includes participation of leaders of the Center-East Region would generate an expectation of improving the quality of politics in Argentina, due to the fact that the current elite has the worst indicators of corruption, of political patronage and management quality, which we verified when we built a regional index that combined these variables. In turn, it is surprising that in the region of residence of the current hegemonic political elite only a little over a third of the people resides, one-quarter of Gross Domestic Product is generated only, and having a low level of per capita GDP and a high rate of poverty, as can be read in the following Annex. That is to say, it is surprising that the hegemonic political elite emerges from the least economically and socially powerful region. For all these reasons, Argentina faces a distortion in the distribution of fiscal resources that transcends the federal issue and harms its economic and human development. The alteration of existing interior geopolitical conditions is required in order to solve this problem. References: Boletín Fiscal (2012). Available online at: http://www.mecon.gov.ar/onp/html/boletin/4totrim11/4totrim11.pdf. Dirección Nacional de Coordinación Fiscal con las Provincias. Ministerio de Economía de la Nación, available online at: http://www2.mecon.gov.ar/hacienda/dncfp/provincial.html. Heller Hermann (1990). Teoría del Estado (2nd ed.), Fondo de Cultura Económica, Buenos Aires. Economic Commission for Latin America (ECLA). Available online at: http://www.cepal.org/cgi-bin/getprod.asp? xml=/argentina/noticias/paginas/4/10424/P10424.xml&xsl=/argentina/tpl/p18f.xsl&base=/argentina/tpl/top-bottom.xsl. Instituto Nacional de Estadísticas y Censos. Ministerio de Economía de la Nación. Available online at: http://www.censo2010.indec.gov.ar/archivos/censo2010_tomo1.pdf. Instituto Geográfico Nacional. Available online at: http://www.ign.gob.ar/NuestrasActividades/Geografia/ DatosArgentina/DivisionPolitica. Imaz José Luis (1973). Los que Mandan, EUDEBA, Buenos Aires. International Monetary Fund (2012). Available online at: http://www.imf.org/external/pubs/ft/weo/2012/01/weodata/. López Bofill Héctor (2000). “Notas sobre la experiencia federal en Argentina”, Revista de Estudios Políticos, Nº 98, Madrid, España. Rojo Salgado Argimiro (2000). “Globalización, integración mundial y federalismo”, Revista de Estudios Políticos, Nº 109, Madrid, España. Musgrave R. and Musgrave P. (1992). Hacienda Pública: Teórica y Aplicada, McGraw Hill Interamericana, México. Garat Pablo M. (2009). “El sistema de de coparticipación federal en la organización constitucional Argentina”, Instituto de Investigaciones Jurídicas. Universidad Autónoma de México. Barberán Ortí R. (2005). “La disparidad de resultados de las balanzas fiscales regionales: Análisis de sus causas y de las posibilidades de convergencia”, En: XII Encuentro de Economía Pública, Palma de Mallorca, 3 y 4 de febrero de 2005. Oficina Nacional de Presupuesto. Ministerio de Economía de la Nación. Available online at: http://www.mecon.gov.ar/onp/html/ Ruarte Bazán R. et al. (2010). “Deformaciones del federalismo fiscal argentino y corrupción política: Un análisis exploratorio”, 43ª Jornadas Internacionales de Finanzas Públicas (Edición en CD.), Universidad Nacional de Córdoba. Ruarte Bazán R. et al. (2011). “Los Saldos Fiscales Regionales argentinos como indicadores políticos y socials: La conformación de élites politico—regionales”, 44ª Jornadas Internacionales de Finanzas Públicas, Córdoba, 2011. Ruarte Bazán R. et al. (2012). “La apropiación regionalizada de la renta pública argentina y la conformación de la emergente élite política”, 45ª Jornadas Internacionales de Finanzas Públicas, Córdoba, available online at:

1835

Matrices of Regional Redistribution of Argentinean Public Revenue and Its Political Appropriation http://blogs.eco.unc.edu.ar/jifp/files/45jifp_t16.pdf. Subsecretaria de Relaciones con las Provincias. Ministerio http://www2.mecon.gov.ar/hacienda/ssrp/informacion/index.php.

de

Economía

de

la

Nación.

Available

online

at:

Appendix Table 6 Relevant Regional info Argentine Socioeconomic Type Territorial Share of Proportion of poor GDP Number of Density GDP per capita 2 3 5 U$S area GDP people 2 inhabitants Inhabitants / km U $ S/inhabitant 2 Billions % Km %

Geographical units ARGENTINA Buenos Aires

1

4

41,261.490

2,886.034

14.3

100

447.644

10.849

12.5

16,070.810

307.571

52.3

33.10

148.175

9.220

11.2

Córdoba

3,403.266

165.321

20.6

7.66

34.287

10.075

8.7

Santa Fe Main districts donors Buenos Aires City Donor Provinces Catamarca

3,285.665

133.007

24.7

7.79

34.853

10.607

9.5

22,759.741

605.899

98

48.55

217.315

9.548

10.6

2,972.596

200

14.863

24.83

111.162

37.396

7.0

25,732.337

606.099

42.4

73.38

328.477

12.765

10.2

378.321

102.602

3.7

0.64

2.848

7.529

14.6

Corrientes

1,020.910

88.199

11.6

1.27

5.697

5.581

19.7

Chaco

1,085.362

99.633

10.9

1.22

5.447

5.018

23.1

523.631

224.686

2.3

1.41

6.327

12.083

10.7

1,271.252

78.781

16.1

2.14

9.594

7.547

11.6

545.286

72.066

7.6

0.57

2.573

4.718

25.2 18.1

Chubut Entre Ríos Formosa Jujuy

692.514

53.219

13.0

0.83

3.727

5.382

La Pampa

328.050

143.440

2.3

0.86

3.867

11.786

5.7

La Rioja

343.160

89.680

3.8

0.50

2.235

6.512

15.5

Mendoza

1,788.534

148.827

12.0

4.17

18.677

10.443

10.3

Misiones

1,133.017

29.801

38.0

1.36

6.078

5.364

19.1

Neuquén

566.992

94.078

6.0

2.10

9.399

16.577

12.4

Río Negro Salta San Juan

656.863

203.013

3.2

1.46

6.524

9.932

11.7

1,249.085

155.488

8.0

1.52

6.808

5.451

23.7

700.483

89.651

7.8

1.04

4.647

6.634

14.0 10.7

San Luis

444.642

76.748

5.8

0.96

4.307

9.687

Santa Cruz

281.779

243.943

1.2

1.18

5.290

18.772

9.7

Santiago del Estero

898.938

136.351

6.6

0.82

3.666

4.078

22.7

1,489.500

22.524

66.1

1.96

8.767

5.886

16.4

Tucumán

Tierra del Fuego 130.834 127.205 1.0 0.60 2.690 20.558 14.5 Recipient Partitions 15,529.153 2,279.935 6.8 26.7 119.168 7.673 16.3 (no B. A. City) Note: 1 Figure estimated by INDEC 2011 and own calculation each territorial unit. 2 Instituto Geográfico Nacional (Argentine). Figures in the table exclude Antarctica, with its total area of 3,745,997 km2 and density 11.0 inhabitants / km2. 3 Own projection according to Economic Commission for Latin America (1997-2005), and the Ministry of Interior’s Office (2002-2007), latest official figures published. 4 Estimated by the International Monetary Fund for 2011. 5

INDEC data, Census 2010.

1836


Stochastic Asset Allocation Models Gavriel Yarmish1, Robert Fireworker2, Harry Nagel2, Joe Thurm1 (1. Department of Computer and Information Science, Brooklyn College, USA; 2. Decision Science, Tobin College of Business, St. John’s University, USA)

Abstract: This paper describes stochastic asset allocation models with a particular focus on the stochastic linear programming paradigm. After discussing the basic one-period risk/return tradeoff model, the concepts are extended to multiple period models. Mathematical solutions and their programming implementations are briefly presented. These programs can be both deterministic and stochastic. An explanation of how a stochastic linear program would be constructed for the asset allocation problem is presented with some real-world applications. These models are important for all sorts of financial companies including insurance companies, pension managers, credit unions and portfolio managers. Knowledge of these techniques will enhance the financial decision making capability of any company. Key words: Linear Programming; Asset Allocation; Stochastic; MonteCarlo JEL codes: C44

1. Overview The general financial Asset Allocation Problem addresses the selection of various securities for a diversified portfolio based on the investor’s objective. A security is an asset, defined and described in a certain legal standardized form. In general, we focus on publicly traded stocks and bonds with well-known risks and return. In fact, any asset class can be used. Investors’ objectives may differ. Older individuals facing imminent retirement usually will select a risk adverse investment portfolio. Younger people usually will select a growth-oriented portfolio since the need for cash withdrawals is well into the future. An insurance company that sells insurance on a long term basis will want to invest the initial money culled from sales to earn a substantial enough return to enable it to charge competitive prices while still maintaining confidence in its ability to meet its obligations. A company facing cash payouts due over many years may want to split their investment. Part of their money will be invested in ‘safe’ vehicles to provide for its immediate disbursements. The rest of their money will be invested in riskier investments that offer the potential for higher Gavriel Yarmish, Associate Professor, Department of Computer and Information Science, Brooklyn College; research areas/interests: optimization, linear programming, parallel/distributed processing. E-mail: [email protected]. Robert Fireworker, Professor, Tobin School of Business, St. Johns University; research areas/interests: business, databases, decision science. E-mail: [email protected]. Harry Nagel, Professor, Tobin School of Business, St. Johns University; research areas/interests: business, databases, decision science. E-mail: [email protected]. Joe Thurm, Associate Professor, Department of Computer and Information Science, Brooklyn College; research areas/interests: business information systems, databases. E-mail: [email protected]. 1837

Stochastic Asset Allocation Models

returns. The purpose of this article is to explain the stochastic linear programming in the context of asset allocation. The full power of utilizing this approach is apparent in the case of large companies that have numerous constraints and obligations that must be kept but without those constraints there is room for the allocation of reserve cash. In order to motivate this we first give precursor models that do not address the case of constraints and then build up to stochastic linear programs. Section 2 gives background on research done in stochastic programming for finance and economics. Section 3 motivates with the simple Markowitz return/risk model. Section 4 describes deterministic (non-stochastic) linear programming. In this case there are liability constraints that must be met. Assuming we can guarantee cash for the outstanding liabilities we would like to invest the remaining cash at a minimum cost. In section 5 we expand the discussion to include stochastic linear programs. In this case we include random variables in our linear programsto model realistic situations where future liability needs and not known with certainty.

2. Background Linear Programming under Uncertainty was first introduced by George Dantzig (Dantzig, 1955) and has since grown to encompass a broad range of applications. A nice survey of financial optimization application is given by Yu et al. (2003). Kouwenberg (2001) develops a scenario generation method for a Dutch pension fund, Castro (2009) applies stochastic programming to cash management and Pflug and Wozabal (2007) deals with portfolio selection when past probability distributions are not known. Another application to portfolio optimization is shown in April et al. (2002). Mortgage-backed security portfolios have been modeled by Zenios(Zenios, 1996). An extensive application of stochastic programming was modeled by the Frank Russel Company for a large insurance Company (Carino et al., 1994; Carino & Ziemba, 1998). Another application for stochastic linear programming relates to the labor input decisions of a typical rural household in an environment with risky agricultural technologies and off-farm employment opportunities (Becker, 1990). Some have attempted to use parallel computers to help solve very large stochastic programs in general (Yarmish & Van Slyke, 2009) and specifically for bond portfolio applications (Moriggia et al., 1998). Mathematical programs have been used to model uncertainty at least as far back as the 60’s and 70’s. Reservoir systems (Houck & Cohon, 1978), Planing models(Tintner & Raghavan, 1970) and other decisions necessary for multi-national firms (Fourcans & Hindelang, 1974; Salmi, 1975) have been modeled. This field has recently received renewed attention (Kall & Mayer, 2011; Miller & Ruszczyński, 2011). Another example of the application of stochastic programming is the Frank Russell Company and The Yasuda Fire and Marine Insurance Co. In this model, an optimal investment strategy was developed using 5 year horizon multistage stochastic programming. The constraints included many legal regulations that apply to insurance companies (Carino et al., 1994; Carino & Ziemba, 1998). Another example is detailed inKusy and Ziemba(1986). This model, developed for the Vancouver City Savings Credit Union was a 5 years multi-period model that included legal, financial and band-related policy considerations and their uncertainties. Mulvey (1996) developed a method of generating scenarios for the Towers Perrin Company. They apply asset-liability planning to the problem of pension management. Golub et al. (1995) developed a multi-period model for portfolio managers that maximize the expected utility of return.

1838


3. Simple Non-constrained Return/Risk Optimization 3.1 Single-period Markowitz Model Markowitz (1959) introduced one approach to tradeoffs between risk and return when constraints are not being modeled. He observed that the amount of return received on an investment is generally dependent on the degree of risk of loss that the investor is willing to assume. This assumes, of course, that knowledge of these investment opportunities is available and known to all so that everyone is able to compete equally for these investments. Here is one way of expressing Markowitz’s approach. Solve n

Maximize

rport  var(rport )

 w E(R ) i 1

n

i

i

n

 wi w j Cov( Ri , R j )

(1)

i 1 j 1

such that the weights wi, for all i’s sum to 1. Where wi = the weight of security i; that is, the fraction of the portfolio value represented by security i. These are the variables for which we must solve. Ri = the return of security i for the one period in question. Ri is a random variable. 3.2 Multi-period Markowitz Model The model just presented is a one period model which works well for an investor who will leave the money invested for the full period. Unfortunately, any current financial obligations must be met from other monies as this model does not include constraints. This may not be realistic. Extending the basic model to handle multiple periods and allowing for the withdrawal of funds is desirable. A simple extension would be to limit the Markowitz model to the period capped by the due date of the first liability. After that liability is paid, rebalance the portfolio using the due date of the next liability as its period. This would lead to the best return: risk ratio for that short time period. Portfolio rebalancing is done via buying and selling assets. Unfortunately, the cost of buying and selling in the market can quickly take away any return received on an investment. What is necessary is a model that would also encompass this cost. A second and more serious problem is that most large investors are companies that have specific obligations that must be met. Examples of such companies are banks and insurance companies. In fear of not meeting their monthly or weekly cash obligations, these companies are likely unable to accept as much risk as the model would indicate. They do not want to entertain the thought of missing payments. What is necessary is a model that matches income with liability obligations.

4. Deterministic Linear Programs: Balancing Income and Liabilities Let us assume that we are a company with cash reserves and a schedule of cash obligations that must be met. The first step is to determine a finite set of i securities that may be part of the portfolio to be constructed. Our goal is to determine the amount of each security to purchase for the portfolio. We are willing to invest as much as possible as long as we know that our cash obligations will be met. In this case we can set up our problem to either maximize expected portfolio return or to minimize the cost of the portfolio securities. In the following example 1839


we minimize the cost of acquiring the portfolio. A simplified linear programming example would be of the form:

Minimize c T x Subject to Ax  b, x  0

(2)

where c, x, A and b are vectors Minimize

cx  a x b j

j

j

Subject to

ij

j

i

i

j

(3)

x j  0 j

th

Where xi would be the number of the i security to purchase, ci is the cost of purchasing the ith security, bj is the sum dollar amount of the liability obligation for year j and aij is the cash flow that security I will give in year j. The idea of this linear program is to seek the minimum total cost sum ci and still satisfy our obligations bj. This objective will be minimized by picking securities with the highest cash flows aij. As an example, suppose we have the information in three tables: Table 1, Table 2 and Table 3. Table 1 shows the percentage return on money invested in two different securities for two different years. Using this table, JP Morgan will return 5.2% in year 1 and 5.4% in year 2 while Morgan Stanley will return 5.8% in year 1 and 5.6% in year 2. Table 1

Returns Return Year 1 5.2 5.8

Security JP Morgan Morgan Stanley

Year 2 5.4 5.6

Table 2 lists the amount of money that needs to be paid in each year. Table 2 Year 1 Year 2

Obligations

Obligations (cash that needs to be paid out) 1245 3540

Table 3 lists the actual cost for each security considered. Table 3

Costs

JP Morgan Bond Morgan Stanley Bond

We would then set up our Linear Program, Minimize c1x1  c2 x2 Subject to a11x1  a12x2  b1 a21x1  a22x2  b2 x1  0, x2  0 as follows:

1840

Cost 102.51 102.86

(4)


Minimize 102.51x 102.86x Subject to 5.2x  5.8x 1245 5.4x  5.6x  3540 x  0, x  0 1

2

1

2

2

2

1

(5)

2

and solve for x1 and x2. These decision variables would tell us, respectively, how many of JP Morgan and Morgan Stanley securities we should purchase. We are seeking to minimize the cost of buying the bonds while ensuring enough revenue is generated in each year to cover that year’s guaranteed payments. When we run this problem the answer is to purchase 632.14 shares of Morgan Stanley at 102.86 per bond for a total cost of $64,984.28 and 0 shares of JP Morgan.

5. Stochastic Linear Programs: Balancing Income and Liabilities. 5.1 Coefficients Are not Known with Certainty In the last section an assumption was made. We assumed that all the numbers given in the three tables were accurate. Indeed this is the working assumption of linear programming models in general and it is the assumption of many optimization models. In fact, this is not the case in general and this is the reason for the performance of sensitivity analysis whenever optimization models are used (Ward & Wendell, 1990). Table 1, cash-flow returns of securities in the portfolio, is certainly not known with certainty in the case of stocks–at most we are forced to rely on expectation values based on past performance. In the case of bonds it is more certain but even then the possibility of default is there. Table 2, obligations, are very often not known ahead of time. A prime example would be the obligations of an insurance company. These obligations are based on insurance claims which are certainly not known before the claims are put in. In Table 3, the cost of the securities, are the only known coefficients in this problem. The reason they are known is that they are the cost now at time 0 when we have to make the decision of purchasing out portfolio. The other two tables refer to returns and obligations that will happen at a later time. Let us analyze the case when the returns aij listed in Table 1 are not known with certainty. This would be the case with the usual stock or commodity. Note that in the case of bonds it might also be wise to include the risk of default into an otherwise deterministic security. 5.2 Stochastic Linear Program We expand our linear programming model to a stochastic version depicted below in Equation (6). Equations (6) and (7), depict a stochastic linear program that corresponds with the deterministic linear program depicted in Equations (3) and (4) respectively. In Equation (6), s stands for scenario. Minimize E [ c x ] j ,s

j ,s

j ,s

Subject to  a x  b j

ij , s

j ,s

i, s

i

(6)

x  0 j, s j ,s

1841


Minimize E [c x  c x 1,1

Subject to

1,1

2 ,1

2 ,1

c x c x 1, 2

1, 2

2, 2

2, 2

c x c x 1, 3

1, 3

2, 3

2, 3

c x c x ] 1, 4

1, 4

2, 4

2, 4

a x a x

b

a x a x

b b

11,1

21,1

1,1

1,1

12,1

22,1

2 ,1

2 ,1

1

2

a x a x 11, 2

1, 2

12, 2

2, 2

1

a x a x 21, 2

1, 2

22, 2

2, 2

b b

2

a x a x 11, 3

1, 3

12, 3

2, 3

1

a x a x 21, 3

1, 3

22, 3

2, 3

a x a x

b b

a x a x

b

11, 4

21, 4

x 0

(7)

1, 4

1, 4

12, 4

22, 4

2

2, 4

2, 4

1

2

j ,s

5.3 Scenario Generation We no longer have a table with known returns. Instead we assume that returns follow a probability distribution based on past performance. We then randomly generate, via a Monte-Carlo style method, a number of scenarios of cash flows. In the following simple example, we only generated 4 scenarios—the more scenarios generated the more realistic the problem but the larger the problem grows. aij,s is the return for security i in year j if scenario s occurs. To be conservative we must make sure that obligations b1 and b2 are met for all these scenarios. Within these constraints, we minimize the expected value of the cost over these scenarios. To make the example concrete, assume, for simplicity, that we have four random variables that historically follow a normal curve with standard deviation (σ) of 1: (1) Cash flow for JP Morgan during Year 1 is denoted by random variable R1 with a mean of 5.2 (µ=5.2, σ = 1). (2) Cash flow for JP Morgan during Year 2 is denoted by random variable R2 with a mean of 5.4 (µ = 5.4, σ = 1). (3) Cash flow for Morgan Stanley during Year 1 is denoted by random variable R3 with a mean of 5.8 (µ = 5.8, σ = 1). (4) Cash flow for Morgan Stanley during Year 2 is denoted by random variable R4 with a mean of 5.6 (µ = 5.6, σ = 1). For this example we generate four scenarios Monte-Carlo style. Let us assume that we have generated the four scenarios depicted in the following four tables: Table 4A Scenario 1 Year 1 5.3 5.4

JP Morgan Morgan Stanley Table 4B

Table 4C JP Morgan Morgan Stanley

1842

Scenario 2 Year 1 5.0 5.5

JP Morgan Morgan Stanley

Year 2 5.7 5.5

Year 2 5.6 5.6


Year 2 5.9 5.4


Table 4D


JP Morgan Morgan Stanley

Year 2 5.7 5.4

5.4Full Stochastic Program Equation (8) is the stochastic counterpart of the linear program depicted in Equation (5). Minimize E [102.51x 102.86x 1,1

Subject to

2 ,1

102.51x 102.86x 1, 2

2, 2

102.51x 102.86x 1, 3

2, 3

102.51x 102.86x ] 1, 4

2, 4

5.3x  5.4x 1,1

1245

2 ,1

5.7x  5.5x 1,1

2 ,1

 3540 1245

5.0x  5.5x 1, 2

2, 2

5.6x  5.6x 1, 2

2, 2

 3540 1245

5.3x  5.3x 1, 3

2, 3

5.9x  5.4x 1, 3

2, 3

5.2x  5.3x

 3540 1245

5.7x  5.4x

 3540

1, 4

x 0

1, 4

2, 4

2, 4

(8)

j ,s

Instead of the two constraints of Equation (5) we have eight constraints, four times as many, corresponding to the four scenarios. We must be sure that our obligations of year j are met no matter which scenario is actualized. Now, at time 0, we don’t know which scenario will be actualized therefore we must prepare four sets of decision variables x1 and x2 that correspond to the number of security 1 and security 2 we must purchase for our portfolio. Whatever our decision, all constraints must be satisfied. We minimize the expectation of the cost based on the given random variables. After running this linear program we take x , x , x , x , and purchase that number of security 1. We also take x , x , x , x , and purchase that amount of security 2.

6. Summary In this article we have illustrated different methods of modeling the asset allocation problem. We began with the one-period Markowitz model and we expanded that model to a multi-period model. We extended the discussion by considering classical Linear Programming which would work for assets that have deterministic or known cash flows. We then discussed Stochastic Linear Programs for assets that have stochastic or unknown cash flows. To give a full appreciation of the applicability of stochastic programming we also provided a number of real-world application examples, illustrating the applicability of these techniques to both past and current real-world applications. These applications are detailed in Section 2 above and include insurance companies, pension managers, credit unions, portfolio managers. A survey of financial applications can be found at Yu et al. (2003). Knowledge of this important field is important for anyone who is serious about making conservative and good investment decisions in the context of a company that operates under constraints. In short, this applies to almost every company. Utilizing mathematical modeling allows a large or moderate size investor to clarify their objectives and will also provide a framework for decision making that will avoid off the cuff decisions and risks that may otherwise be taken.

1843

Stochastic Asset Allocation Models References: April J., Glover F. and Kelly J. (2002). “Portfolio optimization for capital investment projects”, Proc. 34th Conf. Winter Simul. Explor. New Front, pp. 1546-1554. Becker H. (1990). “Labour input decisions of subsistence farm households in southern Malawi”, J Agric Econ, Vol. 41, pp. 162-171. Carino D. R., Kent T. and Myers D. H. et al. (1994). “The Russell-Yasuda Kasai model: An asset/liability model for a Japanese insurance company using multistage stochastic programming”, Interfaces, Vol. 24, pp. 29-49. Carino D. R. andZiemba W. T. (1998). “Formulation of the Russell-Yasuda Kasai financial planning model”, Oper Res, Vol. 46, pp. 433-449. Castro J. (2009). “A stochastic programming approach to cash management in banking”, Eur J Oper Res, Vol. 192, pp. 963-974. Dantzig G. B. (1955). “Linear programming under uncertainty”, ManagSci, Vol. 1, pp. 197-206. Fourcans A. and Hindelang T. J. (1974). “Working capital management for the multinational firm: A simulation model”, Proc. 7th Conf. Winter Simul., Vol. 1,pp. 141-149. Golub B., Holmer M. and McKendall R. et al. (1995). “A stochastic programming model for money management”, Eur J Oper Res, Vol. 85, pp. 282-296. Houck M. H. andCohon J. L. (1978). “Sequential explicitly stochastic linear programming models: A proposed method for design and management of multipurpose reservoir systems”, Water Resour Res, Vol. 14, pp. 161-169. Kall P. and Mayer J. (2011). Stochastic Linear Programming: Models, Theory, and Computation, Springer. Kouwenberg R. (2001). “Scenario generation and stochastic programming models for asset liability management”, Eur J Oper Res, Vol. 134, pp. 279-292. Kusy M. I. andZiemba W. T. (1986). “A bank asset and liability management model”, accessed on 26 Dec 2013, available online at: http://pubsonline.informs.org/doi/abs/10.1287/opre.34.3.356. Markowitz H. (1959). Portfolio Selection: Efficient Diversification of Investments, Yale University Press; John wiley& Sons. Miller N. and Ruszczyński A. (2011). “Risk-averse two-stage stochastic linear programming: Modeling and decomposition”, Oper Res, Vol. 59, pp. 125-132. Moriggia V., Bertocchi M. and Dupaková J. (1998). “Highly parallel computing in simulation on dynamic bond portfolio management”, ACM SIGAPL APL Quote Quad., pp. 215-221. Mulvey J. M. (1996). “Generating scenarios for the Towers Perrin investment system”, Interfaces, Vol. 26, pp. 1-15. Pflug G. andWozabal D. (2007). “Ambiguity in portfolio selection”, Quant Finance, Vol. 7, pp. 435-442. Salmi T. (1975). “Joint determination of trade, production, and financial flows in the multinational firm assuming risky currency exchange rates: A two-stage linear programming model building approach”, Helsinki school of economics Tintner G. andRaghavan N. S. (1970). “Stochastic linear programming applied to a dynamic planning model for India”, Econ Internazionale, Vol. 23, pp. 105-117. Ward J. E. and Wendell R. E. (1990). “Approaches to sensitivity analysis in linear programming”, Ann Oper Res, Vol. 27, pp. 3-38. Yarmish G. and Van Slyke R. (2009). “A distributed, scaleable simplex method”, J Supercomputing, Vol. 49, pp. 373-381. Yu L. Y., Ji X. D. and Wang S. Y. (2003). “Stochastic programming models in financial optimization: A survey”, AMO – Advanced Modeling and Optimization, Vol. 5, No. 1. Zenios S. A. (1993). “Parallel Monte Carlo simulation of mortgage-backed securities”, in: Financial Optimization, Cambridge University Press, p. 325.

1844


Energy Substitution in U.S. Electricity Generation Osei Yeboah1, Saleem Shaik2, Afia Fosua Agyekum1, Julie Melikpor-Lee1 (1. North Carolina Agricultural & Technical State University, USA; 2. North Dakota State University, USA)

Abstract: Rising electricity cost and increasing electricity consumption threatens the ability of businesses to continue in operation by complicating industrial production and operational requirement, thus the need for firms to substitute away from electricity to minimize cost. This paper uses panel seemingly unrelated regression (SUR) model to estimate the substitution between electricity and other forms of energy in U.S electricity generation. The factor share equations for different forms of energy are derived from translog cost function for 48 states from 1970 to 2010. The results from the empirical application suggest limited substitution potential for all the energy inputs. Further, natural gas was found to be the main substitute while wood and waste was a net compliment. Key words: input substitution; energy; elasticity of substitution; factor shares JEL codes: C33, Q4

1. Introduction One of the critical issues in current energy policy debates in both the U.S. and other energy consuming countries is the feasibility of substantially reducing the use of electricity. Issues on electricity have recently dominated the economic decisions of several states across the U.S. economy. In the year 2012, the total amount of electricity produced and the total amount of electricity consumed varied by US regions (National Energy Board of Canada and DOE, 2012). Electricity consumption among states has increased more rapidly on a percentage basis in recent years. Though natural gas and oil are known to occur in certain states, they are not currently produced. Offshore drilling still remains controversial since some of these states often face severe hurricanes and storms. Policymakers, environmentalists, and conservationists in some states admit that drilling for oil or natural gas off shores poses incredible environmental and economic risks to valuable regional resources, including aquatic Osei-Yeboah, Ph.D., Associate Professor and Interim Director, North Carolina A&T State University; research areas/interests: environmental economics and nonmarket valuation; international trade economics and international agribusiness marketing; international agricultural trade; agricultural marketing and cooperatives; applied econometrics; and general equilibrium modeling and policy. E-mail: [email protected]. Saleem Shaik, Ph.D., Associate Professor, North Dakota State University; research areas/interests: agricultural production economics and applied risk analysis of agriculture and food section with special emphasis on farm production (technology, climate and shifting agricultural production systems), financial and institutional (federal farm programs and crop insurance) issues. E-mail: [email protected]. Afia Fosua Agyekum, Ms. Agricultural and Environmental Science, North Carolina Agricultural & Technical State University; research areas/interests: international trade economics; production economics; environmental economics and nonmarket valuation; agricultural marketing and cooperatives; applied econometrics; and general equilibrium modeling and policy. E-mail: [email protected]. Julie Melikpor-Lee, Ms. Agricultural and Environmental Science, North Carolina Agricultural & Technical State University; research areas/interests: international trade economics; production economics; environmental economics and nonmarket valuation; agricultural marketing and cooperatives; applied econometrics; and general equilibrium modeling and policy. E-mail: [email protected]. 1845

Energy Substitution in U.S. Electricity Generation

ecosystems and tourism. Besides prospects for drilling, the different states produce several dry tons of forest, agricultural, urban and mill residues which can potentially generate substantial amounts of electricity each year to adequately supply the annual needs of the residential electricity use of the states in the U.S.A. Majority of the states in U.S. have not engaged in a detailed evaluation of energy in recent years. Currently, apart from few states, most states import virtually all of their fuel resources from other states in the U.S. These imports represent an annual financial diversion of several billions of dollars some of which could be used to develop domestic, alternative energy resources. Growth in electricity consumption for the residences, commercial sectors, transportation sectors and industrial sectors still remain a key focus when it comes to electricity efficiency among the states. Moreover, “clean” electricity for residents has certainly become a critical issue recently. Several states face serious concerns regarding their natural environment. There have been dramatic increases in emissions of air pollutants from electricity use, including nitrogen oxides (NOx), sulfur dioxide (SO2), particulates, mercury, and green house gases, such as carbon dioxide (CO2) and methane. The cost of air pollution in terms of human health alone has been unusual among states in the southeast. The rising electricity cost to certain states further complicates industrial production and operation requirements, often threatening the ability of businesses to continue in operation. In essence, it is high time states and the U.S as a whole considered other types of energy inputs that are environmentally friendly and can adequately substitute for the conventional energy sources at lower costs. The main objective of this paper is to determine the substitution between electricity and other energy input forms in the U.S electricity generation. The specific objectives are to: derive the shares of coal, natural gas, petroleum oil, wood and waste, and electricity as inputs in the energy sector; use a panel econometric model to estimate the system of factor share equations; and construct the elasticity of factor substitution matrix using the estimated parameters to determine the substitutability of energy inputs. Findings of this study will be relevant in the development of a comprehensive energy policy for the region. It will also contribute significantly to the energy policy of the entire U.S. and the regions. The paper is organized into five sections. Section 2 is a brief review of empirical studies on input substitution. Section 3 presents the methodology employed in the analysis. Results of the analysis and its discussions are presented in Section 4 while Section 5 presents the conclusions.

2. Review of Empirical Studies Studies conducted on energy substitution include works by Giffin (1977), Caloghiro et al. (1997), Barnett et al. (1998), Kemfert (1998), Mahmud (2000), Kuper and Soest (2003), Thompson (2005), Roy et al. (2006), Koetse et al. (2007); and Thompson (2011). Most of these studies assessed factor substitution between non-energy inputs specifically labor and capital and energy inputs in different non-energy industries (Caloghiro et al., 1997; Barnett et al., 1998; Kemfert, 1998; Mahmud, 2000; Kuper & Soest, 2003; Thompson, 2005; and Thompson, 2011) while Roy et al. (2006) estimated the substitution and price elasticities of energy and compared across different industries and countries. Most of these studies employed panel data and used log linear and transcendental logarithm production functions to estimate the own price and cross price elasticities of both the energy and non-energy inputs under consideration with the main industries of focus being the non-energy production and manufacturing industries. Caloghiro et al. (1997) and Barnett et al. (1998) found electricity to be a weak substitute for capital and labor, implying electricity subsidies lowered the demand for capital and labor. In Pakistani manufacturing, Mahmud

1846


(2000) employed the partial equilibrium method of analysis and found weak substitution between electricity and gas although there was a slight substitution between aggregate energy and other inputs. Also, Kemfert (1998) employed three different nested CES production function to estimate substitution effects between energy, labor and capital inputs and reported that aggregate energy, capital, and labor are substitutes in West German manufacturing industries in the long run. Roy et al. (2006) also found energy substitution elastcities to vary widely across different industries and countries although they did not consider the factors that accounted for the observed variations. Although these studies have established that energy and capital are substitutes, Hunt (1984), in studying the UK industry sector, found capital and energy to be complements with capital and labor as well as energy and labor being substitutes. To solve this dilemma of conflicting findings, Chichilnisky (1993) employed the general equilibrium model to study energy capital substitution and concluded that whether energy and capital are complements or substitutes depended on the parameters of the model and the price of energy. Alternatively, a study that considered the substitution of energy inputs in the energy industry is the study by Griffin (1997). Employing pooled international sample data, Griffin (1977) used the transcendental logarithm production function model to estimate the inter-fuel substitution relationships between fossil fuels (coal, gas, and residual fuel oil) in the generation of electricity. His findings suggested a greater possibility of substitution among the three energy inputs in the generation of electricity. These reviews reveal that much research has not been conducted on the substitutability of energy inputs within the energy industries as compared to substitution between energy inputs and non-energy inputs in non-energy industries. In terms of energy generation, most studies focus on the environmental impact of switching energy inputs (Goldemberg, 2007; Ogden & Williams, 1989; Chynoweth et al., 2001; and Olah, 2005) with few studies (Giffin, 1977) focusing on the possibility of substitutability between energy inputs in the energy industry. This study therefore addresses this issue by determining the potential substitution between electricity and other energy input forms in the U.S.A electricity generation.

3. Methodology The theory underpinning the study is the theory of production which shows how inputs are combined to produce a given level of output. Energy substitution in the U.S. electricity generation starts with the electricity production function which is given as: , , (1) Where: X = Quantity of electricity produced Z = Energy inputs employed in the electricity production K = Non-energy inputs employed in the electricity production T= Technology The firm is assumed to produce the profit maximizing output X* using the optimal levels of the inputs that minimizes cost of production. The model assumes that the firm is a price taker in both input and output markets. However, the focus of this study will be on the optimal energy input levels that are chosen in other to minimize cost of production. This thus reduces the production function to: , (2) Thus, the basic results concern the comparative static substitution between the various energy inputs

1847


employed in the production of U.S. electricity. The energy inputs (Zi) being considered are coal, natural gas, electricity, wood and waste and petroleum with WC, WN, WE, WW and WP as the prices of each of the energy input, respectively. 3.1 Derivation of the Factor Shares of Coal, Natural Gas, Petroleum Oil, Wood and Waste, and Electricity The translog cost function as developed by Fuss and McFadden (1978) and exemplified by Saicheua (1987) is adopted and is generally given as: ∑ ∑ 0.5 ∑ (3) Specifically, this is written as: 0.5

0.5 (4) From Shephard’s lemma, the partial derivative of total cost function with respect to an input price is that input level thus, the demand for coal (C*) is given as; But

hence

For a perfect competitive firm, total cost (TC) equals total revenue (PX), i.e.,

(5) . It follows that (6)

Where

is the factor share of coal.

Similarly, the factor share for all the other energy inputs are derived using the same formula which is stated generally as:

Where Q = Quantity of energy input used WQ = Unit price of the energy input 3.2 Estimating the System of Factor Share Equations for All the Energy Inputs From the factor share equation in (5) above;

This implies

is the factor share equation for coal.

Similarly, differentiating the TC function in Equation (4) with respect to each of the energy input prices yields the following factor share (i) systems

(7) Where each equation is the factor share equation for energy input, respectively. For the assumption of linear

1848


homogeneity of the cost function in input prices to hold, bi = 1 and bij = 0 for each factor share equation. The returns to scale of the electricity industry can also be calculated from the factor share equations. Returns to scale refers to how much output changes as all inputs are changed by the same proportion. If the sum of the cross coefficients in each factor equation is more than one, then the inputs exhibit increasing returns to scale (IRS), if their sum equals zero, they exhibit constant returns to scale (CRS) and if their sum is less than one, they exhibit decreasing returns to scale (DRS). The estimation of the system of equations is based on the stepwise algorithm using generalized least squares and Maximum Likelihood procedures developed by Bjorn (2004) and implemented in STATA by Nguyen and Nguyen (2010). 3.3 Construction of Elasticity of Factor Substitution Matrix The cross price elasticities of substitution are obtained by taking the second derivative of the TC function in equation (4) with respective to each input. Let’s consider the cross price elasticity of coal for natural gas. = Substituting

(8)

into Equation (8) gives /

=

Since

0

=

(9)

Where However by Shepherd Lemma, =

= N, thus, expanding and simplifying Equation (9) =

Where

is the cross price elasticity of coal with respect to the price of natural gas input, thus: + / Derivation of the other cross price elasticities is similar, and the own price elasticity is also given as / Historical data covering 1970 to 2012 from the U.S. Energy Information Administration (http://www.eia.gov) on total energy expenditure (million dollars), natural gas expenditure (million dollars), petroleum oil expenditure (million dollars), coal expenditure (million dollars), wood and waste expenditure (million dollars) and electricity expenditure (million dollars) were obtained for all the 48 states. Also, historical data covering the same period on prices (dollars/million Btu) of natural gas, petroleum oil, coal, wood and waste and electricity were obtained for the above mentioned states (http://www.eia.gov).

4. Results Petroleum has the highest average factor share value of 0.509 followed by electricity with an average factor share value of 0.303 while wood and waste had the least; 0.005. The average factor share value for coal was 0.056 with the average factor share value for natural gas being 0.126. This implies that petroleum accounts for the most of the energy input cost for production. The factor shares of the five inputs are plotted in Figure 1. The petroleum share has slightly decreased over the period while the other factor shares have slightly shown an increase.

1849


Figure 2 shows the history of factor prices. Electricity prices (WE) rose substantially over the period while the price of coal (WC) remained stationary. Also, the prices of petroleum, natural gas and wood and waste increased slightly over the period.

Figure 1

Trend of Factor Shares (1970-2012)

Figure 2

Trend of Factor Prices (1970-2012)

4.1 Estimated System of Factor Share Equations for all the Energy Inputs The estimated factor share equations are: 0.0783

(10.41) 0.0813

(13.54) 0.2846

(31.3) 0.0143

(17.57)

1850

0.0068

(1.89) 0.0037

(-1.18) 0.0001

(-0.02) 0.0014

(-7.15)

0.0067

(-2.65) 0.0067

(-2.65) 0.0053

(-1.67) 0.0052

(15.08)

0.0053

(-1.67) 0.0098

(3.49) 0.0098

(3.49) 0.0006

(1.62)

0.0052

(15.08) 0.0006

(1.62) 0.0044

(-11.54) 0.0044

(-11.54)

0.0002

()

0.0002

(-1.4)

0.0011

0.0011

()

(9.81)

0.0005

0.0005

()

(3.57)

0.0001

()

0.0001

(5.09)

Energy Substitution in U.S. Electricity Generation 0.5415

0.0015

0.0002

0.0011

0.0005

0.0001

With the exception of the estimated factor share equation for coal, the null hypothesis of continuously improving technology cannot be rejected in any of the other factor share estimates. The estimated coefficients are used for the estimation of the substitution elasticities. The sum of the constant terms in all the factor share equations is 1.00. In addition, the sums of the factor price coefficients for each factor share equation are c = 0.0002, N = -0.0185, E = -0.0189, W = -0.0001 and P = 0.00. Since the sum of the constant terms add up to 1.00 and the sum of the cross coefficients in each factor share equation are approximately 0.00, the conditions for CRS are met thus the U.S electricity industry exhibits constant returns to scale (CRS). CRS implies if the prices of all energy inputs decrease, total cost and output will both fall by the same percentage. 4.2 The Elasticity of Factor Substitution Matrix The derived elasticity matrix is presented in Figure 3 below.

Figure 3 Elasticity of Substitution Matrix

Where: CC = Own price elasticity of coal CE = Elasticity of substitution of coal for electricity CN = Elasticity of substitution of coal for natural gas CW = Elasticity of substitution of coal for wood and waste CP = Elasticity of substitution of coal for petroleum EE = Own price elasticity of electricity EC = Elasticity of substitution of electricity for coal EN = Elasticity of substitution of electricity for natural gas EW = Elasticity of substitution of electricity for wood and waste EP = Elasticity of substitution of electricity for petroleum NN = Own price elasticity of natural gas NC = Elasticity of substitution of natural gas for coal NE = Elasticity of substitution of natural gas for electricity NW = Elasticity of substitution of natural gas for wood and waste NP = Elasticity of substitution of natural gas for petroleum WW = Own price elasticity of wood and waste WC = Elasticity of substitution of wood and waste for coal WE = Elasticity of substitution of wood and waste for electricity WN = Elasticity of substitution of wood and waste for natural gas WP = Elasticity of substitution of wood and waste for petroleum PP = Own price elasticity of petroleum PC = Elasticity of substitution of petroleum for coal PE = Elasticity of substitution of petroleum for electricity PN = Elasticity of substitution of petroleum for natural gas 1851


PW = Elasticity of substitution of petroleum for wood and waste There is limited substitution potential when energy prices rise in electricity production. The own electricity substitution elasticity of -0.697 implies that a 10% increase in the price of electric power will reduce input only 7% and expenditure will rise 3%. There would be weak substitution toward electricity input with coal input rising 2.1 Wood and waste and electricity are compliments. Thus wood and waste and electricity inputs fall with higher wood and waste prices. Wood and waste have the least substitution potential. The own elasticity of substitution of -0.271 implies a 10% increase in the price of wood and waste will reduce input only 2.7% and expenditure will rise 7.3%. However, the energy industry can substitute coal for wood and waste to a higher degree. The substitution elasticity of 1.104 implies a 10% increase in the price of coal will increase the use of wood and waste input by 11.04%; and coal spending will fall by 1.04%. Energy producers respond more to rising natural gas prices. If the current export of natural gas products continues and raises prices by 10% there would be a 9.3% reduction in natural gas input use and expenditures will only go up by 0.7%.

5. Concluding Remarks This paper determines the potential substitution between electricity and other energy input forms in the U.S electricity generation. The present estimates predict electricity producers will spend more on energy as energy prices rise. An increasing price of natural gas only inelastically lowers electricity input while raising wood and waste input. Also, electricity producers are sensitive to wood and waste prices, however, substituting natural gas for wood and waste as their prices rise. The combination of the current U.S. natural gas exports and the rising natural gas prices leaves little room for substitution. The estimated constant returns to scale suggest there is neither under nor over production of electricity. If subsidies on fossil fuels are cut as fuel prices rise over the coming decades, the present model of substitution predicts a proportional decrease in US electricity production. References: Barnett A. H., Reutter K. and Thompson H. (1998). “Electricity substitution: Some local industrial evidence”, Energy Economics, Vol. 20, No. 4, pp. 411-419. Bjorn E. (2004). “Regression system for unbalanced panel data: A stepwise maximum likelihood procedure”, Journal of Econometrics, Vol. 122, pp. 281-291. Caloghirou Y. D., Mourelatos A. G. and Thompson H. (1997). “Industrial energy substitution during the 1980s in the Greek economy”, Energy Economics, Vol. 19, No. 4, pp. 476-491. Chichilnisky G. (1993). “Energy-capital substitution: A general equilibrium analysis”, available online at: http://ssrn.com/abstract=1374615 or http://dx.doi.org/10.2139/ssrn.1374615. Chynoweth D. P., Owens J. M. and Legrand R. (2001). “Renewable methane from anaerobic digestion of biomass”, Renewable Energy, Vol. 22, No. 1, pp. 1-8. Fuss M. and McFadden D. (1978). Production Economics: A Dual Approach to Theory and Applications, North Holland, Amsterdam. Goldemberg J. (2007). “Ethanol for a sustainable energy future”, Science, Vol. 315, No. 5813, pp. 808-810. Griffin J. M. (1977). “Inter-fuel substitution possibilities: A translog application to inter-country data”, International Economic Review, Vol. 18, No. 3, pp. 755-770. Hunt L. C. (1984). “Energy and capital: Substitutes or complements? Some results for the UK industrial sector”, Applied Economics, Vol. 16, No. 5, pp. 783-789. Kemfert C. (1998). “Estimated substitution elasticities of a nested CES production function approach for Germany”, Energy Economics, Vol. 20, No. 3, pp. 249-264. Koetse M. J., de Groot H. L. and Florax R. J. (2008). “Capital-energy substitution and shifts in factor demand: A meta-analysis”, Energy Economics, Vol. 30, No. 5, pp. 2236-2251. Kuper G. and Soest D. P. V. (2003). “Path-dependency and input substitution: Implications for energy policy modeling”, Energy 1852

Energy Substitution in U.S. Electricity Generation Economics, Vol. 25, pp. 397-407. Mahmud S. F. (2000). “The energy demand in the manufacturing sector of Pakistan: Some further results”, Energy Economics, Vol. 22, No. 6, pp. 641-648. National Energy Board of Canada and DOE (2012). “Annual report of international electric export/import data”, Office of Electricity Delivery and Energy Reliability, Form OE-781R. Nguyen Minh and Hoa Nguyen (2010). STATA module: Estimation of system of regression equations with unbalanced panel data and random effects. Working Paper. Ogden J. M. and Williams R. H. (1989). “Solar Hydrogen: Moving beyond Fossil Fuels”. Olah G. A. (2005). “Beyond oil and gas: The methanol economy”, Angewandte Chemie International Edition, Vol. 44, No. 18, pp. 2636-2639. Saicheua S. (1987). “Input substitution in Thailand’s manufacturing sector: Implications for energy policy”, Energy Economics, Vol. 9, pp. 55-63. Thompson H. (2011). “Energy in a physical production function: The US economy from 1951 to 2008”, unpublished paper, Economics Department, Auburn University.

1853


A Comperative Regional Analysis for Housing Demand in Turkey Selahattin Bekmez, Aslı Özpolat (University of Gaziantep, Gaziantep, Turkey)

Abstract: Housing is both good and investment assets so that it plays an important role in the economy. Housing also can be characterized as the most complex economic good because of its durability, heterogeneity, locational fixity, the possibility to raise loans against housing collateral and the effect on well-developed secondary markets. On the other hand, income distribution, socio-economic justice and regional disparities imply that housing market is an important concept for social and cultural transformation as well. When compared with other countries in Europe, housing market has an excellent value in Turkey because of rapid economic growth, reasonable income tax rates, low interest rates, relatively lower risk and many other reasons. So, housing sector in Turkey is very important for those who are looking for new investments. This study aims to compare the dynamic relationship between the housing demand and the variables which determine housing demand for 11 regions of Turkey. Regions have been selected as SRE1 based on Turkish Statistical Institute classification. The model has been estimated using quarterly data from 1992.1 to 2012.4. Building Permits has been considered as housing demand indicator. The other variables used are GDP, Monetary Aggregate, Interest Rate and Inflation. The data used in this study were obtained from Turkish Statistical Institute (TUIK), OECD and Euro Stat. By using the variables mentioned, short term relationship between the series has been analyzed with Granger Causality Test in the first stage of the study. In the second stage of the study, long term relationship has been analyzed with Vector Autoregressive Model (VAR). Key words: housing demand; VAR analysis; Turkey JEL codes: R21, C13

1. Introduction Housing is one of the most complex economic good to analyze because of its durability, heterogeneity, locational fixity, sensitivity to the specific financial and regulatory environment, the possibility to raise loans against housing collateral and a relatively high cost of supply (Renaud, 1996; Iacoviello, 2000, p. 8). Because of these features, unlike many other markets the housing is not only a good asset, but also an investment asset. These features of the housing market imply that it is a collection of loosely connected but segmented market (Iacoviello, 2000, p. 8). Relying on mostly domestic capital, producing high value added, having the potential size of employment and relationship between other sectors—particularly manufacturing—makes Housing a locomotive market. Selahattin Bekmez, Dr., Professor, Faculty of Economics & Administrative Sciences, Department of Economics, University of Gaziantep; research areas/interests: economics. E-mail: [email protected]. Aslı Özpolat, Dr. Lecturer, Oguzeli Vocational School of Higher Education, University of Gaziantep; research areas/interests: economics. E-mail: [email protected]. 1854

A Comperative Regional Analysis for Housing Demand in Turkey

Higher multiplier effect of housing expenditures leads to increase in demand for housing related goods such as furniture and textiles. The housing market has been a target of government fiscal and monetary policy aimed to achieve balanced growth, low inflation and lower rate of unemployment. The economic literature have broadly analyzed housing market and its interaction between the effect of shocks on house prices, economic growth, welfare and the financial position of households (ECBC, 2011). From macroeconomic and microeconomic points of view, the housing market is an important aspect of the whole economy for many countries. In other words, the performance of the housing market has a major impact on the overall performance of the economy (Baffoe, 1998, p. 179). The developments in housing markets influence business cycles, play a key role in the transmission of monetary impulses to the real economy and affect the stability of the financial system. These occur due to variety of reasons: Firstly, housing takes a relatively significant share in economic activity and thus, shocks originating in the housing sector can have significant effects on the macroeconomic variables (Brandt at al., 2010). According to this, Baffoe and Bonnie (1998), has found that labor force, growth rate, inflation rate, interest rate and money supply play an important role on housing prices in the U.S. economy. Apergis (2003) has indicated that a positive shock occurred in housing loan rates reduced the real housing prices and this, real house prices decreased in Greece. Also according to the study, the existence of an increase in the rate of inflation and the labor force increased housing prices. In addition, changes in interest rate have an important role to buy houses and durable goods (Baffoe, 1998, p. 182). While Feldstein (1992) has found that inflation has created a positive impact on the housing demand; Kearl (1979) has indicated that inflation reduces the demand for housing. Secondly, Housing sector has an impact on wealth of people other than economic and financial variables. Non-housing consumption is generated by the changes in house prices. This interaction is related to wealth effect of housing. When increase in housing wealth, Households can raise their consumption in response higher house prices (Rozsavölgyi & Kovacs, 2005, pp. 2-3). Housing and household expenditures are an important part of housing costs. Chetty and Szeidl (2004) has indicated that 20% of household spending occurred from housing expenditures; and housing expenditures did not respond to shocks occurred in the economy in the short term. The third reason is the possibility to raise loans against housing collateral. An increase in property prices raise the value of the collateral available to households, which enables them to borrow more from the credit system, which in turn can be used for financing consumption or investment. The last reason is the effect of house price fluctuations on residential construction. The market value of property may increase its reproduction cost arising from higher house prices. Thus, these increases can be impulsion for the construction of new dwellings (Rozsavölgyi & Kovacs, 2005, p. 2). For the economists believing government interventions, the housing market is an important impact on this process. Their argument is if monetary policy is to be useful as a stabilization policy, housing sector is sensitive to monetary conditions and important to the whole economy (Baffoe, 1998, p. 181). Therefore, taxes or subsidies and other government policy tools affecting the process of the housing sector can produce independent shocks and influence the response of housing markets to economic shocks (ECBC, 2011). There are strong empirical evidences of relationship between economy and the housing sector. Mullbauer and Murphy (2008) have surveyed the multiple interactions between housing markets and economy in UK. According to Mullbauer and Murphy (2008) rise in house prices would lead to a decrease of potential customers and reduce the demand for housing. However, the rent or home ownership will become more restricted than in the past. Therefore, private consumption reduces and ultimately, this causes a reduction in the total growth. Prices in the 1855


housing market do not occur in the short term. For this reason, the demand for housing in any one period is equal to the existing housing stock. Thus, the housing market is not usually an efficient market and bring into equilibrium very slowly (Riddel, 2004, p. 121). However, the efficient market hypothesis has been tested on housing markets and is concluded that housing prices have positive serial correlation in the short-term and have a negative serial correlation in the long term (Hamilton & Schwab, 1985). So, the housing markets are not efficient markets in the long-term. Also in the literature because of the residential markets have higher transaction costs, normal and higher yields are not be obtained continuously (Cho, 1996, p. 146). In this study, the dynamic relationship between the housing demand and the variables determining the housing demand for 11 regions of Turkey will be analyzed. Regions will be selected as SRE1 based on Turkish Statistical Institute classification. The model will be using quarterly data from 1992.1 to 2012.4. Building Permits has been considered as housing demand indicator. The other variables used are GDP, Monetary Aggregate, Interest Rate and Inflation. The data used in this study were obtained from Turkish Statistical Institute (TUIK) and OECD. By using the variables mentioned, short term relationship between the series has been analyzed with Granger Causality Test in the first stage of the study. In the second stage of the study, long term relationship will be analyzed with Vector Autoregressive Model (VAR) as impulse response and variance decomposition.

2. The Model The analysis is estimated by using the VAR Model. The VAR approach makes minimal theorical demands on the structure of the model, and it employs a common lag for all variables in all equations. The method basically involves specifying the set of endogenous and exogenous variables that are believed to interact and hence should be included as part of the economic system that one is trying to model and the largest number of lags needed to capture most of the effects that the variables have on each other (Pindyck & Rubenfeld, 1991, p. 354). The VAR methodology superficially resembles simultaneous-equation modeling in that consider several endogenous variables together. But each endogenous variable is explained by its lagged, or past, values and the lagged values of all other endogenous variables in the model; usually, there are no exogenous variables in the model (Gujarati, 2004, p. 837). The Var models have many applications. They are used to determine how each endogenous variable responds over time to shock in that variable and in every other endogenous variable (Baffoe, 1998, p. 183). For a set of n time series variables y t  ( y 1 t , y 2 t , ..., y nt )' , a VAR model of order p (VAR(p)) can be written as (Enders, 2003, p. 301):

yt  A1 yt 1  A2 yt  2  ...  Ap yt  p  ut

(1)

Certain properties of the variables in the model must be checked in order to determine the appropriate specification for estimation purposes. First, it is necessary to determine whether the variables are difference stationary or trend stationary. A test of stationary (or nonstationary) that has become widely popular over the past several years is the unit root test (Gujarati, 2004, p. 814).This is done by Dickey-Fuller Unit Root test that each variables included in the model contains a unit root. The unit root test involves testing the coefficient of the least square estimate β1 in Δyt = α0+α1t+β1yt-1+∑ni=2βiyt-i, is equal to unity. The unit root test results should be interpreted with caution. Research has shown that the test for unit roots has a low ability to reject the null hypothesis of unit when it is false against 1856


plausible alternatives (Baffoe, 1998, p. 185). The unit roots are tested by using the Augmented Dickey-Fuller (ADF) test, and the results shown in tables. Data in log levels and data in log difference is shown in Table 1. Before VAR analysis, Granger Causality Test must be estimated because of to analyzed short term interaction between variables and for to make Cholesky rank. The VAR model assumes that the current innovations n vector of random variables are unanticipated but become part of the information set in the next period. The implication is that the anticipated impact of a variable is captured in the coefficients of lagged polynomials while the residuals capture unforeseen contemporaneous events. A joint F-test on the lagged polynomials provides information regarding the impact of the anticipated portion of the right-hand side variables (Baffoe, 2004, p. 183). The impact of unanticipated policy shocks on housing demand can be analyzed by employing impulse response functions and variance decompositions. Impulse response functions enable us to analyze the dynamic behavior of the target variables due to a random shock in other variables. The impulse response traces the effect on current and future values of the endogenous variables of one standard deviation shock to the other variables (Baffoe, 2004, p. 188). Variance decompositions tells how much of a change in a variable is due to its own shock and how much due to shocks to other variables. In the SR most of the variation is due to own shock. But as the lagged variables’ effect starts kicking in, the percentage of the effect of other shocks increases over time (Enders, 2003, p. 310). Nevertheless, impulse analysis and variance decompositions (together called innovation accounting) can be useful tools to examine the relationships among economic variables. If the correlations among the various innovations are small, the identification problem is not likely to be especially important (Enders, 2003, p. 280). The model has been estimated using quarterly data from 1992.1 to 2012.4. Building Permits was considered as housing demand. The other variables used as determining the housing demand are GDP, Monetary Aggregate, interest Rate and inflation. Both variables were obtained from Turkish Statistical Institute and OECD. According to analysis; the demand for housing at time (t) is given by: QtD= D( MA,t, GDP,t, R,t ,INF,t) Where: MAt = Monetary Aggregate; GDPt = Gross Domestic Product; Rt = Interest Rate; INFt = Inflation. The model is written as follows: BPt =  1   2 GDPt   3 Rt   4 MAt   5 INFt   t (2)

3. Empirical Results Certain properties of the variables in the model must be checked in order to determine the appropriate specification for estimation purposes. Firstly, it is necessary to determine whether the variables are difference stationary or trend stationary. This is done by testing the null hypothesis that each variable included in the model contains a unit root. If the variables are difference stationary, it is appropriate to estimate the VAR model by using the first difference of the variables. If the variables are trend stationary, the VAR model may be estimated by taking the residuals from the deterministic trend. Secondly, if the variables are difference stationary, it is necessary to establish whether the variables in the model share a common trend. If they do not, estimation of a VAR model

1857


in the first difference is appropriate (Baffoe, 1998, p. s.185). According to this, the unit roots are tested by using the augmented Dickey-Fuller (ADF) test, and the results are shown in Table 1. The results of the test suggest that all variables are different stationary and some variables are different stationary. Table 1 BP

Unit Root Test Result

Level Stationary (T- Stat)* R GDP INF MA

Regions West Marmara -1.022 -1.046 -2.303 -2.334 Aegean -1.709 -1.046 -2.303 -2.334 East Marmara -1.803 -1.046 -2.303 -2.334 West Anatolia -1.509 -1.046 -2.303 -2.334 Mediter. -2.525 -1.046 -2.303 -2.334 Central Anatolia -1.665 -1.046 -2.303 -2.334 West Black Sea -2.021 -1.046 -2.303 -2.334 East Black Sea -1.603 -1.046 -2.303 -2.334 North East Anatolia -1.991 -1.046 -2.303 -2.334 Central East Anatolia -0.588 -1.046 -2.303 -2.334 Southeast Anatolia -2.056 -1.046 -2.303 -2.334 Note: *critical values with constant and trend; -4.20(1%), -3.52 (1%), -3.52 (5%), -3.19 (10%).

1.004 1.004 1.004 1.004 1.004 1.004 1.004 1.004 1.004 1.004 1.004 (5%), -3.19

BP

R

First Stationary (T- Stat)** GDP INF

MA

-3.902 -7.380 -7.267 -4.085 -4.153 -17.008 -7.380 -7.267 -4.085 -4.153 -12.333 -7.380 -7.267 -4.085 -4.153 -17.817 -7.380 -7.267 -4.085 -4.153 -14.886 -7.380 -7.267 -4.085 -4.153 -21.763 -7.380 -7.267 -4.085 -4.153 -6.393 -7.380 -7.267 -4.085 -4.153 -26.205 -7.380 -7.267 -4.085 -4.153 -8.070 -7.380 -7.267 -4.085 -4.153 -21.070 -7.380 -7.267 -4.085 -4.153 -11.752 -7.380 -7.267 -4.085 -4.153 (10%); **critical values with constant and trend; -4.21

Granger Causality test is shown relationship between variables in the short term. The significance levels of the granger causality test provide a summary for analyzing the impact of the anticipated variables on the target level, housing demand. The significance levels of the test, which are based on the hypothesis that all the lags of a given variable in a particular equation are zero, are shown in Table 2. Table 2 Regions

West Marmara

Aegean

East Marmara

West Anatolia

1858

Variables BP R GDP INF MA BP R GDP INF MA BP R GDP INF MA BP R GDP

BP 0.4428 0.8028 0.0448 0.3710 0.9053 0.6091 0.7416 0.4546 0.0866 0.8916 0.2341 0.7453 0.5385 0.4512

Granger Causality Test R 0.8762 0.0162 0.0111 0.3025 0.2542 0.0026 0.0132 0.4287 0.7958 0.0043 0.0580 0.3387 0.4210 0.0011

GDP 0.9101 0.3660 0.0029 0.2724 0.3932 0.2081 0.0217 0.2234 0.8901 0.1483 0.0354 0.3376 0.8056 0.1290

INF 0.8827 0.1760 0.6345 0.9216 0.0225 0.2322 0.8546 0.9296 0.0748 0.0755 0.7763 0.9026 0.5854 0.0764 0.6966

MA 0.6746 0.4425 0.4534 0.0975 0.4126 0.6823 0.4101 0.4389 0.5358 0.5408 0.2675 0.6632 0.2015 0.4571 0.1713 (Table 2 to be continued)

A Comperative Regional Analysis for Housing Demand in Turkey (Table 2 to be continued) INF MA BP R GDP Mediterranean INF MA BP R Central Anatolia GDP INF MA BP R GDP West Black Sea INF MA BP R East Black Sea GDP INF MA BP R Central East Anatolia GDP INF MA BP R Southeast Anatolia GDP INF MA BP R GDP North East Anatolia INF MA

0.5850 0.5084 0.2578 0.5234 0.8053 0.7088 0.0482 0.8497 0.1776 0.2862 0.0656 0.4043 0.2692 0.6331 0.0231 0.9632 0.4394 0.1576 0.2577 0.8230 0.3614 0.1680 0.7939 0.2045 0.4565 0.4455 0.7670 0.7039 0.6740 0.5284

0.0268 0.3242 0.5495 0.0010 0.0648 0.2435 0.0010 0.0048 0.0652 0.1694 0.7700 0.0045 0.1711 0.3428 0.0815 0.0063 0.0479 0.6996 0.4610 0.0023 0.0247 0.1402 0.5255 0.0004 0.0032 0.2026 0.8468 0.0011 0.0015 0.3798

0.1005 0.2219 0.7548 0.2201 0.1217 0.2577 0.2970 0.2415 0.0611 0.1360 0.8013 0.2940

0.3249 0.7075 0.2483 0.0171 0.1004 0.1529 0.4398 0.0192 0.1529 0.0879 0.1397 0.0545 0.5229 0.6088 0.1091 0.0537 0.3698

0.3673 0.7797 0.1457 0.1444 0.8211 0.9620 0.2518 0.0633 0.7226 0.8413 0.7436 0.0336 0.8733 0.0980 0.9711 0.7185 0.0854 0.8632 0.9830 0.8816 0.1532 0.7879 0.9686 0.2159 0.8520 0.8502 0.9702 0.9402 0.9120 0.9465

0.1621 0.7455 0.1915 0.5888 0.0008 0.3577 0.2551 0.4138 0.5657 0.2874 0.1692 0.5592 0.1145 0.6051 0.6041 0.4385 0.4589 0.2029 0.6158 0.4452 0.2363 0.0018 0.0628 0.1876 0.9411 0.0005 0.0985 0.1591

0.9721

The direct and indirect effects can be examined with the help of the table rows and columns. For instance, row 6 and column 6 shows that the impact of change in the inflation rate on the housing demand for Aegean. Accordingly, in the interest rate equation being equal to zero is accepted 98 times out of 100. In the short term, the effects of variables on the housing demand vary according to regions and Cholesky rank is as follow for regions: For West Marmara; Inflation, Monetary Aggregate, Interest Rate, GDP For Aegean; Inflation, Interest Rate, GDP, Monetary Aggregate For East Marmara; Interest Rate, Interest Rate, Interest Rate, GDP For West Anatolia; GDP, Inflation, Monetary Aggregate, Interest Rate 1859


For Mediterranean; Inflation, Monetary Aggregate, Interest Rate, GDP For Central Anatolia; Interest Rate, Monetary Aggregate, Inflation, GDP For West Black Sea; Interest Rate, Inflation, GDP, Monetary Aggregate For East Black Sea; Interest Rate, Monetary Aggregate, Inflation, GDP For North East Anatolia; Interest Rate, Inflation, GDP, Monetary Aggregate For Central East Anatolia; GDP, Interest Rate, Monetary Aggregate, Inflation For South East Anatolia; GDP, Inflation, Monetary Aggregate, Interest Rate 3.1 VAR Analysis As specified before, VAR Analysis have two parts as the impulse response and variance decomposition. (1)Impulse response As a result of the Granger Causality Test is not possible to determine the dynamic structure of the model. The impulse response coefficients provide information to analyze the dynamic behavior of a variable due to a random shock in other variables (Baffoe, 1998, p. 188). Sims (1980) has suggested that the graphs of the impulse response coefficients provide a better device to analyze the shocks. The reactions to shocks of variables in the model to both own and the other variables are important. Cholesky ranking used in the impulse response were taken separately for each regions according to Granger Causality Test. The results for impulse response are shown in Figure 1. West Marmara Response to Cholesky One S.D. Innovations ± 2 S.E. Response of INBP to INBP

Response of INBP to ININF

Response of INBP to INM1

Response of INBP to INFAIZ

Response of INBP to INGDP

.4

.4

.4

.4

.4

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

-.1

-.1

-.1

-.1

-.1

-.2

-.2 2

4

6

8

-.2

10

2

4

6

8

-.2

10

2

4

6

8

-.2

10

2

4

6

8

10

2

4

6

8

10

Aegean Response to Cholesky One S.D. Innovations ± 2 S.E. Response of INYKB to INYKB

Response of INYKB to ININF

Response of INYKB to INFAIZ

Response of INYKB to INGDP

Response of INYKB to LOGM

.6

.6

.6

.6

.6

.4

.4

.4

.4

.4

.2

.2

.2

.2

.2

.0

.0

.0

.0

.0

-.2

-.2

-.2

-.2

-.2

-.4

-.4 2

4

6

8

-.4

10

2

4

6

8

-.4

10

2

4

6

8

-.4

10

2

4

6

8

10

2

4

6

8

10

East Marmara Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB

Response of LYKB to ININF

Response of LYKB to INFAIZ

Response of LYKB to INM1

Response of LYKB to INGDP

.4

.4

.4

.4

.4

.2

.2

.2

.2

.2

.0

.0

.0

.0

.0

-.2

-.2

-.2

-.2

-.2

-.4

-.4

-.4

-.4

2

4

6

8

10

2

4

6

8

10

2

4

6

8

-.4

10

2

4

6

8

10

2

4

6

8

10

West Anatolia Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

-.1

-.1

-.1

-.1

-.1

-.2

-.2 2

1860

4

6

8

10

-.2 2

4

6

8

10

-.2 2

4

6

8

10

-.2 2

4

6

8

10

2

4

6

8

10

A Comperative Regional Analysis for Housing Demand in Turkey Mediterranean Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.4

.4

.4

.4

.4

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

-.1

-.1

-.1

-.1

-.1

-.2

-.2 2

4

6

8

-.2

10

2

4

6

8

-.2

10

2

4

6

8

-.2

10

2

4

6

8

10

2

4

6

8

10

Center Anatolia Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.4

.4

.4

.4

.4

.2

.2

.2

.2

.2

.0

.0

.0

.0

.0

-.2

-.2

-.2

-.2

-.2

-.4

-.4 2

4

6

8

-.4

10

2

4

6

8

-.4

10

2

4

6

8

10

-.4 2

4

6

8

10

2

4

6

8

10

West Black sea Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.4

.4

.4

.4

.4

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

-.1

-.1

-.1

-.1

-.1

-.2

-.2 2

4

6

8

-.2

10

2

4

6

8

-.2

10

2

4

6

8

10

-.2 2

4

6

8

10

2

4

6

8

10

East Black Sea Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





300

300

300

300

300

200

200

200

200

200

100

100

100

100

100

0

0

0

0

0

-100

-100

-100

-100

-100

-200

-200 2

4

6

8

-200

10

2

4

6

8

-200

10

2

4

6

8

-200

10

2

4

6

8

10

2

4

6

8

10

Northeast Anatolia Response to Cholesky One S.D. Innovations ± 2 S.E. Response of DBP to DBP

Response of DBP to DR

Response of DBP to DINF

Response of DBP to DGDP

Response of DBP to DM1

.8

.8

.8

.8

.8

.4

.4

.4

.4

.4

.0

.0

.0

.0

.0

-.4

-.4

-.4

-.4

-.4

-.8

-.8

-.8

-.8

2

4

6

8

10

2

4

6

8

10

2

4

6

8

10

-.8 2

4

6

8

10

2

4

6

8

10

Center East Anatolia Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.4

.4

.4

.4

.4

.2

.2

.2

.2

.2

.0

.0

.0

.0

.0

-.2

-.2

-.2

-.2

-.2

-.4

-.4

-.4

-.4

2

4

6

8

10

2

4

6

8

10

2

4

6

8

10

-.4 2

4

6

8

10

2

4

6

8

10

1861

A Comperative Regional Analysis for Housing Demand in Turkey South East Anatolia Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LYKB to LYKB





.6

.6

.6

.6

.6

.4

.4

.4

.4

.4

.2

.2

.2

.2

.2

.0

.0

.0

.0

.0

-.2

-.2

-.2

-.2

-.2

-.4

-.4 2

4

6

8

10

-.4 2

4

6

8

10

-.4 2

Figure 1

4

6

8

10

-.4 2

4

6

8

10

2

4

6

8

10

Impulse Response Results

For all regions, first columns of Figure show that the response to housing demand shocks occurring by itself. In West Marmara, For example the second column of Figure shows that the effect of one standard deviation of interest rate on housing demand. The third column of figure shows that the effect of one standard deviation of Monetary Aggregate on housing demand. According to results, the most important impacts are respectively occurred by inflation, monetary base, GDP and interest rate. GDP have significantly affected to housing demand in West Marmara, Aegean and south East Anatolia. Housing demand response negative impact to inflation in, but some regions have positive for examples Mediterranean, west black sea and center east Anatolia. Housing demand doesn’t reacts immediately to changes in monetary base in. But, West Black Sea, Center Anatolia and East Black Sea react immediately and interest rate is fluctuating for all regions. (2)Variance Decomposition The variance decompositions show the portion of variance in the prediction for each variable in the model that is attributable to its one shocks to other variables in the model. Variance decomposition divides one of the internal variables change as separate shocks affecting all the internal variables. In this sense, the variance decomposition provides information about the dynamic nature of the system. The results of variance decompositions are reported for each reason in Table 3. The results indicate that impacts of variables are changeable. For examples, in West Marmara GDP has the most important effect on housing demand according to the other variables. But in Center and North East Anatolia GDP has the less effect on housing demand. These differentiations can arise from housing market equilibrium, consumer choices and the other macroeconomic reasons. If the results are evaluated as a general, it can be seen that GDP, interest rate and inflation have important impacts on housing demand. Table 3 Period 1 2 3 4 5 6 7 8 9 10

1862

S.E. 0.442517 0.465253 0.477360 0.504376 0.545327 0.557199 0.573852 0.589523 0.627716 0.642832

LYKB 100.0000 90.49625 86.32171 79.23289 75.84805 72.66515 68.70974 66.16838 63.01706 60.34957

Results for Variance Decomposition West Marmara ININF 0.000000 1.705057 3.097195 3.370729 2.938110 3.390785 4.255334 7.169053 6.605060 8.546119

INMA 0.000000 0.811111 2.932572 3.864067 5.458676 5.523120 6.229938 6.522756 6.913940 7.434340

INR 0.000000 1.418897 2.101140 7.314174 6.446655 8.862218 11.73717 11.20224 9.882671 10.71808

INGDP 0.000000 5.568684 5.547385 6.218142 9.308511 9.558731 9.067816 8.937576 13.58127 12.95189

A Comperative Regional Analysis for Housing Demand in Turkey Aegean Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.387456 0.436261 0.473016 0.477951 0.506058 0.519192 0.523824 0.536715 0.568252 0.573833

INYKB 100.0000 97.28287 82.80662 81.33749 76.86212 73.24668 72.03974 69.32596 69.56884 68.25326

ININF 0.000000 0.945300 7.217586 7.078071 6.391967 7.815190 8.163917 9.251160 8.307408 8.322717

INR 0.000000 0.587065 7.764627 7.903460 7.497997 9.870632 10.77468 10.66139 10.04123 10.92264

INGDP 0.000000 0.080570 0.478890 1.486454 5.586874 5.530960 5.447667 7.299937 8.784939 8.956188

LOGM 0.000000 1.104199 1.732274 2.194521 3.661042 3.536540 3.574000 3.461551 3.297582 3.545196

INR 0.000000 0.863656 3.438122 7.415641 6.527790 9.024659 14.65109 14.95102 14.78416 14.92225

INMA 0.000000 0.436521 0.735862 2.101979 2.051629 2.347874 2.777538 4.311969 4.334510 4.440822

INGDP 0.000000 3.571478 4.365268 4.650834 7.079918 6.872516 6.431834 7.565907 8.794218 8.567375

ININF 0.000000 0.190829 1.154253 4.055197 3.974217 3.906764 7.184166 7.178358 7.124942 7.204717

INMA 0.000000 0.291341 0.341346 6.377915 5.677342 5.674378 5.543991 5.578282 7.359454 8.144053

INR 0.000000 0.920415 1.992872 2.578234 2.324527 5.429950 10.81724 11.24373 10.80569 12.22210

INMA 0.000000 0.041238 2.757352 7.631186 6.503702 6.180500 7.962568 8.659395 8.042321 7.781518

INR 0.000000 0.396370 0.974525 1.316902 1.337374 3.476911 3.946922 4.538128 4.379053 6.330363

INGDP 0.000000 3.645293 3.478370 3.465225 4.766197 5.282853 5.380942 6.032702 7.005732 7.650300

East Marmara Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.325032 0.359712 0.371067 0.384422 0.416777 0.431167 0.447375 0.456420 0.472277 0.478780

LYKB 100.0000 91.90029 86.36472 80.54554 77.40863 72.36151 67.26993 64.64624 63.01788 61.51852

ININF 0.000000 3.228054 5.096026 5.286011 6.932035 9.393443 8.869615 8.524859 9.069227 10.55102 West Anatolia

Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.239514 0.241763 0.249411 0.272339 0.289062 0.294408 0.312324 0.317502 0.334907 0.340702

LYKB 100.0000 98.59727 92.67126 83.69026 83.53570 80.54162 71.62558 71.20689 69.26212 66.93014

INGDP 0.000000 0.000141 3.840273 3.298398 4.488215 4.447288 4.829023 4.792741 5.447792 5.498996 Mediterranean

Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.295165 0.321098 0.338569 0.359015 0.393590 0.405207 0.412514 0.421565 0.438313 0.446077

LYKB 100.0000 90.67172 81.55594 72.91022 73.41825 69.28068 67.48318 65.89592 65.57739 63.42165

ININF 0.000000 5.245381 11.23381 14.67647 13.97447 15.77905 15.22639 14.87385 14.99550 14.81617

1863

A Comperative Regional Analysis for Housing Demand in Turkey Center Anatolia Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.252749 0.279688 0.291325 0.337219 0.356734 0.371280 0.386749 0.399929 0.410088 0.420312

LYKB 100.0000 89.17769 82.77038 65.49773 67.24703 64.34731 62.54950 58.84825 57.64748 57.69801

INR 0.000000 0.194898 4.643840 18.55574 17.05957 18.55513 19.58955 20.39850 19.46421 18.98066

INMA 0.000000 10.47582 10.35777 13.38362 12.83659 11.97333 11.04492 13.23062 14.11512 13.54054

ININF 0.000000 0.087248 0.309559 1.131104 1.515174 1.399110 3.251150 3.602040 5.009082 4.987244

INGDP 0.000000 0.064337 1.918456 1.431811 1.341641 3.725130 3.564873 3.920597 3.764105 4.793553

ININF 0.000000 0.849125 3.614946 3.799196 2.835797 4.291280 4.884713 4.737666 4.951192 5.138473

INGDP 0.000000 0.152033 0.895570 0.898334 0.856076 2.558661 2.533709 3.589768 3.707019 4.471895

INMA 0.000000 0.211476 0.691927 0.676461 2.679088 5.849749 5.794048 5.636587 5.328259 5.582160

INMA 0.000000 0.063855 8.854145 9.319413 6.442294 8.871282 9.556477 9.117686 7.861010 8.030891

ININF 0.000000 1.226918 2.504952 2.368553 1.665952 1.750812 2.288626 2.241989 1.875845 1.769514

INGDP 0.000000 1.153801 1.204365 1.344898 4.354331 6.005028 5.874175 5.818663 6.335236 7.469745

DINF 0.000000 0.791524 0.930147 1.703113 1.731734 2.671773 3.271045 3.509725 3.205778 3.063601

DGDP 0.000000 1.819682 1.780637 2.009522 1.983397 2.842804 2.851461 3.203198 2.836548 2.923856

DMA 0.000000 0.001136 0.402554 0.408888 0.409467 2.266010 3.097323 3.846068 3.270197 3.366075

West Black Sea Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.305054 0.323645 0.331235 0.335593 0.389906 0.406429 0.411198 0.417605 0.442476 0.446976

LYKB 100.0000 96.82916 92.78538 91.75404 91.49908 84.51810 83.30143 82.65406 82.35196 80.98708

INR 0.000000 1.958208 2.012180 2.871971 2.129961 2.782214 3.486098 3.381915 3.661572 3.820388 East Black Sea

Period 1 2 3 4 5 6 7 8 9 10

S.E. 170.1166 179.3682 192.8125 202.5562 244.1255 255.3007 261.3557 267.5918 293.0738 302.9510

LYKB 100.0000 91.86295 82.05373 75.08889 74.96196 68.61618 66.98392 64.36375 67.38913 63.07090

Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.498146 0.606850 0.622354 0.645489 0.649821 0.671720 0.680068 0.689390 0.748031 0.765326

DBP 100.0000 97.24134 94.65562 90.31477 89.11930 85.83011 84.20517 82.80981 84.90352 85.08392

INR 0.000000 5.692477 5.382803 11.87825 12.57546 14.75670 15.29680 18.45792 16.53878 19.65895 Northeast Anatolia

1864

DR 0.000000 0.146318 2.231041 5.563703 6.756105 6.389305 6.575005 6.631204 5.783955 5.562546

A Comperative Regional Analysis for Housing Demand in Turkey Center East Anatolia Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.323758 0.336323 0.341215 0.347291 0.406048 0.437780 0.445355 0.448722 0.470837 0.481175

LYKB 100.0000 93.15582 92.23045 89.27785 88.55997 76.31877 75.70259 74.86587 74.33970 71.21545

INGDP 0.000000 2.008498 2.623713 2.558958 3.674217 7.049124 6.811421 6.916986 8.083183 10.34645

INR 0.000000 4.319770 4.224484 7.016467 5.788663 11.28786 11.46214 11.79410 10.94664 11.74285

INMA 0.000000 0.184293 0.575133 0.581773 1.556187 4.836706 4.977492 4.903781 4.874083 4.990616

ININF 0.000000 0.331623 0.346224 0.564956 0.420960 0.507539 1.046351 1.519255 1.756395 1.704625

ININF 0.000000 5.248605 5.066788 4.911280 4.157989 4.129466 3.967471 4.035018 4.019917 4.426050

INMA 0.000000 0.004093 0.962493 0.929769 1.398427 2.216884 2.318850 2.284822 2.557578 2.995978

INR 0.000000 0.206936 0.561683 3.447662 2.879841 3.177273 3.200448 4.578122 4.797074 5.735887

Southeast Anatolia Period 1 2 3 4 5 6 7 8 9 10

S.E. 0.450989 0.469506 0.478220 0.486669 0.532503 0.542642 0.556954 0.565313 0.581120 0.589531

LYKB 100.0000 92.26945 89.79292 87.14241 83.52905 80.50062 76.51627 74.69179 72.69925 70.64029

INGDP 0.000000 2.270912 3.616112 3.568877 8.034694 9.975753 13.99696 14.41025 15.92618 16.20180

4. Summary and Conclusion The main point of the study is to use non-structural dynamic model to determine the effects of macro variables and by making country comparisons, to determine the differences on the determinants of the demand for housing. The determinant of housing demand generated temporary and permanent shocks in the economy and government policy such as income tax rates, land use regulations, monetary and fiscal policy, and cost of residential. The evidence presented in this paper overall suggests that the component of consumer choice and regional GDP can play a significant role in housing demand fluctuations. Difference between regions and deviations from general housing market equilibrium can be also occurring from these reasons. Income distributions, immigrations, dwelling for foreign and their demand change between according to regions, and the last one is support to building company in emerging regions. References: Apergis N. and Rezitis A. (2003). “Housing prices and macroeconomic factors in Greece: Prospects within the EMU”, Applied Economics Letters, Vol. 10, pp. 561-565. Baffoe-Bonnie J. (1998). “The dynamic impact of macroeconomic aggregate on housing prices and stock of houses: A national and regional analysis”, Journal of Real Estate Finance and Economics, Vol. 17, pp. 179-197. Brandt O., Knetsch T., Penalosa J. and Zollino F. (2010). Housing Markets in Europe: A Macroeconomic Perspective, Spring Publisher. Chetty and Szeidl (2004). “Consumption commitments and asset prices”, The Quarterly Journal of Economics, MIT Pres., Vol. 122,

1865

A Comperative Regional Analysis for Housing Demand in Turkey No. 2, pp. 831-877. Cho M. (1996). “House price dynamics: A survey of theoretical and empirical issues”, Journal of Housing Research, Vol. 7, No. 2. ECBC (2011). European Covered Bond FactBook, European Mortgage Federation Enders W. (2003). Applied Econometric Time Series (2nd ed.), Haboken NJ: J Wiley. Feldstein M. S. (1992). “Comment on James M. Poterba’s paper, tax reform and the housing market in the late 1980s: Who knew what, and when did they know it?”, Federal Reserve Bank of Boston Conference Series, Vol. 36, pp. 252-257. Gujarati D. N. (2004). Basic Econometrics, New York: McGraw Hill, Inc. Kearl J. H. (1979). “Inflation, mortgages and housing”, Journal of Political Economy, Vol. 87, pp. 1-29. Hamilton B. and Schwab R. (1985). “Espected appreciation in urban housing markets”, Journal of Urban Economics, Vol. 18, No. 1, pp. 103-118. Iacoviello M. (2000). “House prices and the macroeconomy in Europe: Results from a structural VAR analysis”, European Central Bank, Working Paper No.18. Iacoviello M. (2010). “Housing in DSGE models: Finding and new directions”, in: Brandt O. & Knetsch T. (Eds.), Housing Markets in Europe: A Macroeconomic Perspective, Spring Publisher. Muellbauer J. (2008). “Housing, credit and consumer expenditure”, CEPR Disccussion Papers 6782. Renauld B. (1996). “Housing finance in transition economics”, The World Bank, Policy Research Working Paper, No.1565. Riddel M. (2004). “Housing-market disequilibrium: An examination of housing-market price and stock dynamics”, Journal of Housing Economics, Vol. 13, pp. 120-135. Pindyck R. and Rubinfeld D. (1991). Econometric Models and Economic Forecasts (3nd ed.), New York: McGraw-Hill, Inc.. Rozsavölgyi B. andKovacs V. (2005). “Housing subsidies in Hungary: Curse or blessing?”, ECFIN Country Focus, Vol. 2, No. 18, pp. 1-5

1866


Reducing the Administrative Burden in Bulgaria: Single Entry Point for Reporting Fiscal and Statistical Information* Diana Yancheva, Karmen Iskrova  (National Statistical Institute of Bulgaria (NSI), Bulgaria)

Abstract: This paper performs the results of the implementation of a “Single Entry Point” project in Bulgaria, which main aims were to reduce the response burden and to improve the quality of statistical information. The project started with the signing of an agreement between the National Statistical Institute and the National Revenue Agency for avoiding the duplication in the collection of microdata and for using the processed information by both institutions. A new online information system was developed and implemented by NSI of Bulgaria as a pillar of the project. Key words: data collection; microdata; online data transmission JEL code: C81

1. Introduction One of the main priorities of the National Statistical System in Bulgaria is to reduce the response burden and at the same time to improve the quality of statistical information. A great achievement in this area was the implementation of the project “Single Entry Point for Reporting Fiscal and Statistical information”. This project is an element of NSI strategy for reducing the administrative burden as well of governmental program for simplification and reducing the administrative burden in the country. Following the Portugal experience, the National Statistical Institute (NSI) and the National Revenue Agency (NRA) signed a bilateral agreement for implementation of the project “Single Entry Point” for reporting Fiscal and Statistical information. The project was realized on the basis of the developed by National Statistical Institute a new Information System “Business statistics” (ISBS) which provides online collection of annual reports of all economic active enterprises, containing a set of accounting and statistical questionnaires.

2. What Is “Single Entry Point” According to the Accounting law and the Law on statistics the enterprises were obliged to provide in both institutions (NSI and NRA) with similar annual reports based on the annual accounting systems of the enterprises. This was considered as an aggravating factor for the business. After starting the project, the organization for *

Presented in 2011, BLUE-ETS Conference, Heerlen. Diana Yancheva, Deputy President, National Statistical Institute of Bulgaria; research areas/interests: business statistics. E-mail: [email protected]. Karmen Iskrova, State Expert, National Statistical Institute of Bulgaria; research areas/interests: business statistics. E-mail: [email protected]. 1867

Reducing the Administrative Burden in Bulgaria: Single Entry Point for Reporting Fiscal and Statistical Information

submission of annual reports was changed. Now the respondents can submit their annual reports in one place - on paper, in the NRA offices (along with their paper annual tax return) or using ISBS. In accordance with the signed agreement, the paper reports are submitted by NRA offices to the Regional Statistical offices for entering into the ISBS and obtaining an integrated database. In this way, it is possible to build a so-called “Data warehouse” and the gathered information can be used for the purposes of both institutions. During the period 2009-2013 more and more respondents assessed the advantages of electronic submission and the number of provided online annual reports had increased significantly.

3. Main Objectives of the Realization of the Project    

Reducing the response burden; Multipurpose use of the collected data by the NRA and NSI; Improving the quality of the collected information; Improving the timeliness of the results;

4. Conditions for the Realization of the Project 4.1 Identification Number of Enterprises The identification number of each enterprise from the Trade Register is identical to that in the Statistical Business Register. This identification code is used in all information systems, which is guaranteed by the national legislation. The uniform coding enables micro-data linking of one and the same enterprise from different information systems. 4.2 Electronic Signature The electronic submission of annual reports and annual tax returns are possible through using the electronic signature (digital certificate). The connections between user’s PCs and the servers of the system are encrypted (via SSL—Secure Sockets Layer). 4.3 Related European and National Legislation The project is developed in accordance with the European legislation and is conformed to the requirements of the following European Regulations:  Regarding Industrial production statistics (Prodcom)—Regulations EC №3924/91, 912/2004, 937/2007, 163/2010, 860/2010.  Regarding Structural Business Statistics (SBS)-Regulations EC №58/97, 1893/2006, 295/2008, 250/2009, 251/2009, 275/2010.  Regarding Foreign Affiliates Statistics (FATS)—Regulations EC №716/2007, 364/2008, 747/2008, 837/2009.  Regarding Statistics on Science and technology-Regulation EC №753/2004.  Regarding Innovation Statistics—Regulation EC №1450/2004.  Regarding Business Register—Regulation EC №177/2008, 192/2009, 1097/2010. The national legal framework includes: the Law on statistics, the National Statistical Program, the Law on Corporate Income Tax, the Law on Income Tax of Individuals and the signed agreement for cooperation between the NSI and the NRA.

1868


4.4 Statistical Confidentiality In accordance with the national and EU legislation it is necessary to ensure the protection of the collected micro-data. Compliance with and calculation of primary and secondary confidentiality in the dissemination of data is required.

5. The Steps of Building the “Single Entry Point” To define the scope and the content of the data that has to be submitted;  To ensure that definitions and concepts used in the reports are identical for both institutions—NRA and NSI;  To introduce amendments in the legal acts related to the fiscal and statistical obligations of business;  To build up the concept of SBS data warehouse ensuring the common use of the data that fits the specific purposes of each institution;  To build up the Information System “Business Statistics”;  To organize public awareness campaign and training sessions for accountants and business associations;  To promote the electronic data submission instead of the paper based one. All this steps were realized by establishing a working group with leading experts from the NSI and the NRA. The main aim of the group was to harmonize the definitions and concepts of used indicators in accordance with the requirements of EU and national legislation. The proposals are coordinated to Ministry of Economy, Ministry of Finance, National insurance institute, professional business associations, Association of Professional Accounting firms and other interested institutions. 

6. ISBS as a Tool of Implementing the Project “Single Entry Point” The system provides online collection of annual reports of all economic active enterprises, containing a set of accounting and statistical questionnaires. The system was released on 1st January 2009, as its building took 12 months. 6.1 ISBS Is a Web-based Application Using Free of Charge Standard Web-browsers and Programs without Need of Installation of Any Paid Software The system is available on the Internet through the websites of NSI and NRA. For the proper functioning of the ISBS is necessary to have installed the following software:  Web-browsers-Microsoft Internet Explorer (7 or 8); Mozilla Firefox (2.0.0, 2.3.0, or 3.5.6) or Google Chrome;  Adobe Reader (8.0 or newer) for reading the pdf forms. 6.2 ISBS Is a Highly Efficient System Using the Modern Data Base Servers and Application Servers Which Allows Simultaneous Work of Thousands of Users The user application is built using the most efficient and modern Java technologies (Oracle ADF Rich Faces + ADF Business Components). The powerful ICT infrastructure guarantees the best quality of Internet access and the operation under high pressure. Many users can introduce data from their annual reports in the system at the same time. 6.3 The System Can Be Upgraded and Modified at Any Time When the Data Requirements Are Changed or Simplified The system is open and could be changed at any time if it is necessary. As there is permanent feedback 1869


provided by respondents who are mostly graduated accountants, some changes have been made as a result of their constructive suggestions. The connections with the respondents are done by phones or e-mail addresses published on the website of the NSI. Everybody can obtain methodological and/or technical assistance from qualified experts. Upon completion of the campaign, the system will be updated and prepared for the next year. 6.4 Automatic Determination of the Principal Activity by Using the “Top-down” Method The enterprises indicate any performed activities and their respective shares in the volume of the turnover. On this basis, the system calculates their principal activity for the reference year by using the “top-down” method. 6.5 The System Could Be Easily Integrated with Other Information Systems Online integration is realized with the NRA system for submitting tax declarations by using a common technology Platform Oracle (Oracle Fusion Middleware Package), which allows the two systems to be integrated with other information systems, including those of other technology platforms (Informix etc.). 6.6 The System Allows Obtaining Statistical Outputs in Various Formats—XLS, PDF, XML, HTML, RTF and CSV For the convenience of users of the collected information and its future use for statistical analysis, the system allows to obtain statistical outputs in various formats—XLS, PDF, XML, HTML, RTF and CSV.

7. Advantages of the Realization of the Project “Single Entry Point” 7.1 It Saves Time and Resources of the Enterprises The annual reports could be still submitted on paper but the business already has a choice—queues in front of the desks or online access from the office. The information system is developed in accordance with the national accounting standards and the data is obtained at the end of the year. Data is entered manually but only primary cells—all amounts are automatically calculated and could be compared with the results of the accounting programs. The identification data and those referring to the previous year appear automatically with a possibility to be changed, if necessary. If there are any errors they are displayed after the validation of the reports with the appropriate cells and the conditions which are not met. Each item has a unique code in the system. 7.2 Multipurpose Use of the Collected Data by the NRA and the NSI Collecting the same data from different institutions has been ceased. The data from ISBS is processed and made available to the NRA in all required aggregate levels. The system calculates the data for Structural Business Statistics (SBS), Foreign Affiliates Statistics (FATS), Industrial production (Prodcom), Statistics on Science and Technology, Innovation Statistics, Labour Market Statistics, Business Demography, Environment statistics etc. The final aggregate data is published on the website of the NSI and is available for everyone. 7.3 The Enterprises Can Submit Their Annual Reports throughout the Day and Night with Uninterrupted Online Access According to the national legislation the system is available for submission of reports from 1st January to 30th April each year, as the respondents fill out their reports for the preceding year, so called reference year. The online access in this period is uninterrupted, 24 hours, including weekends and holidays. The reports can be completed without limit of time, with the possibility of breaks and partial filling of different forms or parts of them. After the successful completion of the work each enterprise obtains a unique ID code.

1870


7.4 The Enterprises Can Access Their Already Submitted Reports at Any Time, Including the Reports from the Previous Years After 30th April the system is locked for submission of the data but the respondents can see and print their reports for each year since 2008 at any time. They can not change the data for the reference year after 30th April or the data for the previous years. This can be done only by an authorized expert in NSI if there is a valid reason for a revision of the data. 7.5 The Quality of the Information Increases due to the Automatic Arithmetic and Logical Controls of the Input Data. The data quality is ensured by the system of arithmetic and logic controls among the items in each form and between the forms in the report (micro data linking). The rules are two types—mandatory and recommended. The mandatory controls have to be met for successful completion of the report. The recommended controls remind for potential flaws, but the report can be finished with them. It is made automatically and there are messages describing the problems and associated indicators and forms, which should be corrected. 7.6 The Time for Data Processing Is Reduced and the Information Productivity Is Increased by Using the Modern Information Technologies After successful completion of the report for each enterprise, the system calculates the full set of SBS indicators. The output tables with required indicators and aggregate levels are defined once and can be obtained immediately after closing the system or when any adjustments of the data are made. In this way the time for production of preliminary SBS data will be reduced from t+11 months to t+7 months. This is achieved by using the modern Java technologies. 7.7 Data Can Be Submitted by the Enterprise or by Chartered Accountants Duly Authorized on Behalf of Enterprise (Especially Important for SMEs)

8. Ongoing Work and Future Plans 8.1 Developing the ISBS The system is constantly evolving. It is forthcoming to include the six sets of annual reports for all kind of financial enterprises. In this way the system will cover all businesses and their results will be received in a short time and with guaranteed quality. 8.2 Improving Cooperation and Communication between Statistical and Administrative Authorities The National Statistical Institute of Bulgaria is continuing its efforts to improve the cooperation with other authorities, which collect or need the similar information. Other governmental institutions are also interested in the opportunity to obtain the processed data they need, without having to collect it. The NSI plays the leading methodological and coordinating role in this process.

9. Other Measures for Reducing Administrative Burden in Statistics Better use of ICT is the main tool for lowering the burden;  Harmonizing the definitions and classifications used in different information systems; It is essential to ensuring that enterprises feel the impact of the reduction efforts. All requests for statistical information include a short explanation of why the information is important. The enterprises have an option to receive feedback on the results of the surveys. 

1871


Experiences and Challenges in Public Information Centers in the State of Morelos, Mexico Roque López Tarángo, Crisóforo Álvarez Violante, Silvia Cartujano Escobar, Selene Viridiana Pérez Ramírez, Paula Ponce Lázaro (State University of Morelos, México)

Abstract: Democratic states are characterized by the participation of their citizens in the management of public affairs, getting institutions and administrations be considered as their own, close to them and open to their needs and aspirations, but above all, they acquire social legitimacy of their work. To achieve it, it is essential to establish mechanisms that enable institutions to open to the public, and help to gain citizen confidence. In the State of Morelos, the Public Information Act, Statistics and Data Protection was enacted on the 27th August, 2003. Derived from it, it is mandatory for all state and municipal public entities to ensure the exercise of the right to have access to public information, through its corresponding Public Information Center. Adecade after the enactment of the Act aforementioned, it is relevant to assess the progress achieved in this area, since we must keep in mind that the implementation of a public policy of transparency has an economic cost, as it is necessary to invest in administrative structures, human resources and materials, staff training, information dissemination, and other that must be anticipated and assumed, in order to prevent that the content of the law remains a rhetorical speech or as a programmatic content. Key words: democratic state; transparency; public information; public policy; public information unit JEL code: K00

1. Introduction The organization of modern society is based on the internationally recognized human rights, which most contemporary states have undertaken to respect, protect, ensure and promote by signing the Universal Declaration of Human Rights of 1948. Article XIX of the Universal Declaration of Human Rights establishes the right of citizens to access the information, which in the past fifteen years has been developing in national legislation and to date, 60 countries Roque López Tarango, Masters Degree in Law, Associate Research Professor, State University of Morelos; research area/interests: human rights. E-mail: [email protected]. Crisóforo Álvarez Violantehas, Masters Degree in Taxes, Associate Research Professor, State University of Morelos; research area/interests: the social security in Mexico. E-mail: [email protected]. Silvia Cartujano Escobar, Masters Degree in Taxes, Associate Research Professor, State University of Morelos; research area/interests: Mexican tax system. E-mail: [email protected]. Selene Viridiana Pérez Ramírez, Masters Degree in Business, Associate Research Professor, State University of Morelos; research area/interests: new technologies. E-mail: [email protected]. Paula Ponce Lázaro, Master Degree in Public Administration, Associate Research Professor; State University of Morelos; research area/interests: organizational behavior. E-mail: [email protected]. 1872

Experiences and Challenges in Public Information Centers in the State of Morelos, Mexico

have passed laws thereon. In the case of Mexico, the right of access to information is relatively new, the Federal Law of Transparency and Access to Public Government Information entered into force in 2002. This law is a federal responsibility, so the states have had to enact their respective state laws. The Congress from our entity incorporated the aforementioned article in Articles 2 and 23-A of the Constitution of the State, then, on 27th August 2003, the Law on Public Information, Statistics and Data Protection was promulgated. Derived from the above legislation, the obligation for all state and local public entities to ensure the exercise of the right of access to public information is created, through its corresponding Public Information Center (UDIP for its acronym in Spanish).

2. Literature Review The legislation of the State of Morelos establishes the existence of a guarantor body of the right to access to public information: The Institute of Public Information and Statistics (IMIPE by its acronym in Spanish), which has an index of indicators (Integral Index of Transparency in Morelos) by which assesses the degree of compliance of the Public Information Units in each obligated entity. Periodically, the IMPE performs evaluations in each UDIP to determine the degree of progress in fulfilling its obligations relating to transparency and accountability.

3. Methodology This work includes the analysis of assessments made on Public Information Centers in the last three periods of the last municipal administrations, that is, during the triennium 2003-2006, 2006-2009 and 2009-2012. The sample of the obligated is composed by UDIPS of the 33 municipalities of the State of Morelos, so no other entities are part of the study.

4. Results Taking into consideration the above-mentioned approach, we can observe the progress of the UDIP’s, regarding their obligation to guarantee the right of citizens to access public information found in defense of local governments, in the periods before indicated: (1) Respect to the establishment of the Public Information Centers, in all municipalities have had errors of origin. For example, in the creation agreement of one of the municipalities, twenty starters of Public Information were established; in addition they noted names of public servants and not the charges, so by not being working the people referred to this council, such agreement is not legally valid. The same situation prevailed with respect to the establishing Agreement of the Council of Classified Information. (2) As for the integration of Classified Information List and the Personal Data catalogue, even though most of the UDIP's have them, in many cases they are not updated. Added to this, there is the tendency of local authorities to integrate to the Classified Information List, information that for some reasons do not want to make available to individuals, even if they are not meeting the legal requirements to be classified as confidential or private. (3) It is important to point out that the IMIPE evaluates the facilities and citizen service, so there are several

1873


items that need to be addressed, such as: (a) The office must have sufficient space for the user to write his/her application. In this regard, we can say that in most cases, the UDIP's do not have optimal conditions. (b) There must be forms for requesting information and available to public view. It is usually the IMIPE who provides formats for each of the regulated entities. (c) There must also be visible signs to locate the Public Information Unit within the institution. Most UDIP's do not meet this obligation. (d) It is mandatory to put in the City Hall as well as in the office of Public Information, material that promotes the right of access to information, such as posters, banners, leaflets, brochures, etc. In this regard, the advertising used is developed, disseminated and provided by the IMIPE. (4) To ensure the exercise of the right to public information, it is necessary to have a system of organization and systematization of information, through which this information is visible, which is why on 31st December, 2008, the Archival Guidelines for those regulated entities were published in the Official Gazette number 4671 of the State of Morelos “Tierra y Libertad”, in terms of what is stated in the Law on Public Information, Statistics and Personal Data Protection of the State of Morelos. In its article 12 establishes the obligation on public bodies to form a File Coordinator Area determined by the head of the entity. The Archive Coordinator Area is responsible for implementing the archival control system of the town hall, which must contain three sections:  Process file.  Concentration file.  History file. Most municipalities do not have an area that is specifically responsible for managing the municipal archives, a situation that has a negative impact on various aspects, it is therefore necessary to carry out the establishment of Archives Coordinator Area. The agreement on the appointment must be published in the Official Gazette of the State of Morelos “Tierra y Libertad”. The archival system, understood as that which concerns the integral management of public documents, constitutes the indispensable premise for the operation and effectiveness of the right of access to information, since the conditions to guarantee the right of access to information are:  The document exists.  That is reachable by the public entity.  That is identifiable by the population. The problem which confronts the citizen to get information from a UDIP in the State of Morelos, in some cases, is not the refusal of public servants, but the impossibility of the administration itself to locate documents due the lack of organization and systematization of the municipal archive. (5) As for the daily functioning of the UDIP’s, it is noteworthy that they have no adequate monitoring implemented to public information requests submitted by the population, both received in writing at the office, as the received through INFOMEX system. Due to the above, most municipalities have disagreement actions brought against them in the IMIPE. One of the most important aspects that arise from obligations of transparency, is the one related to the liability of the obligated parties to have an internet portal, in which information be published in accordance with the indicators set out in the Act of Public Information, Statistics and Personal Data Protection of the State of 1874


Morelos, the Rules of such Act and the Integrated Index of Transparency in Morelos. In this regard, it is noted that a high percentage of municipalities do not have updated information; besides they do not publish it in terms of the provisions of the aforementioned documents. It is also essential to consider that the designation of the UDIP’s heads is the consequence of political commitments or campaign promises that are accompanied by serious drawbacks, such as the server does not have the appropriate profile, with all that it represents. Besides, the Public Information Center is sometimes the “cumbersome” from the municipal administration, which is not provided with the necessary financial, material or human resources to perform the function. The greatest challenge is in the implementation of the law on transparency and access to public information and the consolidation of the independency of the guarantor bodies, and consequently, it is essential to rethink the legal nature of the UDIPS as autonomous bodies from the municipal administration. A public servant, head of the Public Information Center faces a dilemma when there is interest from the City officials to hide certain information. In this regard, to ensure the effectivity of the right of access to information, it would be useful to think that Public Information is not subject neither functionally, nor organically to the Municipal Civil Service, so that the operator can act independently. Its design and concept must be designed to erect them as specialized organs, technicians, whose opinions and operation must ensure fairness and commitment, not with whom it is appointed and holds the office, but with the compliance of the law. Another challenge is to solve the boundary problem, that is to say, to establish the border: to what extent advertising should be expanded and from where preserve secrecy, confidentiality. On the problem of the boundary between public and private, I believe that the answer lies in balance or harmonization of reality, because each case has its own peculiarities which must be taken into account, so the jurisprudence will be the one responsible to set the criteria to guide legal practitioners.

5. Conclusions The right of access to public information is a fundamental feature of any democratic government, in Mexico has its basis in Article 6 of the Political Constitution of the United Mexican States. Consistent with this, the Congress of our entity incorporated it in Articles 2 and 23-A of the Constitution of the State, and later on 27th August 2003, the Law on Public Information, Statistics and Personal Data Protection was promulgated. A society is democratic when its citizens are better informed, that is, are guaranteed the right of access to information and freedom of expression; in this regard, transparency becomes a tool to prevent the use of privileged information. Municipal governments have failed in their attempt to comply with the transparency that is required by law, the main obstacles they face are: lack of citizen participation, secrecy and concealment of public information by municipal public servants, bad practice or resistance to comply with the rulings of the guarantor bodies, there is no real political will to make transparency a reality and poor dissemination of the right of access to information. It is important to note that a Law on Access to Information, by very liberal and progressive that is designed will not guarantee individuals the right to access public information, but in reality should be a real tool for instrumentation of public policy and effective real transparency, coupled with a genuine political will, impeller of the principle of transparency. On this new basis, it is going to be necessary a lot of will, resources, infrastructure and capacity to build

1875


plenty of transparency policies, as many as the diversity of our municipalities. The implementation of a policy of transparency has an economic cost, since it should be invested in administrative structures, human and material resources, staff training, information dissemination, and others, that must be foreseen and assumed to avoid that the contents of the law remains a rhetorical speech or program content. Public entities do not provide complete information, and in general, the one that is offered is not detailed, nor presented to public with the opportunity and timing necessary, so what prevails is the spirit of corporatism or marketing over that of accountability. Therefore, the challenge is the implementation of a political culture so that there will be rulers committed to transparency. There should be attractive incentives that encourage ongoing training for those responsible to safeguard, manage, and release information. References: Domínguez González L. A. (2006). El acceso a la información pública en los municipios de México” en López–Ayllón, S. (coord.), Democracia, Transparencia y Constitución, Propuestas para un debate necesario, México, IFAI-IIJ/UNAM 2006, México. Escobedo F. (2010). La invención de la transparencia, Serie políticas públicas, Porrúa, México. Merino M (2005). “El desafío de la transparencia. Una revisión de las normas de acceso a la información pública en las entidades federativas de México”, en Democracia y transparencia, México, Instituto Electoral del Distrito Federal,. Navarro F. and Villanueva E. (2008). Medios de servicio público y transparencia: análisis para su medición de su desempeño, Senado de la República–Instituto de Investigaciones Jurídicas de la UNAM, México.

1876


Integrating Sustainability Educationin to Business Curriculum: An Analysis of Existing Syllabi Zabihollah Rezaee1, Saeid Homayoun2 (1. The University of Memphis, Memphis, TN 38152-3120, USA; 2. University of Gävle, Gavle, Sweden)

Abstract: This study examines the coverage of business sustainability education (BSE) in the business curriculum and presents pedagogical suggestions for the curriculum design of BSE. A study of 147 syllabi from courses taught at business schools worldwide provides evidence regarding: (1) the scope and nature of BSE; (2) educational approaches to BSE; (3) business sustainability topics that could be taught as part of BSE; and (4) methods of coverage of BSE. Pedagogical issues addressed in this study pertaining to the importance, delivery, and coverage of BSE should help business schools develop and redesign their curriculum to prepare students for the challenges awaiting them in the area of emerging global moves toward business sustainability as well as sustainability reporting and assurance. Keywords: assurance; business sustainability; curriculum design; sustainability education; sustainability guidelines; reporting JEL codes: M14, M41, M48

1. Introduction In today’s business environment corporations worldwide are under profound pressure of high competition to improve their long-term and enduring performance. Business sustainability of focusing on the multiple bottom line (MBL) of economic, governance, social, ethical, and environmental (EGSEE) performance is also gaining considerable attention (Brockett & Rezaee, 2012). Business sustainability in general and sustainability reporting and assurance in particular are gaining momentum in the business and investment communities as more than 3,000 companies worldwide are disclosing their sustainability information in all or some areas of EGSEE sustainability performance (Ernst & Young, 2011). In the aftermath of the 2007-2009 global financial crisis, business organizations are focusing on sustainability as a strategic imperative to achieve not only quarterly results but also long-term EGSEE performance. Despite the importance of business sustainability and sustainability disclosures to corporations and investors, there is limited research on the status of business sustainability education (BSE). This study attempts to fill the gap in literature by examining the coverage of BSE in the business curriculum and presents pedagogical suggestions for curriculum design of BSE. The purpose of this study is to explore the coverage and delivery of BSE by: (1) analyzing the range of Zabihollah Rezaee, Ph.D., Professor of Accounting, Fogelman College of Business and Economics, The University of Memphis; research areas/interests: business sustainability, corporate governance, financial reporting. E-mail: [email protected]. Saeid Homayoun, Ph.D., Assistant Professor, Faculty of Education and Economics, Department of Business and Economic Studies, University of Gävle; research areas/interests: corporate governance, auditing. E-mail: [email protected]. 1877

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi

curriculum content found in existing sustainability courses; (2) examining sustainability course descriptions, objectives, and content; and (3) presenting pedagogical suggestions for curriculum design of BSE. This objective is achieved by performing an exploratory review and content analysis of a sample of 147 academic courses/degrees currently being offered in business schools worldwide. Pedagogical issues addressed in this study pertaining to the importance, delivery, and coverage of BSE should assist business schools in the effective integration of sustainability into their curriculum.

2. Business Sustainability and Sustainability Reporting and Assurance 2.1 Business Sustainability The concept of business sustainability and corporate accountability has become an overriding factor in successful strategic planning for many business organizations worldwide. Business sustainability is a process of increasing the positive impacts and reducing negative effects of operations on sustainable economic, social, and environmental performance. The idea is that an organization must extend its focus beyond achieving short-term profit targets by considering the impact of its operation on long-term financial performance, the community, society, and the environment. Business sustainability can be defined as a process of focusing on business activities that generate long-term economic performance of creating shareholder value as well as voluntary activities that result in the achievement of social, ethical, and government performance that concern all stakeholders (Brockett & Rezaee, 2012). This view is shared by other researchers (Wempe & Kaptein, 2003; van Marrewijik, 2003; Lo & Sheu, 2007). Sustainability as defined above suggests ways in which business, resources, and people can be integrated into a sustainability business model. It emphasizes corporate citizenship, environmental stewardship, compliance with national and local laws, product design and packaging, energy efficiency, and minimization of toxic releases. The existence and persistence of global financial crisis and corporate failures have encouraged companies to pay more attention on their sustainability performance and regulators to require more transparent disclosures. The 2012 survey conducted by the MIT Sloan Management Review-Boston Consulting Group (BCG) indicates that 31% of surveyed companies report that sustainability is contributing to their profits whereas 70% have considered sustainability permanently on their management agenda (BCG, 2012). Byus et al. (2010) compared a matched sample of US firms in the Dow Jones Sustainability Index with firms not in the index and found that firms in the index were found to have a higher gross profit margin and higher return on assets than the firms not on the index. The concept of the “triple bottom line” is often used to assess an organization’s sustainability performance from three perspectives—profit, people, and planet (ISO 26000, 2011). This expansion of determining an organization’s sustainable performance and long-term value-adding strategies has driven a need for new reporting and accountability structures which extend beyond financial statements into non-financial key performance indicators (KPIs) based on environmental impact and social responsibility. 2.2 Sustainability Reporting and Assurance In the post-financial crisis and new regulatory framework era, corporations are under more pressure from a variety of stakeholders to provide accurate, reliable, and relevant financial and non-financial information pertaining to their economic performance, strategic objectives, governance, risk, and social responsibility. The existing financial reporting process is already complicated due to the ever-evolving regulations, rules, and standards as well as extensive volumes of transactions and complexity in the business processes. Capturing,

1878


analyzing, classifying, measuring, recognizing all relevant events and transactions, and disclosing related information can be costly and complicated. Business organizations are searching for an effective, efficient, and feasible way to improve the quality of their financial reporting while ensuring compliance with all applicable rules, laws, regulations, and standards. Sustainability reporting can offer solutions to emerging and widening corporate reporting challenges faced by business firms today. Accounting for business sustainability is the process of identifying, classifying, measuring, recognizing, and reporting performance in all areas of EGSEE (Brockett & Rezaee, 2012). Accounting for sustainability is often referred to as “green accounting” or “green reporting”. Nonetheless, it covers all areas of economic viability, ethical culture, corporate governance, social responsibility, and environmental awareness. Sustainability reporting is gaining considerable attention as more than 3,000 global companies are now issuing stand-alone sustainability reports and it is expected this trend will continue as corporate stakeholders including investors demand more sustainability information (GRI, 2012). Objectivity, reliability, transparency, credibility, and usefulness of sustainability reports are important to both internal and external users of reports and can be enhanced by providing assurance on sustainability reports. Sustainability assurance can be provided internally by internal auditors or external assurance providers. While internal auditors are well-qualified to assist management in the preparation of sustainability reports and providing assurance on them, external users of sustainability reports may demand more independent and objective assurance on sustainability reports. This type of assurance can be provided by certified public accountants (CPAs), professional assurance providers or equivalently accredited individuals, groups, or bodies. Current auditing standards are intended to provide reasonable assurance on financial and internal control reports prepared by management. However, the degree of reliance placed on non-financial information such as sustainability reporting is not completely clear. Assurance standards on different dimensions of sustainability performance reports vary in terms of vigorousness and general acceptability. For example, auditing standards governing reporting and assurance on economics activities presented in the financial statements are well-established and widely accepted and practiced. Assurance standards on other dimensions of sustainability including governance, ethics, social, and environmental standards are yet to be fully-developed and globally accepted. Brockett and Rezaee (2012) point out there are currently two primarily global standards providing assurance guidance on auditing sustainability reports. The International Standard on Assurance Engagements (ISAE) 3000 provides guidance for assurance on non-financial dimension of sustainability. The other global sustainability assurance standard is AA1000AS issued in 2008 by the Account Ability (AA), a global nonprofit organization that established management, reporting, and assurance guidance for nonfinancial dimensions of sustainability performance. In addition to these globally related sustainability assurance standards, there are national sustainability standards including the American Institute of Certified Public Accountants (AICPA) AT 101 and the Canadian Institute of Chartered Accountants (CICA) Handbook Section 5025. The emerging area of business sustainability and sustainability reporting and assurance demands training and education in business sustainability and integration of BSE in the business curriculum.

3. Business Sustainability Education The global investment community is holding public companies responsible and accountable for their business activities and their financial reporting process. Business schools worldwide play an important role in preparing the most ethical and competent future business leaders who understand business sustainability. The public, regulators, the accounting profession, and the academic community are also taking a closer look at

1879


colleges and universities in order to find ways to hold these institutions more accountable for achieving their mission of providing higher education with relevant curriculum. A recent study of 1700 public and privet institutions of higher education indicates that about one-third of colleges and universities have been on an unsustainable financial path and another 28 percent are at the risk of become unsustainable (Blumenstyk, 2012). Long-term sustainability of colleges and universities is vital to the economic growth and prosperity of our nation in preparing the next generation of human capital. The next generations of business leaders must understand the importance of ethical conduct, business sustainability and corporate governance, social responsibility, and environmental matters. Thus, business curriculum must reflect promotion of ethical behavior, professional accountability, and personal integrity taught to business students. Corporate governance, business sustainability, and ethics including accountability, integrity, and transparency must be integrated throughout the business curriculum. Business sustainability education and research has recently been addressed by the global community and accreditation bodies. For example, the Association of Advanced Collegiate School of Business (AACSB) International has established an Ethics/Sustainability Resource Center and on its website asked this question of “Do you think business schools should conduct more research on sustainability and how business can contribute to it?” All respondents as of July 17, 2012 have responded “Yes” to this question suggesting there is an urgent need to do research on sustainability (AACSB, 2012). The proposed integrated framework for defining the post-2015 UN development agenda (United Nations, 2011) suggests a vision built based on the core values of human rights, equality, and sustainability for the entire world’s present and future generations. Four key dimensions of the UN integrated framework are inclusive social development, economic development, environmental sustainability, and peace and security. The Institute of Chartered Accountants in England and Wales (ICAEW) has developed the Business Sustainability Program to promote business sustainability and corporate accountability (ICAEW, 2011). Despite all progress in BSE development, it appears that research and books in business sustainability are fragmented with a lack of an integrated approach covering all EGSEE and many universities have yet to integrate BSE into their curriculum. North American and European business schools typically offer their sustainability education works at the graduate level (Bridges & Wilhelm, 2008). Christensen et al. (2007) examine how deans and directors at the top 50 global MBA programs respond to questions about the inclusion and coverage of the topics of ethics, corporate social responsibility, and sustainability at their respective institutions. They find that a majority (about 84 percent) of business schools require one or more of the three topics (ethics, corporate social responsibility, and sustainability) be covered in their MBA curriculum and there is a trend toward the inclusion of sustainability-related courses. More students are interested in these topics and experiential learning and immersion techniques are used to teach these topics. Rundle-Thiele and Wymer (2010) report that only 27% of universities in Australia and New Zealand required students to take one or more ethics, social responsibility, and/or sustainability core course. Rusinko and Sama (2009) also find that the institutions of higher education are exploring means to integrate sustainability into curricula. The area of sustainability is often discussed in isolation by business academic disciples with a focus on green reporting in accounting, sustainability development in management, and supply chain sustainability strategies in marketing and finance. It is also covered in other disciplines such as social sciences, engineering, and biological science in the content of the global move toward sustainability. Business sustainability practices require an integrated approach to systematically address EGSEE and BSE demands knowledge-base in business 1880


sustainability. Despite the importance of sustainability disclosures to corporations and investors and the move toward integrated sustainability reporting and assurance, there is limited research on the integration of BSE into the business curriculum. This study attempts to fill the gap in literature by examining the status of BSE and its infusion into business curriculum in training the most competent and ethical future business leaders.

4. Methodology Business schools worldwide were searched to determine which ones are currently offering sustainability courses. Their syllabi were then obtained using key search words such as “sustainability”, “business sustainability”, and “sustainability education”. A total of 184 syllabi were identified and only 178 that were directly pertaining to “sustainability education” and providing sufficient course-outline information in the English language were used. Thirteen syllabi were excluded because of insufficient information. Data for this study is based on content analysis of the identified and useable 147 sustainability course syllabi1. Table 1 shows that more than 82 percent of sustainability courses were offered in business schools, with 16 percent in sustainability institutions and centers and the remaining two percent in law schools. Table 1

Business Sustainability Education (BSE) School of Course Offering (n = 147) Number 121 23 3 147

Business School Institute/Centers Others* Total

Percentage 82% 16% 2% 100%

Note: *Others include Law School and Engineering

Table 2 reveals that a majority of these syllabi (44 percent) are from the United States, about 24 percent from Europe, a small portion from Canada, United Kingdom, and Austria (6, 5, and 5 percent respectively) and the remaining 24 percent from other countries including Asia, South America, and the Middle East. Table 2

Business Sustainability Education (BSE) Country of Offering (n = 147)

Number Percentage United States 64 44% Europe* 36 24% Canada 9 6% United Kingdom 7 5% Australia 7 5% Others** 24 16% Total 147 100% Note: *Europe include Austria, Belgium, Check Republic, Denmark, Finland, Hungary, Germany, France, Romania, Switzerland, Sweden, The Netherland, Spain, Scotland, Ireland, Italy. **Others include Abu-Dhabi, Brazil, China, Egypt, India, Japan, Malaysia, New Zeeland, online, Philippines, Singapore, Tanzania, United Arab Emirate.

1

Due to the space limitation, a list of 147 selected sustainability syllabi showing institutions, countries of origin, and URL websites is not included and can be obtained from authors by request. 1881


Table 3 indicates that 87 percent of the selected syllabi are from sustainability-related courses and the remaining 13 percent are from sustainability programs and degrees offered by institutions of higher education worldwide. Table 3

Course Degree/Programs Total

Business Sustainability Education (BSE) Course vs. Degree (n = 147) Number 96 51 147

Percentage 65% 35% 100%

The content analysis method is used to analyze the content of existing sustainability syllabi to determine cognitive knowledge and related topics taught in these courses. This method of content analysis has been widely used in prior research (e.g., Celsi & Wolfinbarger, 2001; Rezaee et al., 2006, 2011) regarding the examination of auditing education, e-commerce and corporate governance, and ethics courses. The selected sustainability-related syllabi were examined on five major attributes including course title, description and objectives, structure, content/topics, and performance evaluation criteria.

5. Results Results of the analysis of the 147 sustainability-related syllabi are presented in the following sections: (1) course titles; (2) course description and objectives; (3) course structure and design; (4) course content and topics; and (5) basis for course grading. (1) Course Title Business sustainability is not globally well-defined in business literature. It is commonly defined as a process of focusing on business activities that generate long-term economic performance of firm value maximization as well as voluntary activities that result in the achievement of social, ethical, and government performance that concern all stakeholders (Brockett & Rezaee, 2012). In this definition, corporate social responsibility (CSR) is viewed as an integral component of business sustainability. The terms “environmental, social, and governance (ESG) reporting”, “risk compliance and governance (RCG)”, “corporate social responsibility (CSR) reporting”, “integrated reporting”, and “sustainability reporting” have been used interchangeably in the literature to describe sustainability reports. This diversity of the definition of business sustainability and sustainability reporting is also reflected in sustainability courses as a wide range of course titles is adopted by universities worldwide in offering sustainability-related courses. As shown in Table 4, the most commonly and frequently used course titles are: (1) sustainability/business sustainability; (2) environmental sustainability; (3) business and society; (4) leadership and sustainability/Sustainable development practices; theory and science of sustainability; (5) corporate social responsibility; (6) ethics, social responsibility and corporate governance; (7) diploma and certification in sustainability; and (8) sustainable products and services. The least frequently used sustainability course titles are: (1) management practices for sustainable business; (2) economics of sustainability: theory and practice; and (3) sustainable product and services. There is a need for a more uniform and globally accepted title for sustainability-related courses. One suggestion is to use “Business Sustainability: Integrated Strategies, Reporting, and Assurance.”

1882

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi Table 4

Business Sustainability Education (BSE) Course/Degree Title (n = 147)

Description Sustainability/Business Sustainability Environmental Sustainability Corporate Social Responsibility Green Energy and Sustainable Business Ethics, Social Responsibility and Corporate Governance Leadership and Sustainability/Sustainable Development Practice Business and Society Diploma/MBA/EMBA/Certificate in Sustainability Business Strategies for Sustainability/Sustainability Reporting Sustainability Sciences Management Practices for Sustainable Business Economics of Sustainability: Theory and Practice Sustainable Product and Services Total

Number 31 20 18 14 10 10 9 9 8 6 6 4 2 147

Percentage 21% 14% 12% 10% 7% 7% 6% 6% 5% 4% 4% 3% 1% 100%

(2) Course Objectives A wide range of objectives are specified in the analyzed 147 syllabi in describing sustainability-related courses. Table 5

Business Sustainability Education (BSE) Degree/Course Objectives (n = 147)*

Objectives Number Identify business social/ethical responsiveness to society 71 Learn how to seek solutions for financial viability, ecological sustainability, and social equity. 69 Understand the role of business in society 66 Understand sustainability value adding activities 64 Learn how to contribute to the future with an understanding of commerce that integrates the economic, 59 social and environmental impacts of business activities Learn the value of environmental auditing 59 Learn the Sustainability as the responsible choice 52 Understand dimensions of business sustainability performance 51 Learn practical examples of ethical and social responsibility 48 Engage students in a learning community and collaboratively develop a deeper understanding of the 44 emerging field Encourage integration of ethical awareness and social responsibility into managerial decision making— 43 application objective Learning Subject competence, Leading change skills, Project experience 41 Design manuals for sustainable development 39 Analyze and evaluate the principles of CSR and Sustainability 38 Integrate sustainability principle and concepts into organizational projects, processes, systems and culture 31 Learn about environmental security and uncertainty 29 Enhance interpersonal and team working skills 26 Develops understanding of new models of sustainability measurement, reporting and assurance 16 Learn the importance of environmental due diligence as a tool for minimizing acquired liabilities in business 15 mergers and acquisitions

Percentage 48% 47% 45% 44% 40% 40% 35% 35% 33% 30% 29% 28% 27% 26% 21% 20% 18% 11% 10%

Note: *Multiple objectives are listed for the analyzed 147 Sustainability-related syllabi

1883


Table 5 shows a total of 19 course objectives are used. The ten most commonly used course objectives are: (1) identify business social/ethical responsiveness to society; (2)learn how to seek solutions for financial viability, ecological sustainability, and social equity; (3) understand the role of business in society; (4) understand sustainability value adding activities; (5) learn how to contribute to the future with an understanding of commerce that integrates the economic; (6) understand social and environmental impacts of business activities; (7) learn the value of environmental auditing; (8) learn the Sustainability as the responsible choice; (9) understand dimensions of business sustainability performance; and (10) learn practical examples of ethical and social responsibility. Course Structure and Design Business sustainability education (BSE) courses can be structured in many ways. An analysis of the 147 sustainability-related syllabi indicates that a combination of lecture, presentation, discussion, analysis, reading assignments, and case methods is used to present sustainability topics in class. Table 6 shows the frequency and percentage of a wide variety of teaching methods used in the examined syllabi. Table 6 also indicates that in terms of structure, there are three fundamental and interrelated components of the course. The first component requires students to actively discuss and present assigned readings from the professional and academic literature on business sustainability. Each student should be prepared to lead the class lecture and discussion of any of the readings and cases assigned for a particular class meeting. The second component involves submission of written analyses of a number of research projects, case studies, and online projects that emphasize business sustainability-related topics. The last component of the course is the requirement of a term paper, oral presentations by students, guest speakers, project pitch, idea blog, and web-related sustainability topics. In the majority of the analyzed syllabi, the instructor’s role is not that of a lecturer, but more of a facilitator and mentor in assisting students with classroom open discussion, exchange of ideas, critical thinking, and development of the individual skills and understanding of sustainability-related topics. Table 6

Business Sustainability Education (BSE) Course Structure (n = 147)*

Structure

Frequency

Percentage of total

Lectures and Discussions

69

47%

Analysis

54

37%

Reading Assignment

53

36%

Case Studies

46

31%

Individual/Independent study

43

29%

Term Papers

36

24%

Oral Presentation by Students

27

18%

Guest Speakers

26

18%

Project Pitch

11

7%

Others**

61

41%

Note: *More than one course structure was mention in the analyzed 147 sustainability related syllabi. **Other course structure/method consisting in online courses, idea blog, group presentation and web-related approaches.

(4) Topical Content of Business Sustainability Education (BSE) Table 7 presents the frequency and percentage of the BSE topics identified in the examined 147 sustainability-related syllabi. Twenty-three topics were identified in all syllabi, and some were prevalent. The

1884


following first nine topics listed were common in at least one-third of all the syllabi analyzed: (1) green energy and strategy; (2) sustainable development/business;(3) business responses to sustainability challenges; (4) sustainability challenges and business impacts; (5) business ethics; (6) global sustainability; (7) corporate social responsibility; (8) management and communication for global sustainability; and (9) ethical and social basis for sustainability including environmental ethics and the demographic transition. Table 7

Business Sustainability Education (BSE) Topics Covered (n = 147)*

Topics

Frequency

Green Energy and Strategy

70

Percentage of total 48%

Sustainable Development/Business

67

46%

Business responses to sustainability challenges

64

44%

Sustainability challenges and business impacts

60

41%

Business Ethics

58

39%

Global Sustainability

57

39%

Corporate Social Responsibility

54

37%

Management & Communication for Global Sustainability

52

35%

Ethical and social basis for sustainability including environmental ethics and the demographic transition

50

34%

Strategic Implications–innovative strategies

39

27%

The big picture of sustainability challenges

33

22%

Sustainability Mindset

31

21%

Economic Environment, Risk and Security of the Environment

30

20%

Sustainability Frameworks and Tools

30

20%

Fundamentals in Sustainable Development

24

16%

Sustainable Leadership

23

16%

Environmental Law

21

14%

Measuring Sustainable Performance

21

14%

Sustainable Operation Management

21

14%

Research in Sustainability

18

12%

Best practices in sustainability strategies

15

10%

Principles Of Sustainability

15

10%

Sustainability Metrics and Reporting

13

9%

Accountability and Sustainability Reporting

10

6%

Models of Sustainability

10

6%

Note: *Multiple topics are listed for the analyzed 147 Sustainability-related syllabi

(5) Course Evaluation Criteria The criteria for evaluating students’ performance are shown in Table 8. Nine different grading factors were identified. Homework assignments, case studies/topic presentation, class participation, research papers, and exams make up the largest proportion of course grades, followed by class attendance and quizzes. More than 20% of the analyzed syllabi did not make available (on the web) their basis for determining student grades, and many provided multiple bases for evaluating student performance.

1885

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi Table 8

Topics Homework Assignments Case Study/Topic Presentation Research Paper Class Participation Term Paper Quizzes Final Exam Attendance Mid Term Exam

Business Sustainability Education (BSE) Basis for Grading (n = 147)* Frequency 74 60 55 53 39 24 20 14 9

Percentage of total 50% 41% 37% 36% 27% 16% 14% 10% 6%

Note: *More than one grading base was mention in the analyzed 147 Sustainability -related syllabi

6. Curriculum Design of BSE The coverage of BSE topics in a separate course, or the integration of them into business courses, requires the classification of related topics into teaching modules. The use of the module approach to BSE enables instructors to customize their syllabus by promoting critical thinking and the flexibility to cover all or selected modules in their courses. An analysis of the 147 syllabi, along with the author’s teaching and research experience in business sustainability, suggests that universities either offering or planning to offer a stand-alone sustainability course or infusing sustainability-related topics into the business curriculum should consider a module approach in designing their BSE coverage. Appendix A presents a module approach to the development of BSE. This approach and suggested modules in Appendix A should be useful to business schools in designing a stand-alone course in business sustainability. It consists of all ten modules that can be integrated into a variety of business and accounting courses which can be found as following: BSE Structure, Design and Content Module 1: Introduction to Business Sustainability and Sustainability reporting and Assurance  The Importance and Relevance of Business Sustainability  Nature and Role of Business Entities in Our Society  Current Status of Sustainability and Accountability  Drivers of Sustainability Initiatives and Practices  Business Sustainability Framework  Sustainability Key performance Indicators  Five Dimensions of Sustainability Performance (1) Economic Sustainability Performance (2) Governance Sustainability Performance (3) Social Sustainability Performance (4) Ethical Sustainability Performance (5) Environmental Sustainability Performance Module 2: Fundamentals of Business Sustainability  Evolution of Business Sustainability  Recent Developments and Initiatives in Sustainability  Status of Business Sustainability and Sustainability Reporting and Assurance  Going Forward  Stakeholder Theory of Business Sustainability 1886

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi  Sustainability Information Needs of Investors Module 3: Sustainability Reporting and Assurance  Usefulness of Sustainability Information  Sustainability Reporting Process  Sustainability Reporting Principles  Sustainability Reporting in Action  Promotion of Sustainability Reporting  Future of Sustainability Reporting  Mandatory versus Voluntary Sustainability Reports  Sustainability Assurance  Types of Assurance Opinions  Internal controls Relevant to Sustainability Performance  Sustainability Risk Management  Research in Sustainability Reporting and Assurance Module 4: Economic Dimension of Sustainability Performance  Long-term and Enduring Financial Performance  Economic KPIs  Public Trust and Investor Confidence in Public Financial Information  Promoting Transparency in Financial Reporting  Global Financial Reporting Language  Forward-Looking Financial Reports  Internal control reporting  Benefits of Mandatory Internal Control Reporting  Internal Control over Business and Operations (ICBO)  Integrated financial and internal control reporting  Integrated Sustainability Reporting Module 5: Corporate Governance Dimensions of Sustainability Performance  Corporate Governance Definition  Drivers of Corporate Governance  Global Convergence in Corporate Governance  Sources of Corporate governance (Sarbanes-Oxley Act of 2002 and Dodd-Frank Act of 2010)  Corporate Governance in the Post-crisis  Corporate Governance Functions (1) Oversight Function (Board Committee (2) Managerial function (3) Compliance Function (4) Internal audit function (5) Legal and financial advisory function (6) External audit function (7) Monitoring function  Proxy Voting for Sustainability  Corporate Governance KPIs  Corporate governance Emerging Issues and Challenges  Corporate Governance Reporting and Assurance Module 6: Corporate Social Responsibility (CSR) Dimension of Sustainability Performance  Social KPI’s  Corporate Social Responsibility (CSR)  International Organization for Standardization (ISO) 26000  Globalization-related CSR  Employee-related CSR  Product and marketing-Related CSR

1887

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi Supply Chain-related CSR Stakeholder-related CSR  CSR and Financial Performance  CSR Performance Measurement  CSR Programs  CSR Reporting Module 7: Ethical Dimension of Sustainability  Business Ethics  Ethics and Law  Ethics KPIs  Workplace Ethics  Training and Ethics Education  Corporate Culture  Corporate Codes of Ethics Rules and Best Practices  Financial Reporting Integrity  Ethics Reporting Module 8: Environmental dimension of Sustainability Performance  Environmental Key Performance Indicators  Environmental Regulations in the United States  Climate Risk Disclosure  Global Environmental Initiatives and Regulations  Kyoto Protocol and mechanisms  Monitoring emission targets  European Union Emissions trading system( EU – ETS)  Carbon Reduction Commitment (CRC )  The World Bank Carbon Finance Unit’s (CFU)  Societal Factors influencing corporate environmental behavior  World Wildlife Fund (WWF)  UN Global Compact  Clean Air Initiative  International Organization for Standardization (ISO) 14000  Environmental management systems  Environmental Management Concepts  Environmental reporting  Environmental Assurance and Auditing Module 9: Business Sustainability and Sustainability Reporting and Assurance in Action  Global Initiatives on Business Sustainability  Global Initiatives on Sustainability Reporting and Assurance  Global Reporting Initiative (GRI)  The International Integrated Reporting Committee (IIRC)  Relevance of ISO Standards to Business Sustainability  Best Practices of Business Sustainability  Best Practices of Sustainability Reporting  Best Practices of Sustainability Assurance  Sustainability Reporting and Extensible Business Reporting Language (XBRL)  Sustainability Assurance and Continuous Auditing Module 10: Business Sustainability in Transition and Business sustainability Education (BSE)  The Emergence of Business Sustainability  The Emergence of Sustainability Reporting and Assurance  Teaching pedagogical Approaches in Business Sustainability  Research in Business Sustainability  

1888

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi  

Case studies in Business Sustainability Future of Business Sustainability

The first three modules introduce business sustainability and sustainability reporting and assurance. Sustainability topics included in these modules are fundamentals and dimensions of sustainability performance as well as sustainability reporting and assurance initiatives and guidelines. Modules four through eight discuss the five dimensions of economic, governance, social, ethical, and environmental (EGSEE) sustainability performance in details. These modules are built on the fundamentals of business sustainability and sustainability reporting and assurance from the first three modules. Module nine discusses the current initiatives of business sustainability, sustainability reporting and assurance issues, and future of business sustainability. The last module focuses on business sustainability education including pedagogical approaches in teaching, research, and cases in business sustainability as well as sustainability reporting and assurance.

7. Limitations and Suggestions for Future Studies This study is subject to limitations of typical content analysis research and thus its findings should be applied with care due to the sample selection and sample size. The findings, however, should be of interest and use by universities that are planning to offer BSE. Other caveats about the results include the possibility of misclassification of BSE learning objectives and topical coverage in conducting a content analysis of the analyzed sample. The results of this study should be interpreted with caution primarily because BSE is emerging and evolving. BSE programs and courses are growing, causing the structure, content, and delivery of BSE to evolve. While it is not possible to identify all the detailed topics covered in a BSE course by analyzing a syllabus, due diligence was made to identify sustainability-related topics in all 147 syllabi. These limitations and evolving business sustainability and sustainability reporting and assurance practices suggest that considerable flexibility should be built into BSE courses and programs. As business sustainability evolves and more business schools consider integrating BSE into their curriculum, more in-depth research on the content and delivery of BSE should be conducted. Future research should provide insights into BSE programs and course developments by conducting surveys to obtain views from both academicians and practitioners. This study presents the existing supply side of BSE. A survey or interview in gathering opinions and insights from both practitioners and academicians provide relevant information about demand for and interest in BSE and can capture what can and should be covered in BSE courses and not just what is covered.

8. Conclusions Recent global financial crises and resulted economic meltdown have turned the spotlight on the issue of business sustainability. Business sustainability is a process that affects strategic decision to focus on long-term sustainable economic performance as well as achievement of governance, social, ethical, and environmental performance. Business sustainability initiatives and sustainability reporting and assurance best practices are at the center stage of corporate strategies, accountability, and compliance programs. Business and accounting curricula are being reassessed worldwide in light of evolving business sustainability. This study examines the coverage and delivery of business sustainability education (BSE), including course offerings, the topical coverage of BSE in the courses, and the methods of delivery of BSE. Sustainability-related topics range from an introduction to

1889


fundamentals and dimensions of business sustainability to sustainability reporting and assurance. The analysis of the selected 147 syllabi indicates that BSE is evolving with business schools and that their faculty are designing sustainability courses based on their interests, knowledge-base, skills, philosophies, and demands. The suggested module approach to BSE should benefit faculty currently teaching sustainability, as well as those planning to offer sustainability courses in the future. The analysis of the sampled 147 sustainability course syllabi provides evidence regarding the current status of BSE. Given the evolution of BSE from coverage of special topics in environmental issues, green economy, and CSR in the early 2000s, this study suggests that a BSE course be offered as part of the general requirement in business education. This approach for the integration of BSE into the business curriculum should enable students to understand all five EGSEE dimensions of sustainability performance. The global continuous growth of business sustainability and sustainability reporting and assurance has encouraged business faculty to find proper ways to deliver BSE. An integrated module approach in teaching BSE should provide a framework for business schools in training future business and professional leaders who understand the importance of all five EGSEE dimensions of sustainability performance. References: AACSB International (2012). “Ethics/sustainability resource center”, available online at: http://www.aacsb.edu/resources/ethics-sustainability/. Blumenstyk G. (2012). “One-third of colleges are on financially ‘unsustainable’ path: Bain study finds”, The Chronicle of Higher Education, available online at: http://chronicle.com/article/One-Third-of-Colleges-Are-on/133095/?cid=at&utm_source =at&utm_medium=en#top. Boston Consulting Group (BCG) (2012). “Press releases: Business sustainability survey”, January 24, 2012, available online at: http://www.bcg.com. Bridges C. M. and Wilhelm W. B. (2008). “Going beyond green: The ‘why and how’ of integrating sustainability into the marketing curriculum”, Journal of Marketing Education, Vol. 30, pp. 33-46. Brockett A. and Rezaee Z. (2012). Corporate Sustainability: Integrating Performance and Reporting, John Wiley and Sons, New Jersey, the United States. Byus K., Deis D. and Ouyang B. (2010). “Doing well by doing good: Corporate social responsibility and profitability”, SAM Advanced Management Journal, Vol. 75, No. 1, pp. 44-55. Celsi R. and Wolfinbarger M. (2001). “Creating renaissance employees in an era of convergence between information technology and business strategy: A proposal for business schools”, Journal of Education for Business, (July/August), pp. 308-317. Christensen L. J., Peirce E., Hartman L. P., Hoffman W. M. and Carrier J. (2007). “Ethics, CSR, and sustainability education in the Financial Times top 50 global business schools: Baseline data and future research directions”, Journal of Business Ethics, Vol. 73, No. 4, pp. 347-368. Ernst &Young LLP. (2011). “TBL Technology considers sustainability reporting”, available online at: http://www.ey.com. Global Reporting Initiative (GRI) (2012). “A complete listing of organizations currently providing sustainability reports”, available online at: http://www.globalreporting.org/ReportServices/GRIReportsList/. Institute of Chartered Accountants in England and Wales (ICAEW) (2011). “Business sustainability program”, available online at: http://www.icaew.com/businesssustainability. International Organization for Standardization (ISO) (2010). “ISO 26000, Social Responsibility”, available online at: http://www.iso.org/iso/iso_catalogue/management_and_leadership_standards/social_responsibility/sr_iso26000_overview.htm#sr-1. Lo S. H. and Sheu H. J. (2007). “Is corporate sustainability a value-increasing for business?”, Corporate Governance, Vol. 15, No. 2, pp. 345-358. Rezaee Z., Lambert K. R. and Harmon W. K. (2006). “Electronic commerce education: Analysis of existing courses”, Accounting Education: An International Journal, Vol. 15, No. 1, pp. 73-88. Rezaee Z., Zhang R., Saadullah S. and Ziegenfuss D. E. (2011). “Corporate governance education: An analysis of existing syllabi”, Journal of Governance and Public Policy, Vol. 6, No.1, pp. 62-92.

1890

Integrating Sustainability Education into Business Curriculum: An Analysis of Existing Syllabi Rundle-Thiele S. R. and Wymer W. (2010). “Stand-alone ethics, social responsibility, and sustainability course requirements: A snapshot from Australia and New Zealand”, Journal of Marketing Education, Vol. 32, No. 5, pp. 5-12. Rusinko C. A. and Sama L. M. (2009). “Greening and sustainability across the management curriculum: An extended journey”, Journal of Management Education, Vol. 33, No. 3, pp. 271-275. United Nations (UN) (2011). “UN system task team on the post-2015, UN Development Agenda: Realizing the Future We Want for All”, available online at: http://dev.un.org/millenniumgoals/beyond2015.shtml. Van Marrewijk M. (2003). “Concepts and definitions of corporate social responsibility and corporate sustainability: Between agency and communication”, Journal of Business Ethics, Vol. 44, pp. 95-105. Wempe J. and Kaptein M. (2002). The Balanced Company: A Theory of Corporate Integrity, Oxford: Oxford University Press.

1891


Toward an Integrated Theory of Sustainability Dilip Mirchandani1, Theodoros Peridis2 (1. Rowan University, Glassboro, NJ 08028, USA; 2. York University, Toronto, Canada)

Abstract: The strategic management literature is combined with a cross-national and cross-cultural perspective to offer a broad conceptual framework that guides the development of a research roadmap that begins with an inductive theory building phase which is followed by an empirical testing phase. The potential benefits, of such an approach, include movement toward an integrated theory of sustainability that will provide useful insights to multiple constituents including managers and policy makers. Key words: strategy; sustainability; cross-cultural JEL codes: M14, M16, M19, Q56 This paper proposes a conceptual framework by drawing together different research streams that have addressed sustainability from different perspective. In addition, a research design is proposed that will be viable in a cross-national and cross-cultural context and, as a first step, engage in inductive theory building. The primary goal of the conceptual model and research roadmap is: Advancement of knowledge: build and test a theoretically grounded model that informs the role of culture and external factors in shaping organizational strategies for sustainability. Specifically, the proposed research will seek to understand: (a) What is the relationship between structural factors such as economic, technological, social, political and regulatory conditions and firms’ sustainability strategies? How do these relationships vary across different countries? (b) How does national culture influence the worldview of firms and inform their sustainability strategies? (c) How do the relationships between structural factors, cultures and strategies vary across different types of firms, namely small firms, large corporations, and multinationals? (d) What are the dynamic elements that govern these relationships and how do they change over time?

1. Context Consider the strategic choices of firms in four Mediterranean countries with similar sun exposure, yet very different responses to the benefits of solar energy. In Spain, long term government incentives are in place to promote solar energy solutions and numerous alternatives compete in the retail market, ranging from the most basic rooftop panels to expensive parabolic dishes (Whelan, 2008). The larger players in the market sell products Dilip Mirchandani, Ph.D., Professor, Rohrer School of Business, Rowan University; research areas/interests: strategic management, international business, sustainability, management education. E-mail: [email protected]. Theodore Peridis, Doctor, Professor, Schulich School of Business, York University; research areas/interests: strategic management, international business. E-mail: [email protected]. 1892

Toward an Integrated Theory of Sustainability

primarily sourced from China and Israel, as well as locally. In Greece, while financial incentives are in place, there isn’t as intense a push for alternative sources of energy, even though the country critically depends on expensive fossil fuel imports for its needs (Melander, 2009). Just a bit further east though, with substantively similar structural factors, the island of Cyprus has enthusiastically embraced solar energy (Jensen, 2000). Across the sea, residents of Israel are required by law to get their hot water from solar energy; consequently, a rather inexpensive, if not unattractive solar panel and water tank adorns every rooftop; domestic firms dominate the market (Maple, 2009; Rabinovitch, 2009; Thomas, 2008). Israel and Cyprus are the world’s leaders of solar water heaters with over 90% penetration among the countries’ households, yet Cyprus has little local production (Maxoulis, Charalampous, & Kalogirou, 2007). Moreover, dozens of Israeli start-ups are investing in research and development to develop new solar energy capabilities (Kloosterman, 2009). In all four countries, the penetration of solar panels is much higher compared to North America, even in sun drenched Southern United States and Mexico. Moreover, while both large and small firms are active players in the green energy movement, one observes a broad range of different strategies pursued by local and international firms. Contrast the responses from firms in the above countries to those in Canada. Canadians are genuinely concerned about the natural environment and are quite informed about issues of environmental stewardship and sustainability (Bord, Fisher, & O’Connor, 1998; Dunlap, Gallup, & Gallup, 1993; Franzen, 2003). The effects of global warming are well understood, yet, on a per capita basis, Canadians are the highest users of energy on the planet (International Energy Agency, 2010) and international environmental conservation organizations take issue with Canada’s energy intensive exploration of the oil sands in Alberta and Saskatchewan (Prebble et al., 2009; Yakabuski, 2011). In spite of the large landmass and suitable sites, Canada’s wind farms are still rather underdeveloped (Liming, Haque, & Barg, 2008). Traditional energy producers dominate the landscape. By contrast, recent reports from China suggest the emergence of a major drive toward green energy spearheaded by capital availability, incentives and regulation (Schreurs et. al., 2007; Bradsher, 2010); small and large firms alike are investing heavily in wind and other clean energy projects that are promising to make China the largest green energy producer in the world.

2. Literature Review This proposed research aims to move sustainability research from simply describing what organizations do toward a systematic understanding of the factors that guide organizations’ sustainability choices and strategies. Oriented in the strategic management paradigm, which originated from the industrial organization economics theory of structure-conduct-performance (s-c-p), this research initiative will identify the structural factors that guide firms’ sustainability choices as well as consider the role of culture in influencing these choices. Undeniably, regulations and financial incentives compel firms to undertake sustainability initiatives. A number of studies have compared regulatory frameworks of different countries ((King & King, 2005; J. I. Lewis & Wiser, 2007; Lund, 2009; Schreurs et al., 2009). Little though, has been done to connect regulatory frameworks with business strategies and substantiate potential connections between environmental regulations and corporate performance (Barnett, 2007; McGee, 1998; Rugman & Verbeke, 1998; Salazar, Husted, & Biehl, 2011). Industry structure has been shown to determine firm conduct but the research on sustainability strategies has not been able to link any traditional structural variables to sustainability choices (Husted, Allen & Kock, 2011). And although conceptual work has looked at the results of individual firms’ sustainability efforts, the results have been mixed. Available

1893


frameworks have not been able to convincingly connect sustainability strategies to performance offering alternatively positive and negative explanations (Margolis & Walsh, 2001; 2003; Husted & Allen, 2007; 2009). Barnett (2007) argues that firms’ ability to improve stakeholder relationships is instrumental in converting socially responsible behaviors to financial performance. While conceptually appealing, this line of work has not yet been followed by empirical research nor has it gone beyond the proverbial “it depends” to provide useful advice to managers and public policy makers with respect to choices of CSR strategies and their expected payoffs. Similarly, studies of environmental attitudes have attempted to capture individuals’ diverse predispositions toward the natural environment and their corresponding behaviors (Felonneau & Becker, 2008; Grob, 1995; Kaiser, Wolfing, & Fuhrer, 1999; Scott & Willits, 1994; Vining & Ebreo, 1992; Vitouch, 1996). A fair number of studies have focused on individuals’ values and their attitudes toward sustainability (Corraliza & Berenguer, 2000; Grunert & Juhl, 1995; Poortinga, Steg, & Vlek, 2004; Stern, Dietz, Kalof, & Guagnano, 1995) and projects have even attempted to conduct cross-national surveys (Ester, Vinken, Simões & Aoyagi-Usui, 2003; Leiserowitz, Kates & Parris, 2006; Oreg & Katz-Gerro, 2006; Schultz et al., 2005; Schultz & Zelezny, 1998). At the individual level, frameworks such as the New Ecological Paradigm (Dunlap & Vanliere, 1978; Stern, Dietz, & Guagnano, 1995), the Nature Relatedness Scale (Nisbet, Zelenski, & Murphy, 2009), and the environmental attitudes inventory (Milfont & Duckitt, 2009) have been developed to assess individuals’ tendencies toward the environment. Moreover, theoretical models under various names such as “value-attitude-behavior” or “value-belief-norms” have been adapted to link individuals’ environmental values to behaviors (Guagnano, Stern, & Dietz, 1995; Homer & Kahle, 1988; Kaiser, Hubner, & Bogner, 2005; Oreg & Katz-Gerro, 2006; Stern, Dietz, Abel, Guagnano, & Kalof, 1999; Vaske & Donnelly, 1999). These conceptual links have been admittedly complex and results have been mixed, although a stream of literature has attempted to explain the observed value-behavior gap and identify the obstacles to pro-environmental behavior despite environment friendly values (Bardi & Schwartz, 2003; Blake, 1999; Kennedy, Beckley, McFarlane & Nadeau, 2009; Kollmuss & Agyeman, 2002; Stern, 2000; Tilley, 1999; Torelli & Kaikati, 2009). What is more, these models describe individuals’ daily routines such as composting food scraps and walking or bicycling, but are inherently inconsistent with an understanding of the factors that guide organizations’ sustainability choices and strategies. This research proposal and design reflects a conceptual departure from prior perspectives which look for parallels between the individual and firm levels. Here, it is proposed that macro factors are salient influencers to sustainability behavior. Moreover, values, embedded within national cultures or what some call “collective mental programming” (Hofstede, 1980) decisively influence individual and organizational behaviors and as such, they play a dominant role in the making of these sustainability choices. Different models have been proposed and wide-scale comparisons of different cultures have been compiled (Hofstede, 1980, 2001; Inglehart, 1977; Schwartz, 1994a, 1994b) in attempts to codify human values. Cultures have been seen to incorporate individuals’ values and norms as they are informed by their perception of power distance, individualism, masculinity, uncertainty avoidance, and long term orientation (Hofstede, 1990). Altruistic and egocentric values are or ought to be reflected in these dimensions and shape environmental attitudes. In contrast, referring to ecological issues, Stern (2000) argues that social structures such as pertinent legislative configurations and financial incentives shape individuals’ values regarding environmental issues. Both positions support the presence of some relationships between structures, cultures and sustainability strategies but advocate very different relationships. Moreover, despite all this work, researchers have been unable to transfer many of the insights across different levels of analysis, particularly from individuals to organizations and across countries. The dependent variables 1894


have stubbornly remained at the individual level and when researchers have focused on the implications of cultural differences on organizations (Barney, 1986; Hofstede, 1990; Kogut & Singh, 1988; R. D. Lewis, 2000; Newman & Nollen, 1996; Schein, 1990), they have not considered these implications for sustainability issues. The proposed research explicitly addresses these weaknesses and is designed to explore and explain the structure, culture and conduct relationships that govern organizations’ sustainability strategies. Critically, even under similar economic, technological, social, political, and regulatory conditions substantial differences seem to arise among the behaviors of organizations, as well as their performance related to sustainability. Virtually nothing has been done to appreciate these differences, while arguably among all factors, the largest impact on the environment and society potentially arises from the actions of corporations. Academic fields such as management, political science, environmental studies, sociology, international business, as well as law and economics are asking similar questions (Barnett, 2007; Margolis & Walsch, 2003) highlighting the importance of the proposed research program and the potential contributions of its findings. The overarching network of macro factors may be depicted as follows: Structure Performance

Sustainability Strategies

Culture

Conduct

The literature has documented the impact of many aspects of the macro structure within which firms operate including industry configurations as well as legal and regulatory pressures. The inclusion of culture in the proposed research roadmap is significant given the impact of culture on underlying assumptions and values, world views, and mental models. The role of conduct is also important given the numerous and varied drivers of the choices firms make based on intentional and/or realized strategies, industry recipes, behavioral inertia, competitive imperatives, and types of engagement with stakeholders. All three macro factors should influence the sustainability strategies chosen by the firm and help explain the evolution of these strategies especially as they move from simple to deeply embedded. The performance (economic, environmental, and social) of the firm is a function of the specific sustainability strategy choices made within the larger context of the macro factors.

3. Methodology The first research question focuses on the structural factors and potentially different sets of factors that serve to inform the sustainability choices of firms. Given that neither existing knowledge nor available frameworks provide a parsimonious answer here, the proposed research needs initially to address theory building rather than theory testing; appropriately, we will first carry out a series of intensive case studies in four different countries with the aim of providing the basis for the development and calibration of the conceptual model. Using an

1895


inductive approach, the research design will follow traditional multiple case studies methodology (Eisenhardt, 1989; Yin, 1994). The design will allow us to draw on the strategic management and the sustainability literatures integrating and extending insights and constructs already developed within that body of work into the theory developed in the course of this research roadmap (cf. Fischer et al., 2007). The choice of countries is both opportunistic and fortunate. The researchers have extensive contacts in India, Israel and Mexico, and Canada and these countries will provide a point to departure for the first phase of the proposed research. The four countries exhibit substantially divergent scores in the World Bank`s development index, in regulatory, political and social structures, and in Hofstede’s cultural dimensions, affording the research design the requisite variety in structural factors and cultures. The cases to be developed will be chosen so as to represent a range of firms that engaged in substantial decision processes that led to the adoption of noteworthy sustainability strategies. We will use three sources of data to identify potential cases: (a) available annual sustainability reports, typically published by publicly traded firms; (b) the Corporate Knights annual survey of responsible business; as well as (c) scan the local public press for suitable candidates. Initially the focus will be on cases in the manufacturing and traditional energy sectors to ensure comparability, as well as diversity. The two sectors represent broad elements of any economy, and in particular the countries at the focus of the first stage of this research, and are typically identified as both major causes and potential solutions to sustainability. We plan to develop twelve case studies. This number is consistent with other case based research (e.g., Eisenhardt & Graebner, 2007; Zott & Huy, 2007), and we believe it is warranted given the sources of variation between cases and a need for considerable attention to inter-case differences in order to map out a theory that identifies the relationships between factors, cultures and strategies, as well as afford us sufficient base cases for subsequent longitudinal studies. Cases will explore firms’ sustainability strategies, analyzing the decisions that led to the “low hanging fruit” of pollution prevention, energy efficiency, and process improvements, as well as more complex strategies such as adoption of product stewardship, life cycle analysis and clean technology. Consistent with the literature on managerial cognition and decision theory (Peterson, 2009; but also Pink, 2006; Weitzner & Peridis, 2011), we will examine the decision making processes that lead to these strategic choices, including establishing objectives, search processes, criteria setting, evaluation of alternatives, alternative selection, as well as group think, and cognitive biases in search and selection. We will utilize both written material and interviews with the protagonists involved in the decisions. The recollections of managers and decision makers as well as prospective tracking will allow reasonable inferences about the strategic intent, and expose the cognitive and value based judgments that led to the adoption of specific strategies. The cases will allow a temporal view in order to capture the evolution of sustainability strategies over a period of time and study the dynamics of the decisions; and provide insight into the role of external conditions such as the introduction of new regulatory frameworks on firms’ subsequent actions. The cases will explore and document the richness of the phenomenon and provide insights into the underlying relationships that link structural conditions, cultures, conduct, and strategic responses. The use of explanation building modes of qualitative data analysis (Yin, 1994; Fischer & Reuber, 2004), to inform the conceptual model articulating companies’ decisions about sustainability strategies, will enable the first cut of theory building. An initial set of propositions about sustainability strategy formulation will be derived from this inductive phase. The comparison of the results of initial case studies with derived propositions, will occur iteratively going between case data and theoretical propositions. After completing the within-case analysis of each individual case, the next step will be cross-case analyses comparing cases based both on individual dimensions 1896


suggested by the literature as well as new dimensions that emerge through the process. Finally, once theoretical saturation is achieved, the resulting conceptual model will need to be operationalized to move to the quantitative research stage to systematically study the sustainability decisions and strategies of firms across countries and ascertain the impact of structural factors and cultural differences in sustainability choices. This second phase will entail testing the conceptual model and assessing the strength and direction of the relationships between structure, culture, conduct, strategy, and performance within the realm of business and sustainability. The main activities here will involve data collection and analysis. In transitioning to the quantitative phase of this project, one of the first steps will entail the development of reliable and valid measures of the five sets of constructs. Given the present miscellany of metrics (Barnett, 2007; Lund, 2009), it will be useful to look to the cases to also illuminate measures that will be consistent across countries and across sectors. The frame of reference for conceptualizing performance outcomes will originate from the dimensions of the Environmental Performance Index developed by Yale University, Columbia University and the World Economic Forum (Yale.org, 2010) and those of the Global Reporting Initiative (GRI, 2010), which collectively measure impact on environmental health, ecosystem vitality, and environmental protection applying a lens that effectively captures each firm’s triple bottom line. The case analyses will shed light on necessary adjustments to convert these dimensions to measures that can be utilized across countries and sectors. Data will be collected through survey instruments (mainly to capture strategy and performance) and public data (mainly to capture structure and culture). To correct for reporting biases, the collection of secondary data for the former (such as reports and financial statements) and primary data for the latter (for example awareness of policies and individual values assessments) will provide better reliability of key measures. To ensure adequate variation, the research will consider firms and situations that are and are not pursuing sustainability strategies. As such surveys are more likely to elicit responses from environmental stewards than laggards, it will be essential to address non-response bias by ensuring a broad representation and a diverse population in this study. Similar steps will be taken in each country. Moreover, econometric analyses will allow the use of Heckman (1976, 1979) corrections to control for potential selection biases. The analyses will utilize detailed data on different external factors that were present during the time that decisions were made including regulatory, technological, social, political, and economic macro variables and detailed data on values and cultures including altruism, uncertainty avoidance, masculinity, collectivism, and long term orientation. While the measures of strategic choices are yet to be determined, these will likely include utilizing metrics that capture firms’ choices across different dimensions; for example, the type and the intensity of different adopted strategies, the level of entrenchment and their longevity. Also, the necessary control variables will be included so that the analyses allows for meaningful comparisons across countries and industries.

4. Future Directions and Constituents The primary purpose of the proposed research roadmap is to develop an understanding of a set of related questions in a way that is useful to multiple key constituents. Some of these key questions and constituents are:  What are the reasons for such diversity in firms’ actions?  What are the implications?  Why do companies respond so differently and pursue dramatically diverse sustainability strategies across different countries?

1897


What could governments do to accommodate and respond to the diversity?  What conditions would better facilitate different government programs to achieve desirable results? Policy makers, practitioners, executives of multinational corporations and their subsidiaries, officers of intergovernmental organizations, as well as those of not-for-profit agencies could benefit enormously from answers to these questions as they would allow them to make informed choices about programs, investments, and strategies so as to pursue their objectives more effectively. Management practice: convert the insights gained from this research program to inform managers’ decision models with respect to sustainability strategies and shed light on the question of the relationship between sustainability strategies and firms’ overall performance, especially addressing the role of structural factors and cultures. Public Policy: inform the effectiveness of public policy initiatives that relate to sustainability and associated macro objectives; understand the influence of culture and external factors in the uptake of sustainability strategies and the usefulness of pertinent government policies. 

References: Bardi A. and Schwartz S. H. (2003). “Values and behavior: Strength and structure of relations”, Personality and Social Psychology Bulletin, Vol. 29, No. 10, pp. 1207-1220. Barnett M. L. (2007). “Stakeholder influence capacity and the variability of financial returns to corporate social responsibility”, Academy of Management Review, Vol. 32, No. 3, pp. 794-816. Barney J. B. (1986). “Organizational culture-Can it be a source of sustained competitive advantage?”, Academy of Management Review, Vol. 11, No. 3, pp. 656-665. Blake J. (1999). “Overcoming the ‘value-action gap’ in environmental policy: Tensions between national policy and local experience”, Local Environment, Vol. 4, No. 3, pp. 257-278. Bord R. J., Fisher A. and O’Connor R. E. (1998). “Public perceptions of global warming: United States and international perspectives”, Climate Research, Vol. 11, No. 1, pp. 75-84. Bradsher K. (2010, January 31). “China leading global race to make clean energy”, The New York Times, p. A1(L), available online at: http://www.nytimes.com/2010/01/31/business/energy-environment/31renew.html. Corraliza J. A. and Berenguer J. (2000). “Environmental values, beliefs, and actions—A situational approach”, Environment and Behavior, Vol. 32, No. 6, pp. 832-848. Dess G., Lumpkin G., Eisner A. and Peridis T. (2011). Strategic Management: Creating Competitive Advantages (3rd ed.), McGraw Hill Ryerson. Dunlap R. E., Gallup G. H. and Gallup A. M. (1993). “Of global concern-Results of the health of the planet survey”, Environment, Vol. 35, No. 9, pp. 7-39. Dunlap R. E. and Vanliere K. D. (1978). “The new environmental paradigm: A proposed measuring instrument and preliminary results”, Journal of Environmental Education, Vol. 9, No. 4, pp. 10-19. Eisenhardt K. M. (1989). “Building theories from case study research”, Academy of Management Review, Vol. 14, pp. 532-550. Eisenhardt K. H. and Graebner M. E. (2007). “Theory building from cases: opportunities and challenges”, Academy of Management Journal, Vol. 50, pp. 25-32. Ester P., Vinken H., Simões S. and Aoyagi-Usui M. (2003). Culture and Sustainability: A Cross-national Study of Cultural Diversity and Environmental Priorities Among Mass Public and Decision Makers, Amsterdam Dutch University Press. Felonneau M. L. and Becker M. (2008). “Pro-environmental attitudes and behavior: Revealing perceived social desirability”, Revue Internationale De Psychologie Sociale-International Review of Social Psychology, Vol. 21, No. 4, pp. 25-53. Fischer E., Otnes C. C. and Tuncay L. (2007). “Pursuing parenthood: Integrating cultural and cognitive perspectives on persistent goal striving”, Journal of Consumer Research, Vol. 34, No. 4, pp. 425-440. Fischer E. and Reuber R. (2004). “Contextual antecedent and consequences of relationship between young firms and distinct types of dominant exchange partners”, Journal of Business Venturing, Vol. 19, No. 5, pp. 681-706. Franzen A. (2003). “Environmental attitudes in international comparison: An analysis of the ISSP surveys 1993 and 2000”, Social

1898

Toward an Integrated Theory of Sustainability Science Quarterly, Vol. 84, No. 2, pp. 297-308. Grob A. (1995). “A Structural model of environmental attitudes and behavior”, Journal of Environmental Psychology, Vol. 15, No. 3, pp. 209-220. Grunert S. C. and Juhl H. J. (1995). “Values, environmental attitudes, and buying of organic foods”, Journal of Economic Psychology, Vol. 16, No. 1, pp. 39-62. Guagnano G. A., Stern P. C. and Dietz T. (1995). “Influences on attitude-behavior relationships—A natural experiment with curbside recycling”, Environment and Behavior, Vol. 27, No. 5, pp. 699-718. Heckman J. (1976). “The common structure of statistical models of truncations, sample selection, and limited dependent variables and a simple estimator for such models”, Annals of Economic and Social Measurement, Vol. 5, pp. 475-492. Heckman J. (1979). “Sample selection bias ans a specification error”, Econometrica, Vol. 47, pp. 153-161. Hofstede G. H. (1980). “Motivation, leadership, and organization—Do American theories apply abroad?”, Organizational Dynamics, Vol. 9, No. 1, pp. 42-63. Hofstede G. H. (1990). Cultures and Organizations: Software of the Mind, New York: McGraw-Hill. Hofstede G. H. (2001). Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations across Nations (2 nd ed.), Thousand Oaks, Calif.: Sage Publications. Homer P. M. and Kahle L. R. (1988). “A structural equation test of the value attitude behavior hierarchy”, Journal of Personality and Social Psychology, Vol. 54, No. 4, pp. 638-646. Husted B. W. and Allen D. (2007). “Strategic corporate social responsibility and value creation among large firms: Lessons from the Spanish experience”, Long Range Planning, Vol. 40, No. 6, pp. 594-610. Husted B. W. and Allen D. (2009). “Strategic corporate social responsibility and value creation: A study of multinational enterprises in Mexico”, Management International Review, Vol. 49, No. 6, pp. 781-799. Husted B. W., Allen D. and Kock N. (2011). “Value creation through social strategy”, Business & Society (forthcoming). Inglehart R. (1977). The Silent Revolution: Changing Values and Political Styles among Western Publics, Princeton, N.J.: Princeton University Press. International Energy Agency (2010). Energy Balances of OECD Countries, Paris: OECD. Jensen M. C. (2001). “Value maximization, stakeholder theory, and the corporate objective function”, Journal of Applied Corporate Finance, Vol. 14, No. 3, pp. 8-21. Jensen T. L. (2000). Renewable Energy on Small Islands, Copenhagen, Denmark: Forum for Energy and Development. Kaiser F. G., Hubner G., and Bogner F. X. (2005). “Contrasting the theory of planned behavior with the value-belief-norm model in explaining conservation behavior”, Journal of Applied Social Psychology, Vol. 35, No. 10, pp. 2150-2170. Kaiser F. G., Wolfing S. and Fuhrer U. (1999). “Environmental attitude and ecological behaviour”, Journal of Environmental Psychology, Vol. 19, No. 1, pp. 1-19. Kennedy E. H., Beckley T. M., McFarlane B. L. and Nadeau S. (2009). “Why we don’t “walk the talk”: Understanding the environmental values/behaviour gap in Canada”, Human Ecology Review, Vol. 16, No. 2, pp. 151-160. King N. J. and King B. J. (2005). “Creating incentives for sustainable buildings: A comparative law approach featuring the United States and the European Union”, Virginia Environmental Law Journal, Vol. 23, No. 3, pp. 397-459. Kloosterman K. (2009). “Seven solar technologies from Israel that could change our planet”, available online at: http://israel21c.org/environment/seven-solar-technologies-from-israel-that-could-change-our-planet. Kogut B. and Singh H. (1988). “The effect of national culture on the choice of entry mode”, Journal of International Business Studies, Vol. 19, No. 3, pp. 411-432. Kollmuss A. and Agyeman J. (2002). “Mind the gap: why do people act environmentally and what are the barriers to pro-environmental behavior?”, Environmental Education Research, Vol. 8, No. 3, pp. 239-260. Leiserowitz A. A., Kates R. W. and Parris T. M. (2006). “Sustainability values, attitudes, and behaviors: A review of multinational and global trends”, Annual Review of Environment and Resources, Vol. 31, pp. 413-444. Lewis J. I. and Wiser R. H. (2007). “Fostering a renewable energy technology industry: An international comparison of wind industry policy support mechanisms”, Energy Policy, Vol. 35, No. 3, pp. 1844-1857. Lewis R. D. (2000). When Cultures Collide: Managing Successfully across Cultures (2nd ed.), London; Naperville, Ill.: Nicholas Brealey. Liming H., Haque E. and Barg S. (2008). “Public policy discourse, planning and measures toward sustainable energy strategies in Canada”, Renewable & Sustainable Energy Reviews, Vol. 12, No. 1, pp. 91-115. Lund P. D. (2009). “Effects of energy policies on industry expansion in renewable energy”, Renewable Energy, Vol. 34, No. 1, pp.

1899

Toward an Integrated Theory of Sustainability 53-64. Maple S. (2009). “Solar energy—Why all eyes are on Israel”. Ezine Articles, available online at: http://ezinearticles.com/?Solar-Energy---Why-All-Eyes-Are-on-Israel&id=2700489. Margolis J. D. and Walsh J. P. (2001). People and Profits: The Search for A Link Between A Company’s Social and Financial Performance, Mahwah, N.J.: Lawrence Erlbaum Associates. Margolis J. D. and Walsh J. P. (2003). “Misery loves companies: Rethinking social initiatives by business”, Administrative Science Quarterly, Vol. 48, No. 2, pp. 268-305. Maxoulis C. N., Charalampous H. P. and Kalogirou S. A. (2007). “Cyprus solar water heating cluster: A missed opportunity?”, Energy Policy, Vol. 35, No. 6, pp. 3302-3315. Melander I. (2009). “Greece lags on renewable energy: The Boston Globe”, available online at: http://www.boston.com/news/world/europe/articles/2009/08/02/greece_lags_on_renewable_energy. Milfont T. L. and Duckitt J. (2009). “The environmental attitudes inventory: A valid and reliable measure to assess the structure of environmental attitudes”, Journal of Environmental Psychology. Newman K. L. and Nollen S. D. (1996). “Culture and congruence: The fit between management practices and national culture”, Journal of International Business Studies, Vol. 27, No. 4, pp. 753-779. Nisbet E. K., Zelenski J. M. and Murphy S. A. (2009). “The nature relatedness scale linking individuals’ connection with nature to environmental concern and behavior”, Environment and Behavior, Vol. 41, No. 5, pp. 715-740. Oreg S. and Katz-Gerro T. (2006). “Predicting proenvironmental behavior cross-nationally—Values, the theory of planned behavior, and value-belief-norm theory”, Environment and Behavior, Vol. 38, No. 4, pp. 462-483. Peterson M. (2009). An Introduction to Decision Theory, Cambridge: Cambridge Univeristy Press. Peridis T. (2007). Competing Ethical Standards: CCC and the Sale of Verifox in Africa, SSB Cases, Toronto. Peridis T. (2010). Hudson Beaver Creek’s Certified Farmed Salmon, SSB Cases, Toronto. Pink D. H. (2005). “A whole new mind”, New York: Riverhead, Penguin Group. Poortinga W., Steg L. and Vlek C. (2004). “Values, environmental concern, and environmental behavior—A study into household energy use”, Environment and Behavior, Vol. 36, No. 1, pp. 70-93. Prebble P., Coxworth A., Simieritsch T., Dyer S., Huot M. and Walsh H. (2009). “Carbon copy: Preventing oil sands fever in Saskatchewan”, available online at: http://pubs.pembina.org/reports/sask-carbon-copy-report.pdf. Rabinovitch A. (2009). “New solar farm adds hot water to cheap electricity”, Reuters, available online at: http://www.reuters.com/article/idUSTRE53J0H220090420. Salazar J., Husted B. W. and Biehl M. (2011). “Thoughts on the evaluation of corporate social performance through projects”, Journal of Business Ethics (forthcoming). Schein E. H. (1990). “Organizational culture”, American Psychologist, Vol. 45, No. 2, pp. 109-119. Schultz P. W., Gouveia V. V., Cameron L. D., Tankha G., Schmuck P. and Franek M. (2005). “Values and their relationship to environmental concern and conservation behavior”, Journal of Cross-Cultural Psychology, Vol. 36, No. 4, pp. 457-475. Schultz P. W. and Zelezny L. C. (1998). “Values and proenvironmental behavior—A five-country survey”, Journal of Cross-Cultural Psychology, Vol. 29, No. 4, pp. 540-558. Schwartz S. H. (1994a). “Are there universal aspects in the structure and contents of human-values”, Journal of Social Issues, Vol. 50, No. 4, pp. 19-45. Schwartz S. H. (1994b). “Beyond individualism/collectivism: New cultural dimensions of values”, in: U. Kim, H. C. Triandis, C. Kagitcibasi, S. C. Choi & G. Yoon (Eds.), Individualism and Collectivism: Theory, Method, and Applications, Thousand Oaks, CA: Sage, pp. 85-119. Scott D. and Willits F. K. (1994). “Environmental attitudes and behavior—A Pennsylvania survey”, Environment and Behavior, Vol. 26, No. 2, pp. 239-260. Stern P. C. (2000). “Toward a coherent theory of environmentally significant behavior”, Journal of Social Issues, Vol. 56, No. 3, pp. 407-424. Stern P. C., Dietz T., Abel T., Guagnano G. A. and Kalof L. (1999). “A value-belief-norm theory of support for social movements: The case of environmentalism”, Research in Human Ecology, Vol. 6, No. 2, pp. 81-97. Stern P. C., Dietz T. and Guagnano G. A. (1995). “The new ecological paradigm in social-psychological context”, Environment and Behavior, Vol. 27, No. 6, pp. 723-743. Stern P. C., Dietz T., Kalof L. and Guagnano G. A. (1995). “Values, beliefs, and proenvironmental action—Attitude formation toward emergent attitude objects”, Journal of Applied Social Psychology, Vol. 25, No. 18, pp. 1611-1636.

1900

Toward an Integrated Theory of Sustainability Thomas J. (2008). “90% of Israeli homes have solar water heaters”, MetaEfficient, available online at: http://www.metaefficient.com/hot-water-systems/90-of-israel-homes-have-solar-water-heaters.html. Tilley F. (1999). “The gap between the environmental attitudes and the environmental behaviour of small firms”, Business Strategy and the Environment, Vol. 8, No. 4, pp. 238-248. Torelli C. J. and Kaikati A. M. (2009). “Values as predictors of judgments and behaviors: The role of abstract and concrete mindsets”, Journal of Personality and Social Psychology, Vol. 96, No. 1, pp. 231-247. Vaske J. J. and Donnelly M. P. (1999). “A value-attitude-behavior model predicting wildland preservation voting intentions”, Society & Natural Resources, Vol. 12, No. 6, pp. 523-537. Vining J. and Ebreo A. (1992). “Predicting recycling behavior from global and specific environmental attitudes and changes in recycling opportunities”, Journal of Applied Social Psychology, Vol. 22, No. 20, pp. 1580-1607. Vitouch O. (1996). “Ecological attitudes predict political behavior in environmental affairs”, International Journal of Psychology, Vol. 31, No. 3-4, pp. 2835-2835. Weitzner D. and Peridis T. (2011). “Corporate governance as part of the strategic process: Rethinking the role of the board”, Journal of Business Ethics (forthcoming). Whelan C. (2008). “Is the sun setting on solar power in Spain?”, Scientific American, available online at: http://www.scientificamerican.com/article.cfm?id=is-the-sun-setting-on-solar-power-in-spain. Yakabuski K. (2011, October 5). “Keystone XL: More about the politics than the petroleum”, The Globe and Mail, RB1. Yin R. K. (1994). Case Study Research: Design and Methods, Thousand Oaks, CA: Sage.

1901


Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises Xin Jin1, 2, Song Chen1, Jie Wang2, Jinghua Xia3 (1. Tongji University, Shanghai, China; 2. Stanford University, Stanford, California, USA; 3. Landray Research Institute, Shenzhen, China)

Abstract: Knowledge Management (KM) has become widely employed as a new but promising management tool for optimizing business management and operations in China. In order for enterprises to develop a mature KM process capable of demonstrable benefits, they must assess their adoption of KM with regard to certain benchmarking practices. This paper, using a systematic approach, attempts to design and build a novel KM maturity (KMM) evaluation mechanism that combines and adapts existing models developed in the context of western management practices with new features from China’s unique culture and social background. First, the key criteria are extracted via empirical analysis of the results of an extensive questionnaire, completed and returned by hundreds of Chinese enterprises in 2012. Next, based on a quantitative assessment of KMM, we further analyze the KM benchmarking practices in China. Finally, we demonstrate that KM practice in China is in a transitional stage from a content-oriented approach to a community-practice-oriented and employee-growth-oriented approach; we also observe that the Chinese enterprises with better management practices and more advanced information technology infrastructure are experimenting with integrated application-oriented KM approaches that can better foster technology and management innovations in enterprises. Key words: knowledge management; knowledge management maturity; knowledge management maturity model JEL code: M19

1. Introduction Research on knowledge management (KM) and its applications in enterprise management can be traced back to the early 1960s (Davenport & Grover, 2001). However, it is generally recognized that knowledge management Xin Jin, Ph.D. Candidate, School of Economics and Management, Tongji University; research areas/interests: knowledge management and technical innovation. E-mail: [email protected]. Song Chen, Ph.D., Professor, School of Economics and Management, Tongji University; research areas/interests: technical innovation and management. E-mail: [email protected]. Jie Wang, Ph.D., Consulting Professor, Executive Director of Center for Sustainable Development and Global Competitiveness, Stanford University; research areas/interests: interdisciplinary research in information and knowledge management for innovation driven sustainable development, cross-culture and cross-disciplinary knowledge systems for smart infrastructures and smart city, and computational learning and data analytics for social science and engineering systems. E-mail: [email protected]. Jinghua Xia, Ph.D., General Manager of Shanghai Landray Consulting Co.; research area/interests: knowledge management. E-mail: [email protected]. 1902

Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises

as a well-established discipline commenced only after the mid-1980s (Lambe, 2011). Management guru Peter F. Drucker proposed that knowledge is a key aspect of modern enterprise development and a critical resource in the globalized economic environment (Drucker & Drucker, 1993). Obviously, in a knowledge economy, knowledge has become an important asset in any organization (Santanen et al., 2004). However, various survey results indicate that, for most enterprises, only 20% of corporate experience and knowledge is recorded explicitly (Zhang & Guo, 2009). Internal and external observations of many organizations show that a large amount of knowledge related to corporate core competencies and competitiveness becomes largely tacit knowledge, i.e., the knowledge is basically hidden and undocumented, and sometime even unrecognized. With the arrival of KM, it is believed that by optimizing the implementation of KM tools, enterprises can often capture a majority of the missing 80% of knowledge, systematically improving the competitiveness of these enterprises through improved use of KM as a critical weapon in a knowledge-based economy. In China, it has just been over a decade or so since the introduction of knowledge management (KM) as a major tool for sustainable enterprise competitiveness; retrospectively, a dramatically increasing interest in the study of KM and its applications, both in academia and industry, has been observed. Currently, KM has become widely employed as a new but promising management tool for optimizing business management and operations in China. But to be more objective, the standards and methodology of knowledge management in China are still in their infancy. In order for Chinese enterprises to develop a mature and reliable KM process capable of providing demonstrable benefits, they must assess their adoption of KM with regard to certain benchmarking practices. Fortunately, we can employ the idea of knowledge management maturity (KMM), a key concept for evaluating KM, to help Chinese enterprises better apply KM in their businesses. While most existing research has focused on theoretically developed descriptive models of KMM in China, there are only a very few studies based on realistic empirical data. To close this research gap in describing realistic KMM for Chinese enterprises and to promote more empirical studies of the success and failure of KM applications for them, we attempt to design and build a novel KMM evaluation mechanism that combines and adapts existing models developed in the context of western management practices with new features from China’s unique culture and social background. We hope that through this paper on KMM surveys and benchmarking studies, which uses a more systematic approach, we can shed light on the best practices for adopting KM within Chinese enterprises and provide a reference on experience and criteria for KM applications, mapping out a path that Chinese enterprises can follow.

2. Constructing Chinese KMM Model (KMMM) 2.1 Literature Review In recent years, KM has gradually become a hot topic in academia for management research and also in the business community. Various scholars and professional institutions put forth their views and understanding of KM. For example, Ponelis and Fair-Wessels assert that KM is a new dimension of strategic information management (Ponelis & Fairer-Wessels, 1998). Davenport and Prusakclaim that Knowledge Management is the process of capturing, distributing, and effectively using knowledge (Davenport & Lawrence, 1998). Skyrme suggests that Knowledge Management is the explicit and systematic management of vital knowledge along with its associated processes of creating, gathering, organizing, diffusing, using, and exploiting that knowledge (Skyrme, 2000). Charnell Havens and Ellen Knapp advocate that knowledge is rooted inhuman experience and social context, and

1903


managing it well means paying attention to people, culture, organization structure, as well as to information technology (Havens & Knapp, 1999). Through different perspective studies, one can observe that there are some similarities in the personalized comprehension of knowledge management. These similarities can be mainly summarized from two aspects: the knowledge process and knowledge management factors. From a practical viewpoint, knowledge management frameworks from the American Productivity & Quality Center (APQC) and McKinsey & Company, which represent state-of-the-art thinking on KM, provide a more comprehensive description of the concepts of knowledge process and knowledge management factors. APQC developed a Road Map to Knowledge Management Results (Berztiss, 2002). It includes two rings: knowledge processes and support factors. The Road Map to Knowledge Management Results includes four essential support factors: leadership and strategy, corporate culture, information technology and infrastructure, and performance evaluation. Only an optimized combination of these factors may contribute to the achievement of better knowledge management. Meanwhile, the knowledge process is composed of collection, organization, transformation, use, creation, identification, sharing process, etc., for various types of knowledge. The inner ring represents the value-added process of knowledge circulating within the organization, while the outer ring is the support factors that ensure knowledge sharing within the organization. The two rings complement each other. As a consulting leader and knowledge management pioneer, McKinsey & Company establishes a complete framework for Knowledge Management (Koenig & Neveroski, 2008). Its Knowledge Management framework consists of three levels: knowledge asset level—including individual and organizational knowledge assets; knowledge cycle level-including the initial accumulation and consolidation, integration, sharing, learning, use, innovation, feedback, etc. of knowledge; and critical success factors-including leadership and goals, organizational structure, organizational culture, and systems and infrastructure of information technologies. 2.2 Chinese KMM Model Construction Although there are a number of KMM studies, the research on how to design a KMM model that is suitable for Chinese enterprises is still largely missing. Therefore, this article focuses on analyzing the two dimensions of the knowledge process and knowledge management support elements for Chinese enterprises. Through its unique methodologies, this paper develops a roadmap model for knowledge management maturity that integrates a model of knowledge processes called “Wheel of Knowledge” (Bettoni & Schneider, 2003) to better deal with complex KMM contexts. In order to make a KMM model suitable for representing and analyzing China’s scenarios, this study interviewed 10 well-known knowledge management experts and scholars using the Degree Centrality method of network analysis to define and analyze the specific content contained on two key dimensions of the knowledge process and the knowledge management support elements. The Degree Centrality refers to the number of direct links to the nodes in the network. The number of links indicates the influence and importance of the node (Jun, 2009). In the dimension of the knowledge process, this study sorts and codes the key links based on the Degree Centrality method, then it establishes a relationship network matrix, processes the data using social network analysis software Ucinet 6, and finally uses the Net Draw function to draw a network diagram (see Figure 1). In Figure 1, circles represent the key links in the knowledge process and blocks represent the choices of the 10 experts. It can be seen that there are five major network nodes (shown as the circles) in the entire network center: they are knowledge precipitation, knowledge sharing, knowledge learning, knowledge application, and knowledge innovation. 1904


Correspondingly, this reveals that Culture, Management, and Technology constitute the three key elements of knowledge management support in China. In this paper, Culture refers to the trust and sharing knowledge culture in China; Management refers to management tools of standardized behaviors, knowledge assessment, motivation system and others; Technology refers to the means of IT support.

Figure 1 Network Nodes of the Knowledge Process

Figure 2

Network Nodes of the Knowledge Management Support Elements

Based on these two dimensions, this paper analyzes the level of Knowledge Management in Chinese enterprises. Enterprise Knowledge Management Maturity can be set in five levels. From low to high, there are Level One (Initial Level), Level Two (Awareness Level), Level Three (Basic Level), Level Four (Optimal Level), and Level Five (Innovative Level). On the basis of long-term research and practice of enterprise knowledge management, this paper summarizes some of the salient features of each level in Table 1:

1905

Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises Table 1 KMM level Initial Level 15-30points Awareness Level 31-45 points Basic Level 46-60 points Optimal Level 61-70 points Innovative Level 71-75 points

Relevant Features of KMM Level

Relevant features •Relatively closed, conservative, resistant to new things, lack of awareness of management knowledge •Lack of knowledge precipitation and sharing mechanisms and disordered knowledge production and use •Lack the necessary technical tools •Aware of the need for knowledge sharing, executives actively promote Knowledge Management •Consciously precipitate knowledge and prepare the appropriate documentation respective of the various categories of business •Establish document management and office automation systems •Mutual trust between members of the organization, good communication and knowledge can be shared •Precipitation of knowledge can be applied to all business practices •Have a basic Knowledge Management system, which is not effectively integrated in the enterprise business •Members of the organization work closely to seek progress and innovation, contributions to knowledge become a habit •Formed a continuously self-learning organization, best practice guides business efficient execution •Building knowledge portals, and effective integration throughout the business •The enterprise has become an efficient learning organization, each member works efficiently and lives happily •Continuous innovation, application of knowledge constantly optimizes business operations •Advanced IT System with knowledge mining and intelligent decision support

2.3 KMM Questionnaire Design Based on the five aspects consisting of precipitation, sharing, learning, application, and innovation, and the three support elements of technology, management, and culture, this study designed the questionnaire and analyzed the level of knowledge management to measure an enterprise’s knowledge management. To distinguish between “good businesses” and “general businesses,” this research collected a large amount of reliable data from many enterprises by repeated screenings and constant revisions. It finalized 15 major questions to design the questionnaire. These 15 questions constitute the basis of the Knowledge Management Maturity questionnaire, and use a Linkert5-point scale and forward scoring method design. Table 2 Knowledge process Five aspects Knowledge precipitation Knowledge sharing Knowledge learning Knowledge application

Questionnaire Design

Support elements Culture Management Technology Q1 Q6 Q11 Q2 Q7 Q12 Q3 Q8 Q13 Q4 Q9 Q14 Knowledge innovation Q5 Q10 Q15 Note: Q1: Staffs develop the habit of timely summarizing and reporting the experiences and lessons learned Q2: Staffs are willing to share experiences and knowledge with other colleagues Q3: Staffs learn and act with high spirits Q4: Staffs take full advantage of past experience and knowledge in work Q5: Staffs often actively put forward innovative solutions to solve the problems encountered in work Q6: Have appropriate management mechanisms and organizational models to ensure effective accumulation of knowledge Q7: Have the appropriate management mechanisms to ensure that staff to use a variety of formal and informal methods to share their experiences and knowledge Q8: Often organize staff to participate in internal training, book clubs, and other activities, and establish a mentoring mechanism Q9: Have the appropriate processes and systems to ensure that the accumulation of knowledge is fully utilized Q10: Have the appropriate processes and systems to ensure knowledge innovation Q11: Have (or be able to use) IT systems to effectively accumulate knowledge Q12: Have (or be able to use) IT systems to effectively share knowledge Q13: Have (or be able to use) IT systems to effectively support staff learning Q14: Have (or be able to use) IT systems to retrieve, access knowledge, and work together Q15: Have (or be able to use) IT systems to effectively support innovation

1906


2.4 Data Collection and Reliability This survey started in the first half of 2012 and lasted for half a year and involved more than a hundred domestic enterprises. The questionnaires were mainly distributed in provinces of higher informatization and developed economies in China, such as Beijing, Guangdong, Shanghai, Jiangsu, Zhejiang, and Shandong. A total of 385 questionnaires were distributed and 318 valid questionnaires were collected; valid questionnaires accounted for 82.5% of the distributed questionnaires. We believe that the survey can meet the needs of the study and the final research presents the current situation and development trends of knowledge management in Chinese enterprises. This questionnaire has three parts: basic information, KMM evaluation, and knowledge management trend survey. The industry distribution of this research is shown in Figure 3. IT, Manufacturing, Construction & Real Estate, and Finance are the top four industries. The positions of staff completed the survey are depicted in Figure 4.

Figure 3

Industry Distribution of This Research

Figure 4 Positions of Those Completing Survey

To assess the reliability of the KMM evaluation part of the questionnaire, this study uses Cronbach’s alpha approach (see Table 3). Since the Cronbach α coefficients of three single facets and total scales are greater than 0.8, the KMM evaluation part has attained a high level of reliability. In addition, the questionnaire had been pre-tested and amended several times before the final release. Therefore we believe the validity of the questionnaire is relatively high. 1907

Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises Table 3 Elements Culture

Management

Technology

Total

Cronbach α Coefficients Cronbach  0.847

Items Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q1-Q15

0.866

0.864

0.932

3. Status and Challenges of KMM in China 3.1 The Overall Status of KMM in China (1) The overall survey score of Knowledge Management Maturity is 50.4 points, which falls in the initial phase of Level Three. The overall survey score of organizational Knowledge Management Maturity is 50.4 points, which is in the middle stage of Level Three (Basic Level). Due to a certain degree of bias for the application of KM in the investigated organizations, we believe that the general Knowledge Management Maturity in China is not as optimistic as the survey indicates. The KMM may actually be in the initial stages of Level Three.

Figure 5

Questionnaire Score

From the scores of the 15 issues, it found that: organizations give Q4 “employees take advantage of and fully use past experience and knowledge” the highest score, which reflects the realistic and pragmatic enterprise culture; organizations give Q15 “have (or be able to use) IT systems to effectively support innovation” the lowest score, which reflects that organizations’ investments in implementing IT tools that support knowledge innovation lags behind.

1908


(2) Within the five aspects of knowledge process (precipitation, sharing, learning, application, and innovation), knowledge precipitation gets the highest score, and the lowest score is for knowledge innovation. Among those five aspects of knowledge process, knowledge precipitation gets the highest score, an average of 3.49 points. This indicates that Chinese companies have generally recognized the importance of knowledge precipitation and more are concerned about how to precipitate the organization’s core knowledge to reduce the risk to business. In contrast, in innovation, organizations get the lowest score, an average of 3.14 points. Therefore, Chinese enterprises need to strengthen the focus and practices in this area. Regarding the order of the five links’ scores, we find that the level for precipitation is higher than for application, application is higher than sharing, sharing is higher than learning, and learning is higher than innovation.

Figure 6 Scores of the Five Aspects of Knowledge Process

(3) In the three support factors of knowledge management (culture, management, and technology), technology is at the lowest level, which in turn affects the facilitation of the culture of the management and the implementation of the management mechanism. In the three support factors of knowledge management (culture, management and technology),organizations get the highest score in the culture dimension, an average of 3.42 points, which illustrates Chinese organizations generally recognize the dynamic role of the advocacy of “culture” for the construction of knowledge management. In comparison, in management, organizations get the second highest score, an average of 3.38 points, indicating that organizations fully intend to support implementation of knowledge management systems. Organizations get the lowest score in the technical dimension, an average of 3.27 points. There is knowledge management and the low level of technology, in turn, will affect the facilitation culture of knowledge management and implementation of the management mechanism. 3.2 The Important Findings of the KM Survey (1) K_S_ Asset and KM_ Input are the main factors for improving KM, although there is no correlation between KM_SP and KMM. The knowledge management trend survey portion of the questionnaire contained 12 questions. On the basis of a descriptive analysis of all the questions, this study conducts an exploratory analysis of some of the questions, and obtains the following findings. The questions included in the exploratory analysis are listed as follows: (a) What changes in your enterprise regarding knowledge management concerns and input have been made during the past 2-3 years? The response can be “the concerns and input have increased”, “the concerns and input

1909


have not changed”, “the concerns and input are decreasing”, or “do not know”. (b) Does your enterprise regard knowledge as a strategic asset? The response can be “Yes”, “No”, “Unclear”, or “Other”. (c) Has your enterprise developed knowledge management strategic planning? The response can be “Yes”, “No”, or “Unclear”. To simplify the description, here KMM represents enterprise Knowledge Management Maturity; KM_ Input represents enterprise knowledge management input; K_S_ Asset represents an enterprise regarding knowledge as a strategic asset; KM_SP represents an enterprise developing KM strategy planning. In this study, it uses these three questions’ response results to analyze the correlation of KMM and KM_ Input, KMM and K_S_ Asset, and KMM and KM_SP. This study applies the two-tailed significant test by Pearson simple correlation coefficient through SPSS18.0. The verification results are shown in Table 4.

KMM

Table 4 Variables Pearson correlation Sig. (2-tailed)

Correlation Coefficient of Variable KM_Input K_S_Asset 0.237** 0.367** 0.003 0.000

KM_SP -0.027 0.738

Note: ** Correlation is significant at the 0.01 level (2-tailed)

To further validate the causal relationship of KMM and KM_ Input, and KMM and K_S_ Asset, respectively, this study performs a regression analysis on them. The regression results are shown in Table 5.As it can be observed, K_S_ Asset and KM_ Input are the significant effects of predictor variables in KMM. In contrast, K_S_ Asset has greater impact on the effect of KMM (Beta = 0.332, p < 0.001), followed by KM_ Input (Beta = 0.170, p = 0.024), and the constant term also enters the regression equation. R2 value is not high, due to the fact that the KMM is influenced by many factors not included in this regression analysis. In addition, we find F = 15.189 and p < 0.001, indicating that the associated explanatory variables have strong interpretation powers, and the significance levels are acceptable. It can be concluded that there are causal relationships between K_S_ Asset and KMM, and also with KM_ Input and KMM. If an enterprise considers knowledge a strategic asset, its KMM can be improved by an impressive score of 8.486. If an enterprise knowledge management input has been increased, its KMM can be improved by a score of 4.259. The result also shows that there is no correlation between KMM and KM_ SP. The reason may be due to knowledge management already being embedded into the business processes of the enterprises and so it is already reflected in the overall strategic planning of the enterprises. Table 5 Variables Constant KM_Input K_S_Asset

B 41.982 4.259 8.486

Regressive Analysis Results

Unstandardized coefficients Std. error 1.785 1.875 1.906

Standardized coefficients Beta 0.170 0.332

t

Sig.

23.522 2.272 4.452

0.000 0.024 0.000

Note: R2 = 0.162, Adj. R2 = 0.151, F = 15.189, Sig. = 0.000

(2) In the question: “If your enterprise has or deems it necessary to formulate a knowledge management strategy, which department is responsible for the formulation process?” From the response analysis, it finds that: enterprises which establish KM departments to promote knowledge management gained the highest average score 1910


of KMM; enterprises which develop knowledge management strategies via the general manager or CEO gain the second-highest average score; the cross-sector decision-making committee gains a higher score than the information technology department. Table 6

Different Responsible Departments Gain Different Average Scores

Responsible department KM department General manager or CEO Cross-sector decision-making committee Information technology department Other

Average score of KMM 53.60 48.71 47.16 46.86 45.33

3.3 The Major Challenges of KM in China Currently, domestic industry in China does not have a unified standard definition of Knowledge Management. Different organizations vary on basic comprehension of knowledge management and positioning. This reality causes some obstacles in the promotion of knowledge management throughout China. This investigation finds that: 78.75% of the respondents believe that Knowledge Management is closely related to “knowledge-based corporate culture”; 75.63% of the respondents think that Knowledge Management is to “share excellent practice experience”; 73.75% of the respondents deem that Knowledge Management and “learning organization” are closely associated; in addition, 64.38% of the respondents believe that Knowledge Management and “innovation management” are very related. This indicates that more enterprises understand Knowledge Management from the perspective of management and culture, expecting that knowledge management can shape the culture of knowledge-based enterprises, strengthen enterprises’ sharing capabilities, develop learning organizations, and enhance their ability to innovate. A few years ago most people in China still considered Knowledge Management to be document management and IT systems, but results of this study reveal that the common understanding of KM has changed greatly just after a few years. However, there are still 51.25% of respondents who consider KM an IT tool, indicating that half of Chinese enterprises still lack a fundamental understanding of the critical role of KM. In addition to the non-uniform comprehension of Knowledge Management found in this survey, this research also finds that in the implementation of knowledge management, there are four major challenges that deserve special attention: (1) Lack of understanding and promotion by senior leadership; (2) Lack of incentives to encourage knowledge sharing; (3) knowledge management is not prioritized; (4) Due to time constraints, project participation by staff is insufficient. For “Lack of understanding and promotion by the senior leadership,” enterprises face the choice of “want or not want”. This issue needs to be regarded as a priority. If senior executives do not really understand and support knowledge management, knowledge management will lack the needed strategic driving force. The second challenge of “Lack of incentives to encourage knowledge sharing” makes enterprises face the choice of “willing or not willing”. Whether employees are willing to participate and contribute their knowledge is critical to the long-term viability of knowledge management. Therefore, the construction of performance mechanisms and incentives to support knowledge management is also a key factor for enterprises to consider. Knowledge management to some extent is the management of knowledge workers, to find management techniques which knowledge workers are willing to accept. For the third and fourth challenges of “Knowledge management is not prioritized” and “Due to time constraints, project participation by staff are insufficient”, they represent “able or not able” options for enterprise workers. If knowledge management is not a high priority,

1911


employees will not allocate time for knowledge management when they are busy with other work assignments. Therefore, how to integrate knowledge management into business processes rather than viewing it as a separate and additional task is the challenge for enterprises that want to design KM-enabled efficient and optimized business processes.

Figure 7

Differing Comprehension of Knowledge Management

Figure 8

Challenges of Knowledge Management Construction

4. Benchmarking Research of KM in China In the 1996 report of “The knowledge-based economy”, OECD clearly stated that distinctions can be made between different kinds of knowledge which are important in the knowledge-based economy: know-what, know-why, know-how, and know-who (OECD, 1996). Based on this classification, this research categorized the realistic knowledge management practices of Chinese enterprises, or the Chinese enterprise “knowledge capital”, into the following categories: Information Capital (know-what): Electronic management practices including structured information such as 1912


business data and unstructured information such as documents, records, messages, and others. Human Capital (know-who): People-oriented practices including staff skills, expert resources, relationships, etc.. Process Capital (know-how): Process-centric practices, including processes, systems, and methods on strategy, business, and functions. Strategy Capital (know-why): The core competitiveness of enterprise practices including proprietary technology, methodology, and business model, as well as issue-specific innovation mechanisms.

Figure 9

Constructing KM Practice Mode

The companies surveyed believe that the essence of Knowledge Management is “knowledge capital”. That means improved information capital, human capital, process capital, and strategic capital in the organizations, making this capital precipitate, share, learn, apply, and innovate in the organizations and create value for the enterprises. Based on the concept of knowledge capital and by combining the best practices of knowledge management in China, this paper summarizes the typical scenarios of knowledge management practice in Chinese enterprises as six different types of KM practice. They are knowledge content-oriented, community practice-oriented, employee growth-oriented, business process-oriented, integrated application-oriented, and innovation driven-oriented. And it defines the main features of the six typical practice modes of KM for Chinese enterprises as follows: 

Knowledge Content-oriented: Content-oriented KM that mainly employs scientific classification



Systems to build knowledge warehouses, to realize information capital effective precipitation and

management. It may also achieve the goals of avoiding intellectual capital loss and the accumulation of the core competencies. 

Community Practice-oriented: Community practice-oriented KM that emphasizes sharing knowledge

between people. It is based on the idea of virtual organizations and social networks, building platforms for knowledge exchange and interaction within the enterprise, mining and precipitating the tacit knowledge of the core staff, and shaping the open and innovative knowledge-sharing culture. 

Employee Growth-oriented: Employee growth-oriented KM is concerned with effective integration of the

knowledge management mechanism and human capital management system. It is designed to help employees continue enhancing their abilities to work. Combined with key positions to establish a complete system of 1913


scientific training to accelerate the upgrading of staff capacity, this is an important application of knowledge management in the field of human resource development. 

Business Process-oriented: Process-oriented KM is the application of knowledge in business processes. By

identifying the company’s core processes and effectively streamlining these core processes related to input, output, and knowledge support, it can realize the perfect combination of processes and knowledge to improve the efficiency of the implementation of the process and the quality of results. 

Integrated Application-oriented: Integrated application-oriented KM is an approach that integrates multiple

information systems to build a unified knowledge portal with unified search functionality across systems. It may also establish associated channels for ready-to-use on-demand knowledge resources so that it puts the right person at the right place at the right time to get the right knowledge. 

Innovation Driven-oriented: Innovation driven-oriented knowledge management aims to help enterprises to

explicate their own needs for achieving core competencies when necessary. It can identify, build, maintain, and create the core intellectual assets, and form a continuously optimized operational management mechanism for the enterprise. Based on the above six models, and with an analysis of the Most Admired Knowledge Enterprises (MAKE)award-winning enterprises of China in 2011 and 2012, this study sums up the main activities of the Chinese KM benchmarking enterprises in Table 7. Based on this summary, this paper draws the following conclusions on Chinese knowledge management characteristics: (1) For the MAKE award-winning enterprises, most of the domestic enterprises have completed knowledge content-oriented practice. Half of the enterprises are in the KM process of promoting communities of practice, employee growth, and business process-orientation. Some industry leaders with better management and information technology infrastructure are trying to practice integrated application-oriented and innovation driven-oriented KM. (2) As for the current situation of Chinese enterprises’ knowledge management practices, most of the Chinese enterprises’ knowledge management are still in the KM1.0 stage, mainly embodied in practice keywords including document management, content management, knowledge classification, knowledge warehouse, and knowledge map. Some enterprises’ knowledge management practices have entered KM2.0 stage. Its core idea for KM2.0is that to fulfill the KM2.0scenarios, the KM system will need the help of Web2.0, Communities of Practice, SNS, Mobile Network, knowledge push systems and other means to strengthen the level of knowledge application. Forward-looking enterprises have been trying to practice KM3.0. The core requirements of the practice is that in the fields of knowledge intelligence and on-demand data, knowledge mining, Semantic Web, knowledge cloud and open innovation, they need to be ready for more real-life applications (Carbone et al., 2012). (3) In conclusion, the next 3-5 years will be a rapid growth period of knowledge management in China. The field of knowledge management in China is in a transitional period from the start-upstage to the rapid growth stage. Knowledge management will become the choice of increasing numbers of enterprises to enhance their management practices and to guide their IT infrastructure optimization. During this process, more practical approaches of knowledge management will certainly emerge, making enterprise knowledge management more pragmatic and efficient.

1914

Investigation of Knowledge Management Maturity and Benchmarking Practices in Chinese Enterprises Table 7

Enterprise China Merchants Securities

China Asset

Kingdee

Ctrip China Merchants Property Baosteel Group China Merchants Bank Siemens China

NetDragon Tsingtao Brewery

Chinese KM Activities: Make Award-winning Enterprises in 2011 and 2012 Integrated Communities Employee Business Content-oriented application-o practice-oriented growth-oriented process-oriented riented Knowledge center, Experts network, Knowledge push Knowledge maps, Lecture room system Classic case Si Ku QuanShu (customer service CSR integrated library, service with the script library, knowledge training&learning warehouse library, office document library) Industry solutions think tank, Kingdee Enterprise Communities, management think Staff Microblog tank Knowledge Expert pool, Knowledge service warehouse, Knowledge center Knowledge map associations Management assets library, Product Post knowledge Working assets library, map knowledge map Project assets library Expert pool, Baosteel Technology sharing Knowledge technology transfer platform associations system 95555 Staff knowledge System, map, RoadShow Digital library Case management, platform University of CMB integration Knowledge Portal associations, Online training integration, Knowledge center Expert pool, system Knowledge Wikipedia mining Mobile learning platform, Study system Work circles NetDragon University Knowledge Online virtual Marketing warehouse university activities of KM

Innovation driven-oriented

Innovative studio, Baosteel professors

“3i” Innovation Management, Semantic Web Innovation value chain management

5. Conclusion With the continuous application of the maturity model in the field of knowledge management, how to build a Knowledge Management Maturity Model for Chinese enterprises has caused widespread concern in academia and industry. This article first analyzes and summarizes the theory and practice of knowledge management at home and abroad. Then, through a comprehensive application of two methods using expert interviews and social network analysis, it constructs a Knowledge Management Maturity Model from the two dimensions of knowledge process and knowledge management support elements, which reveals the basic structure and operational mechanism of Chinese knowledge management (JIN et al., 2013).

1915


Based on the questionnaire survey of 318 enterprises, it finds that overall, the Chinese enterprise KMM is at the initial phase of Basic Level. Through linear regression analysis, it is determined that “continued investment in knowledge management” and “regarding knowledge as a corporate strategic asset” will significantly enhance enterprises’ KMM, and enterprises which establish specialized departments to promote knowledge management have gained the highest average score for KMM. From the questionnaire survey, it also finds that Chinese enterprises’ knowledge management in general is facing these challenges: “want or not want”, “willing or not willing”, and “able or not able”. On the basis of quantitative analysis of the questionnaires, this paper analyzes the MAKE award-winning enterprises’ knowledge management practices in China in 2011 and 2012, and constructs six modes for knowledge management in Chinese enterprises. Through analyzing the award-winning enterprises’ knowledge management practices, this paper has established Chinese enterprises’ progressive development path from KM1.0, KM2.0, to KM3.0. It finds most of the Chinese enterprises’ knowledge management practices are still in the KM1.0 stage, some have entered the KM2.0 stage, and forward-looking enterprises have been trying to practice the KM3.0.It provides benchmarking practices and maps a path for other Chinese enterprises to promote knowledge management. References: Berztiss A. T. (2002). “Capability maturity for knowledge management”, 23rd International Workshop on Database and Expert Systems Applications, IEEE Computer Society. Bettoni M. C. and Schneider S. (2003). “The essence of knowledge management: A constructivist approach”, in: Proc. of the Fifth Intern. Conf. on Enterprise Information Systems, ICEIS, Citeseer. Carbone F. and Contreras J. et al. (2012). “Open innovation in an enterprise 3.0 framework: Three case studies”, Expert Systems with Applications, Vol. 39, No. 10, pp. 8929-8939. Davenport T. H. and Grover V. (2001). Special Issue: Knowledge Management, ME Sharpe. Davenport T. H. and Lawrence P. (1998). How Organizations Manage What They Know, Boston, MA, Harvard Business School Press. Drucker P. F. (1993). Post-Capitalist Society, New York, USA, Routledge. Havens C. and Knapp E. (1999). “Easing into knowledge management”, Strategy & Leadership, Vol. 27, No. 2, pp. 4-9. Jin X. and Chen S. et al. (2013). “Research on co-innovation pattern based on enterprise 2.0”, Science & Technology Progress and Policy, Vol. 29, No. 24, pp. 5-8. Jun L. (2009). Overall network analysis handout: UCINET software Practical Guide, Shanghai, China: Gezhi Press. Koenig M. and Neveroski K. (2008). “The origins and development of knowledge management”, Journal of Information & Knowledge Management, Vol. 7, No. 4, pp. 243-254. Lambe P. (2011). “The unacknowledged parentage of knowledge management”, Journal of Knowledge Management, Vol. 15, No. 2, pp. 175-197. OECD (1996). The Knowledge-Based Economy, Paris, France. Ponelis S. and Fairer-Wessels F. A. (1998). “Knowledge management: A literature overview”, South African Journal of library and Information Science, Vol. 66, No. 1, pp. 1-9. Santanen E. L. and R. O. Briggs et al. (2004). “Causal relationships in creative problem solving: Comparing facilitation interventions for ideation”, Journal of Management Information Systems, Vol. 20, No. 4, pp. 167-198. Skyrme D. J. (2000). “Developing a knowledge strategy: From management to leadership”, in: Knowledge Management: Classic and Contemporary Works, pp. 61-84. Zhang C. H. and Guo L. (2009). “Knowledge sharing model in group enterprise”, Studies in Science of Science, Vol. 3, p. 14.

1916


Managing the Virtual University: A Real Experience Yannet Liliana Mesa Medina (North Catholic University Foundation, Antioquia, Colombia)

Abstract: The North Catholic University Foundation is a virtual university, characterized by its virtual education model which obviously impacts its functions of teaching, extension, research, internationalization/outreach. This institution is private in nature, recognized by the Colombian government in 1997 by Resolution 1761 adopted by the Ministry of Education, and belongs to the Diocese of Santa Rosa de Osos, in the northern department of Antioquia. E-learning as an institutional mark, and ratified by the persistence of its policies, has over 15 years of institutional life, necessarily demands administrative and financial management to ensure sustainability, a point of balance and fulfilling the founding mission: to bring comprehensive education and contexts to houses where students live and work, with the mediation of information and communication technologies, ICT. In addition, it adapted, without overlapping or ignoring the Colombian education policy framework and business despite the natural gaps that still persist in these areas. In short, they are facing administrative and educational innovation. Socializing the administrative experience and financial management of the virtual school as a way of transferring knowledge is what motivates this paper. Key words: e-learning; virtual workers; virtual enterprise’ manage the unseen and planning phase JEL code: M190

1. Context: Virtual Educational Enterprise When ending the last decade of the twentieth century, the founders and directors of the nascent North Catholic University Foundation made the decision to create the first virtual college of Colombia, they also faced the challenge of another way to manage an institution. Indeed, 15 years after that decision, the experience of managing this virtual educational enterprise, matches with the reflections of Rojas (2006, p. 82) quoted—Boilers, Gonzalez de Celis, Barcia & Chacon (2010, p. 129)—in terms of tackling a new form of management characterized by understanding and responding to social and human change. This is characterized by a world embedded within the rapid development of information technologies and telecommunications. The North Catholic University Foundation is an institution of higher education in Colombia, belonging to the Diocese of Santa Rosa deOsos in northern Antioquia. It is private university, approved and recognized by the Ministry of Education, MEN. The organization is characterized by the virtual education model supported naturally in information and communication technologies, ICT, pedagogical-didactic take on meaning and communicatively—creative in the teaching-learning process. Consequently, the methodology is defined as the creation of virtual learning Yannet Liliana Mesa Medina, Business Administrator, Specialist, Financial Management Master in Economics and Finance, North Catholic University Foundation; research areas/interests: administrative and financial. E-mail: [email protected]. 1917

Managing the Virtual University: A Real Experience

environments, using ICT agreements to generate applied relevant knowledge, impacts and transform the contexts of interest of students, teachers and administrators. This model is defined as open, flexible and relevantly focused on identified needs and the learning pace of the learner, and allowing deliberate processes of interaction, collaboration and cooperation in the acquisition of learning and knowledge. In addition to the above, the virtual education model of this institution responds naturally to the substantive functions of teaching, research, extension and internationalization, as well as the demands of society and community. In three decades of institutional life there are students and teachers present in 91 municipalities in the department of Antioquia, 27 departments of Colombia, and in 17 countries (Angola, Argentina, Brazil, Cape Verde, China, Ecuador, Egypt, Spain, United States, Guatemala, Israel, Italy, Mexico, Peru, United Kingdom, Dominican Republic and Venezuela). Based on the above, from the experience of the Northern Catholic University Foundation, the virtual educational enterprise with a human response to the demand and need for training people in skills and abilities required for the globalized world for insertion into the new economy as Epper states (2004, p. 13) is driven by technology, telecommunications and the advancement of science. It is a staging of competent persons creatively measured, understand and manage forms, flexible and horizontal organizational structures which are based necessarily on the educational possibilities of communication and management of customized technological infrastructure that responds to the needs of the virtual education organization. Specifically, the virtual education model branding transversely determines the administration as to respond effectively to the challenge of ensuring the sustainability and equilibrium (Molina Restrepo, Jimenez Medina & Múnera Bureau, 2012) and fulfillment of the founding mission, with based on institutionalized processes, human talent experts, and the scope of the technology used. This approach is consistent with boilers, Gonzalez de Celis, Barcia & Chacon (2010, p. 123) in arguing that the virtual enterprise demands major changes in the organizational context that claim to be made systematically, and further characterized and focused on the client, from the definition of value and technology integrator (Rojas, 2006, p. 136) that enables such virtual enterprise.

2. Manage the Unseen The virtual workers cannot physically see those who are doing it gathered in one place, nor does it allow one to measure the amount of time they are present in a specific work site, wrote Jairo Jimenez Múnera (2005, p. 111), who exercised as to the date Chief Executive of the North Catholic University Foundation. This statement confirms the challenge of managing an organization of decentralized higher education in each workstation manager, administrator and teacher, who are in different geographic locations and contexts, but connected by the technologies used. He adds that Jimenez Múnera the characteristics of this institution motivated a series of questions to analyze any human, when sizing the implications of the administrative process applied to a virtual educational enterprise, these approaches and possible responses from experience are presented in the following Table 1:

1918


Table 1 Approaches and Possible Responses From Experience • The virtual nature of the Northern Catholic University Foundation allows its teachers to be fully tele-workers, like 70 percent of employees with administrative functions. • It is not by meeting timetable requirements, but validated by the results of the projects undertaken and contained in the action plan address registered with the appropriate planning. How do you measure the time • It adopts the principle of flexible working for employees, according to planning each working in the virtual? agency and institution of the people who make up the teams. • It raises awareness and enables the employee and teacher at the effective time management, communication culture, appropriation and use of utilities, services and technology tools endowed by the institution. • It has regulated and institutionalized an intranet as a disposal of institutional activity and action plans, reports, previews, results of projects that develop each area. Quarterly review of progress and achievements of the projects contained in the action plans How to supervise an employee who in accordance with the targets. has no fixed schedule or date on the • Processes of horizontal communication among dependencies and individuals for calendar? collaborative and cooperative work. • Performance evaluation and improvement plan • Programs and strategies for welfare support designed to ensure satisfaction and quality of life of employees (telecommuters, teachers, administrators).

According to the above, the fact remains that the organization adopts and adapts the classic administrative process stages of Planning, Organization, Implementation, Coordination, Control and Evaluation in the virtual enterprise (Molina Restrepo, Jimenez Medina & Múnera Bureau, 2012). Experience has taught us to conclude that the administration as a science or art has universal principles adaptable to any company or institution. In addition to the above, the different management models—Strategic Management, Management by objectives, among others—are also adaptable to a virtual university type educational enterprise. It is therefore possible to demonstrate the achievement of mission objectives and ensure the sustainability and growth with any of these models without differentiating the effectiveness of either. The planning phase is considered key in the virtual educational enterprise, because it claims anticipated and budgeted demand for services and products required by the university community in developing the substantive functions of teaching, research, extension and internationalization. Additionally, at this stage it is considered compliance and respect for education and business regulatory framework in Colombia, without flooding them or ignoring them, despite the natural gaps that still exist in these areas. In connection with the phase of organization is the immediate consequence of the virtual educational service planning and administrative institutions. The organization then takes the form of shares of enrollment and academic record, formalization of hiring teachers with the appropriate educational placement (teaching and research), and finally, access to training platform for students and teachers. According Múnera Jimenez (2005, p. 114) the organization also refers to the use of communication channels and loyalty technology infrastructure for educational and administrative management of the virtual educational organization, which also requires to quantified preferentially. Phases of Implementation, Coordination, Control and Evaluation in the virtual enterprise are specified in the zeal to articulate the performance, developments, products and developments from all units in accordance with the Development Plan, strategic objectives and goals. Central to the administration of virtual educational enterprise is the care required to manage financial resources just to ensure sustainability and breakeven (Molina Mesa Restrepo, Jimenez Medina & Munera, 2012) referred to above. No wonder Múnera Jimenez (2005, p. 117) stresses that. The resources needed that develop a virtual educational enterprise are very significant at first, because the required equipment and software licenses are expensive, especially if it is gigantic in scope. On this basis, the

1919


Northern Catholic University Foundation proposed vision and mission as a leader in virtual training, to carry and maintain eight higher education programs of undergraduate, specialization and virtual high school. This was (even still) a huge challenge that could not be paid by the university community or by the Diocese of Santa Rosa de Osos as a pioneer and patron of the project. Who, then, could sponsor such a special and benign idea of a virtual university? There were two options: the state and private enterprise. At first, the Colombian government supported the idea through a grant, but it was the private company which continued to support this feat, and that the work is in place today, have been instrumental in the aid and grants from various institutions (... .) But sustainability is not fully guaranteed even when it counts on reliable technology infrastructure, skilled human talent, and a portfolio of educational products and virtual services, because the natural dynamics of education in any country is that the retention of students is a five-year project. Additionally, it has the aspect of defection which also occurs in virtual education. So, decisions and strategies are required to contribute to enrich and update the portfolio of services and virtual product quality. From the experience of this institution will address the following strategies and lines of work:  Strengthening of university extension through agreements and alliances with other universities or companies.  Emphasis on quality and continuous improvement in the governance, performance and teaching staff, quality of virtual learning environments.  Partnerships and programs with the state for education and training  Programs with curricula containing courses and common areas.  Training programs and tailored training and business needs.  Applied research in different lines derived from virtual education.  Design and production of digital content and virtual learning environments for training of different audiences. In this section, we consider the differential educational materials for education and training and those for commercial demand (e-learning).  Enrollment of students quarterly or permanently.  Educational marketing strategies through various means. 2.1 Profile of Human Talent to Manage the Educational Enterprise Virtual The profile, skills and abilities ideal of the administrator and other human talent that is now calling for a virtual university, begins with having people with an open mind (attitude) to labor from anywhere in the world based on ICT (computer and mobile devices with access internet) without attachment to bricks and cement offices. That is, people capable of working with the option of tele-working, which contribute to achieving corporate objectives (boilers, Gonzalez de Celis, Barcia & Chacon, 2010, p. 138). In the administrative structure professionals who know very well the operation of educational enterprises are required, and preferably with appropriate experience as educators in various positions and at various levels of training (Table Molina Restrepo, Jimenez Medina & Munera, 2012). Additional capabilities, skills and abilities to be managers, in all matters concerning the systematization, information technology and the use of ICT, both corporately and in the educational training methods. Therefore, the human talent required is an interdisciplinary team of people from their disciplines, knowledge and responsibilities that add value and build true virtual university daily. Required, and then is a list comprised of systems engineers, business administrators, teachers, education graduates, journalists, ICT experts, psychologists, social workers, among other disciplines. 1920


In general, then presents the skills demanded and skills profile of the manager and employee of the virtual educational enterprise:  Skills demonstrated in the use and appropriation of information and communication technologies.  Be creative and proactive in adding value to individual performance.  Work collaboratively and cooperatively with the (s) of person (s) dependence, or team assigned  Communication skills (oral, scriptural) demonstrated to interact quickly, timely, efficient, and effective.  Leadership skills in planning, development and evaluation.  Ability to solve problems.  Respect for intellectual property, copyright section.  Ability to work in the workplace assigned by the institution or in the form of telework.

3. By Way of Conclusion: Lessons Learned, Challenges This journey through the administrative experience of the virtual educational organization leaves the academic community and those interested in the topic, a set of reflections that the Northern Catholic University Foundation delivered as the transfer and social appropriation of knowledge. In this regard, a general outline of these reflections by way of conclusions: 3.1 Lessons Learned  The virtual university is more accessible in the economic and logistical for the profile of young-adult student. Not so easy to finance as a company.  Adoption of alternative employment such as teleworking given the nature of the virtual university, Case Foundation Catholic University of the North, the whole plant of teachers’ works under this figure contract. Additional teleworkers are also 70 percent of the clerks or staggered days and times (twice a week move to the office).  The virtual education model determines the administrative transversely as to respond effectively to the challenge of ensuring sustainability and balance point of the educational organization.  The infrastructure and technological equipment is expensive and becomes obsolete in a short time, why it is essential to decide on a model (educational and administrative) to acquire the technology that best meets the needs identified.  Modern management of virtual educational enterprise cannot be separated from a strategic marketing focus on the student and education and training needs for life and throughout life.  The virtual university goes where the student lives and works nearby in context, therefore, the methodology avoids some displacement and migration to urban centers and large cities has the largest supply of education.  The virtual university has to focus on their educational and administrative management that makes it viable and sustainable over time, so it is advisable to outsource the technology (platforms, data center, housing, servers, among other services).  The virtual university is real, it is the university's present and future.  The quality of training of the virtual university is similar to that offered by educational programs and services and distance learning methodologies.  Training and continuous training of human talent in the organization to update or improve the skills and abilities (competencies) own topics of education and management of virtual educational organization.

1921


Desertion is the greatest enemy of development and maintenance of virtual educational organization.  It is necessary that the curriculum of virtual education programs offered provide for common courses with the appropriate requirements and quality.  The design and production of virtual learning materials are different when education and training purposes, that when the objective is commercial or business response type e-learning.  The premise of quality has to be the strategic objective of the virtual university cross. Research and apply or to update and innovate new ways and creating learning environments mediated by ICT. The walls and columns of the virtual university are its virtual model, the processes of interaction between areas and community of scholars who make up the organization. All supported by a robust and reliable technology infrastructure that allows fully implement functions of teaching, research, extension/outreach and internationalization. 3.2 Challenges  To change of culture and mentality of a broad academic and state that considers only the classroom agora of knowledge and methodology based on traditional transmission teacher-student. In short, break the paradigm of more than 2000 years classroom teaching and learning.  A proper administration of virtual educational enterprise has to propose innovative and successful strategies in an effort to retain and engage the student and other human talent skilled in the necessities of virtuality.  The virtual educational enterprise as the University of inclusion and opportunity for people with disabilities or living in places where traditional institutions fail for reasons of physical distance or because these contexts are not profitable for most of these institutions.  Fully virtual university to form human beings that impact and solve problems in contexts where living and working nearby.  The virtual university as a platform for training of human beings for inclusion in the so-called new economy characterized by technology and telecommunications art. Literacy and the new “digital illiteracy” of this century according to the trends and convergence of digital and globalized world. 

References: Calderas J. R., González de Celis G., de Barcia E. and Chacón R. (2010). “De la empresa tradicional a la empresa virtual: valores transformativos. Negotium: Revista Científica Electrónica Ciencias Gerenciales”, Fundación Miguel Unamuno y Jugo Venezuela, Vol. 6, No. 17, pp. 122-153, available online at: http://bit.ly/wpuLvk. Epper R. M. (2004). “La torre de marfil de la nueva economía”, in: Enseñar al profesorado cómo utilizar la tecnología: buenas prácticas de instituciones líderes Barcelona, España: Editorial UOC: Colección Educación y Sociedad Red. Ferrer A. (2010). “El cambio y su incidencia en la gestión tecnológica”, El Universal, pp. 3-4. Jiménez Múnera J. A. (2005). “La gestión administrativa de la educación virtual”, in: A. A. Sánchez Upegüi (Ed.), Educación virtual: reflexiones y experiencias, Medellín, Colombia: Fundación Universitaria Católica del Norte, pp. 111-118. Molina Restrepo M. L., Mesa Medina Y. L. and Jiménez Múnera J. A. (2012, marzo). “Interview by N. D. Roldán López: Personal Interview—Cómo se ha administrado la Fundación Universitaria Católica del Norte?”, Santa Rosa de Osos, Antioquia. Rojas Luis Rodolfo. (2006). “Los retos de la gerencia en la sociedad de información”, Revista NEGOTIUM: Ciencias Gerenciales, Vol. 2, No. 5.

1922


Convergence or Divergence between European HRM and American HRM* Gurhan Uysal (School of Business, Ondokuz Mayıs University, 55139 Samsun, Turkey)

Abstract: American HRM has shareholder perspective, and European HRM has stakeholder perspective. This creates divergence between HRM of both markets. This divergence makes social context important in European HRM. European HRM aims to protect interests of society; therefore, there is involvement of social partners in European HRM. American HRM aims to protect interests of shareholders in firms. Thus, American HRM is performance-oriented. Divergence may appear due to economic system difference between two markets. USA has liberal market economy view, and EU has coordinated market economy view, and EU is regulated economy. Thus, firms are autonomous in American HRM due to liberal market view. EU firms are restricted in HRM due to social awareness of regulated economy. However, firms are becoming deregulated in EU economy similar to USA economy. There is debate in academic literature, if there is European HRM or HRM in Europe. European firms apply European HRM recently. European HRM covers American HRM and social partners. Social variables make European HRM divergent from American HRM. European firms are adopting firm performance orientation in HRM. This makes European HRM convergent with American HRM. Key words: European HRM; American HRM; social context; convergence; divergence issue JEL codes: M12, M16, M21

1. Introduction Firms in global economy apply of HRM practices in their organizations. Definition of HRM is, HRM is implement of HRM practices in firms. This is current stage of HRFM in global firms to apply HRM practices. Practices of HRM are, for example, compensation, career planning, performance appraisal, reward, HR planning, staffing, etc. For example, reward is important practice in English HRM. They reward prominent employees in their organizations. American HRM has two basic objectives: performance and behaviours. American HRM aims to increase performance of their employees; and they aim to have an impact on firm performance. Secondly, American HRM aim to develop positive organizational behaviours in employees. American HRM aims to obtain expected employee behaviours such as motivation, satisfaction, commitment, citizenship behaviour, not intention to leave behaviour. Because American HRM perceives their employees as human capital. And they don’t want to lose out *

This paper is accepted for presentation in conference at 9th Silkroad International Conference, held by International Blacksea University, Tbilisi, Georgia on May 23-25th, 2014. Gurhan Uysal, Ph.D., Associate Professor, School of Business, Ondokuz Mayıs University; research field/interests: human resource management. E-mail: [email protected]. 1923

Convergence or Divergence between European HRM and American HRM

their human capital. Because human capital requires long time to possess. In addition, positive behaviours also have an impact on individual performance. It is expected that positive employee behaviours increase individual performance of employees. Because employees that have positive behaviours are expected to perform greater efforts at job. So, this study discusses American HRM, and it argues convergence and divergence debate between American and European HRM.

2. American HRM: Performance Focus American HRM is based on configurational perspective. It builds HR system in management of HR in firms. For example, Prof. Huselid advocates HPWS, HR systems and HR architecture in HRM theory. Secondly, American HRM has shareholder perspective. Because USA is shareholder economy. Therefore, American HRM has firm performance targets to provide more value to shareholders. Thirdly, American HRM considers employees as a resource. This is aligned with resource-based theory of Barney (1991). In addition, American HRM has two priorities. They are resource-based view and positive organizational attitude and behaviours. American HRM aims to develop positive organizational attitude and behaviours in employees, and it perceives employees as resources. Secondly, American HRM aims to increase firm performance to provide more value to shareholders. Therefore, American HRM is involvement of HRM with corporate strategy. That is defined by Devanna et al. in 1981. Stakeholders are important in European HRM. Therefore, European HRM has stakeholders perspective. Stakeholders are such as involvement of state, institutions, their regulations, trade unions and social partners. Therefore, European firms consider stakeholders in management of HR. Secondly, European HRM head for adopting American HRM but they have social context and environment. Therefore, European HRM can be formulated as European HRM= (American HRM) + (social context).American HRM is to apply individual HRM practices in firms. In addition, they also consider HR system notion. State regulations, social partners, institutions make up social context in European HRM. Calculative HRM is in favor of performance variations in HR. Collaborative HRM is based on psychological contracts between employees and firm. And it aims to develop positive organizational attitude and behaviours. Collaborative model is called as Harvard model. Calculative model is called as Michigan model. Collaborative model is applied in European HRM. HR is becoming strategic partner in American firms since 1990s. HR becomes strategic partner via employer of choice, high performance work systems, and a set of incentives. American HRM is based on performance focus both individual and firm. European HRM is becoming more performance oriented. So, convergence between European and American HRM may appear with performance variables. American firms are more autonomous than their European counterparts in HRM. European firms are restricted in HRM to protect stakeholders. However, European economies become market deregulations recently.

3. European HRM First paper in HRM in Europe is published in 1987 in France. It questions that HRM is applied in Europe. European HRM has contextual perspective, Prof. Brewster disagrees with universalistic perspective of HRM in 1924


European HRM. American HRM has configurational perspective. Therefore, this perspective variations may result in divergence between American and European HRM. But European HRM goes to deregulations in HRM that may decrease impact of contextual perspective on HRM. HRM is developing field in Europe, and American MNCs deploy American HRM in European markets. European HRM is adopting firm performance target. Thirdly, European HRM is adopting firm performance target similar to American HRM. Major difference between American and European HRM is state deregulations such as regulations in staffing, dismissal, industrial relations. Fourthly, European HRM has contextual perspective. Role of state, institutions, trade unions, social partners, labor legislations build social environment in European HRM. Chris Brewster, Paul Sparrow, A. Hegewisch, Wolfgang Mayrhofer, and Paul Gooderham are leading researchers in Europe in European HRM. Most citations are given to Prof. Chris Brewster in European HRM research. Chris Brewster claims that there is a European HRM. On the other hand, Paul Sparrow says that there is “HRM in Europe” notion. European HRM is between American HRM and social context. HRM in Europe concept adopts American HRM practices. Major difference between American and European HRM is organizational autonomy. Free economy culture in USA economy results in organizational autonomy in American HRM. American HRM is more autonomous in management of HR. While European HRM is restricted by stakeholders and regulations in HRM. This difference may emerge with differences between economic systems in both markets. USA is liberal market economy, and firms are free in management of HRM. EU economy is coordinated market economy, and EU is regulated economy. Therefore, firms have restrictions in HRM by state and regulations. For example, institutional’s regulations in Europe reduce flexibility of firms in management of HR. Major disadvantage of social context in European HRM is reduced employment flexibility. Firms are autonomous in American HRM to protect state. Americans aim to protect their state; therefore, firms are aoutonomous in HR. However, Europeans aim to protect their society and social rest. Therefore, firms have restrictions in HR. Involvement of state makes European HRM distinct from American HRM. However, European HRM goes to state deregulations in HRM recently, which is similar to American HRM. A German practitioner stresses that major difference between American and European HRM is state regulations in staffing, dismissal, industrial relations. For example, French labor law strongly affects training practices in France. However, Germany recently adopts deregulations in HRM instead of strict employment rules. That may increase flexibility in HR in European HRM similar to American HRM.

4. Convergence or Divergence Convergence between American and European HRM may be achieved through MNCs. Because American MNCs insist of applying headquarter’s HRM practices in local subsidiaries. This spread American HRM across European market. Divergence may be achieved through social context. Because there is economic system difference between two markets. USA is shareholder economy, and EU is stakeholder economy. EU aim to protect stakeholders in HRM. Therefore, they have state interventions, institutional impact and regulations in European HRM. American HRM has greater centralization in HRM in MNCs’ subsidiaries. MNCs force subsidiaries to apply headquarter’s HRM practices. American HRM resists against institutional requirements in host countries. And this insist deploys American HRM worldwide.

1925


5. Conclusion To conclude, EU is coordinated market economy and regulated market economy. Regulated market economy considers social partners and legislative framework in market decisions. European HRM has restrictions in HRM due to coordinated market economy. Because firms have to consider stakeholders and social partners in HRM. American HRM has organizational autonomy due to liberal market economy. Secondly, European HRM is becoming similar to American HRM. It is convergence between European and American HRM. There may be three effects for this convergence. European HRM apply of American HRM practices in firms. And secondly, European HRM adopts firm performance objective in management of HR. In addition, European firms apply American HRM practices. Because MNCs deploy HRM practices across Europe. Thirdly, Europe is becoming deregulated economy, and firms are becoming autonomous in HR. In conclusion, MNCs lead to convergence in global HRM, and national institutions and their regulations lead to divergence in global HRM. Rising of MNCs in global economies unify HRM; therefore, it leads to convergence in HRM. MNCs force local subsidiaries to adopt their HRM practices. On the other hand, national institutions force firms to comply with their law obligations.

6. Discussion: Summary To conclude, this paper argues convergence or divergence debate between American and European HRM. Basically, American HRM has 4 priorities: First, American HRM is autonomous. American HRM has organizational autonomy in management of HR. It is based on liberal market economy view on USA economy. Secondly, American HRM aims to increase individual performance of their employees. Thus, they apply HRM practices in their organizations to increase performance of their employees. Those practices are, for example, training, performance appraisal, career planning, HR planning, recruitment etc. Thirdly, American HRM is based on configurational perspective. It is based on HR system. Most of American HRM scholars advocate advantage of HR system in management of HR. These scholars are, for example, J.B. Arthur, M. Huselid, and J. MacDuffie. HR system represents interrelationship of individual HRM practices, and it has greater impact on individual performance than individual HRM practices. Furthermore, English HRM is based on individual HRM. They apply individual HRM practices in their organizations. Fourthly, American HRM perceives their employees as resources. Employees are resources in HRM. Therefore, HRM is not dismissal of employees. There should not be lay offs in firms in management of HR. Because employees are distinctive resources. Because American HRM considers this resource as a human capital. If firms fire out their employees, they fire out their human capital. European HRM is becoming more performance oriented, and European HRM apply of HRM practices in firms. Therefore, European HRM is becoming similar to American HRM. In addition, European HRM adopts state deregulations in management of HR. However, divergence exists between two markets due to social awareness in European market. European HRM has stakeholder perspective, and American HRM has shareholder perspective. Because Europeans consider social partners in management of HR. and so, European HRM has social context in management of HR. Thus, European HRM has contextual perspective in HRM, while American HRM has configurational perspective in HRM.

1926


Finally, EU is regulated market economy. Therefore, European firms witness restrictions in management of HR. On the other hand, American firms have organizational autonomy in HR because of liberal market view of USA economy. References: Barney J. B. (1991). “Firm resources and sustained competitive advantage”, Journal of Management, Vol. 17, pp. 99-120. Becker B. E. and Huselid M. A. (1992). “Direct estimates of SD, and the implications for utility analysis”, Journal of Applied Psychology, Vol. 77. No. 3, pp. 227-233. Boselie P. and Paauwe J. (2004). “Human resource function competencies in European companies”, ERIM Report Series Reference No. ERS-2004-069-ORG. Boselie P., Paauwe J. and Jansen P. (2000). “Human resource management and performance: Lessons from the Netherlands”, ERIM Report Series, ERS-2000-46-ORG, Erasmus Research Institute of Management. Brewster C. and Larsen H. H. (2000). “The Northern European dimension”, in: Brewster C. & Larsen H. H. (Eds.), Human Resource Management in Northern Europe, Oxford, Blackwell Business. Brewster C. (2007). “A European perspective on HRM”, European Journal of International Management, Vol. 1, No. 3, pp. 239-259. Brewster C. (1995). “Towards a European model of HRM”, Journal of International Business Studies, Vol. 26, No. 1, pp. 1-21. Brewster C. (1995). “Industrial relations and human resource management: A subsersive European model”. Industrielle Berziehungan, Vol. 2, No. 4, pp. 395-413. Brewster C. (1992). “European human resource management: Reflection of, or challenge to, the American concept?”, Cranfield School of Management, Cranfield Institute of Technology, Cranfield, UK. Burbach R. and Royle T. (2010). “Global integration versus local adaptation of E-HRM system in a US MNC”, 3rd European Academic Workshop on Electronic Human Resource Management, pp. 249-306. Cesyniene R. (2008). “Globalization and human resource management”, Ekonomika, Vol. 82, pp. 41-56. Claus L. (2003). “Similarities and difference in human resource management in the European union”, Thunderbird International Business Review, Vol. 45, No. 6, pp. 729-755. Damer D. (2002). “The Japanese vs the European approach to human resource management in China”, MCD term paper, Research Paper. Devanna M. A., Fombrun C. and Tichy N. (1981). “Human resource management: A strategic perspective”, Organizational Dynamics, Vol. 9, No. 3, pp. 51-68. Devanna M. A., Fombrun C. and Tichy N. (1984). “A framework for strategic human resource management”, in: C. J. Fombrun, N. M. Tichy & M. A. Devanna (Eds.), Strategic Human Resource Management, New York: John Wiley and Sons, pp. 33-51. Fenton-O’Creevy M., Gooderham P. and Nordhaug O. (2008). “Human resource management in US subsidiaries in Europe and Australia: Centralization or automomy?”, Journal of International Business Studies, Vol. 39, No. 1, pp. 151-166. Ferreira P. (2012). “Is there A European convergence in HRM Practices? A cluster analysis of the high-performance paradigm across 31 countries”, in: Silva R. & Tome (Eds.), UFHRD 2012-13th International Conference on HRD Research and Practice across Europe: The Future of HRD–2020 and beyond: Challenges and Opportunities, Famalicao, CLEGI, pp.140-153. Freed A., Hyatt J., Papachristou A. and Papalexandris N. (...). “Greek HRM: Building the critical competencies”, The RBL Working Paper Series, available online at: http://www.rbl.net, pp. 1-7. Gooderham P., Parry E. and Ringdal K. (2008). “The impact of bundles of strategic human resource management practices on the performance of european firms”, The International Journal of Human Resource Management, Vol. 19, No. 11, pp. 2041-2056. Hendry C. and Pettygrew A. (1990). “HRM: An agenda for the 1990s”, International Journal of Human Resource Management, Vol. 1, No. 1, pp. 17-25. Ignjatovic M. and Svetlik I. (2003). “European HRM clusters”, EBS Review, Autumn, pp. 25-39. Karoliny Z., Farkas F. and Poor J. (2009). “In Focus: Hungarian and central eastern European characteristics of human resource management: An international comparative survey”, JEEMS, Vol. 1, pp. 9-47. Katou A. A. and Budhwar P. S. (2009). “Causal relationship between HRM policies and organizational performance: Evidence from the Greek manufacturing sector”, European Management Journal. Poutsma E., Lighthart P. E. and Veersma U. (2006). “The diffusion of calculative and collaborative HRM practices in European firms”, Industrial Relations, Vol. 45, No. 4, pp. 513-525. Scholz C. and Müller S. (2010). “Human resource management in Europe: Looking again at the issue of convergence”, 11th

1927

Convergence or Divergence between European HRM and American HRM International Human Resource Management Conference, Birmingham, UK, June 9-12th. Schuler R. S. (2000). “The internationalization of human resource management”, Journal of International Management, Vol. 6, pp. 239-260. Springer Beverly (…). “US HRM and the EU social policy: A case study of the works council directive”, American Graduate School of International Management. Stavrou E., Brewster C. and Charalambous C. (2004). “Human resource management as a competitive tool in Europe”, IIRA HRM Study Group Working Papers in Human Resource Management, No.5, September. Takei H. and Ho Y. (2007). “Human resource management and governance in the Central and Eastern Europe”, 21st Century Center of Excellence Program, Policy and Governance Working Paper Series, No. 119. Zupan N. and Kase R. (2005). “Strategic human resource management in European transition economies: Building a conceptual model on the case of Slovenia”, International Journal of Human resource Management, Vol. 16, No. 6, pp. 882-906.

1928


The Intervening Effects of Whistleblowing in Reducing the Risk of Asset Misappropriation Mohamad Afzhan Khan Bin Mohamad Khalil1, Anuar Bin Nawawi2, Nurmazilah Dato’ Mahzan3 (1. Open University Malaysia, Malaysia; 2. Universiti Teknologi MARA, Malaysia; 3. University of Malaya, Malaysia)

Abstract: The main objective of this study is to examine the intervening role of whistleblowing in preventing and detecting fraud in order to reduce the risk of asset misappropriation. A review of the literature was undertaken with the aim of crafting the research instrument based on the fraud triangle theory. Qualitative pretesting and quantitative pilot testing were conducted prior to the final data collection exercise. Out of the 553 questionnaires distributed in the final data collection regime, 334 were usable replies and this merits a structural equation modelling to be performed. A confirmatory factor analysis was undertaken on the independent, dependent and mediating variables prior to the assessment of the final model. The results of the structural equation model (Chi square = 2.17; RMSEA = 0.06; CFI = 0.85; TLI = 0.84; NFI = 0.76; PGFI = 0.64) were convincing based on the suggestions from previous studies. This study found that whistleblowing has a partial mediating effect on the relationship between the independent variables (internal control system and professional scepticism) and the risk asset misappropriation. Moreover, it was discovered that there is a full mediation effect of whistleblowing on the relationship between code of ethics and the risk of asset misappropriation. Key words: asset; fraud; whistleblowing JEL codes: G3, G5, M42

1. Motivation and Objective of the Study The gravity of fraud cases can be seen by the sheer statistics and the adverse consequences of fraud. For example, the Association of Certified Fraud Examiners (2012) in their latest study reported that asset misappropriation is getting from bad to worse. Skimming (203 cases), cash larceny (152 cases), billing (346 cases), expense reimbursement (201 cases), cheque tampering (165 cases), payroll (129 cases) and cash register disbursement (50 cases) were among the types of asset misappropriation reported in that global survey. This study is motivated by the need to shed more light on the risk of asset misappropriation, a major form of fraud in many organizations, and the reluctance to whistleblow by those few capable of doing so (Buckley et al., 2010; Lee & Fargher, 2012; Read & Rama, 2003). Asset misappropriation can severely affect organizations, in particular those

Mohamad Afzhan Khan Bin Mohamad Khalil, MBA, Business School, Open University Malaysia; research areas/interests: fraud, auditing, accounting, corporate governance and business ethics. E-mail: [email protected]. Anuar Bin Nawawi, Ph.D., Faculty of Accountancy, Universiti Teknologi MARA; research areas/interests: fraud, auditing, accounting, corporate governance and business ethics. E-mail: [email protected]. Nurmazilah Dato' Mahzan, Ph.D., Faculty of Business and Accountancy, University of Malaya; research areas/interests: fraud, auditing, accounting, corporate governance and business ethics. E-mail: [email protected]. 1929

The Intervening Effects of Whistleblowing in Reducing the Risk of Asset Misappropriation

operating within the banking sector. If whistleblowing is ignored, there will be problems of theft of asset, falsification of sales data, inflated claims from suppliers, manipulation of inventories and overstatement of revenues (Somers & Casal, 2011). Shareholders will suffer from financial losses which will lead to erosion of confidence from potential investors (Bierstaker et al., 2006). Pieces of empirical evidence have shown that internal control system (Sengur, 2012), code of ethics (Wesel, 2009; Yekta et al., 2010), professional scepticism (Masruri & Zahro, 2013) and technology tools (Grabosky, 2007) solely are not sufficient to prevent and detect fraud. The mediated role of whistleblowing is necessary. Organisations must be aware of the roles of whistleblowing and how it can help other mechanisms to reduce fraud. The main objective of this study is to examine the intervening role of whistleblowing in preventing and detecting fraud in order to reduce the risk of asset misappropriation. Information from this study can be helpful in minimizing the allocation of time, resources utilized and cost incurred to combat asset misappropriation.

2. Review of Empirical Studies and Research Gaps Poor internal control is one of the top factors contributing to fraud in Malaysia (KPMG, 2009). The weaknesses of internal control system have been researched in a very large scale by previous studies (Dorminey et al., 2010; Hermawati, 2013; Kassem & Higson, 2012; Rae et al., 2008). There are few studies that have analyzed the relationship between internal control systems and fraud without the effect of whistleblowing (Bierstaker et al., 2006; Coram et al., 2008). Thus, the first research gap identified is the need to include whistleblowing. Numerous researchers have advocated the use of code of ethics as a means to prevent fraud (Bierstaker et al., 2006; Domoro & Syed Agil, 2012; Masruri & Zahro, 2013; Rae et al., 2008; Sengur, 2012; Wesel, 2009; Yekta et al., 2010). A number of researchers (Okpara, 2003; Yekta et al., 2010) have examined the relationship between ethics and whistleblowing. The causal relationship between code of ethics and reporting wrongdoings could be extended to examining fraud to create the second research gap of this study. The mediating effect of whistleblowing on the relationship between professional scepticism and fraud is the third gap that is to be narrowed down by this study. Haugen & Selin (1999) discussed the common technological controls such as passwords, firewalls and encryptions to detect fraud through their qualitative study. Their study was extended by Bierstaker et al. (2006) by conducting mean analysis to explain the effectiveness of technology tools in preventing fraud. In the banking environment, Usman & Shah (2013) through their review of literatures have elaborated on the significant effect of technology in reducing fraud. Subsequently, results from the questionnaires, personal interviews, and document review conducted in a study by Njanike et al. (2009) on 13 commercial banks showed that ineffective technological controls may lead to unethical behaviour. However, all of the studies above have not linked technology controls and whistleblowing.

3. Research Methodology A literature review was first done to develop the initial research questionnaire. Subsequently, the questionnaire was pretested in a series of interviews with six experts. Thematic analysis was then conducted to analyze the information gathered from the interviews. Thereafter, a quantitative pilot study among 55 bankers was conducted. From the pilot study, factor analysis and reliability tests were performed on all the respective constructs. Several items were removed from the measurement of the six constructs in the pretest and pilot test stages. The final survey instrument was then administered to 553 bankers out of which there were 334 usable 1930


responses. Systematic simple random sampling and face to face data collection methodology were used in this study. The data were collected with the help of nineteen research assistants. Prior to the structural equation modelling, a Mann Whitney test was conducted to ensure that there was no non-response bias. Cronbach alphas were calculated on all the six constructs and all the indices were above 0.80, implying reliability (Zikmund et al., 2010). The normality of distribution of the constructs was determined using the Kurtosis measurements and the indices for all the constructs were within the recommended ranges of Lei & Lomax (2005). Finally, confirmatory factor analysis (CFA) was performed on all the constructs and the results were convincing. Structural equation modelling (SEM) was then performed. During the assessment of non response bias, reliability, normality and confirmatory factor analysis, several items were deleted in every construct to ensure that the remaining items were valid and reliable for a model assessment. The respective constructs employed in this study are disclosed and summarized in Table 1 below. Table 1 Independent Variables Mediating Variable Dependent Variable

Constructs Employed in this Study (1) Internal Control System (2) Code of Ethics (3) Professional Scepticism (4) Technology Tools Whistleblowing Risk of Asset Misappropriation

4. Review on the Principles of Mediation Table 2

Principles of Full Mediation by Kenny & Baron (1986)

First test : Explanation: The first test between independent (X) and dependent variable (Y) must be significant (c) Second test :

Explanation: The following must take place in the second test which is the final model: (1) There is evidence of a significant linear relationship (a) between the independent variable (X) and the mediator (M) in equation (1). (2) There is evidence of a significant linear relationship (b) between the mediator (M) and the dependent variable (Y) in equation (2). (3) Equation (3) is no longer significant (c’) in the model. The relationship of X and Y diminishes when M is introduced in the model.

The pioneer methodologists who contributed to the literature on the principles of mediation are Kenny & Baron (1986). Their method has been reviewed and discussed in many studies. Several other methodologists (Hair et al., 2010; Iacabocci et al., 2007; Little et al., 2007) have discussed and supported the principles of mediation by Kenny & Baron (1986). There are certain empirical conditions for mediation that must be met. According to Kenny & Baron (1986), two tests must be conducted in the process of mediation assessment. Consistent with their view, Hair et al. (2010) have also supported this methodology. Table 2 will further assist in describing the mediation principles. In light of the discussion on the principles of mediation, it is to be noted that there have been some disagreements between methodologists in recent studies. According to Iacabocci et al. (2007), if there is no relationship between the independent variable (X) and dependent variable (Y), then mediation test should not be performed. In contrast, other methodologists like Rucker et al. (2011) view it differently. Based on the discussion of Rucker et al. (2011), it is justified that focusing on the significance level in the first test can impede research.

1931

The Intervvening Effects of Whistleblow wing in Reduccing the Risk off Asset Misapp propriation

The main reeasons discusssed in their liiterature are (1) evidence for full mediation will disscourage reseearchers from m examining indirect i effeccts which caan hinder theeory develop pment (2) the notion of partial mediiation is lesss impressive than t full mediiation and thiis is unwarrannted in any reesearch.

5. Un nderpinningg Theory an nd Conceptual Frameework The fraaud triangle theory t has been b widely used u by prev vious researchhers in the sstudy of frau ud preventionn (Dorminey et e al., 2010; Hillison H et al.,, 1999; Mironniuc et al., 2012; Sitorus & Scott, 2008;; Skousen et al., a 2009) andd whistleblow wing (Hermaw wati, 2013; Lee L & Farghher, 2012). The T PwC Gloobal Econom mic Crime Su urvey (2009)) reported thatt pressure (688%), opportunnity (18%) annd rationalizaation (14%) arre the factorss of fraud. Wh histleblowingg should be a catalyst c ratheer than enableer in the preveention and dettection of frauud. Whistlebllowing is designed to onlyy improve inddividual and collective c behhaviour of peeople workin ng in a compaany rather thhan directly fighting fi fraudd (Bunget et al., a 2009). Thee following Figure 1 depicts the concep ptual framewoork of this stuudy.

Figure 1 Conceptual Framework F

6. Dirrect Effect Assessmen A t The dirrect effect asssessment betw ween the independent and d dependent variables v is fiirst conducted d. Accordingg to Kenny & Baron (19886), the direcct effect assesssment needss to be conduucted prior to performing g a structurall equation moodelling. Thiis is the firstt test that shhould be perfformed to deetermine wheether mediatiion exists orr otherwise ass explained in i Table 2 eaarlier in this study. The siignificance values v and strrength of relationships inn terms of Beeta are discloosed in Table 3 below. As A could be seen, the diirect effects between the independentt variables (innternal controol system, codde of ethics and a professio onal scepticism m) and the deependent varriable (risk off asset misapppropriation) are a significantt. The p-values of all the significant s asssociations falll below 0.10 (Zikmund ett al., 2010). The T result off the direct annalysis is quite similar to o previous stuudies (Coram m et al., 2008 8; Masruri & Zahro, 20133). Coram ett al. (2008) reported thaat organizatio ons with goood internal ccontrols may detect assett misappropriation. Previoous studies haave performed a number of o analyses on o professionnal scepticism m (Charron & Lowe, 2008; Masruri & Zahro, 2013)). Charron & Lowe (2008 8) explained that t people w who exhibit high h levels off professionall scepticism are a in a betteer position too detect fraud d. In this stuudy, the only association which is nott significant is i between teechnology toools and the risk of assett misapproprriation (p-vallue = 0.46). Nonetheless,, technology tools t were stiill included inn the structurral equation modelling m anaalyses follow wing the view of Rucker ett al. (2011). Moreover, M som me other methhodologists advocate a the inclusion i of all a variables iin the model regardless off

1932


their significance in order to control for confounding (Bursac et al., 2008). According to Pearl (2012), the hypothesized model should be subjected to a test ‘of the entire system of variables,’ irrespective of whether the tested part bears any relationship to the resulting claims. These views seem to depart from the conventional principles developed by Kenny & Baron (1986). However, in this study, the departure from the conventional principles is minimal judging by the fact that it only affects one construct which is technology tools. Thus, despite the rigidity proposed by Kenny & Baron (1986), this study takes the view that it is important to test all the constructs in a simultaneous model to examine relationships. Table 3 No H1(a) H1(b) H1(c) H1(d)

Summary of the First Hypothesis

Hypotheses Statement There is a direct effect of internal control system on the risk of asset misappropriation There is a direct effect of code of ethics on the risk of asset misappropriation There is a direct effect of professional scepticism on the risk of asset misappropriation There is a direct effect of technology tools on the risk of asset misappropriation

Beta estimate

P-value

Interpretation

-0.24

0.01

Supported

-0.19

0.02

Supported

-0.21

0.01

Supported

-0.04

0.46

Not supported

7. Structural Equation Modelling Assessment Figure 2 portrays the SEM assessment performed to satisfy the research objective of this study. Table 4 reports the indices from the assessment. The discussion of the results will follow suit. This study will uphold the view of Zikmund et al. (2010) in providing a cutoff threshold of the significance level of 10% (p < 0.10).

e53

e54

1 ICS24

e55

e56

e57

e58

e59

e60

e61

e62

e63

e64

e65

1

1

1

1

1

1

1

1

1

1

1

1

ICS23

ICS22

ICS21

ICS20

ICS19

ICS17

ICS16

ICS15

ICS14

ICS13

ICS12

ICS10

e66

e67

1 ICS9

e68

1 ICS8

1 ICS7

e69

1 ICS6

e70

1

e71

1

ICS5

ICS4

e72

1 ICS3

e73

1 ICS1

1 INTERNAL CONTROL

1 e40

COE12

1 e39

COE11

1

1 e38

WB1

e41

1

COE10

e37

AM18

1

1

WB3

COE7

e42

1 AM14

1 e36

1 COE6

1 e35

1

CODE OF ETHICS

WB4

1 WB5

e33

1 WB6

1

e45

AM11

1

ASSET MISAPPROPRIATION

1 1 e32

WB7

WHISTLE BLOWING

COE2

e46

WB8

COE1

AM9

e47

1

1 WB9

WB10

1

1 AM2

e50

1 AM1

1

PS8

WB12

1

e2 e3 e4 e5 e6 e7 e8 e9 e10 e11

e51

1

PS6

WB17

PROFESSIONAL SCEPTICISM

1 e26

e49

WB11

1

e27

AM3

1

PS9

e28

e48

1

PS10

e29

AM4

1

1 e30

1 AM10

1

1

1 e31

1 AM12

1 1

COE3

e75

e44

e74

COE4

1

1 AM13

COE5

1 e34

1

e43

e1

e52

PS5

1 e25

PS4

TECHNOLOGY TOOLS

1 1 e24

1

PS3

1 e23

PS2

TT14

TT13

TT11

1

1

1

e22

e21

e20

TT9

1 e19

TT7

TT6

1 e18

1 e17

Figure 2

TT5

1 e16

TT4

1 e15

TT3

1 e14

TT2

1 e13

TT1

1 e12

Structural Equation Model

1933


Table 4

Indices from the Structural Equation Modelling

Equation 1 (cross referenced with Table 2) Mediating Variable


Beta Estimates

P-value

Whistleblowing

Internal Control System

0.70

Whistleblowing

Code of Ethics

0.34

0.01***

Whistleblowing

Professional Scepticism

0.10

0.03**

Whistleblowing

Technology Tools

-0.03

0.55 NS

0.01***

Equation 2 (cross referenced with Table 2) Dependent Variable

Mediating Variable

Beta Estimates

P-value

Asset Misappropriation

Whistleblowing

0.20

0.08*

Equation 3 (cross referenced with Table 2) Dependent Variable


Beta Estimates

P-value


Internal Control System

-0.34

0.01***


Code of Ethics

-0.08

0.28 NS


Professional Scepticism

-0.12

0.05**

Asset Misappropriation *

Note: Significance level < 0.10;

Technology Tools **

Significance level < 0.05; Table 5

No

0.10 ***

Significance level < 0.01;

0.09* NS

Not Significant

Summary of the Second Hypothesis

Hypothesis Statement Results There is a relationship between internal control system and the risk of asset H2(a) Partial Mediation misappropriation when mediated by whistleblowing. There is a relationship between code of ethics and the risk of asset misappropriation H2(b) Full Mediation when mediated by whistleblowing. There is a relationship between professional scepticism and the risk of asset H2(c) Partial Mediation misappropriation when mediated by whistleblowing. There is a relationship between technology tools and the risk of asset H2(d) No Mediation misappropriation when mediated by whistleblowing.

Interpretation Supported Supported Supported Not supported

8. Findings and Discussion The indices for model fit in this study are shown in Table 6. Justifications and benchmarks are also provided from previous studies to defend the structural equation model used in this study. All indices fall within suggested range. The results of the structural equation model (Chi square = 2.17; RMSEA = 0.06; CFI = 0.85; TLI = 0.84; NFI = 0.76; PGFI = 0.64) are convincing based on the recommendations provided by previous studies as could be seen from Table 6. The model fit is first confirmed before any discussion is made on the hypothesis testing. According to Little et al. (2007), there should be at least one of any two situations that must be observed to confirm a partial mediation. In the first situation, the Beta value for the relationship between the independent and dependent should be reduced with the inclusion of the mediating variable for a partial mediation to take place. Alternatively, in the second situation, the p-value must be less significant in the relationship between the independent and dependent when the mediating variable is included in the model. In the structural equation analysis conducted, there are significant relationship between internal control system and whistleblowing (Beta = 0.70, p = 0.01), moderately significant relationship between whistleblowing and the risk of asset misappropriation (Beta = 0.20, p = 0.08) and significant negative relationship between internal control system and the risk of asset misappropriation (Beta = -0.34, p = 0.01) as shown in Table 4. As

1934


such, a partial mediation can be concluded in this case because Beta is reduced from -0.24 (direct test) to -0.33 (mediating test) implying that the strength of internal control system and the risk of asset misappropriation becomes less prevalent when the mediating variable (whistleblowing) is incorporated in the model. These results indicate that although an internal control system could prevent and detect the risk of asset misappropriation, the control system will work better if it is intervened by whistleblowing. Hypothesis H2(a) is thus supported in this study. Proper authorization for wage and overtime, proper documentation and proper authorization system according to limit are forms of internal control that could detect the risk of asset misappropriation such as manipulating payroll records, diverging wages, and overstating hours worked and overcharging banks expenditure for personal gain. It is also recommended that appropriate staff rotation policy, mandatory leave policy and cross referencing for potential employees as part of internal control mechanisms that could prevent the creation of phantom employees, manipulation of payroll records and stealing of cash from the organization. The role of whistleblower as rectifying criminal behaviour at the workplace is important because criminal behaviour such as theft, gambling and drug abuse could serve as red flags of fraud. Table 6

Confirmation of Model Fit

Index Observed Range from this study Desired range recommended by authorities Root Mean Square of Error Approximation 0.06 Hair et al. (2010) suggested indices below 0.10 (RMSEA) Comparative Fit Index (CFI) 0.85 Javali (2011) indicates that 0.82 is acceptable fit. Hooper et al. (2008) have recommended a more Tucker Lewis Index 0.84 lenient viewpoint whereby figures which are as low (TLI) as 0.80 should be acceptable Normed Fit Singh (2009) has recommended that NFI is to be 0.76 Index (NFI) acceptable if it is as low as 0.60 to as high as 0.90 Range of acceptable chi square from as high as 5.0 Chi Square 2.17 (Wheaton et al, 1977) to as low as 2.0 (Tabachnick and Fidell, 2007). Parsimony Goodness of Fit Index (PGFI) 0.64 Hair et al. (2010) recommended indices above 0.50

In the structural equation analysis conducted, there is a significant relationship between whistleblowing and code of ethics (Beta = 0.34, p = 0.01), moderately significant relationship between whistleblowing and the risk of asset misappropriation (Beta = 0.20, p = 0.08) and an insignificant negative relationship between code of ethics and the risk of asset misappropriation (Beta = -0.08, p-value = 0.28) as shown in Table 4. The assessment clearly justifies that there is a full mediation of whistleblowing in intervening the relationship between code of ethics and the risk of asset misappropriation. This is because; the relationship between the code of ethics and the risk of asset misappropriation cease to be significant with the inclusion of whistleblowing in the model assessment as proposed by the authorities (Hair et al., 2010; Kenny & Baron, 1986). However, in the direct effect assessment, without the inclusion of the mediating variable, the relationship between the code of ethics and the risk of asset misappropriation is significant (Table 3). This implies that without a whistleblowing mechanism in place, the code of ethics cannot be used to prevent and detect fraud. In light of the above assessment, hypothesis H2(b) is supported in this study. The structural equation model findings in this study concur with the literature of Wesel (2009) who strictly discussed that code of ethics cannot stand alone to fight fraud and corruption. The whistleblowing role of helping other people and reducing future damage could be considered as an intervention factor between professionalism and fraud detection (e.g., conspiracy between buyers and suppliers, unlawfully permitting special prices and unethically granting business to suppliers for personal gain). This is because the value of professionalism

1935


ensures that employees who whistleblow disregard the relationships, financial interest and elements of collectivism when performing their official duties of uncovering misconduct in the banking sector. The act of being fair and providing quality auditing justifies the behaviour of helping other people. The behaviour of helping other people then ensures that acts of misconduct such as corruption and collusion are brought to the light of justice. The behaviour of being objective impacts the assessment of evidence persuasiveness in auditing for fraud. Objectivity is important in maintaining public confidence and this ensures that there is no personal relationship between auditors, clients, suppliers and buyers which could possibly lead to a potential conflict of interest. That is why having an opinion which is free from bias is a necessity in preventing and detecting asset misappropriation. Only then can collusion between buyers and sellers and giving privilege prices to chosen individuals could be avoided in the banking sector. Whistleblowers must realize that they have a role of alerting top management of observed wrongdoings and should come forward to report wrongdoings. Another contribution of this study is that it has extended the conceptual model of Yekta et al. (2010). However, the findings of this study are not consistent with what has been reported by Yekta et al. (2010) who reported an insignificant relationship between code of ethics and reporting wrongdoings (p value = 0.20). The SEM assessment conducted for this study extends the knowledge gained from previous studies (Domoro & Syed Agil, 2012; Masruri & Zahro, 2013) that directly regressed ethics and fraud without studying the mediation effect of whistleblowing. From Table 4, it could be seen that there is a significant relationship between professional scepticism and whistleblowing (Beta = 0.10, p = 0.03), moderately significant relationship between whistleblowing and the risk of asset misappropriation (Beta = 0.20, p = 0.08) and a significant negative relationship between professional scepticism and the risk of asset misappropriation (Beta = -0.12, p = 0.05). The assessment provides support for partial mediation because when whistleblowing is incorporated in the model as an intervening effect between professional scepticism and the risk of asset misappropriation, the significance level is reduced from p = 0.01 (Table 3) to p = 0.05 (Table 4) indicating a lesser strength between the independent variable and dependent variable when mediating variable is taken into account. Thus, hypothesis H2(c) is supported in this study. Among the possible ways of being sceptical when auditors are in doubtful or curious situations are to conduct unannounced inventory observations and testing some low risk accounts (Mckee, 2006). This is because auditors have a responsibility of reporting or whistleblowing if they find any wrongdoings. This then reduces the risk of asset misappropriation such as falsification of physical inventories, theft of properties, theft of low risk accounts such as tools, supplies, waste and scrap materials. Internal auditors, fraud investigators, risk managers and compliance officers in banks spend so much of their time gathering sufficient information and evidence against fraudulent practices such as forgery of cheques, theft of cash, over-claiming of expenses, falsification of accounting records and managerial reports. All the efforts will be gone to waste if they do not realize that they have a whistleblowing role of alerting the top management. Finally, hypothesis 2(d) is rejected in this study because there is no significant relationship between whistleblowing and technology tools (Beta = -0.03, p = 0.55) based on the principles of Kenny & Baron (1986). The result in this section is consistent with the findings of KPMG (2009). Their survey indicates that information technology can be a liability instead of an advantage to fraud prevention measures. The survey reported abuse of passwords or privileges (71%), manipulation of the weaknesses in the information technology systems (36%), lack of segregation of duties (21%) and hacking (14%). Consistent with the conclusion provided by Behling et al. (2009), this study considers that technology controls are not always sufficient to reduce misappropriations. This is 1936


due to the fact that technology tools are not widely used by the banking sector as they are often expensive and deemed redundant.

9. Implications With the SEM assessment conducted in this study, the main objective of examining the mediating role of whistleblowing in preventing and detecting fraud in order to reduce the risk of asset misappropriation has been met and detailed discussion has been provided to narrow down the literature gaps. This study offers two managerial implications. The study of KPMG (2009) revealed that unethical behaviour in organizations leads to reputation damage (86%) and loss of employees’ morale (78%). The study of Somers & Casal (2011) found out that 45% of employees prefer not to blow the whistle because they believe that no action will be taken after the whistleblowing is done. However, in this study, the opinion of Somers & Casal (2011) is disputed. This study suggests that whistleblowers do play a major role in preventing unethical behaviour. This should be a wakeup call to all auditors, risk managers and compliance officers to whistleblow in preventing losses due to asset misappropriation in the banking sector. One managerial implication in this study is that every bank should provide a channel for whistleblowing. The code of ethics communicated to employees must be accompanied by whistleblowing policy that specifies the roles of and protection of whistleblowers. Another managerial implication of this study is that the banking sector must consider investing in a more robust internal control system and the hiring of people with professional scepticism. This is because greater surveillance and managerial diligence in the internal control system are needed to decrease employee related fraud. References: Association of Certified Fraud Examiners (2012) Survey. Behling S., Floyd K., Smith T., Koohang A. and Behling R. (2009). “Manager’s perspectives on employee information technology fraud issues within companies/organizations”, Issues in Information System, Vol. 10, No. 2, pp. 76-81. Bierstaker J. L., Brody R. G. and Pacini C. (2006). “Accountants’ perceptions regarding fraud detection and prevention methods”, Managerial Auditing Journal, Vol. 21, No. 5, pp. 520-535. Bowlin K., Hobson J. L. and Piercey M. D. (2013). “The effects of auditor rotation, professional scepticism, and interactions with managers on audit quality”, working papers series, Social Science Research Network, retrieved on March 31, 2013, available online at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1914557. Buckley C., Cotter D., Hutchinson M. and O’Leary C. (2010). “Empirical evidence of lack of significant support for whistleblowing”, Corporate Ownership and Control, Vol. 7, No. 3, pp. 275-283. Bunget O. C., David S. and Maria I. (2009). “Ethics and internal audit: Whistleblowing issues”, retrieved on December 31, 2011, available online at: http://mpra.ub.uni-muenchen.de/17312/ Bursac Z., Gauss C. H., Williams D. K. and Hosmer D. W. (2008). “Purposeful selection of variables in logistic regression”, retrieved on August 18, 2013, available online at: http://www.scfbm.org/content/3/1/17. Charron K. F. and Lowe D. J. (2008). “Scepticism and the management accountant: Insights for fraud detection”, Management Accounting Quarterly, Vol. 9, No. 2, pp. 9-16. Coram P., Ferguson C. and Moroney R. (2008). “Internal audit, alternative internal audit structures and the level of misappropriation of assets fraud”, Accounting and Finance, Vol. 48, pp. 543-559. Domoro O. M. O. and Syed Agil S. O. (2012). “Ethics and corruption empirical study in the Libyan police force”, Australian Journal of Basic and Applied Sciences, Vol. 6, No. 6, pp. 353-357. Dorminey J. W., Fleming A. S., Kranacher M. J. and Riley R. A. (2010). “Beyond the fraud triangle”, CPA Journal (July), pp. 16-24. Grabosky P. (2007). “The Internet, technology, and organized crime”, Asian Journal of Criminology, Vol. 2, No. 2, pp. 145-161. Hair J., Anderson R. and Babin B. (2010). Multivariate Data Analysis (7th ed.), Prentice Hall. Haugen S. and Selin J. R. (1999). “Identifying and controlling computer crime and employee fraud”, Industrial Management & Data

1937

The Intervening Effects of Whistleblowing in Reducing the Risk of Asset Misappropriation Systems, Vol. 99, No. 8, pp. 340-344. Hermawati L. M. (2013). “The influence of fraud triangle upon the existence of financial statement fraud”, retrieved on January 11, 2014, available online at: http://www.savepageaspdf.com/c8ac50ce9e0348da989370c9b1e5c92e/PAPER%20FRAUD% 20TRIANGLE%20MURTANTO%20AND%20LUTFIANA%20(Final).htm. Hillison W., Pacini C. and Sinason D. (1999). “The internal auditor as fraud-buster”, Managerial Auditing Journal, Vol. 14, No. 7, pp. 351-363. Hooper D., Coughlan J. and Mullen M. R. (2008). “Structural equation modelling: Guidelines for determining model fit”, Electronic Journal of Business Research Methods, Vol. 6, No. 1, pp. 53-60. Iacobucci D., Saldanha N. and Deng X. (2007). “A meditation on mediation: Evidence that structural equations models perform better than regressions”, Journal of Consumer Psychology, Vol. 17, No. 2, pp. 139-153. Javali S. (2011). “A structural equation model of the determinants of health care in the surveyed households in rural of Dharwad District, Karnataka State, India”, retrieved on June 08, 2013, available online at: http://www.webmedcentral.com/article_view/1698. Kassem R. and Higson A. (2012). “The new fraud triangle model”, Journal of Emerging Trends in Economics and Management Sciences, Vol. 3, No. 3, pp. 191-195. Kenny D. A. and Baron R. M. (1986). “The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51, No. 6, pp. 1173-1182. KPMG (2009). KPMG Fraud Survey 2009 Report, Malaysia Lee G. and Fargher N. (2012). “Companies’ use of whistleblowing to detect fraud: An examination of corporate whistleblowing policies”, Journal of Business Ethics, Vol. 114, No. 2, pp. 283-295. Lei M. and Lomax R. G. (2005). “The effect of varying degrees of non-normality in structural equation modeling”, Structural Equation Modeling: A Multidisciplinary Journal, Vol. 12, No. 1, pp. 1-27 Little T. D., Card N. A., Bovaird J. A., Preacher K. J. and Crandall C. S. (2007). “Structural equation modeling of mediation and moderation with contextual factors”, in: T. D. Little, J. A. Bovaird & N. A. Card (Eds.), Modelling Contextual Effects in Longitudinal Studies, Mahwah, NJ: Lawrence Erlbaum Associates, pp. 207-230. Masruri A. M. and Zahro N. I. (2013). “Corruption awareness, ethical sensitivity, professional scepticism and risk of corruption assessment: Exploring the multiple relationship in Indonesian case”, retrieved on June 10, 2013, available online at: http://wbiworldconpro.com/uploads/malaysia-conference-2013/management/438-Ashari.pdf. McKee T. E. (2006). “Increase your fraud auditing effectiveness by being unpredictable!”, Managerial Auditing Journal, Vol. 21, No. 2, pp. 224-231. Mironiuc M., Robu I. B. and Robu M. A. (2012). “The Fraud Auditing: Empirical study concerning the identification of the financial dimensions of fraud”, Journal of Accounting and Auditing: Research & Practice, Vol. 2, pp. 1-13. Njanike K., Dube T. and Mashayanye E. (2009). “The effectiveness of forensic auditing in detecting, investigating, and preventing bank fraud”, Journal of Sustainable Development in Africa, Vol. 10, No. 4, pp. 405-425. Okpara J. O. (2003). “Can corporate ethical codes of conduct influence behavior? An exploratory study of financial managers in a developing economy”, presented at the Academy of Business & Administrative Sciences (ABAS), pp. 1-12, retrieved on January 11, 2013, available online at: http://www.sba.muohio.edu/abas/2003/brussels/okpara_ethical %20codes%20 of %20conduct%20 and%20employeebehavior.pdf Pearl J. (2012). “The causal foundations of structural equation modeling”, retrieved on August 18, 2013, available online at: http://ftp.cs.ucla.edu/pub/stat_ser/r370.pdf. PwC Global Economic Crime Survey (2009). Rae K., Subramaniam N. and Sands J. (2008). “Risk management and ethical environment: Effects on internal audit and accounting control procedures”, Journal of Applied Management Accounting Research, Vol. 6, No. 1, pp. 11-30. Read W. J. and Rama D. V. (2003). “Whistle-blowing to internal auditors”, Managerial Auditing Journal, Vol. 18, No. 5, pp. 354-362. Rucker D. D., Preacher K. J., Tormala Z. L. and Petty R. E. (2011). “Mediation analysis in social psychology: Current practices and new recommendations”, Social and Personality Psychology Compass, Vol. 5, No. 6, pp. 359-371. Sengur E. D. (2012). “Auditors’ perceptions of fraud prevention measures”, Evidence From Turkey, Vol. 14, No. 1, pp. 128-138. Singh R. (2009). “Does my structural model represent the real phenomenon?: A review of the appropriate use of structural equation modelling (SEM) model fit indices”, The Marketing Review, Vol. 9, No. 3, pp. 199-212. Sitorus T. and Scott D. (2008). “The Roles of collusion, organizational orientation, justice avoidance and rationalization on

1938

The Intervening Effects of Whistleblowing in Reducing the Risk of Asset Misappropriation commission of fraud: A model based test”, Review of Business Research, Vol. 8, No. 1, pp. 132-147. Skousen C. J., Smith K. R. and Wright C. J. (2009). “Detecting and predicting financial statement fraud: the effectiveness of the fraud triangle and SAS No. 99”, Advances in Financial Economics, Vol. 13, pp. 53-81. Somers M. and Casal J. C. (2011). “Type of wrongdoing and whistle-blowing: Further evidence that type of wrongdoing affects the whistle-blowing process”, Public Personnel Management, Vol. 40, No. 2, pp. 151-164. Tabachnick B. G. and Fidell L. S. (2007). Using Multivariate Statistics (5th ed.), Pearson Education, Upper Saddle River, NJ. Usman A. K. and Shah Mahmood Hussain (2013). “Critical success factors for preventing e-banking fraud”, Journal of Internet Banking and Commerce, Vol. 18. No. 2, pp. 1-16. Wesel C. C. (2009). “International business ethics and anti- corruption: A code of ethics cannot stand alone”, Retrieved December 31, 2011, from http://www.gowlings.com/resources/PublicationPDFs/WesselC_CodeofEthics_Aug09.pdf Wheaton B., Muthen B., Alwin D. F. and Summers G. (1977). “Assessing reliability and stability in panel models”, Sociological Methodology, Vol. 8, No. 1, pp. 84-136. Yekta Z. A., Ahmad Z. A. and Kaur A. (2010). “Corporate code of ethics and reporting wrong doing in private sector organizations in Malaysia”, International Journal of Business and Management, Vol. 5, No. 7, pp. 84-91. Zikmund W. G., Babin B., Carr J. and Griffin M. (2010). Business Research Methods (8th ed.), Australia, South-Western, Cengage Learning.

1939


The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs Ricardo Garcia Ramirez, Gonzalo Maldonado Guzman, Maria del Carmen Martinez Serna (Autonomous University of Aguascalientes, Mexico)

Abstract: The constant changes established in the new business environment, the high globalization of markets and the prominent level of competitiveness are forcing enterprises, especially the small and medium-size ones, to transform or modify their business strategies to adequate them to the market requirements. Within this new perspective, market orientation, entrepreneurial orientation and innovation are the three most important topics that are being considered nowadays by an increasing number of important companies for the development of strategies. Therefore, the aim of this research is to analyze the relationship between market orientation, entrepreneurial orientation and innovation by using a sample of 318 SMEs from Aguascalientes State (Mexico). The results obtained show that, on one hand, both market orientation and entrepreneurial orientation have significant positive effect in the innovation of SMEs. On the other hand, market orientation has a significant positive effect in entrepreneurial orientation. Key words: market orientation; entrepreneurial orientation; innovation; small and medium-sized enterprises (SMEs) JEL code: M21

1. Introduction The fast globalization of markets, the assertiveness in the level of competitiveness and the fast technological changes are elements that have characterized the first two decades of this century. They have brought, on one hand, a drive in market and entrepreneurial orientation as it has never been seen before (Kwak et al., 2013). On the other hand, there is the inevitable incorporation of innovation as an inexorable business strategy for firms (Reza & Tajeddini, 2011). Therefore, different researchers and academics consider that the depending on the skills developed by companies to take advantage of the opportunities given by the market will depend mostly on the levels of market and entrepreneurial orientation that their businesses achieve (Miles & Arnold, 1991; Zahra & Covin, 1995; Hurley & Hult, 1998; Wiklund, 1999; Atuahene-Gima & Ko, 2001; Matsuno et al., 2002; Kara et al., 2004; Kumar et al., 2011).

Ricardo Garcia-Ramirez, Ph.D. Candidate, Autonomous University of Aguascalientes; research areas/interests: innovation, marketing, entrepreneurship. E-mail: [email protected]. Gonzalo Maldonado-Guzman, Ph.D., Autonomous University of Aguascalientes; research areas/interests: innovation, marketing, information technology. E-mail: [email protected]. Maria del Carmen Martinez-Serna, Ph.D., Autonomous University of Aguascalientes; research areas/interests: innovation, market orientation, entrepreneurial orientation. E-mail: [email protected]. 1930

The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs

In this regard, in the literature of business and management sciences, some researchers and academics have come to a conclusion on their papers about the need of companies to incorporate market and entrepreneurial orientation in order to achieve a higher return and level of innovation (Deshpandé et al., 1993; Atuahene-Gima & Ko, 2001; Hult et al., 2004; Bhuian et al., 2005). Thus, Deshpandé et al. (1993) proved that firms, which adopted and implemented a culture of market orientation and entrepreneurial orientation, obtained a better performance both inside and outside the organization than those that did not. Similar results were also obtained by Atuahene-Gima and Ko (2001), who concluded in their research that the firms which incorporated a market orientation and an entrepreneurial orientation got a higher performance when compared to other organizations that haven't adopted these two business strategies. Similarly, Hult et al. (2004) established in their paper that the companies which considered market orientation and entrepreneurial orientation as business strategies improved significantly the activities related to business revenue as well as all the activities related to innovation. In a similar trend, Bhuian et al. (2005) also stated that the firms which integrated market and entrepreneurial orientation as part of their daily activities achieved a higher business revenue and a higher level of innovation when compared to those business that did not considered them as their strategies. As a result, the literature has enough theoretical and empirical evidence that demonstrates the existence of a high relationship between market orientation, entrepreneurial orientation and innovation. However, most of the published papers that analyze the relationship among these three constructs have been carried out in developed countries (Hult et al., 2004; Tajeddini, 2010) and only a few have been applied in developing countries or countries with an emerging economy (Reza & Tajeddini, 2011). This is why it is necessary more research work in this type of countries. With this view in mind, the main contribution of this empirical paper is the analysis of the existing relationship between market orientation, entrepreneurial orientation and innovation in small and medium-size enterprises (SMEs) in an emerging country, as it is the case of Mexico. A second contribution is the methodology that has been used in this research paper since it will apply a structural equations modeling of second order to analyze the proposed theoretical model as a whole, which will allow a deeper examination of the relationships among the three selected variables.

2. Literature Review Different researchers, academics and professionals in the field of marketing consider that the current literature has given little attention to market orientation as a strategic and significant area in organizations (Cadogan et al., 2002; Kok et al., 2003; Li, 2005; Li et al., 2006; Schindehutte et al., 2008; Gopal, 2008), since market orientation does not only allow firms to achieve a higher level of revenue (Narver & Slater, 1990; Kohli & Jaworski, 1990; Deshpandé et al., 1993; Matsuno et al., 2002; Renko et al., 2009), but it also facilitates the integration and implementation of innovation activities (Deshpandé et al., 1993; Pelham & Wilson, 1996; Han et al., 1998; Lukas & Ferrell, 2000; Verhees & Meulenberg, 2004; Kirca et al., 2005; Grinstein, 2008). Similarly, there is a broad consensus among researchers and academics that have analyzed and debated market orientation. This construct is considered as a result of the implementation of the marketing philosophy of organizations, which pays special attention not only to customers and consumers but also to the main competitors (Kok et al., 2003). Thus, market orientation can be considered as a skill obtained by firms to analyze both the internal and external environment around them more accurately (Reza & Tajeddini, 2011).

1931


This idea of market orientation demands from the organizations to monitor frequently the markets in which they participate in order to carry out the necessary changes based on the prediction of customers' needs, the likes and preferences of current and potential consumers and the capacity of the firm to bond with customers, suppliers and distributors (Schindehutte et al., 2008). For this reason, Hammond et al. (2006) considered that a high level of market orientation creates a high level of skills, partially or totally, of the organization, which will enable it to achieve the aims and goals established by the management. Currently, the skill that needs to be increased more in firms, especially SMEs, is the one of innovation as it has become an essential business strategy as well as the vital energy that allows enterprises to survive in the market (Renko et al., 2009). Therefore, the integration and implementation of market orientation from firms will help in the development of innovation skills as there is enough empirical evidence in the literature, which matches market orientation and innovation in a positive, significant way (Pelham & Wilson, 1996; Lukas & Ferrell, 2000; Verhees & Meulenberg, 2004; Kirca et al., 2005; Elg, 2005; Sen, 2006; Grinstein, 2008). Thus, by considering the information presented previously, at this point it can be established the following hypothesis: H1: Higher level of market orientation, higher level of innovation On one hand, the entrepreneurial orientation is considered in literature as a construct that enables firms to take advantage of the opportunities given by the market through their processes and operations regardless of their size and economic sector (Kemelgor, 2002). Similarly, Covin and Slevin (1991) considered that business orientation can be easily measured by three dimensions: proactivity, risk-taking tendency and innovativeness. Proactivity is the skill that enterprises develop to take initiative, especially in key moments (Kwak et al., 2013), besides the ability to anticipate opportunities provided by the market and the participation of such opportunities in emerging markets (Lumpkin & Dess, 1996; Dimitratos et al., 2004). Risk-taking tendency is considered in literature as one of the most important operations in enterprises as it produces an appropriate environment for the integration and development of innovation activities (Hult et al., 2004). Lastly, innovativeness involves the implementation and development of creative, unusual activities or new solutions to problems and needs found in businesses which leads firms to integrate new ideas or methods that simplify the operation of the firms (Kwak et al., 2013). Hence, Hult et al. (2004) as well as Hurley and Hult (1998) considered in their corresponding papers that innovativeness is the notion that firms have about accepting new ideas provided by their workers and employees as well as the integration of a new organizational culture. In this regard, entrepreneurial orientation has become nowadays in one of the most important business strategies in enterprises, mostly in SMEs, since it enables businesses to produce a trend in pioneering innovation activities ahead of their main competitor (Miller, 1983). With this perspective, entrepreneurial orientation does not only increase significantly the skills of organizations but it also provides technical knowledge so businesses create technical solutions to satisfy the preferences and needs of customers and current/potential consumers (Workman, 1993; Gatignon & Xuereb, 1997). Similarly, the integration and implementation of entrepreneurial orientation of firms demands that they develop a probing nature and risk-taking which can be considered as essential mechanisms in the improvement of processes in products innovation (Miller, 1983; Slater & Narver, 1995; Lumpkin & Dess, 1996). Consequently, entrepreneurial orientation (defined as an organizational strategy) enables the implementation of effective and efficient actions even in those innovation activities of products or services that require a high level of risk and a high financial content (Atuahene-Gima & Ko, 2001). Likewise, Zhou et al. (2005) came to the conclusion that entrepreneurial orientation has significant positive 1932


effects in innovation. Similarly, Avlonitisa and Salavoub (2007) considered that firms which are entrepreneurial (and, logically, have adopted and implemented entrepreneurial orientation), are commonly identified by the high level of risk and by the proactive, competitive attitude that they take in the development and introduction of innovative products that are novel in the market in which the organizations participate. Therefore, considering the information portrayed earlier, the following hypothesis can be established: H2: Higher level of entrepreneurial orientation, higher level of innovation On the other hand, different researchers and academics consider that both market orientation and entrepreneurial orientation are two philosophical disciplines which are too similar (Hills & LaForge, 1992; Morris et al., 2002). Both of them have as a main goal to satisfy the needs of customers as well as to answer quickly the demands established by the external environment of businesses (Kwak et al., 2013). Furthermore, Webster (1981), Zeithaml and Zeithaml (1984) as well as Wiklund and Shepherd (2005), in their papers, came to the conclusion that entrepreneurial orientation can be considered as a proactive element of market orientation. At the same time, Slater and Narver (1995) concluded that entrepreneurial orientation can be considered as an essential complement of market orientation since enterprises need the integration and implementation of both orientations in order to achieve a higher level of business revenue and innovation. Thus, Slater and Narver (1995) recommended that both market orientation and entrepreneurial orientation can provide a change in the firms’ organizational culture, in a way that organizations can obtain a higher apprenticeship so they have better opportunities of getting a higher level of business revenue and, consequently, a higher additional value for their customers and consumers. Moreover, Hult and Ketchen (2001) as well as Morris et al. (2002) came to the conclusion that market orientation and entrepreneurial orientation are essential resources in the organization. Also, both orientations contribute to get a higher level of financial revenue of the enterprise. Similarly, Atuahene-Gima and Ko (2001) concluded in their paper that both market orientation and entrepreneurial orientation give a better performance of the measurement of market participation, market access and the level of quality of the products made by the organization. This leads to the existence of a positive and significant relationship between both orientations. In this regard, Gonzalez-Benito et al. (2009) showed empirically in their research carried out in Spanish firms the existing relationship between disciplines of market orientation and entrepreneurial orientation. These researchers found a positive, significant relationship between the two orientations up to a point that they concluded that a high degree of market orientation from the organizations implies a high degree of entrepreneurial orientation. This also proves that both orientations “are two common elements that can easily complement one with the other” (Gonzales-Benito et al., 2009, p. 516). Hence, considering the information showed previously, the following hypothesis can be established: H3: Higher level of market orientation, higher level of entrepreneurial orientation

3. Methodology In order to answer the three hypotheses presented in the theoretical framework about the existing relationship between market orientation, entrepreneurial orientation and innovation, an empirical study was carried out in 318 SMEs in Aguascalientes State (Mexico), by taking into account the directory of the Business Information System for Mexico in Aguascalientes State (Sistema de Información Empresarial de México, in spanish), which had 5,194 registered companies on June 2013. For practical purposes of this research, the only enterprises that were

1933


considered were the ones that had between 5 and 250 employees, and for this reason the directory was reduced to 1,261 firms. The sample was selected randomly and considering a reliability level of 96% and a sampling error of ±4.5%, which gives a total of 368 firms. Similarly, the recollection instrument of information was designed to be completed by the managers of SMEs; it was carried out as a personal interview to the 400 selected firms. From these, 318 were validated which is a response rate of 87% and were applied during from August to October 2013. A scale proposed by Kohli and Jaworski (1990) was considered in order to measure market orientation. The scale establishes that market orientation can be measured in three dimensions: customer orientation measured by a six-item scale; competitor orientation measured by a four-item scale and; interfunctional coordination measured by a five-item scale. Entrepreneurial orientation, a scale proposed by Miller (1983) was used with adaptations from Covin and Slevin (1991), Lumpkin and Dess (2001) as well as Dess and Lumpkin (2005) who established that this orientation can be measured in three dimensions: proactivity measured by a six-item scale; risk-taking measured by a six-item scale and; innovativeness measured by a six-item scale. Finally, innovation was measured by a five-item scale and it was adapted from Baker and Sinkula (1999, 2009). All the items of the three scales used were measured by a five-point Likert scale (from 1 = Not very important to 5 = Very important as limits). In order to evaluate the reliability and validity of the measurement scales used in this paper, a Confirmatory Factor Analysis (CFA) was carried out by using the method of maximum likelihood with the software EQS 6.1 (Bentler, 2005; Brown, 2006; Byrne, 2006). The reliability of the theoretical model was evaluated by Cronbach’s alpha and the Composite Reliability Index (CRI) (Bagozzi & Yi, 1988). Additionally, the recommendations made by Chou, Bentler and Satorra (1991) and by Hu, Bentler and Kano (1992) were taken into consideration regarding the correction of statistics of the theoretical model when it is considered that the normalcy of data is present by using also the robust statistics which give a better statistical adjustment of data (Satorra & Bentler, 1988). The CFI results are shown in Table 1 and they indicate that the relationship between market orientation, entrepreneurial orientation and innovation have a good adjustment (S-BX2 = 433.502; df = 394; p = 0.000; NFI = 0.883; NNFI = 0.987; CFI = 0.988; y RMSEA = 0.018). Likewise, all the items of related factors are significant (p < 0.001). The size of all the standardized factorial loads are above the value 0.60 (Bagozzi & Yi, 1988). Cronbach’s alpha and CRI have a value above 0.70 and the Average Variance Extracted (AVE) has a value above 0.50 (Fornell & Larcker, 1981). These values indicate that there is sufficient evidence of convergent validity and reliability, which justifies the internal reliability of the scales (Nunally & Bernstein, 1994; Hair et al., 1995). Table 1 Variable

Customer Orientation (F1)

Competitor Orientation (F2) Interfunctional Coordination (F3)

Internal Consistence and Convergent Validity Evidence of the Theoretical Model Factor Robust Average Factor Cronbach’s Indicator CRI Loading t-value Loading Alpha OC1 0.767*** 1.000a OC2 0.772*** 12.523 OC3 0.656*** 10.300 0.720 0.838 0.844 OC4 0.659*** 9.986 OC6 0.749*** 9.154 OP1 0.732*** 1.000a 0.733 0.774 0.777 OP3 0.757*** 7.977 OP4 0.710*** 8.800 CI2 0.735*** 1.000a 0.721 0.754 0.768 CI3 0.827*** 10.376

AVE

0.521

0.536

0.528

(Table 1 to be continued)

1934

The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs (Table 1 continued) CI4 0.601*** 8.047 F1 0.789*** 11.563 Market Orientation F2 0.919*** 5.094 0.783 0.830 F3 0.642*** 8.736 PR1 0.687*** 1.000a PR2 0.724*** 11.484 Proactivity PR4 0.716*** 11.979 0.723 0.839 (F4) PR5 0.735*** 11.892 PR6 0.751*** 12.378 TR1 0.729*** 1.000a TR4 0.781*** 8.248 Risk-Taking 0.736 0.819 (F5) TR5 0.744*** 8.916 TR6 0.691*** 8.784 IN1 0.725*** 1.000a IN2 0.640*** 13.033 Innovativeness IN3 0.794*** 13.076 0.746 0.861 (F6) IN4 0.756*** 12.169 IN5 0.814*** 15.477 F4 0.907*** 10.318 Entrepreneurial 0.827 0.867 F5 0.780*** 8.535 Orientation F6 0.795*** 9.893 II1 0.683*** 1.000a II2 0.708*** 10.593 Innovation II3 0.773*** 9.857 0.765 0.875 II4 0.869*** 10.982 II5 0.792*** 10.247 S-BX2 (df = 394) = 433.502; p < 0.000; NFI = 0.883; NNFI = 0.987; CFI = 0.988; RMSEA = 0.018 Note: a = Parameters constrained to the value in the identification process; *** = p < 0.01.

0.831

0.627

0.845

0.523

0.826

0.543

0.863

0.560

0.868

0.688

0.877

0.589

Regarding the discriminating validity of the theoretical model, the evidence is shown in two ways which can be observed in Table 2. Firstly, it can be seen the confidence interval test (proposed by Anderson & Gerbing, 1988), which establishes that, with an interval of 95% of reliability, none of the individual elements of the latent factors of the correlation matrix has the value of 1.0. Secondly, it can be seen the extracted variance test (proposed by Fornell & Larcker, 1981) which indicates that the variance extracted between each pair of constructs is higher than their corresponding AVE. Therefore, according to the results obtained from both tests, it can be concluded that both measurements show enough evidence of discriminating validity from the theoretical model. Table 2 Variables Market Orientation Entrepreneurial Orientation Innovation

Discriminant Validity Measuring of the Theoretical Model Market Orientation 0.627 0.228-0.316 0.164-0.264

Entrepreneurial Orientation 0.074 0.688 0.205-0.309

Innovation 0.046 0.066 0.589

The diagonal represents the Average Variance Extracted (AVE), whereas above the diagonal part presents the Variance (the correlation squared). Below the diagonal, it is shown the correlation estimation of the factors with a confidence interval of 95%.

1935


4. Results In order to prove the hypotheses presented in the theoretical model, a structural equations modeling with software EQS 6.1 by means of CFA of second order was applied (Bentler, 2005; Byrne, 2006; Brown, 2006). In it, the nomological validity of the theoretical model was examined through the Chi-square test, which compared the results obtained between the theoretical model and the measurement model. Such results indicate that the differences between both models are not significant which can offer an explanation of the relationships observed among the latent constructs (Anderson & Gerbing, 1988; Hatcher, 1994). Table 3 shows these results in a more detailed way. Table 3

Structural Equation Modeling Results from the Theoretical Model Standardized path Hypothesis Path coefficients H1: Higher level of market orientation, higher level of Market Orientation → Innovation 0.489*** innovation. H2: Higher level of entrepreneurial orientation, higher Entrepreneurial O. → Innovation 0.445*** level of innovation. H3: Higher level of market orientation, higher level of Market. O. → Entrepreneurial O. 0.843*** entrepreneurial orientation. 2 S-BX (df = 395) = 574.249; p < 0.000; NFI = 0.845; NNFI = 0.940; CFI = 0.945; RMSEA = 0.038

Robust t-Value 3.119 2.967 5.417

Note: *** = P < 0.01

Table 3 shows the results obtained from the implementation of the second order structural equations modeling and regarding to the hypothesis H1 the results obtained, β = 0.489, p < 0.01, indicate that market orientation has significant positive effects in the innovation of SMEs in Aguascalientes (Mexico). Regarding to the hypothesis H2 the results obtained, β = 0.445, p < 0.01, indicate that entrepreneurial orientation has significant positive effects in the innovation of SMEs. Regarding to the hypothesis H3 the results obtained, β = 0.843, p < 0.01, indicate that market orientation has a positive and significant impact in the entrepreneurial orientation of SMEs in Aguascalientes.

5. Conclusion and Discussion The results obtained in this empirical study can conclude that, on one hand, both market orientation and entrepreneurial orientation have direct implications in the innovation of SMEs of Aguascalientes State (Mexico). If SMEs adopt and implement activities of market orientation and entrepreneurial orientation as business strategies or as daily activities, the innovation activities of enterprises will increase vastly. On the other hand, if more activities of market orientation are adopted by SMEs, the activities of entrepreneurial orientation will also be increased. Therefore, it is possible to conclude that the innovation of SMEs will have a better performance and it will be more efficient and effective if the organizations develop the activities and actions of market orientation and entrepreneurial orientation. This implies, on one hand, that the managers of SMEs have to carry out the corresponding actions so the organization is capable of adopting and implementing the activities that lead to market orientation. In other words, the necessary adjustments need to take place so enterprises recollect as much information as possible from their customers; so they know more precisely their likes and needs; so they make the necessary changes to their products and services based on the preferences of their customers and carry out innovations on their products and

1936


services based on the demand from their current and potential consumers. To put it another way, so the enterprises adopt a customer orientation. Similarly, SMEs have to carry out a series of activities leading to integrate a competitors orientation, i.e. they have to take steps towards identifying the activities done by their main competitors, the prices and costs that they have in similar products, the advertising and promotional campaigns that are implemented as well as the adjustments or innovations made to products and services that they offer in the market. Moreover, managers of SMEs will have to carry out the required actions into the organization so an interfunctional coordination is adopted in which every department or functional area shares the information about customers and competitors in a way that the actions have a consensus among those departments or areas. In this regard, if managers of SMEs carry out the activities needed by the integration and implementation of market orientation, then the firms will have more opportunities of doing innovation activities in their products and services which will eventually lead them to get a higher revenue, bigger competitive advantages and a greater level of competitiveness. Hence, market orientation allows firms to improve their innovation activities significantly as they have to do a series of action that enables them not only adapt their products and services to current and future consumers but also the activities that provide enough information of their main competitors as well as activities that ease decision making among departments or functional areas of the organization. On the other hand, managers also have to carry out the corresponding activities to integrate and implement the entrepreneurial orientation. This can direct enterprises to take actions to be more proactive than their main competitors by trying to be the first ones in making changes or adjustments to their products and service to be suitable to the needs of their customers. They also have to take higher risks in the design and implementation of business strategies, try to adopt an entrepreneurial attitude in the new businesses demanded by the market which are areas with a higher probability of getting better results than the ones their main competitors could obtain. Similarly, in order for enterprises to be more innovative, managers also have to be innovative. They have to create an environment within the organization that allows both workers and employees to express their idea freely and in a consensus with the other departments or functional areas to look for a solution to the problems faced by the firm. As a consequence, if enterprises want to increase their level of innovation significantly, managers will have to integrate in their business strategies both the market orientation and the entrepreneurial orientation because these two orientations do not only produce positive, significant results in innovation activities but they also promote the innovation in SMEs. It is worth noting that this research has some limitations that are important to be considered. One of them is related to the use of measurement scales in both of the market and entrepreneurial orientation as well as innovation because only three factors or dimensions were considered to measure the two orientations and five items to measure innovation. Further studies will need to incorporate different scales to prove the results obtained here. A second limitation it the obtainment of information since only a small part of it has been considered for both market and entrepreneurial orientation as well as innovation with qualitative variables. More research will be needed to incorporate quantitative variables to demonstrate if the same results are obtained. A third limitation is about the measurement of variables from the three scales that were used as it was used an average of five items to measure each one of the three dimensions of market orientation, six items to measure each one of the three factors of entrepreneurial orientation and only five items to measure innovation. In further studies, it will be necessary to use other items or a higher number of items in order to measure the three constructs. A fourth limitation is that the interviews were applied only to managers and/or owners of SMEs so the results 1937


obtained can vary significantly if a different population is used such as customers and suppliers. Therefore, other future studies should incorporate these people to verify the results obtained. Finally, the last limitation is that only SMEs with 5 to 250 employees from Aguascalientes State (Mexico) were considered. Further investigations will need to consider enterprises with less than five employees as they represent more than 60% of the population in order to prove the results obtained. Also, it is considered wise to go beyond the results obtained in this research in order to analyze and discuss more deeply the following: what effects would innovation have in SMEs if a quantitative scale were used to measure both market orientation and entrepreneurial orientation? What results would be obtained in innovation activities of SMEs if other factors or dimensions were used to measure market orientation and entrepreneurial orientation? What specific activities from market orientation and entrepreneurial orientation have more positive, significant effects in the innovation of SMEs? These and many other questions may be answered in posterior investigations. References: Anderson J. and Gerbing D. (1988). “Structural equation modeling in practice: A review and recommended two-step approach”, Psychological Bulletin, Vol. 13, pp. 411-423. Atuahene-Gima K. and Ko A. (2001). “An empirical investigation of the effect of market orientation and entrepreneurship orientation alignment on product innovation”, Organization Science, Vol. 12, No. 1, pp. 54-74. Avlonitisa G. J. and Salavoub H. E. (2007). “Entrepreneurial orientation of SMEs, product innovativeness, and performance”, Journal of Business Research, Vol. 60, No. 5, pp. 566-575. Bagozzi R. P. and Yi Y. (1988). “On the evaluation of structural equation models”, Journal of the Academy of Marketing Science, Vol. 16, No. 1, pp. 74-94. Baker W. E. and Sinkula J. M. (1999). “The synergistic effect of market orientation and learning orientation on organizational performance”, Journal of the Academy of Marketing Science, Vol. 27, pp. 411-427. Baker W. E. and Sinkula J. M. (2009). “The complementary effects of market orientation and entrepreneurial orientation on profitability in small business”, Journal of Small Business Management, Vol. 47, No. 4, pp. 443-464. Bentler P. M. (2005). EQS 6 Structural Equations Program Manual, Encino, CA: Multivariate Software. Bhuian S. N., Menguc B. and Bell S. J. (2005). “Just entrepreneurial enough: The moderating effect of entrepreneurship on the relationship between market orientation and performance”, Journal of Business Research, Vol. 58, No. 1, pp. 9-17. Brown T. (2006). Confirmatory Factor Analysis for Applied Research, New York, NY: The Guilford Press. Byrne B. (2006). Structural Equation Modeling with EQS, Basic Concepts, Applications, and Programming (2th ed.), London: LEA Publishers. Cadogan J. W., Diamantopolulos A. and Siguaw J. A. (2002). “Export market-oriented activities: Their antecedents and performance consequences”, Journal of International Business Studies, Vol. 33, No. 3, pp. 615-626. Chou C. P., Bentler P. M. and Satorra A. (1991). “Scaled test statistics and robust standard errors for nonnormal data in covariance structure analysis”, British Journal of Mathematical and Statistical Psychology, Vol. 44, pp. 347-357. Covin J. G. and Slevin D. P. (1991). “A conceptual model of entrepreneurship as firm behavior”, Entrepreneurship Theory and Practice, Vol. 16, No. 1, pp. 7-25. Deshpandé R., Farley J. U. and Webster F. E. (1993). “Corporate culture, customer orientation and innovativeness in Japanese firms: A quadrad analysis”, Journal of Marketing, Vol. 57, No. 1, pp. 23-37. Dess G. G. and Lumpkin G. T. (2005). “The role of entrepreneurial orientation in stimulating effective corporate entrepreneurship”, Academy of Management Executive, Vol. 19, No. 1, pp. 147-156. Dimitratos P., Lioukas S. and Carter S. (2004). “The relationship between entrepreneurship and international performance: The importance of domestic environment”, International Business Review, Vol. 13, No. 1, pp. 19-41. Elg U. (2005). “A study of inter-firm market orientation dimensions on Swedish, British and Italian supplier-retailer relationships”, Land University, Institute of Economic Research, Working Paper Series No. 2005/6. Fornell C. and Larcker D. (1981). “Evaluating structural equation models with unobservable variables and measurement error”, Journal of Marketing Research, Vol. 18, pp. 39-50. Gatingnon H. and Xuereb J. M. (1997). “Strategic orientation of the firm and new product performance”, Journal of Marketing

1938

The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs Research, Vol. 34, No. 1, pp. 77-90. Gonzalez-Benito O., Gonzalez-Benito J. and Muñoz-Gallegos P. A. (2009). “Role of entrepreneurship and market orientation in firms’ success”, European Journal of Marketing, Vol. 43, No. 3/4, pp. 500-522. Gopal R. (2008). “Effect of customer services efficiency and market effectiveness on dealer performance”, Department of Marketing, Business Division, Monterrey Institute of Technology and Higher Education, ITESM. Grinstein A. (2008). “The effect of market orientation and its components on innovation consequences: A meta-analysis”, Journal of the Academy of Marketing Science, Vol. 36, No. 2, pp. 166-173. Hair J. F., Anderson R. E., Tatham R. L. and Black W. C. (1995). Multivariate Data Analysis with Readings, New York, NY: Prentice-Hall. Hammond K. I., Webster R. L. and Harmon H. A. (2006). “Market orientation, top management emphasis and performance within university schools of business: Implications for universities”, The Journal of Marketing Theory and Practice, Vol. 14, No. 1, pp. 69-85. Han K. J., Kim N. and Srivastava R. K. (1998). “Market orientation and organizational performance: Is innovation a missing link?”, Journal of Marketing, Vol. 62, No. 4, pp. 30-45. Hatcher L. (1994). A Step by Step Approach to Using the SAS System for Factor Analysis and Structural Equation Modeling, Cary, NC: SAS Institute Inc. Hills G. and LaForge R. W. (1992). “Research at the marketing interface to advance entrepreneurship”, Entrepreneurship Theory and Practice, Vol. 16, No. 3, pp. 33-59. Hu L. T., Bentler P. M. and Kano Y. (1992). “Can test statistics in covariance structure analysis be trusted?”, Psychological Bulletin, Vol. 112, pp. 351-362. Hult G. T. M. and Ketchen D. J. (2001). “Does market orientation matter? A test of the relationship between positional advantage and performance”, Strategic Management Journal, Vol. 22, No. 9, pp. 899-906. Hult G. T. M., Hurley R. F. and Knight G. A. (2004). “Innovativeness: Its antecedents and impact on business performance”, Industrial Marketing Management, Vol. 33, No. 5, pp. 429-438. Hurley R. F. and Hult G. T. M. (1998). “Innovation, market orientation and organizational learning: An integration and empirical examination”, Journal of Marketing, Vol. 62, No. 3, pp. 42-54. Kara A., Spillan J. E. and DeShields O. W. (2004). “An empirical investigation of the link between market orientation and business performance in non-profit service providers”, Journal of Marketing Theory and Practice, Vol. 12, No. 2, pp. 59-72. Kemelgor B. H. (2002). “A comparative analysis of corporate entrepreneurial orientation between selected firms in the Netherlands and the USA”, Entrepreneurial & Regional Development, Vol. 14, No. 1, pp. 67-87. Kirca A. H., Jayachandran S. and Bearden W. O. (2005). “Market orientation: A meta-analytic review and assessment of its antecedents and impact on performance”, Journal of Marketing, Vol. 69, No. 1, pp. 24-41. Kohli A. K. and Jaworski B. J. (1990). “Market orientation: The construct, research positions and managerial implications”, Journal of Marketing, Vol. 54, No. 4, pp. 1-18. Kok R. A. W., Hillebrand B. and Biemans W. G. (2003). “Market-oriented product development as an organizational learning capability: Findings from two cases”; SOM Research Institute. Kumar V., Jones E., Venkatesan R. and Leone R. P. (2011). “Is market orientation a source of sustainable competitive advantage or simply the cost of competing?”, Journal of Marketing, Vol. 75, No. 1, pp. 16-30. Kwak H., Jaju A., Puzakova M. and Rocereto J. F. (2013). “The connubial relationship between market orientation and entrepreneurial orientation”, Journal of Marketing Theory and Practice, Vol. 21, No. 2, pp. 141-161. Li J. J. (2005). “The formation of management networks of foreign firm in China: The effect of strategic orientation”, Asia Pacific Journal of Management, Vol. 22, No. 4, pp. 423-443. Li L., Ching-Yick-Tse E. and Yan-Gu B. (2006). “The relationship between strategic planning and entrepreneurial business orientation”, The Chinese Economy, Vol. 39, No. 6, pp. 77-82. Lukas B. A. and Ferrell O. C. (2000). “The effect of market orientation on product innovation”, Academy of Marketing Science Journal, Vol. 28, No. 2, pp. 239-247. Lumpkin G. T. and Dess G. G. (1996). “Clarifying the entrepreneurial orientation and linking it to performance”, Academy of Management Review, Vol. 21, No. 1, pp. 135-172. Lumpkin G. T. and Dess G. G. (2001). “Linking two dimensions of entrepreneurial orientation to firm performance”, Journal of Business Venturing, Vol. 16, pp. 429-451. Matsuno K., Mentzer J. T. and Ozsomer A. (2002). “The effects of entrepreneurial proclivity and on business performance”, Journal

1939

The Relationship between Market Orientation, Entrepreneurial Orientation, and Innovation: Evidence from Mexican SMEs of Marketing, Vol. 66, No. 7, pp. 18-32. Miles M. P. and Arnold D. R. (1991). “The relationship between marketing orientation and entrepreneurial orientation”, Entrepreneurship Theory and Practice, Vol. 15, pp. 49-65. Miller D. (1983). “The correlates of entrepreneurship in three types of firms”, Management Sciences, Vol. 29, No. 7, pp. 770-791. Morris M. H., Schindehutte M. and LaForge R. W. (2002). “Entrepreneurial marketing: A construct for integrating emerging entrepreneurship and marketing perspectives”, Journal of Marketing Theory and Practice, Vol. 10, No. 4, pp. 1-19. Narver J. D. and Slater S. F. (1990). “The effect of market orientation on business profitability”, Journal of Marketing, Vol. 5, pp. 20-35. Nunnally J. C. and Bernstein I. H. (1994). Psychometric Theory (3rd ed.), New York: McGraw-Hill. Pelham A. M. and Wilson D. T. (1996). “A longitudinal study of the impact of marketing structure, firms structure, strategy, and marketing orientation culture on dimensions of small-firm performance”, Journal of the Academy of Marketing Science, Vol. 24, No. 1, pp. 27-43. Renko M., Carsrud A. and Brännback M. (2009). “The effect of a market orientation, entrepreneurial orientation, and technological capability on innovativeness: A study of young biotechnology ventures in the United States and Scandinavia”, Journal of Small Business Management, Vol. 47, No. 3, pp. 331-369. Reza M. A. and Tajeddini K. (2011). “Effect of market orientation and entrepreneurial orientation on innovation: Evidence from auto parts manufacturing in Iran”, Journal of Management Research, Vol. 11, No. 1, pp. 20-30. Satorra A. and Bentler P. M. (1988). “Scaling corrections for chi square statistics in covariance structure analysis”, American Statistics Association 1988 Proceedings of the Business and Economic Sections, pp. 208-313. Schindehutte M., Morris H. and Kocak A. (2008). “Understanding market-driving behavior: The role of entrepreneurship”, Journal of Small Business Management, Vol. 46, No. 1, pp. 4-26. Sen B. (2006). “Market orientation: A concept for health library”, Health Information and Libraries Journal, Vol. 23, No. 1, pp. 23-31. Slater S. R. and Narver J. C. (1995). “Market orientation and the learning organization”, Journal of Marketing, Vol. 59, No. 7, pp. 63-74. Tajeddini K. (2010). “Effect of customer orientation and entrepreneurial orientation on innovativeness: Evidence from the hotel industry in Switzerland”, Tourism Management, Vol. 31, pp. 221-231. Verhees F. J. H. M. and Meulenberg M. T. G. (2004). “Market orientation, innovativeness, product innovation and performance in small firms”, Journal of Small Business Management, Vol. 42, No. 2, pp. 134-154. Webster F. E. (1981). “Top management’s concern about marketing: issues of the 1980’s”, Journal of Marketing, Vol. 45, No. 3, pp. 9-16. Wiklund J. (1999). “The sustainability of the entrepreneurial orientation-performance relationship”, Entrepreneurship Theory and Practice, Vol. 24, No. 1, pp. 37-47. Wiklund J. and Shepherd D. (2005). “Entrepreneurial orientation and small business performance: A configurational approach”, Journal of Business Venturing, Vol. 20, No. 1, pp. 71-91. Workman J. P. (1993). “Marketing’s limited role in new product development in one computer systems firm”, Journal of Marketing Research, Vol. 30, pp. 405-421. Zahra S. A. and Covin J. G. (1995). “Contextual influences on the corporate entrepreneurship-firm performance relationship: A longitudinal analysis”, Journal of Business Venturing, Vol. 10, No. 1, pp. 43-58. Zeithaml C. P. and Zeithaml V. A. (1984). “Environmental management: Revising the marketing perspective”, Journal of Marketing, Vol. 28, No. 3, pp. 46-53. Zhou K. Z., Yim C. K. and Tse D. K. (2005). “The effects of strategic orientations on technology and market-based breakthrough innovations”, Journal of Marketing, Vol. 69, No. 4, pp. 42-60.

1940


Financial and Economic Factors Affecting the Entrepreneurial Characteristics of Small and Medium Family Enterprises Roberto Espíritu Olmos, Héctor Priego Huertas, Alejandro Rodríguez Vázquez (University of Colima, Colima, México)

Abstract: The small and medium enterprises (SMEs) constitute the most important maintenance inside the economy of the municipalities in Mexico, thus his high impact affecting the generation of both, employment and production. According with statistical information, in Mexico there are approximately 4 million 15 thousand enterprises, which 99.8% is MSMEs that generate 52% of the Gross Domestic Product (GDP) and 72% of the employment in the country. For the importance of the SMES it is important to do actions to improve the economic environment and to support directly to the companies, with the intention of creating the conditions that can contribute to their establishment, development and consolidation. Nowadays it is possible to observe that one does not see the opening of new companies due to a series of obstacles which the businessmen meet for the lack of knowledge of how to perform the necessary procedures for their entrance into the entrepreneurship and local economic market. Key words: economics; finance; entrepreneurship; SME JEL code: M2

1. Introduction Today small and medium family businesses are struggling to stay in the commercial and industrial market. For this, they must be further strengthened in the commercial sector as competition is one of the major factors affecting the way that companies are located and to increase or decrease their growth within the market. In Mexico the lack of funding is regretted but still they continue to survive. In the present research an analysis of the Small and Medium Enterprises (SMEs) of a Mexican city was made in 2013, using questionnaires to experts in the area to know which is the cause that stops the growth of SMEs in the city. The concept of the company according to the elements that compose it and the unit given to them is as follows, is the economic and social unit where capital and manpower are coordinated to achieve production of goods and services that satisfy a market. Jay (1967), defined as institutions for the effective use of resources by a government (board) to maintain and increase shareholder wealth and provide security and prosperity for

Roberto Espíritu Olmos, Ph.D., University of Colima; research areas/interests: enterprising behavior and financial analysis of enterprises. E-mail: [email protected]. Héctor Priego Huertas, Master in Finance, University of Colima; research areas/interests: enterprising behavior and financial analysis of enterprises. E-mail: [email protected]. Alejandro Rodríguez Vazquez, Master in Taxes, University of Colima; research areas/interests: enterprising behavior and financial analysis of enterprises.E-mail: [email protected]. 1941

Key Determinants of University Selection among International Students in Ghana

employees. Classification published in the Official Gazette of June 30, 2009, according to Table 1, establishes that firm size is determined from the obtained number of workers multiplied by 10%; plus the amount of annual sales by 90%. This figure should be at or below the Maximum Combined Top of each category, ranging from 4.6 in the case of micro, to 250 for medium. Table 1 Stratification Size Sector Micro All trading Small Industry and services Trading Medium Services Industry

Stratification Sized Enterprises in Mexico

Range of number of workers Until 10 From 11 to 30 From 11 to 50 From 31 to 100 From 51 to 100 from 51 to 250

Range of annual sales amount (mp) Until $4 From $4.01 to $100 From $4.01 to $100 From $100.01 to $250 From $100.01 to $250 From $100.01 to $250

*Maximum limit combined 4.6 93 95 235 235 250

Note: *Maximum Combined Top = (Employees) X10% +(Annual Sales) X 90%

The firm size is determined from the score obtained according to the following formula: Score company (number of employees)X10% +(Annual Sales Amount) X 90%, which must be equal to or less than the Top Maximum Combined category. It was considered as the initial hypothesis that employers would point as the main cause for good growth the opportunity of financing to entrepreneurs and not knowing the requirements that need to be met to carry out the start of operations. The study was applied to employers whohave11-30workers and from30-100workers into their hands respectively, according to the stratification published in the Official Journal of the Federation (2009). Problem statement: This research aims to identify the problems that entrepreneurs face when they wish to establish a business and how to solve them. All these problems arise from lack of information on what procedures to follow for the registration with the corresponding authorities. Problems related to the lack of vision for what they want to implement, lack of planning on the business to be undertaken will also rise. Hypothesis: According to the foregoing, the following hypotheses are proposed: H1: Lack of funding is one of the main causes that stops the growth of SMEs. H2: entrepreneurs are not aware of the procedures to be carried out to register their company with the appropriate authorities for funding. According to all these problems the research question arises: What is the problem with the entrepreneurs of small and medium businesses interested in establishing their business face? Objectives:The overall objective is to identify problems that entrepreneurs of small and medium enterprises who wish to settle in the municipal market are facing. And as specific objectives are: Display problem before which call forth Hakkar entrepreneurs face when wanting to start a business and promote economic growth of SMEs. As specific objectives are: Display problem with which these entrepreneurs face when wanting to start a business and promote economic growth of SMEs. Justification: The justification of this research is to find the cause of the decrease that SMEs have recently had in the town, this way one can find that companies do not have adequate procedures to the tax authorities. That is the reason why they have great conflicts within them, which means that the company does not have a good end. 1942


Conflicts cannot be listed because they are too many to be enlist, but this research tries to present the most important or those who are.

2. Literature Review Family businesses are formed by members of the same family, being these who also have enough capacity to guide them to success, values that predominate in these businesses are just the product of beliefs, expressions, culture and social relations between its members. Values are also a very important point for the family business to continue in the market, since most of them are similar to the life of a human being are born, grow, mature and die (Gallo, 2008). It is therefore important to analyze the study of family businesses, because they are very vulnerable to failure over time as reflected in the survival rates of these companies. Gallo (2008) mentions that there are records which indicate that only the first and second generation have a 30% survival and that only 10% to 15% of family businesses reach a third generation. The changes that take place over time and generations are vulnerable points in the life cycle of these companies, as they tend to be the main reasons why family businesses tend to fail in the growth and continuity. Family businesses have certain advantages as already been mentioned as having a greater vision of the company in the long term, greater commitment, dedication, greater participation in decisions and faster solution and staff turnover is reduced staff and therefore agency costs are reduced. Also this type of organization have certain characteristics that distinguishes them from non-family businesses such as (1) these companies have greater economic participation in most countries. (2) but also they are more likely to disappear (Barugel, 2005). For Mexico, family businesses represent an opportunity to create new jobs, and there by supports the economy of the community, and are also involved in increasing the economic structure of the nation (Neubauer & Lank, 1999). This is why we must support; promote these businesses, and its continuity, since they are the future of many young people and there is a possibility that they can continue with the family business. In other words, the closeness that young people have with family organizations and their efforts to contribute to the strengthening are very important factors for the survival of this kind of business. The concept of enterprise in general, according to the elements that compose it and the unity to be given to it is this. It is the economic and social unit where capital and manpower are coordinated to achieve the production of goods or services that satisfy a market. Jay (1967) defines it as Jay (1967) defines it as “Institutions for the effective use of resources by a government (the board) to maintain and increase shareholder wealth and provide security and prosperity to the employees”. According to the Dictionary of the Royal Spanish Academy is defined as “An entity composed of capital and labor, as factors of production and devoted to industrial, commercial or services, for profit and the consequent responsibility”. Valdivia (1963), defined it as the economic and social unit in which capital, labor and management are coordinated to achieve a production that meets the requirements of the human environment in which the company operates. Fernández Arena (1977), mentions that it is the productive unit or service, constituted as practical or legal issues, by integrating resources and uses management to achieve its objectives. In its commercial or economic meaning the company is composed of capital and labor as factors of production devoted to industrial, commercial or services for profit and with responsibility (Rodriguez, 1996). In economics, is defined as the economic operator or autonomous unit of control and decision, using inputs

1943


or factors of production transforms them into goods and services or other inputs. This is not a legal entity but an organization that has defined objectives, such as profit, the common good or charitable purposes, and to which it uses factors of production, producing goods and services (Sepúlveda, 1996). From a fiscal point of view in his article 16° last paragraph of the Tax Code of the Mexican Federation defines the company as “the natural or legal person performing the activities referred to in this Article, either directly through financial trust or through third parties; per establishment means any place of business that, partially or totally, the above business activities are developed”. “The industrial organization, commercial, service, professional, agricultural or any other measures, generated for the regular exercise of an economic activity based on the production, extraction exchange of goods or service delivery, used as fundamental to the fulfillment of that purpose element of capital investment and/or the amount of labor, assuming the benefit in obtaining the own risk of the activity carried out” (Tax Theory and Technique, 2012).

Small and medium sized enterprises better known as SMEs are the largest generators of employment worldwide, as well as increase the economy of all the continents of the world, which is why SMEs are one of the most important variations in the growth of a state or country. According to the Economic Commission for Latin America and the Caribbean (ECLAC), the presence of SMEs in the economic structure of the region is significant, representing a significant percentage of variables such as production, employment and amount of enterprises (Martinez, 2012). In Mexico SMEs are major generators of jobs and contribute to improving the distribution of income. Small and medium enterprises (SMEs) number about 350,000, accounting for about 95 percent of all domestic enterprises, generate 70 percent of national employment (Negrete, 2012). Given the importance of SMEs, it is necessary to implement measures to improve the economic environment and directly supporting companies with the aim of creating the conditions that contribute to the establishment, development and consolidation (SME Observatory, 2012). The family is considered in our country, as the core on which the life of the person is rotated, but in fact this term is longer, because not only parents and children living in the same environment is limited, but it also includes more generations, as grandparents and grandchildren, these at the second and third generation. According to the term family, this comes also apply in family firms where participation is achieved second and third generations. Family businesses are considered and, when administered by a family, as what mentioned Gallo (2008), cited by De la Garza et al. (2011) defined family businesses as enterprises in which: The family has the power of decision and also one family owned the majority of the ownership of the business. The duties and responsibilities of government and management bodies exercising power rests only in some family members. Not all members of the second generation are involved in the family business. Neubauer and Lank (1999) argue that family businesses have important features such as the sharing with family members a vision. The qualities that characterize these businesses are the power, experience and culture. This conceptualization of family businesses has not had an integrated and comprehensive definition, as for the author these distinctive features that make these businesses are the factors that affect the objectives, strategy, business structure, and these factors are management, human resources, ownership pattern and succession. Family businesses are characterized by cultural and traditional narrow that occur within families who run them. In family business the most energy, money and people come into the group, the main feature is the actual relationship that exist within it, on the other hand, not family businesses, are created for a business plan that is represented by shareholders, financial matters, which are in charge of providing the capital for the establishment of the business. Planning and succession must be grounded in ideas transcending new generations. According to Gallo (2008), 1944


states that the succession is the transfer of leadership to the next generation, the reason is that it must worked as a process and not as an event. In family businesses own members are allowed to keep control over the activities that are present within it. Most companies are similar to the existing, the sense that arises is that the firstborn is setting the rules, and the rest of the family is the one that assumes the business work (Basco, 2005).

3. Methodology This research was conducted in a municipality in Mexico during the months of January to June 2013 by means of a questionnaire applied to small and medium entrepreneurs who had a maximum of 100 workers. Measuring instruments were questionnaires to business owners. To carry out this work only a multiple-choice questionnaire was applied for business owners. Once the instrument was formulated, analyzed its structure and content, as well as how to grade the answers, a pilot test was applied to small and medium businesses in the city to analyze its reliability. It was applied to owners of SMEs in order to gather the required information. The population size is the number of elements to be considered for the study, in this case being 789 SMEs Affiliated to the National Chamber of Commerce (CANACO) belonging to a city of Mexico on 2013. Were applied 278 to SMEs entrepreneurs in the city, previously selected according to the formula applied:  PQN n PQ  N 1 Where: N: Is the size of the population. N: Es el tamaño de la población, siendo un total de 11590 estudiantes matriculados. n: Sample size studied. P and Q=1-P are unknown. For maximum uncertainty is considered: P=Q=0.5. are unknown. For maximum uncertainty is considered: () prefijado: for a  = 0.05 representing a confidence level of 95%  = 1.96 Para un = 0.1 representing a confidence level of 90%  = 1.64 εis the permissible error, usually handled values of 2% o 3% Confidence level 90% n = 278

4. Results Below are graphically presented the results related to entrepreneurial activities of family businesses, according to the questionnaire applied to the own owners and relating to economic and financial factors. OWN RESOURCES

FAMILY CAPITAL

CREDIT BANK

OTHERS 0%

35%

40%

25% Figure 1

How Did You Start Your Business? Source: The authors 1945


The result is that most family businesses start operations with a good deal of own and family resources. This means they are the base of the opening leaving far behind the granting of a bank loan. Once the information on how to start their financial activities was obtained, they were asked whether the control activities of the business, are mainly familiar, as shown in Figure 2. YES

NO

25%

75% Figure 2

Is the Main Control of the Business Familiar? Source: The authors

The result shows that the main control is concentrates in the own family, which is concentrated in 75%, delegating a few other activities to the rest of the staff.

YES

NO

25%

75% Figure 3 Are the Main Positions Predominantly Occupied by Members? Source: The authors

The result to that question shows, according to the figure above, coincides with the control activities taken in the family business, so this is consistent with the previous data. Regarding the form of budgeting the family business, they were questioned as follows. YES

NO

20%

80% Figure 4 Do You Perform A Budget According to the Needs of Your Business? Source: The authors

1946

Key Determinants D o University Selection of S amon ng Internationaal Students in G Ghana

The ressult presentedd in the figuree above show ws that, budgeeting is done according too the businesss basic needss required, whhich is consistent with the inherent requuirements of family busineess. YES

NO O

45% 55%

Figure 5

Doo Family Mem mbers Participa ate in Designing the Budget? Soource: The auth hors

It is im mportant to invvolve the mem mbers of the family in thee family businness budgetinng, that most of the familyy businesses taakes them intto account, ass shown in the figure abov ve. What do o you think is the most important factor f or cause ho olding back the g growht of the bussiness?

Everithing iss expensive 17% % Do nott know anythinng about busineess7%

lack of funding 33%

C Competence 43% Figure 6 Causes Hold ding Back the Growth G of the Company Soource: The auth hors

43% off businesses surveyed s saidd that competiition is very big b today, 33% % pointed to the lack of funding fu creditt by financiall institutions is another faactor that hoolds back the growth of SMEs S and enntrepreneursh hip, and 20% % remaining annswers are vaaried. Regardding the nextt question, thhey were askked if they have h any finaancial record about all th heir activitiess during the period in whicch they were working, w whiich is presentted in the following figure. Do yoy haave any financiaal record about all the activites during the period in which w you weree working?

No 20%

Yes 80%

Figure 7 Financial Records R hors Soource: The auth

19477


80% replied yes, what benefits them keep track of their financial activities, although only 20% said no to. This comes to give an advantage to those who perform them, because they will realize the cost that they spend for each of the activities carried out by the business, particularly in developing their budget projections. The main result of the investigation it was found that one of the biggest problems of the owners of family SMEs is competition, one of the main causes holding back the growth of enterprises. Likewise, financing is one of the factors that affect the creation of new business as it is not easy for them to obtain a financial loan or credit, so many are turning to family capital.

5. Conclusions Applied research instruments provided extensive information. This work focuses on the most significant results slowing the growth of SMEs and entrepreneurship in SMEs in local families during the year 2013. Competition was identified as the main cause that slows the growth of SMEs in the city, said by the SME business owners. However, the economic situation is a major factor holding back growth. As an initial hypothesis was considered that the lack of external financing was the main cause that held back the growth of the business for the lack of an adequate accounting system to allow them to develop reliable financial statements to get a loan in a financial institution. As an end-product of this research, part of the hypothesis was fulfilled as the lack of funding, is a cause that affects the growth of SMEs and entrepreneurship in new business in the city. However, it was found that survey respondents considered competition as one of the main factors holding back the growth of SMEs and entrepreneurship, since there are now mostly Microenterprises which capture the attention and meet the needs of the general population. References: Barugel E. A. (2005). “Un código de buenas prácticas para la supervivencia de la Empresa de Familia”, Universidad del CEMA. Basco R. (2005). “Comportamientos en la Dirección y el Gobierno de la Empresa Familiar, Análisis Empírico de la Profesionalización como Garantía de Continuidad”, Tesis doctoral leída en el Departamento de Organización de Empresas, Faculta de Ciencias Económicas y Empresariales de la Universidad de Complutense de Madrid. De la Garza Ramos M. I., Medina Quintero J. M., CheínSchekaibán N. F., Jimenez Almaguer K. P., Ayup González J. and Díaz Figueroa J. G. (2011). “Los valores familiares y la empresa familiar en el nordeste de México”, Cuad. Adm. Bogotá (Colombia),pp. 315-333. Diario Oficial de la Federación (30 de Junio de 2009). Gallo M. A. (2008). Ideas básicas para dirigir la empresa familiar,Barañain: Ediciones Universidad de Navarra. Fernández Arena J. A. (1977). El Proceso Administrativo (11 ed.), México: Diana. Martinez M. (2012). BBC MUNDO.COM. Recuperado el 23 de ABRIL de 2012, de BBC MUNDO.COM: http://news.bbc.co.uk/hi/spanish/business/barometro_economico/newsid_4349000/4349686.stm Negrete M. J. (20 de febrero de 2012). “Tiempo Logístico”, Recuperado el 28 de abril de 2013, de El sistema Informativo de Comercio Exterior, available online at: http://www.tiempologistico.com/. Neubauer F. and Lank A. G. (1999). “La Empresa Familiar”, Como dirigirla para que perdure, Deusto, Bilbao. Observatorio PYME (2012). Recuperado el 5 de Marzo de 2012, available online at: http://www.observatoriopyme.org/encuestas-y-estudios/cifras-de-pymes/. Rodriguez V. J. (1996). Cómo Administrar Pequeñas y Medianas Empresas, México: Edita International Thomson Editores. Teoria y Tecnica tributaria (agosto de 2012). Recuperado el 20 de mayo de 2013, available online at: http://umtributaria1.wordpress.com/2011/03/27/concepto-fiscal-de-empresa/. Jay A. (1967). Management and Theory, Machiavelli. Sepúlveda C. (1996). Diccionario de TerminosEconomicos. Valdivia I. G. (1963). La Sociologia de la Empresa, México: Editorial Jus. 1948