Corporate Ownership & Control / Volume 13, Issue 1 ...

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10

CORPORATE OWNERSHIP & CONTROL

Postal Address:

Postal Box 36 Sumy 40014 Ukraine

Tel: +380-542-698125 Fax: +380-542-698125 e-mail: [email protected] www.virtusinterpress.org

Journal Corporate Ownership & Control is published four times a year, in September-November, DecemberFebruary, March-May and June-August, by Publishing House “Virtus Interpress”, Kirova Str. 146/1, office 20, Sumy, 40021, Ukraine.

Information for subscribers: New orders requests should be addressed to the Editor by e-mail. See the section "Subscription details".

Back issues: Single issues are available from the Editor. Details, including prices, are available upon request.

Advertising: For details, please, contact the Editor of the journal.

Copyright: All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing of the Publisher.

Corporate Ownership & Control ISSN 1727-9232 (printed version) 1810-0368 (CD version) 1810-3057 (online version) Certificate № 7881

Virtus Interpress. All rights reserved.

759


CORPORATE OWNERSHIP & CONTROL Volume 13, Issue 1, 2015, Continued - 10

CONTENTS

THE PERFORMANCE OF AUDIT COMMITTEES IN JORDANIAN PUBLIC LISTED COMPANIES

761

Modar Abdullatif, Hala Ghanayem, Rand Ahmad-Amin, Saleen Al-shelleh, Lara Sharaiha THE EFFECT OF FAMILY CONTROL AND MANAGEMENT ON PERFORMANCE, CAPITAL STRUCTURE, CASH HOLDING, AND CASH DIVIDENDS

775

Márcio Telles Portal, Leonardo Fernando Cruz Basso U.S. COMMUNITY BANKING PERFORMANCE

781

Kabir Hassan, William J. Hippler, Walter Lane DETERMINANTS OF CAPITAL ADEQUACY RATIO: AN EMPIRICAL STUDY ON EGYPTIAN BANKS

802

Osama A. El-Ansary, Hassan M. Hafez INVESTIGATING THE RELATIONSHIP BETWEEN CORPORATE GOVERNANCE AND BANKS’ FINANCIAL PERFORMANCE: EGYPT CASE

810

Amr Youssef, Mohamed Bayoumi THE IMPACT OF CAPITAL STRUCTURE AND CERTAIN FIRM SPECIFIC VARIABLES ON THE VALUE OF THE FIRM: EMPIRICAL EVIDENCE FROM KUWAIT

830

Ahmad Mohammad Obeid Gharaibeh, Adel Mohammed Sarea CORPORATE GOVERNANCE ON FINANCIAL PERFORMANCE OF INSURANCE INDUSTRY

840

Gardachew Worku Fekadu IMPACT OF SUPERVISORY BOARD MEMBERS’ PROFESSIONAL BACKGROUND ON BANKS’ RISK-TAKING

849

Dennis Froneberg, Florian Kiesel, Dirk Schiereck GUIDING CRITERIA FOR OPERATIONAL RISK REPORTING IN A CORPORATE ENVIRONMENT J. Young

760

881

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 THE MARKET CONCENTRATION AND BANKING INDUSTRY PERFORMANCE

897

Mohammed Salameh Anasweh THE LEVEL AND STABILITY OF INSTITUTIONAL OWNERSHIP AND ITS INFLUENCE ON COMPANY PERFORMANCE IN SOUTH AFRICA

905

Waldette Engelbrecht IMPACT OF STRUCTURE ON ORGANISATIONAL PERFORMANCE OF SELECETED TECHNICAL AND SERVICE FIRMS IN NIGERIA Ann I. Ogbo, Nwankwere F. Chibueze, Orga C. Christopher, Igwe A. Anthony

761

918


THE PERFORMANCE OF AUDIT COMMITTEES IN JORDANIAN PUBLIC LISTED COMPANIES Modar Abdullatif, Hala Ghanayem, Rand Ahmad-Amin, Saleen Al-shelleh, Lara Sharaiha

Abstract This study aimed to explore the degree to which audit committees in Jordanian public listed companies possess the necessary characteristics needed to enable them to perform their duties, and the level of their actual performance of these duties. To do so, the study used a questionnaire survey of the views of external auditors with experience in auditing Jordanian public listed companies. The main findings of the study show that audit committees do possess the necessary characteristics needed, but only to a limited extent. In addition, their performance of their duties was also to a limited extent. The study concluded that these findings can be attributed to the family business model dominant in most Jordanian companies, where the demand for effective audit committees is considered low due to limited agency costs involved. Keywords: Jordan; Audit Committees; Corporate Governance; Public Listed Companies; Agency Costs; Performance Jel Classification: M42, G3 * Princess Sumaya University for Technology, Amman, Jordan

1.

Introduction

Audit committees are an important tool of corporate governance. Their perceived importance has been witnessed by their inclusion and increased roles given to them in numerous international regulations related to corporate governance. The importance of the effectiveness of audit committees has increased in the wake of the financial scandals that occurred in the last two decades, and culminated in the enactment of regulations in developed countries (such as the Sarbanes-Oxley Act in the USA) that dealt with corporate governance issues, including audit committees. The trend towards regulating corporate governance has witnessed an increase in developing countries as well as in developed countries, in an attempt to reduce the chances of financial scandals and company failures, given the negative consequences such events have on the national economies of developing countries. According to Shehata (2015), Jordan was an early reformer in terms of adopting corporate governance regulations in the Middle East and North Africa region. However, Jordan tended to adopt corporate governance regulations that are generally very similar to those adopted by developed countries, while the nature of corporate governance systems in its companies is very different. This is due to the Jordanian companies being significantly smaller than those in developed countries, and the majority of them being closely-held family businesses. It is known that in family

businesses, agency costs between owners and managers are generally low, due to the overlap and relations between owners and managers (Songini and Gnan, 2015). This arguably affects the level of demand for effective corporate governance mechanisms, including audit committees (see Abdullatif and Al-Khadash, 2010). Therefore, this study covers the performance of audit committees in practice in the context of a developing country, Jordan, one of the earlier countries in its region to adopt audit committees in its instructions to public listed companies. However, as Jordanian public listed companies are in most cases closely-held and in many cases family businesses, the demand for effective audit committees may be limited due to low agency costs between owners and managers. Therefore, this study aims to cover the actual practice of audit committees in Jordan when performing their duties, in order to assess the level of their effectiveness and relate it to the nature of Jordanian businesses and their corporate governance systems. The study covers the characteristics and duties of audit committees in detail, and aims to answer the following questions:  To what extent do audit committees in Jordanian public listed companies possess the necessary characteristics that enable them to perform their duties effectively?  To what extent do audit committees in Jordanian public listed companies perform their duties regarding financial reporting,

762


general controls, dealing with internal auditors, and dealing with external auditors? The remainder of this study is organised as follows. First, literature on audit committees is reviewed (including an analysis of the status of audit committees in Jordan). Second, the research design is discussed. This is followed by presenting the research findings, and the study conclusions and implications for practice. 2. Literature review 2.1. Audit committees and their roles An audit committee is a corporate governance mechanism that started to appear significantly in the USA and Canada in the 1970s (Porter et al., 2014). Several definitions were given for an audit committee. For example, Rezaee (2009, p. 119) defines it as “a standing committee of the company’s board of directors to act as a liaison between management and the external auditor”. Verschoor (2008, p. 228) defines an audit committee as “A standing committee of the board of directors organized under the by-laws of the corporation. Duties of the committee are prescribed by statute, regulation, and best business practice. They involve oversight of financial reporting, auditing, ethics and compliance, and risk management processes”. Arens et al. (2014, p. 135) define an audit committee as “a selected number of members of a company’s board of directors whose responsibilities include helping auditors remain independent of management”. Most audit committees consist of members of the board of directors who are not members of the company’s executive management (Arens et al., 2014). From these definitions, it can be concluded that an audit committee is a corporate governance tool that uses non-executive directors as a means of control and oversight over several managerial roles such as internal auditing, risk management, compliance, and financial reporting, and that this role includes intervention when a conflict occurs between executive management and the external auditor over financial reporting matters. For an audit committee to effectively operate and achieve its goals, it members have to be independent from the executive management, have financial knowledge, and meet frequently under a well-defined agenda (Rezaee, 2009). Several regulations enacted around the world included issues related to audit committees. Such regulations include the Sarbanes-Oxley Act in the USA, and the Cadbury Report in the UK. Roles of the audit committee regarding the internal control process include oversight of the reliability of the financial reporting, the effectiveness and efficiency of operations, and the compliance with laws and regulations. This includes understanding the processes of establishing and maintaining sufficient and effective internal controls, assessing the internal

control operation and evaluating any weaknesses in it, understanding the documentation of compliance, and reviewing management and auditor reports regarding the effectiveness of internal control (Rezaee, 2009). In particular for financial reporting, the audit committee can improve the quality of financial information through oversight of the financial reporting process (Bedard and Gendron, 2010), including reviewing annual and interim statements and the main accounting principles and estimates used in preparing them (Rezaee, 2009). As for the role of audit committees in oversight on the internal auditing function, this includes overseeing the internal audit plan and its resources and its status in the organisation, and assessing the quality and effectiveness of the internal audit function in fulfilling its duties (Verschoor, 2008). Roles of the audit committee related to external auditing include (based on the Sarbanes-Oxley Act) appointing the external auditor and oversight on the audit firms’ work, including preapproval of any services to be provided by the audit firm, and resolving any financial reporting disagreements between the external auditor and the company’s management (Arens et al., 2014). 2.2. Audit committees in Jordan While there are thousands of companies of different types and legal formations operating in Jordan, audit committees are legally associated with public listed companies. According to the Amman Stock Exchange (ASE) website (accessed on 23/6/2015) there are 236 companies publicly listed on the ASE. These companies belong to different business sectors including financial services, other services, manufacturing, and real estate. According to their financial strength, they are classified into three different levels of listing. Since 1998, Jordanian public listed companies have been required to prepare their financial statements using International Financial Reporting Standards (IFRS), and get them externally audited under International Standards on Auditing. The Big Four audit firms, in addition to many other firms with or without international affiliations, are involved in auditing the financial statements of Jordanian public listed companies. Audit committees were first introduced into the Jordanian legislation in 1998, when the Jordan Securities Commission (JSC) instructions (JSC, 1998) required public listed companies to establish audit committees that consist of three non-executive members of the board of directors. These committees were to meet at least four times annually, and were responsible for discussing the work of external and internal auditors and the annual and interim financial statements, and compliance with the required laws and regulations. Updated legislation with some more details was enacted in 2004 (JSC, 2004), but these instructions did not add any new responsibilities for audit committees (Abdullatif, 2006).

763


In 2009, the JSC issued a code of corporate governance for public listed companies. This code (JSC, 2009) added more details regarding audit committees, including the requirement of financial knowledge for all audit committee members (and financial expertise of at least one member in accounting or finance). The code required audit committees to meet at least four times in a year, and to meet with the external auditor independently from the company’s management at least once in a year. The code added more detail to the nature of the duties of audit committees, but these duties were not significantly expanded, and remained in the areas of oversight over the financial reporting function, and reviewing the work of the external auditor on financial reporting and evaluating internal controls. Unlike the situation in the USA, audit committees in Jordan are not authorised to select the company’s external audit firm, dismiss it, or determine its fees. Jordanian public listed companies are in general closely-held with the family business model dominant. There is limited separation between the management and ownership of companies, and audit committee members, while not executive managers, may be of direct relations with some executive managers, thus limiting the effectiveness of these committees (Abdullatif and Al-Khadash, 2010). It is likely that the traditional agency costs (between owners and managers) in Jordanian public listed companies are low due to the limited separation between ownership and management of companies, and therefore the demand for an effective audit committee may be limited, especially with the weakness of the minority shareholders (see Naciri, 2008). Therefore, audit committees in Jordan might be mainly established as a legally required corporate governance mechanism, while their level of actual performance of their duties is limited, and this is the main issue this study aims to explore. 2.3. Empirical evidence Empirical evidence shows early adoption of audit committees by a significant number of companies in the USA (Mautz and Neumann, 1977) and the UK (Chambers and Snook, 1979) even before the establishment of audit committees became mandatory. In the USA, voluntary formation of audit committees was found to be associated with larger firm size, lower ownership of the company’s shares by managers, higher percentage nonexecutive directors, and the use of a Big Eight audit firm (Pincus et al., 1989). Collier (1993) found relatively similar results in the UK. In addition, audit committees in the USA were found to be associated with fewer fraud allegations, fewer illegal acts, and fewer SEC enforcement actions (McMullen, 1996). Defond and Jiambalvo (1991) found similar results related to associating the effectiveness of audit committees with a lower likelihood of fraud.

Numerous studies of the characteristics and roles of audit committees were conducted worldwide. In developed countries, a large portion of these studies included relating audit committee characteristics (such as independence, financial and/or industry experience, and frequency of meetings) with some output measure of the implementation of their roles (such as improving the processes of internal auditing and external auditing, and improving the quality of financial reporting). Regarding audit committees’ effects on internal auditing and internal control, Mat Zain et al. (2006) found that audit committees that are independent and financially knowledgeable positively affect internal auditors’ assessment of what they contribute to financial statement auditing. Similarly, Sarens et al. (2013) found that more knowledge and experience of audit committee members is associated with more informal interaction with internal auditors in Australia. In the UK, Alzeban and Sawan (2015) found that audit committees characterised by more expertise and frequency of meetings lead to better implementations of internal audit recommendations. Similar results were found by Naiker and Sharma (2009), who reported that external audit experience of audit committee members is associated with better monitoring of internal control and financial reporting. Prigden and Wang (2012) found that audit committees, combined with the use of Big Four audit firms, lead to more internal control quality. Regarding the effects of audit committees on external auditing, Beattie et al. (2000) and DeZoort and Salterio (2001) found that experience of audit committees leads to better support for external auditors in cases of dispute with managers. Similar results were found in Malaysia by Salleh and Stewart (2012a) and Salleh and Stewart (2012b). Hoitash and Hoitash (2009) found that stronger audit committees give better support for external auditors and less likelihood of dismissing them as a result of an unfavourable audit opinion. They found that audit committee independence leads to better auditor independence and audit quality. Finally, Sultana et al. (2015) found that the independence and financial expertise of audit committees is associated with shorter audit report lag. As for the effects of audit committees on financial reporting, audit committees were found to lead to reduction of earnings management (Baxter and Cotter, 2009). Beasley et al. (2000) found that audit committee effectiveness leads to a lower likelihood of fraud. They report that companies that committed fraudulent financial reporting was less likely to have independent audit committees and that their audit committees held fewer meetings. Cohen et al. (2014) found that industry and financial expertise improves the audit committees’ monitoring of financial reporting, while Sultana (2015) found that financial expertise of audit committee members and frequency of audit committee meetings are positively associated with accounting conservatism.

764


In Arab countries, Kamel and Elkhatib (2013) surveyed auditors, senior accountants, and academics on the roles of audit committees in Egypt. They found that the most important roles of audit committees were reviewing accounting policies and practices, and evaluating internal control. When selecting audit committee members, their independence and accounting experience were considered as very important. In Saudi Arabia, Alzeban (2015) surveyed chief internal auditors of Saudi listed companies. He found that compliance with International Standards for the Professional Practice of Internal Auditing is positively affected when audit committee members are independent, meet with the chief internal auditor, and have accounting and auditing knowledge, and when the chief internal auditor has a long tenure. AlTwaijry et al. (2002) interviewed academics, internal auditors, and external auditors about the role of audit committees in Saudi Arabian corporations. They found views questioning the independence and accounting expertise of audit committee members, and that the audit committees have limited working relations with external and internal auditors. Finally, in Tunisia, Adel and Maissa (2013) found a positive relation between audit committee member experience and audit committee frequency of meetings, and more interaction with internal auditors. Few empirical studies were conducted in Jordan regarding audit committees. An early study by AlFarah (2001) found that external auditors view audit committees as generally ineffective, while internal auditors view them more favourably in terms of effectiveness, especially in dealing with internal auditors and discussing the financial statements. In his survey of external auditors, Abdullatif (2006) found that the perceived benefits of audit committees were generally limited in terms of increasing auditor independence, the quality of financial statements, and the probability of detecting fraud and internal control weaknesses. As for the effects of Jordanian audit committee characteristics, Hamdan et al. (2013a) studied the effects of audit committee characteristics on earnings quality. They found that earnings quality is positively affected by financial experience of audit committee members and by the number of their meetings, negatively affected by the size of the audit committee and the level of ownership of its members of the company’s common shares, and unaffected by the independence of audit committee members. However, Hamdan et al. (2013b) found that size of the audit committees and independence and financial expertise of their members are positively related to company financial performance and share performance, but not operational performance. Al-Akra et al. (2010) found that audit committees in Jordan significantly affect the level of disclosure compliance with IFRS, while AlSa’eed and Al-Mahamid (2012) found a positive effect for the understanding of audit committee members of the functions of the audit committees on

financial reporting. Finally, Aljaaidi et al. (2015) found that more frequent meetings of audit committees are associated with a shorter audit report lag. 2.4. The contribution of this study As mentioned above, studies on audit committees in Jordan are limited in their number and in their coverage of relevant issues regarding audit committees. There are very few studies that covered the actual performance of audit committees in Jordanian public listed companies, and this study is, to the best of the researchers’ knowledge, by far the most detailed study to cover this topic in Jordan. This study therefore has the potential to be useful and contribute extensively to our knowledge about the performance of audit committees, and be therefore useful to Jordanian companies, legislators, and policy makers. Internationally, the contribution of this study is likely to be high in terms of expanding our understanding of how audit committees operate in different contexts, taking into account the specific characteristics of these contexts. Most of the legislation and previous studies on audit committees covered contexts of developed countries (mainly Anglo-American countries), and these countries differ significantly from the Jordanian context in terms of the nature of their companies and corporate governance systems. Therefore, studying the performance of audit committees in a very different business environment is likely to significantly contribute to our knowledge on audit committees, and be useful for strengthening their role as a useful corporate governance tool. 3. Research method The research method used in this study is the questionnaire survey. It was selected in order to receive as many responses as possible from qualified respondents. The questionnaire asked about the actual performance of audit committees in Jordanian public listed companies. The respondents were told to answer the questions based on their own experience in practice, and not necessarily what is required by laws or governance codes. The questionnaire used a fivechoices Likert-scale, with value 1 being assigned for strong disagreement, and value 5 for strong agreement. In addition to personal background questions, the questionnaire consisted of five groups of questions. These groups included the characteristics of audit committees, audit committee responsibilities regarding financial statements, general supervision, internal auditing, and external auditing. The statements were selected from the review of local and international literature, and related regulations determining audit committee responsibilities, especially those related to Jordan, such as JSC (2009).

765


A significantly large number of statements in the questionnaire are, to the researchers’ best knowledge, not used in any previous study in Jordan, and the level of detail included in this study’s research method is unprecedented in any Jordanian study (as the precise study topic is very under-researched in Jordan), the fact that gives this study a high potential in terms of its contribution to knowledge. The study population was defined as external auditors with ample experience who work for audit firms that audit at least three Jordanian public listed companies. The choice of external auditors was motivated by their perceived general knowledge about audit committees in Jordanian public listed companies, and therefore audit firms that audit less than three Jordanian public listed companies were excluded when defining the study population. Members of audit committees of Jordanian public listed companies were excluded from the study population due to factors such as their names not being publicised, their possibility of providing biased information about their own performance, and (most importantly) their knowledge being generally limited to only one company, while the study asked about the performance of audit committees in general. Internal auditors were also excluded due to the same last reason. External auditors were therefore seen as the more suitable target study population, due to their more general knowledge about the study topic. Based on the Amman Stock Exchange website, only eleven audit firms met the above mentioned population definition (auditing at least three Jordanian public listed companies). Of these, one firm refused to participate in the study, leaving the researchers with only ten firms. These firms included all firms associated with the Big Four, and six other firms, five of which have international affiliations. Questionnaires (in Arabic) were distributed and collected by hand by the researchers in April and May 2015. The researchers gave the questionnaires to responsible individuals in the firms, who were told to distribute the questionnaires to auditors with ample experience (ideally being middle or highly ranked) to be qualified to answer the questions. The

questionnaires were collected by the researchers about one to three weeks after their administration to the firms. This method of questionnaire distribution “is likely to generate a significantly higher response rate, compared to mail or email distribution, without significantly impairing the reliability of the responses, since the researcher did not interfere with the respondents or otherwise affect them when they completed the questionnaire” (Abdullatif, 2013, p. 65). Based on the sizes of the firms involved, and the degree of willingness of the firms to participate in the study, 131 questionnaires were distributed, and 93 usable responses were returned, giving a 71% response rate. This rate is generally considered very good compared to similar Jordanian studies, especially when targeting middle and high-ranked external auditors. 4. Findings 4.1. Personal background of the respondents Table 1 shows a summary of the personal background of the respondents in the study sample. It can be seen from Table 1 that the vast majority of the respondents had job titles higher than “junior”, that all but one of them had a university degree, that all but five majored in accounting, that about half of them had professional certificates, and that about half of them had more than five years of audit experience. It can also be seen that the majority of respondents worked for relatively large audit firms with international affiliations, with most of them working for a Big Four audit firm. In general, this sample is arguably suitable for this type of study, and compares very favourably to other Jordanian studies surveying views of external auditors. The possible limitation in terms of a minority of the sample being less experienced is offset by the fact that views of the less experienced auditors did not statistically differ significantly from the views of the more experienced auditors (see subsection 4.5).

Table 1. The respondents’ personal background Variable Job Title Junior auditor Semi-senior auditor Senior auditor Supervisor or assistant manager Manager Partner Total Highest academic qualification ofrespondent Below first university degree Bachelor degree Master degree

766

Frequency

Percentage

26 19 25 10 9 3 92

28.2 20.6 27.2 10.9 9.8 3.3 100

1 81 9

1.1 88.0 9.8


Variable Doctoratedegree Total Professional certificates of respondent JCPA CPA Other More than one certificate None Total Scientific specialization of respondent Accounting Finance Other Total Respondent's auditing experience Below 5 years 5-10 years 11-20 years Above 20 years Total Number of auditors in the auditing firm where the respondent works Below 5 auditors 5-9 auditors 10-19 auditors Above20 auditors Total International affiliation for auditing firm where the respondent works Yes, one of the Big Four. International, But not Big Four No international affiliation Total 4.2. Characteristics of audit committees As a starting point, the questionnaire covered the degree to which audit committees in Jordan possess qualities necessary for them to perform their required roles according to Jordanian regulations and any other international-recognised best practices. These audit

Frequency 1 92

Percentage 1.1 100

4 28 4 6 50 92

4.4 30.4 4.4 6.5 54.3 100

87 3 2 92

94.5 3.3 2.2 100

53 26 9 4 92

57.6 28.3 9.8 4.3 100

1 5 17 69 92

1.1 5.4 18.5 75.0 100

61 28 3 92

66.3 30.4 3.3 100

committee characteristics include financial and industry knowledge and experience, independence, board and managerial support and cooperation, and desire to perform duties effectively. Table 2 summarises the respondents’ views on these issues.

Table 2. Views on characteristics of audit committees Statement number S1 S2 S3 S4 S5 S6 S7 S8 S9

Statement

Mean

Audit committees have knowledge about the nature of the company's business. Audit committees have knowledge about accounting matters. Audit committees have at least one person who has experience in accounting and finance. Audit committees have ability to address and evaluate the risk of fraud. Audit committees have ability to assess the risks the company faces and its responses to these risks Audit committees have a high degree of independence. Boards of directors show interest in and support audit committees. Executive management of companies cooperate with audit committees. Audit committees have the desire to achieve effectiveness in their performance. Average mean for all statements

4.26 4.09 4.02

Standard Deviation 0.530 0.789 0.859

3.88 3.82

0.862 0.955

3.82 3.99 3.98 3.91 3.97

0.932 0.684 0.642 0.706

767


It can be seen from table 2 that audit committees in Jordan are perceived to generally possess the necessary qualities to perform their work effectively, but that the level of such possession is not very high, since most statements had an average response of less than 4 out of 5, indicating that the degree of agreement with the statements was moderate, rather than high. It is notable that audit committee members were seen as knowing about the general business of their companies, but less knowledgeable about the related accounting issues, causing caution since audit committees have significant oversight of the financial reporting function. The least supported statement by respondents was S6 regarding audit committee independence, something arguably expected given the perceived close relations between audit committee members and executive managers of their companies. This could likely limit the effectiveness of Jordanian audit committees, especially with levels of their managerial and board cooperation and support also not perceived as being very high. 4.3. Performance of audit committees’ duties In this subsection, the performance of audit committee’s duties regarding financial reporting, general supervision, and relations with internal and external auditors is discussed in detail. Summary findings of the study on these issues are reported in

Tables 3,4,5, and 6. It can be seen from these tables that the general views of respondents are that audit committees perform their roles only to a moderate level (most statements show averages below 4 from 5), rather than to a high level. In detail, it can be inferred from the respondents’ perceptions about the performance of audit committees regarding financial reporting (table 3) that this performance level is not very high, and that audit committees seem to mainly review financial statements, but their tendency to do more than that, such as evaluating internal controls over financial reporting, or providing recommendations on accounting issues or suggesting the use of financial experts is limited. These results are in agreement with those reported in table 2 as to the fact that the financial knowledge of audit committees is somewhat limited and that the support they find from company boards and management is also relatively limited. This can arguably be associated with the low agency costs between managers and owners in Jordanian companies, which lead to lower demand for highquality audit committee performance, given the strong relations between members of the committee and company managers. This would probably lead to audit committees performing the main required role towards financial statements, which generally is reviewing them, but being less inclined to do much more than that.

Table 3. Views on audit committees’ oversight over financial reporting Statement number S10 S11 S12 S13 S14 S15

Statement Audit committees review financial statements before submitting them to the board of directors. Audit committees provide recommendations about accounting policies. Audit committees study and evaluate internal audit and internal control procedures over financial reporting. Audit committees provide recommendations about obtaining expert assistance to evaluate some financial statements items Audit committees provide recommendations about modifying the financial statements based on external audit results. Audit committees study the quality of the financial statements regarding transparency of disclosure. Average mean for all statements

The results regarding audit committees’ roles regarding financial reporting are relatively similar to their roles regarding general supervision. As shown in table 4, it seems that the roles regarding general supervision are also performed only to a moderate level, and also emphasise explicit requirements, such as holding meetings, compared to the more detailed issues to be considered in these meetings, such as dealing with compliance, related party transactions, and quality control. Reasons for such findings can also arguably be associated to low agency costs between owners and managers.

768

Mean

Standard Deviation

4.18

0.824

3.85

0.896

3.86

0.859

3.71

0.939

3.95

0.761

3.87

0.863

3.90

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Table 4. Views on Audit committees’ role in general supervision Statement number S16 S17 S18

S19 S20 S21

Statement

Mean

Audit committee members meet at least four times annually. Audit committee members hold unscheduled meetings when necessary. Audit committees review related party transactions before their approval by the company to avoid any conflict of interest. Audit committees oversee the company’s implementation of corporate governance. Audit committee oversee the company’s compliance with laws, regulations and common standards Audit committee oversee the company’s implementation of quality control standards Average mean for all statements

3.97

Standard Deviation 0.777

4.16

0.601

3.73

0.817

3.79

0.719

3.91

0.798

3.89

0.777

Tables 5 and 6 report on the performance of audit committees’ duties regarding their relations with internal and external auditors. Similar to findings reported in tables 3 and 4, it seems that the roles of audit committees regarding internal and external auditing are only performed to a moderate degree, and that the emphasis is mainly on explicit legal requirements, such as holding meetings. The higher mean responses regarding roles with external auditors, compared to relatively similar roles with internal auditors, can also be arguably attributed to the contact with external auditors being more explicit than the contact with internal auditors. Again, reasons for such findings can arguably be associated to low agency costs between owners and managers. This finding is relatively similar to that of Al-Twaijry et al. (2002) in their study on Saudi Arabian data, where they found that the working relations of audit committees with internal and external auditors are limited. Finally, two issues are worth attention. First, as reported in table 6, the roles of audit committees in

3.91

appointing external auditors and determining their fees and scope of work are the least performed issues by Jordanian audit committees. This can be justified by these roles not being legally required for Jordanian companies, thus supporting the conclusion that legally required explicit responsibilities are significantly performed to a higher degree than other responsibilities considered by some foreign regulations as good practice. It can be arguably concluded that to some extent, audit committees in Jordan are seen as a legal burden that has to be fulfilled rather than an effective governance mechanism. Second, while audit committees are established by Jordanian companies, their level of intervention in resolving accounting-related conflicts between the company’s management and its external auditor is alarmingly low (3.7 mean response). This can arguably also be related to the low demand for highly effective audit committees due to low agency costs involved.

Table 5. Views on audit committees’ relation with internal auditors Statement number S22 S23 S24 S25 S26 S27 S28 S29

Statement

Mean

Audit committee effectively influence selecting the internal auditors. Audit committees provide recommendations to the board of directors about the internal audit and its relation with the external audit function. Audit committees inquire about the internal auditors’ experience and provide related recommendations. Audit committees regularly meet with the internal auditors. Audit committees hold unscheduled meetings with the internal auditors when necessary. Audit committees review the internal audit plan at least once annually. Audit committees discuss suggestions and reservations of the internal auditors. Audit committees discuss the level of management’s response to suggestions and reservations of the internal auditors. Average mean for all statements

3.77

Standard Deviation 0.973

3.99

0.763

3.87

0.788

3.84

0.745

4.04

0.648

3.92

0.774

3.96

0.755

3.89

0.769

769

3.91

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Table 6. Views on audit committees’ relation with external auditors Statement number S30

S31 S32 S33 S34 S35

S36

S37 S38 S39 S40 S41 S42 S43

Statement Audit committees discuss issues related to the appointment of the external auditor and their compliance with the company’s regulations and the degree of their independence. Audit committees review the external audit plan at least once annually. Audit committees meet the external auditor at least once annually. Audit committees hold unscheduled meetings with the external auditor when needed. Audit committees discuss suggestions and reservations of the external auditor. Audit committees discuss the level of management’s response to suggestions and reservations of the external auditor. Audit committees study the external auditors’ plan and ensure that the company provides them with all the documents, evidence and facilities they need. Audit committees review the external auditor’s evaluation of internal control Audit committees inquire about the external auditor’s experience. Audit committees review the company correspondence with external auditors and provide recommendations on this. Audit committees have an effective role in appointing the external auditor. Audit committees have an effective role in determining the external auditor’s fees Audit committees determine the scope of the work of the external auditor. Audit committees intervene for resolving conflicts between the external auditor and the executive management on accounting matters. Average mean for all statements

4.5. Potential effects of personal backgrounds of respondents on their views The researchers tested whether there is a statistically significant effect for the personal backgrounds of the respondents on their views regarding the issues raised in this study. This was performed through splitting the respondents into categories (see Table 1) and repeating that for each personal background variable (job title, academic qualification, professional certificates, academic specialisation, audit experience, number of auditors in audit firm, and international affiliation of audit firm) separately. In some cases, the researchers had to merge some categories due to the very small frequency in some categories. The Kruskal-Wallis test was then run to compare means for the categories of each personal background variable. In general, very few statements for each personal background variable produced statistically significant differences in views, and these differences seemed

Mean

Standard Deviation

4.11

0.836

3.96

0.838

4.10

0.780

4.17

0.636

4.15

0.607

3.91

0.717

3.89

0.744

3.82

0.838

3.78

0.942

3.80

0.905

3.81

0.936

3.56

0.983

3.39

1.243

3.70

0.976

3.87

random, without following any particular pattern. Thus, the researchers concluded that there are generally no differences in views of the respondents that can be attributed to their personal backgrounds [1]. These results enhance the robustness of the findings, and alleviate to some extent the effects of the potential limitation of a few respondents having limited audit experience. Of particular importance in this analysis is whether different Jordanian public listed companies have different characteristics or performance levels of audit committees that may be associated to the difference in size or business nature among these companies. In general, there is no specialisation trend observed in Jordanian audit firms in terms of the business nature of their clients, and (as shown in Table 1), audit firms with international affiliations dominate the market of auditing Jordanian public listed companies. Given that larger companies (notably banks) are generally likely to be audited by Big Four audit firms, the Kruskal-Wallis test mentioned above was applied to compare views of

770


auditors from Big Four firms with views of auditors from other audit firms involved in the study. The views were very similar between the two groups, with no notable trends of statistically significant differences between the views of both groups. This implies that the findings of this study can, to some extent, be generally associated with audit committees of all Jordanian public listed companies, regardless of their size or business nature.

4.6. Reliability of the results To test for the reliability of the findings, the Cronbach’s alpha method was used for each group of statements. As shown in Table 7, all of the Cronbach’s alpha values reported were significantly higher than the minimum acceptable value of 0.7 (Suanders et al., 2012). Therefore, the findings are acceptable in terms of reliability.

Table 7. Cronbach’s alpha reliability test results Cronbach’s Alpha 0.846 0.903 0.843 0.821 0.878

Variable Qualifications of audit committees Audit committees’ oversight over financial reporting. Audit committees’ role in general supervision Audit committees’ relation with internal auditors Audit committees’ relation with external auditors 5. Conclusions This study explored the performance of audit committees of their roles and responsibilities in Jordanian public listed companies. It utilised a questionnaire survey designed to explore the issues of whether audit committees in Jordanian public listed companies do have the necessary characteristics needed for them to effectively perform their duties, and the extent to which they perform their duties regarding oversight of financial reporting, general control and supervision, relations with internal auditors, and relations with external auditors. Given their perceived knowledge about audit committees in Jordanian companies in general (rather than specific knowledge about one company only), external auditors from audit firms that audit the financial statements of at least three Jordanian public listed companies were used as the study population. The main findings of the study were that audit committees in Jordanian public listed companies do generally possess the necessary characteristics to enable them to perform their duties, but only to a limited degree, thus limiting their ability to perform their duties effectively. When exploring the level of performance of detailed responsibilities and duties, similar results were found, as the performance generally happened also to a limited level, and generally emphasised explicit legal requirements (such as holding meetings, or reviewing financial statements) over more-detailed issues. That is, audit committees in Jordan were perceived to be less likely to go into details of financial reporting (such as using experts for valuation, or assessing disclosure transparency), or dealing with internal and external auditors (such as influencing the selection of internal and external auditors and dealing with their recommendations). Comparing these findings to those of the most similar previous study to this study in Jordan, Abdullatif (2006), the results are generally same in that audit committees in Jordan are not

Number of Statements 9 6 6 8 14

sufficiently effective, thus suggesting that things have not improved significantly after about a decade. Suggested reasons for these findings arguably include that, unlike the situation in many developed countries, the demand for effective performance from audit committees in Jordan is not high, given the corporate governance system applied in most Jordanian public listed companies. Family businesses generally have low agency costs between owners and managers, and therefore board members may not demand high-quality practices of corporate governance, especially if they come at a cost they perceive to be higher than the related benefits. This may end up with corporate governance in general, and audit committees in particular, being generally viewed by boards and managers as a legal burden that has to be fulfilled as simple as possible, rather than a useful tool for the company’s success and sustainability. It is argued that in family businesses run as public listed companies, the significant agency costs are not between owners and managers, but rather between majority and minority shareholders (Shaohua, 2010). Minority shareholders will in many cases be weak (see Naciri, 2008), and therefore need intervention by regulatory authorities to enact laws and regulations that are aimed to guard the interests of minority shareholders and other stakeholders involved. With this being the case in Jordan, this study recommends more intervention by the regulatory authorities in Jordan not only to regulate corporate governance issues, including audit committees, but also to ensure the actual implementation of these laws and regulations to an acceptable level by the companies. Audit committees in Jordan should be required and encouraged to adopt a more effective role in the corporate governance process. Their members should be selected with more care, and given more support and cooperation from boards and managers of companies. Audit committees should also be more involved in general oversight of the financial reporting process, and apply their duties

771


towards internal control, dealing with internal auditors, and dealing with external auditors to a larger and more effective degree. This will likely lead to better governance of public listed companies, and therefore potentially improve their share prices and lead to better financial results and business sustainability. This study used a questionnaire survey to explore the performance of audit committees in Jordan. Avenues for future research on audit committees in Jordan may include detailed case studies on audit committee performance in Jordan, and better understanding of the findings of this study through, for example, performing more studies on the association of audit committee characteristics with financial results and share prices. Endnotes 1. Detailed tables of findings for this subsection are omitted because they are too large in size, and generally did not show statistically significant differences in views among respondents. References 1.

2.

3.

4.

5.

6.

7.

8.

Abdullatif, M. (2006), “The effectiveness of audit committees in Jordanian public shareholding companies and potential company characteristics affecting it: Perceptions from auditors in Jordan”, Dirasat: Administrative Sciences, Vol. 33 No. 2, pp. 450-467. Abdullatif, M. (2013), “Fraud risk factors and audit programme modifications: Evidence from Jordan”, Australasian Accounting, Business and Finance Journal, Vol. 7 No. 1, pp. 59-77. Abdullatif, M. and Al-Khadash, H.A. (2010), “Putting audit approaches in context: The case of business risk audits in Jordan”, International Journal of Auditing, Vol. 14 No. 1, pp. 1-24. Adel, B. and Maissa, T. (2013), “Interaction between audit committee and internal audit: Evidence from Tunisia”, The IUP Journal of Corporate Governance, Vol. 12 No. 2, 59-80. Al-Akra, M., Eddie, I.A., and Ali, M.J. (2010), “The influence of the introduction of accounting disclosure regulation on mandatory disclosure compliance: Evidence from Jordan”, The British Accounting Review, Vol. 42 No. 3, pp. 170-186. Al-Farah, A.M.S. (2001), “The effectiveness of the audit committees in Jordanian public shareholding companies – An analytical field study”, Master thesis, The University of Jordan. Aljaaidi, K.S., Bagulaidah, G.S., Ismail, N.A., and Fadzil, F.H. (2015), “An empirical investigation of determinants associated with audit report lag in Jordan”, Jordan Journal of Business Administration, Vol. 11, No. 4, pp. 963-980. Al-Sa’eed, M.A. and Al-Mahamid, S.M. (2012), “The role of effective audit committee in strengthening the financial reporting: Evidence from Jordanian listed companies”, Corporate Ownership and Control, Vol. 9 No. 3, pp. 59-68.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

772

Al-Twaijri, A.A., Brierley, J.A., and Gwilliam, D.R, (2002), “An examination of the role of audit committees in the Saudi Arabian corporate sector”, Corporate Governance: An International Review, Vol. 10 No. 4, pp. 288-297. Alzeban, A. (2015), “Influence of audit committees on internal audit conformance with internal audit standards”, Managerial Auditing Journal, Vol. 30 No. 6/7, pp. 539-559. Alzeban. A. and Sawan, N. (2015), “The impact of audit committee characteristics on the implementation of internal audit recommendations”, Journal of International Accounting, Auditing and Taxation, Vol. 24 No. 1, pp. 61-71. Arens, A.A., Elder, R.J., and Beasley, M.S. (2014), Auditing and Assurance Services: An Integrated Approach, 15th edition, Pearson Education Ltd. Harlow, UK. Baxter, P. and Cotter, J. (2009), “Audit committees and earnings quality”, Accounting and Finance, Vol. 49 No. 2, pp. 267-290. Beasley, M.S., Carcello, J.V., Hermanson, D.R., and Lapides, P.D. (2000), “Fraudulent financial reporting: Consideration of industry traits and corporate governance mechanisms”, Accounting Horizons, Vol. 14 No. 4, pp. 441-454. Beattie, V., Fearnley, S., and Brandt, R. (2000), “Behind the audit report: A descriptive study of discussions and negotiations between auditors and directors”, International Journal of Auditing, Vol. 4 No. 2, pp. 177-202. Bedard, J. and Gendron, Y. (2010), “Strengthening the financial reporting system: Can audit committees deliver?”, International Journal of Auditing, Vol. 14 No. 2, pp. 174-210. Chambers, A.D. and Snook, A.J. (1979), “1978 survey of audit committees in the United Kingdom: A summary of findings”, Working paper No. 10, City University Business School. Cohen, J.R., Hoitash, U., Krishnamoorthy, G., and Wright, A.M. (2014), “The effect of audit committee industry expertise on monitoring the financial reporting process”, The Accounting Review, Vol. 89 No. 1, pp. 243-273. Collier, P. (1993), “Factors affecting the formation of audit committees in major UK listed companies”, Accounting and Business Research, Vol. 23, Supplement 1, pp. 421-430. DeFond, M.L. and Jiambalvo, J. (1991), Incidence and circumstances of accounting errors”, The Accounting Review, Vol. 66 No. 3, pp. 643-655. DeZoort, F.T. and Salterio, S.E. (2001), “The effects of corporate governance experience and financialreporting and audit knowledge on audit committee members’ judgments”, Auditing: A Journal of Practice and Theory, Vol. 20 No. 2, pp. 31-47. Hamdan, A.M.M., Mushtaha, S.M.S., and Al-Sartawi, A.M. (2013a), “The audit committee characteristics and earnings quality: Evidence from Jordan”, Australasian Accounting, Business and Finance Journal, Vol. 7 No. 4, pp. 51-79. Hamdan, A.M., Sarea, A.M., and Reyad, S.M.R. (2013b), “The impact of audit committee characteristics on the performance: Evidence from Jordan”, International Management Review, Vol. 9 No. 1, pp. 32-42.

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 24. Hoitash, R. and Hoitash, U. (2009), “The role of audit committees in managing relationships with external auditors after SOX: Evidence from the USA”, Managerial Auditing Journal, Vol. 24 No. 4, pp. 368397. 25. Jordan Securities Commission (JSC) (1998), Instructions on Disclosure, Accounting and Auditing Standards and Conditions to be Met by Auditors of Parties Under the Supervision of the Jordan Securities Commission (Number 1 for the Year 1998). 26. Jordan Securities Commission (JSC) (2004), Instructions of Issuing Companies Disclosure, Accounting and Auditing Standards for the Year 2004. 27. Jordan Securities Commission (JSC) (2009), Corporate Governance Code for Shareholding Companies Listed on the Amman Stock Exchange. 28. Kamel, H. and Elkhatib, S. (2013), “The perceptions of audit committees’ role in an emerging market: The case of Egypt”, Journal of Economic and Administrative Sciences, Vol. 29 No. 2, pp. 85-98. 29. Mat Zain, M., Subramaniam, N., and Stewart, J. (2006), “Internal auditors’ assessment of their contribution to financial statement audits: The relationship with audit committee and internal audit function characteristics”, International Journal of Auditing, Vol. 10 No. 1, pp. 1-18. 30. Mautz, R.K. and Neumann, F.L. (1977), “Corporate audit committees: Policies and practices”, Ernst and Ernst, New York, USA. 31. McMullen, D.A. (1996), “Audit committee performance: An investigation of the consequences associated with audit committees”, Auditing: A Journal of Practice and Theory, Vol. 15, No. 1, pp. 87-103. 32. Naciri, A. (2008), “The MENA countries national systems of corporate governance”, in Naciri, A. (editor), Corporate Governance Around the World, Routledge, London, UK, pp. 303-322. 33. Naiker, V. and Sharma, D. (2009), “Former audit partners on the audit committee and internal control deficiencies”, The Accounting Review, Vol. 84 No. 2, pp. 559-557. 34. Pincus, K., Rusbarsky, M., and Wong, J. (1989), “Voluntary formation of corporate audit committees among NASDAQ firms”, Journal of Accounting and Public Policy, Vol. 8, No. 4, pp. 239-265.

35. Porter, B., Simon, J., and Hatherly, D. (2014), Principles of External Auditing, 4th edition, John Wiley & Sons Ltd., Chichester, UK. 36. Pridgen, A. and Wang, K.J. (2012), “Audit committees and internal control quality: Evidence from nonprofit hospitals subject to the Single Audit Act”, International Journal of Auditing, Vol. 16 No. 2, pp. 165-183. 37. Rezaee, Z. (2009), Corporate Governance and Ethics, John Wiley & Sons, Inc., Hoboken, USA. 38. Salleh, Z. and Stewart, J. (2012a), “The role of the audit committee in resolving auditor-client disagreements: A Malaysian study”, Accounting, Auditing and Accountability Journal, Vol. 25 No. 8, pp. 1340-1372. 39. Salleh, Z. and Stewart, J. (2012b), “The impact of expertise on the mediating role of the audit committee”, Managerial Auditing Journal, Vol. 27 No. 4, pp. 378-402. 40. Sarens, G., Christopher, J., and Zaman, M. (2013), “A study of the informal interactions between the audit committee and internal audit in Australia”, Australian Accounting Review, Vol. 23 No. 4, pp. 307-329. 41. Saunders, M., Lewis, P., and Thornhill, A. (2012), Research Methods for Business Students, 6th edition , Pearson Education Ltd., Harlow, UK. 42. Shaohua, H. (2010), “Auditor choice, audit fees, and internal governance in family firms”, Master thesis, Lingnan University, Hong Kong. 43. Shehata, N.F. (2015), “Development of corporate governance codes in the GCC: An overview”, Corporate Governance: The Journal of Business in Society, Vol. 15 No. 3, pp. 315-338. 44. Songini, L. and Gnan, L. (2015), “Family involvement and agency cost control mechanisms in family small and medium-sized enterprises”, Journal of Small Business Management, Vol. 53 No. 3, pp. 748-779. 45. Sultana, N. (2015), “Audit committee characteristics and accounting conservatism”, International Journal of Auditing, Vol. 19 No. 2, pp. 88-102. 46. Sultana, N. Singh, H., and Van der Zahn, J-L.W.M. (2015), “Audit committee characteristics and audit report lag”, International Journal of Auditing, Vol. 19 No. 2, pp. 72-87. 47. Verschoor, C.C. (2008), Audit Committee Essentials, John Wiley & Sons, Inc., Hoboken, USA.

773


THE EFFECT OF FAMILY CONTROL AND MANAGEMENT ON PERFORMANCE, CAPITAL STRUCTURE, CASH HOLDING, AND CASH DIVIDENDS Márcio Telles Portal*, Leonardo Fernando Cruz Basso** Abstract This study investigates the effect of family firm on corporate performance and financial policy (capital structure, cash holding, and cash dividends). Using a sample of Brazilian firms, the study uses a treatment effect model to address self-selection and endogeneity problems. The results show that family firm has a negative net effect on performance. Family control has an effect on financial policies that indicate a aversive behavior to preserve control. The results indicate less problem of free cash flow and more risk-taking behavior in family-manage companies, suggesting that such aversion behavior is reduced when the family controls and manages the firm. This is the first study that takes into account the effect of family firm behavior through multiple financial policies. Keywords: Family Firm, Performance, Capital Structure; Cash Holdings; Dividends; Risk Aversion; Risk Taking JEL classification: G12; G15 * Rio Grande do Sul Federal Institute of Education, Science and Technology, Management and Business Department, Rua Domingos Zanella, 104, ZIP 99.713-028, Erechim, Rio Grande do Sul/RS, Brazil ** Mackenzie Presbyterian University, Department of Finance, Rua da Consolação, 930, ZIP 01302-907, São Paulo/SP, Brazil

1. Introduction Existing literature extensively addresses risk perception differences (Hiebl, 2013) between familyowned and managed firms and non-family-owned firms worldwide (La Porta et al., 2000) and the implications on investment decisions (Anderson et al. 2012; Croci et al., 2011; Lee, 2006). However, finance literature provides limited insight into financial outcomes and why any differences between the two types of companies affect corporate performance and their financial policies. The research results are currently inconclusive concerning the effect of family-run firms on financial outcomes: empirical tests and relationship theories of ownership and financial outcomes reveal and predict positive, negative, or zero, relationships depending on the trade-off between alignment and entrenchment (King and Santor 2008). The ambiguity of the empirical results is attributed to two factors. First, the empirical evidence is concentrated in countries such as the US and the UK, which are characterized by firms with dispersed ownership and that comply to the rule "one share-one vote" (Gamma and Galvão, 2012), that is, the characteristics differ from most companies worldwide (La Porta et al., 2000). The literature indicates that the trade-off between the alignment of interests and expropriation by family control is a function of the institutional environment in which the company does business (Gamma and

Galvão, 2012) because the right of minority shareholders and how they are protected depends significantly on the law and quality of enforcement (Shleifer and Vishny, 1986). Additionally, in a weak protective legal environment, the controlling shareholder is dominant (Gamma and Galvão, 2012), but these are environments where controlling families are also more able to expropriate minority shareholders (Faccio et al., 2001). Therefore, the question of how family firms perform in different institutional settings arises. Second, research on the outcomes of ownership, control, and family management has not adequately addressed the potential endogeneity problems (Himmelberg et al., 1999) and isolated the effects of the use of "controlenhancing mechanisms" (Gamma and Galvão, 2012; King and Santor, 2008). Because the empirical research on this topic is currently in the development stage, this study evaluates the effect of family control and management on corporate performance and financial policies such as capital structure, cash dividends, and cash holdings. The Brazilian institutional environment is characterized by a high ownership concentration and low protection to creditors and shareholders. Therefore, Brazil is a favorable environment to expand the evidence on family control/management and its financial outcomes in emerging economies. The results suggest that the family effect on performance and financial policies depends on the

774


nature of the family involvement in control and management. Family firms negatively impact corporate performance. Taken together, the results suggest that firms with family control exhibit risk aversion that is reflected in financial policies, however, that aversion behavior is reduced when the family controls and manages the firm. In contrast, firms with founding family management exhibit risktaking behavior. This study contributes to the literature by investigating the family effect on a set of financial policies, which allows further exploration of the implications of family control/management on the financial behavior of firms. Additionally, this study adequately addresses selection bias and endogeneity problems and increases the evidence from emerging economies. This article is organized as follows. Section 2 reviews the theories and evidence concerning family ownership and management and its impact on performance and corporate financial policies. Section 3 describes the methodological procedures. Section 4 reports the analysis and discusses the results. Section 5 presents the final considerations. 2.

Literature review

2.1 The effect of family firms on corporate performance The effect of family ownership on corporate performance and value remains unclear (Gamma and Galvão, 2012). The theoretical and empirical studies document ambiguous results. Studies report that family ownership structure leads to better, worse, or has no effect on performance. Overall, this literature has been categorized into two axes. The first, led by Berle and Means (1932), suggests a causal effect of governance variables on corporate performance. The other axis, led by Demsetz and Lehn (1985) and Demsetz and Villalonga (2001), suggests that governance variables respond endogenously to firm and industry characteristics and, therefore, without necessarily inducing an observable causal effect on performance. Investigations following Berle and Means (1932) adopt two directions for the causal effect of family on firm performance. The first direction assumes a positive effect from the mitigating agency problem type I (principal-agent) by the alignment mechanism of interest (Gamma and Galvão, 2012). The arguments for this line suggest that the positive effect is associated with two factors. First, family firms, as insiders or blockholders controlling and operating in well-regulated and transparent markets, reduce agency costs (Anderson and Reeb, 2003) and gain advantages from the monitoring and disciplining of agents' decisions (Fama and Jensen, 1983) even when they are not involved in management (Shleifer and Vishny, 1986). Second, family firms are less myopic and have a longer investment horizon, rendering them optimal

investment decision makers (King and Santor, 2008; Stein, 1989). Evidence found by Anderson and Reeb (2003) and McConaughy et al. (1998) supports this view. The other perspective assumes that the family firm may have a negative effect on performance because of type II agency problems (minority shareholder as the controlling shareholder) associated with the entrenchment mechanism (King and Santor, 2008). Entrenchment often uses "control-enhancing mechanisms" (Lease et al., 1984) and family members in management rather than more qualified external professionals (Anderson and Reeb 2003; Schulze et al., 2001; Morck et al., 1988; Shleifer and, Vishny, 1986). This entrenchment reduces market discipline by reducing the effectiveness of the board (Jensen and Ruback, 1983) and access to the managerial labor market (Holmström and Tirole, 1993). When associated with the prevalence of major shareholder family drivers in countries with weak investor protection (Gamma and Galvão, 2012), such entrenchment allows the expropriation of minority shareholders (Faccio et al., 2001) either by consumption of perquisites and excessive salaries or through the loss of lucrative business opportunities because of excessive risk aversion (Morck et al., 2000). Evidence found by Holderness and Sheehan (1988) supports this view. The second line assumes that governance variables are endogenously determined by current and potential shareholders in the profit maximization process given observable and unobservable firm characteristics (Himmelberg et al., 1999; Demsetz, 1983). Therefore, a systematic relationship should not be observable between family control/management and firm performance (Gamma and Galvão, 2012; Demsetz and Villalonga, 2001; Demsetz and Lehn, 1985;). The central argument is that an efficient market for corporate control leads to optimal control structures in accordance with the corporate characteristics to penalize inefficient structures (King and Santor, 2008). Evidence found by Demsetz and Lehn (1985) support this view. For King and Santor, 2008, given the opposition of theories, it is not surprising that the empirical literature has produced mixed results. The research has suggested that the benefits and costs of the family firm on performance depend on the institutional environment that firms operate (Gamma and Galvão, 2012). Given the theoretical ambiguity and inconclusive evidence on the existence and direction of the effect, the relationship between family control/management becomes an empirical subject. 2.2. The impact of family firms on financial policies The effect of control/family management on financial policy remains largely unexplored. The literature has used the capital structure as a proxy for risk control propensity or risk aversion in family firms (Hiebl,

775

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 2013; Gamma and Galvão, 2012). The direction of the effect of ownership on the capital structure depends on the risk aversion of the manager, the monitoring and bankruptcy costs, the threat of takeovers, and growth opportunities for the firm (King and Santor 2008). From one perspective, a positive effect is expected when families prefer the use of debt to maintain control and ownership by avoiding the issuance of new shares (Stulz, 1988). From another perspective, families tend to be risk averse to the loss of control and concentrate wealth in the firm (Hiebl, 2013; Fama and Jensen, 1983) to prevent lender monitoring that reduces private control benefits (King and Santor, 2008 ). This risk aversion implies a reduction in leverage, increased self-financing (Gamma and Galvão, 2012), and dual-class shares (King and Santor, 2008). Some empirical evidence indicates a positive (Bianco et al., 2013; Fitzsimmons and Douglas, 2006), negative (Fitzsimmons and Douglas, 2006; Gallo and Vilaseca, 1996; Holderness and Sheehan 1988), or zero (Anderson et al., 2003) effect of family control/management on the capital structure. Existing empirical studies focus on the effects of family firms on risk aversion, control risk propensity, expropriation of minority shareholders, and other agency problems. These studies observe behavior through the capital structure choice of firms. However, we consider it more appropriate to incorporate the decisions on cash holdings and cash dividends for corroborative evidence. The capital structure and dividends can be used to reduce the free cash flow agency costs, for example, the private benefits of control in family firms (Gonzalez et al., 2014; Pindado et al., 2012; Wei et al., 2011; Setia-Atmaja, 2010). This behavior would imply a reduction in managers’ cash holdings (Steijvers and Niskanen, 2013; Jensen, 1986). However, the interplay between debt, cash holdings, and dividends may reflect the risk profile of the controlling/management family or difficulty in accessing external financing. In these cases, firms can maintain high levels of cash holdings, low dividends, and low leverage to maintain control and reduce the need to dilute its control rights (Anderson and Hamadi, 2009). Similarly, firm family control/management facing financial constraints may have low levels of leverage, cash dividends, and high levels of cash holdings (Almeida et al., 2004;. Fazzari et al., 1988.). Firms with founding family control/may be at risk because of overconfidence or optimism (Hiebl, 2013), which implies a greater propensity towards leverage without the corresponding need to maintain liquidity by increase cash holdings and reducing cash dividends. However, there is evidence that family firms may be overly risk averse, which implies a higher level of cash holdings and a lower level of leverage (Anderson et al., 2003; Anderson and Hamadi, 2009).

The advantages and disadvantages of the family firm co-exist (Wei et al., 2011), and the ultimate effect of family control/management on the capital structure, cash holdings, and cash dividends depends on the extent of family involvement in management, which dictates the costs and net benefits that are dominant in the family firm. Thus, the prevalence of certain capital structures and the factors that determine those structures is an empirical question (Gonzalez et al., 2014;. Anderson and Hamadi, 2009). King and Santor 2008 found that if a direction for the effect of family control/management on capital structure cannot be established, empirical test results should be used as evidence. 3. Methodology 3.1 Empirical strategy Consistent with King and Santor (2008) and Miller et al. (2007), and following the best practices for ownership structure estimation and performance relationships to mitigate omitted variable bias and problems of endogeneity (Wang and Shailer 2013), we adopted panel data specification with robust standard errors clustered by firm. The specification adopted is the most appropriate for this study because (1) the ownership structure, corporate performance, capital structure, cash holdings, and cash dividends can be determined by unobservable characteristics, and (2) we can use time invariants or variables that exhibit low variation over the study period (e.g., industry dummies, control rights/cash flow, and control-cash flow wedge), or variables close to timeinvariants (e.g., dummies that identify control/family management). Families are potentially in a position to determine ownership structure and financial policies to maintain control and/or management of the firm. The decision to maintain control and/or management can be determined by corporate characteristics such as performance, capital structure, cash holding, and dividends. Therefore, there may be a potential endogeneity problem from self-selection in the study of the effect of control/family management on performance and corporate financial policy. The panel data model addresses endogeneity associated with specific effects of the unobservable firm but does not adequately address the bias of self-selection. Considering that the explanatory effect of the interest variable in this study (family control or management) is binary, consistent with Miller et al. (2007), we adopted the treatment effect model (TEM) estimated by maximum likelihood and standard errors clustered at the firm level to manage the potential self-selection bias. The treatment group is identified by a dummy variable equal to one for family control/management companies. The outcome variables are the same, that is, Tobin's Q, return on assets (ROA), total and longterm debt, cash holding, and cash dividends.

776


The TEM is expressed by two equations defined in two stages (estimation of selection model and outcome model). In the first stage of the model, we

applied a probit regression to a selection model where family control/management is determined by a selected set of variables as follows:

𝐹𝐴𝑀𝐼𝐿𝑌 𝐹𝐼𝑅𝑀𝑖,𝑡 = 𝛼 + 𝛽 ` 𝑥𝑖,𝑡 + ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝐷𝑢𝑚𝑚𝑦 + ∑ 𝑌𝑒𝑎𝑟𝐷𝑢𝑚𝑚𝑦 + 𝜀𝑖,𝑡 𝑗

(1)

𝑡

where (𝐹𝐴𝑀𝐼𝐿𝑌 𝐹𝐼𝑅𝑀𝑖,𝑡 ) is a binary variable with a value of one if the company has family control and/or management and zero otherwise. The vector 𝑥𝑖,𝑡 of the variable selection considers all the predictors that meet significant components of the determinants of family control/management, such as cash holding (Miller et al., 2007), growth sales (Villalonga and Amit, 2006), age, and size of the firm (Zhou, 2012; Anderson and Reeb, 2003). As a complement, we added three other determinants of family control/management as selection variables not included in the outcome model but considered important in the literature (Zhou, 2012; Villalonga and Amit, 2006; Anderson and Reeb, 2003; Demsetz and Villalonga, 2001; Demsetz and Lehn, 1985); government regulation ( 𝑅𝐸𝐺𝑈𝐿𝑖,𝑡 ); volatility of stock price ( 𝑉𝑂𝐿𝑖,𝑡 ); asset size squared ( 𝑆𝐼𝑍𝐸 2 𝑖,𝑡 ), and a dummy for companies with a founding age above/below the median sample ( 𝑂𝐿𝐷 𝐹𝐼𝑅𝑀𝑖,𝑡 ).

The second stage of TEM captures the effect of family control/management on performance and corporate financial policy (outcome models) according to equations 2 to 4. Because of the multidimensional nature of corporate performance, to examine the impact of family control/management on performance, we adopt two proxies: Tobin's Q and ROA. The proxy Tobin's Q is a forward-looking perspective to reflect the market value while the proxy ROA adopts a backward-looking perspective by reflecting profitability and productivity. The ROA measure is susceptible to manipulation and managerial accounting differences while Tobin's Q may reflect the market sentiment (Wang and Shailer, 2013). Therefore, we use both measures. To examine the effect of family control/management on performance, we estimate the following outcome model using ordinary least squares:

𝑃𝐸𝑅𝐹𝑂𝑅𝑀𝐴𝑁𝐶𝐸𝑖,𝑡 = 𝛼 + 𝛽 ` 𝑥𝑖,𝑡 + 𝛿𝐹𝐹𝑖,𝑡 + ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝐷𝑢𝑚𝑚𝑦 + ∑ 𝑌𝑒𝑎𝑟𝐷𝑢𝑚𝑚𝑦 + 𝜀𝑖,𝑡 𝑗

where performance (𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒𝑖,𝑡 ) is defined operationally as Tobin's Q ( 𝑄𝑖,𝑡 ) or return on assets ( 𝑅𝑂𝐴𝑖,𝑡 ). 𝑥𝑖,𝑡 is a set of corporate features documented in the literature as determinants of corporate performance that serve as control variables. Among the control variables are size ( 𝑆𝐼𝑍𝐸𝑖,𝑡 ), age ( 𝐴𝐺𝐸𝑖,𝑡 ), annual sales growth ( 𝐺𝑅𝑂𝑊𝑇𝐻𝑖,𝑡 ), total financial debt ( 𝑇𝑂𝑇𝐴𝐿 𝐷𝐸𝐵𝑇𝑖,𝑡 ), quality of corporate governance ( 𝐺𝑂𝑉𝐸𝑅𝑁𝐴𝐶𝐸 𝑄𝑈𝐴𝐿𝐼𝑇𝑌𝑖,𝑡 ), capital expenditures ( 𝐶𝐴𝑃𝐸𝑋𝑖,𝑡 ), operational business risk ( 𝑂𝑃𝐸𝑅𝐴𝑇𝐼𝑂𝑁𝐴𝐿 𝑅𝐼𝑆𝐾𝑖,𝑡 ), control rights ( 𝐶𝑂𝑁𝑇𝑅𝑂𝐿 𝑅𝐼𝐺𝐻𝑇𝑆𝑖,𝑡 ), cash flow rights ( 𝐶𝐴𝑆𝐻 𝐹𝐿𝑂𝑊 𝑅𝐼𝐺𝐻𝑇𝑆𝑖,𝑡 ), and wedge between cash flow and control rights ( 𝐶𝑂𝑁𝑇𝑅𝑂𝐿 − 𝐶𝐴𝑆𝐻 𝐹𝐿𝑂𝑊 𝑊𝐸𝐷𝐺𝐸). The firm

(2)

𝑡

family (𝐹𝐹𝑖,𝑡 ) is a dummy variable that identifies the type of family control/management of the firm (one for family control/management and 0 otherwise). We adopt a family control ( 𝐹𝐴𝑀𝐼𝐿𝑌 𝐶𝑂𝑁𝑇𝑅𝑂𝐿𝑖,𝑡 ) and two family management settings, a wide ( 𝐹𝐴𝑀𝐼𝐿𝑌 𝑀𝐴𝑁𝐴𝐺𝐸𝑀𝐸𝑁𝑇𝑖,𝑡 ) and restricted to founding family ( 𝐹𝑂𝑈𝑁𝐷𝐸𝑅 𝐹𝐴𝑀𝐼𝐿𝑌 𝑀𝐴𝑁𝐴𝐺𝐸𝑀𝐸𝑁𝑇𝑖,𝑡 ). Table 1 lists the definitions of the variables. ε_ (i, t) is the residual mean-zero adjusted to the specific heterogeneity of the firm. To examine the effect of family control/management on the capital structure, we estimate the following outcome model:

𝐿𝐸𝑉𝐸𝑅𝐴𝐺𝐸𝑖,𝑡 = 𝛼 + 𝛽 ` 𝑥𝑖,𝑡 + 𝛿𝐹𝐹𝑖,𝑡 + ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝐷𝑢𝑚𝑚𝑦 + ∑ 𝑌𝑒𝑎𝑟𝐷𝑢𝑚𝑚𝑦 + 𝜀𝑖,𝑡 𝑗

(3)

𝑡

where 𝐿𝐸𝑉𝐸𝑅𝐴𝐺𝐸𝑖,𝑡 is defined as total financial debt ( 𝑇𝑂𝑇𝐴𝐿 𝐷𝐸𝐵𝑇𝑖,𝑡 ) and financial long-term debt ( 𝐿𝑂𝑁𝐺 − 𝑇𝐸𝑅𝑀 𝐷𝐸𝐵𝑇𝑖,𝑡 ). The vector 𝑥𝑖,𝑡 of control variables is the same as that in equation 1, except we

exclude total financial debt and include cash holding ( 𝐶𝐴𝑆𝐻 𝐻𝑂𝐿𝐷𝐼𝑁𝐺𝑖,𝑡 ) and tax shield ( 𝑇𝑆𝑖,𝑡 ). To examine the effect of family control/management on cash holding, we estimate the following outcome model:

𝐶𝐴𝑆𝐻 𝐻𝑂𝐿𝐷𝐼𝑁𝐺𝑖,𝑡 = 𝛼 + 𝛽 ` 𝑥𝑖,𝑡 + 𝛿𝐹𝐹𝑖,𝑡 + ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝐷𝑢𝑚𝑚𝑦 + ∑ 𝑌𝑒𝑎𝑟𝐷𝑢𝑚𝑚𝑦 + 𝜀𝑖,𝑡 𝑗

𝑡

777

(4)

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 where 𝐶𝐴𝑆𝐻 𝐻𝑂𝐿𝐷𝐼𝑁𝐺𝑖,𝑡 is defined as cash and short-term investments to total assets. The vector 𝑥𝑖,𝑡 of variables is the same as that of equation 1, except we exclude cash holding and non-equity tax shield

and include short-term financial debt ( 𝑆𝐻𝑂𝑅𝑇 − 𝑇𝐸𝑅𝑀 𝐷𝐸𝐵𝑇𝑖,𝑡 ) and other liquid assets (𝑂𝐿𝐴𝑖,𝑡 ). To examine the effect of family control/management on dividends, we estimate the following outcome model:

𝐶𝑎𝑠ℎ𝐷𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑖,𝑡 = 𝛼 + 𝛽 ` 𝑥𝑖,𝑡 + 𝛿𝐹𝐹𝑖,𝑡 + ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝐷𝑢𝑚𝑚𝑦 + ∑ 𝑌𝑒𝑎𝑟𝐷𝑢𝑚𝑚𝑦 + 𝜀𝑖,𝑡 𝑗

where 𝐶𝑎𝑠ℎ𝐷𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑖,𝑡 is defined as dividends and interest on shareholders' equity to sales. The vector 𝑥𝑖,𝑡 of control variables is the same as that of equation 3, except we exclude the short-term financial debt variable and add total financial debt. 3.2 Sample, data, and definition of variables

operational

The population of this study is composed of all publicly traded non-financial Brazilian companies. All companies have annual financial information and a market, control/ownership structure with data available from the Economática® database, annual reports (IAN), and reference forms published by the

(5)

𝑡

Securities and Exchange Commission (CVM) and the Stock Exchange, Commodities and Futures Exchange (BMandFBovespa) from the year 2005 to the year 2012. We exclude the firm-year observations with at least one of the following restrictions: (a) annual net sales and/or total asset growth over 100% to eliminate observations of companies with changes in business fundamentals; (b) the variable Tobin’s Q with a value above 10 to avoid potential measurement errors. Continuous variables were winsorized adopting the 2.5% limit of observations in each tail. Table 1 shows the definitions of the variables used in the empirical tests.

Table 1. Variable definitions Variables Age Capex Cash Dividend Cash Flow Rights Cash Holding Control Rights Family Control Family Management Founder Management

Family

Governance Quality Long-term Financial Debt Operational Risk Other Liquid Assets ROA Sales Growth Short-term Financial Debt Size Tangibility Tax Shield Tobin’s Q Total Financial Debt Wedge Control-Cash Flow Rights

Definition Natural logarithm of firm age since its foundation. Ratio of capital expenditures to total assets. Ratio of cash dividend plus interest on equity to sales. Percentage of shares owned by the controlling shareholder. Ratio of cash and short-term investments to total assets. Percentage of shares with voting rights owned by the controlling shareholder. Dummy variable that takes the value of one if the individuals or individual of the same family has 50% or more of shares with voting rights and zero otherwise. Dummy variable that takes the value of one if, in a familiar controlling firm, the same family member is the CEO and/or the President of the Board of Directors and zero otherwise. Dummy variable that takes the value of one if, in a familiar controlling firm, the same founder family member is the CEO and/or the President of the Board of Directors and zero otherwise. Dummy variable that takes the value one if the firm is listed on the three high-governance listing of BM&Fbovespa and zero otherwise. Ratio of long-term financial debt to total assets. Standard deviation of ROA during the sample time series. Ratio of inventories and accounts receivables to total assets. Ratio of EBITDA to total assets. Annual sales growth rate. Ratio of short-term financial debt to total assets. Natural logarithm of total assets. Ratio of fixed assets to total assets. Ratio of the difference between EBIT and the ratio of income tax payments to corporate tax rate to sales. Ratio of the market value of assets to total assets. The market value of assets is defined as total assets minus equity plus the market value of equity. Ratio of total financial debt to total assets. Difference between control rights and cash flow rights minus one.

778


Table 2 reports the industrial distribution of observations for the full sample and between different definitions of companies with family control/management. We observe a high variation in the occurrence of family control/management between the industries classified by NAICS-1. Family businesses are more concentrated among the manufacturing and trade industries and less

concentrated among utility industries such as energy and telecommunications. This distribution pattern corresponds with international studies (King and Santor, 2008). The prevalence of asymmetric family control/management companies across industries suggests the need to control the industry in the multivariate analysis.

Table 2. Distribution of observations by industry Industries

Construction and Engineering Diversified Education Services Farming Health Hotels and Restaurants Manufacturing Mining Real Estate Retail and Distribution Services Telecommunications and Media Travel, Entertainment, and Leisure Transportation Utilities Total

Whole Sample

Family Control Nº % Obs.

Nº Obs.

%

104

6..6%

52

73 20 23 14 4 717 25 41 79 27

4..6% 1..3% 1..5% 0..9% 0.3% 45.6% 1.6% 2.6% 5.0% 1.7%

99

Family Management

Founder Family Management

Nº Obs.

%

Nº Obs.

%

7.7%

92

10.5%

53

24.0%

27 6 17 5 0 422 3 11 49 15

4.0% 0.9% 2.5% 0.7% 0.0% 62.3% 0.4% 1.6% 7.2% 2.2%

55 15 18 7 4 499 5 27 50 17

6.3% 1.7% 2.1% 0.8% 0.5% 57.1% 0.6% 3.1% 5.7% 1.9%

21 3 8 4 0 89 3 19 1 14

9.5% 1.4% 3.6% 1.8% 0.0% 40.3% 1.4% 8.6% 0.5% 6.3%

6.3%

12

1.8%

20

2.3%

0

0.0%

2

0.1%

0

0.0%

0

0.0%

0

0.0%

65 278 1571

4.1% 17.7% 100

27 31 677

4.0% 4.6% 100

35 30 874

4.0% 3.4% 100

3 3 221

1.4% 1.4% 100

4. Analysis and discussion of results Table 3 shows the descriptive statistics and univariate tests for the variables used for the analysis period from the year 2005 to the year 2012. We use a parametric test of mean differences to detect differences in these variables according to family control/management and other types of control or management. We identified systematic differences between companies with family control/management compared to other ownership structures, regardless of the criteria used in the classification. Companies with family control/management exhibit the following characteristics: a lower Tobin’s Q, except for firms with founder management; a lower ROA; lower total financial debt except for firms with founder family management; lower shortterm financial debt; lower long-term financial debt except for firms with founder management; greater cash holding; fewer control rights except for familycontrolled firms; fewer cash flow rights; greater wedge between control and cash flow right except for firms with founder family management; smaller size; greater other liquid assets; greater maturity except for

firms with founder family management; fewer dividends on sales; fewer tangible assets, and lower volatility of cash flows. The level of sales growth and corporate governance was significantly higher for companies with founder family management.

779


Table 3. Descriptive Statistics

Age Capex Cash Dividend Cash Flow Rights Cash Holding Control Rights Governance Quality Long-term Debt Operational Risk Other Liquid Assets ROA Sales Growth Short-term Debt Size Tangibility Tax Shield Tobin`s Q Total Debt Wedge Control-Cash Flow Firm-years Firms

[A] [B] [A]≠[B] [C] [D] [C]≠[D] [E] [F] [E]≠[F] FC Others FM Others FFM Others 1.561 1.416 0.00*** 1.527 1.417 0.00*** 1.270 1.512 0.00*** (0.39) (0.42) (0.39) (0.43) (0.43) (0.40) 0.143 0.149 0.561 0.139 0.156 0.08* 0.153 0.145 0.60 (0.21) (0.17) (0.21) (0.17) (0.25) (0.18) 0.035 0.070 0.00*** 0.034 0.080 0.00*** 0.034 0.058 0.00*** (0.07) (0.09) (0.07) (0.11) (0.07) (0.09) 0.528 0.571 0.00*** 0.51 0.605 0.00*** 0.514 0.559 0.00*** (0.19) (0.23) (0.20) (0.22) (0.16) (0.22) 0.111 0.090 0.00*** 0.105 0.090 0.00*** 0.113 0.096 0.03** (0.12) (0.09) (0.12) (0.08) (0.11) (0.11) 0.725 0.685 0.00*** 0.669 0.743 0.00*** 0.628 0.714 0.00*** (0.15) (0.22) (0.19) (0.20) (0.20) (0.19) 0.447 0.521 0.00*** 0.480 0.500 0.427 0.683 0.457 0.00*** (0.49) (0.49) (0.49) (0.50) (0.46) (0.49) 0.160 0.174 0.03** 0.160 0.177 0.01** 0.169 0.168 0.85 (0.14) (0.12) (0.14) (0.12) (0.14) (0.13) 0.057 0.067 0.02** 0.061 0.066 0.257 0.046 0.066 0.00*** (0.08) (0.08) (0.08) (0.08) (0.03) (0.09) 0.184 0.124 0.00*** 0.191 0.098 0.00*** 0.237 0.136 0.00*** (0.16) (0.15) (0.17) (0.13) (0.18) (0.15) 0.094 0.127 0.00*** 0.091 0.141 0.00*** 0.090 0.117 0.00*** (0.09) (0.13) (0.11) (0.12) (0.07) (0.12) 0.099 0.113 0.249 0.107 0.107 0.959 0.190 0.093 0.00*** (0.23) (0.24) (0.25) (0.21) (0.30) (0.22) 0.114 0.083 0.00*** 0.114 0.075 0.00*** 0.112 0.094 0.01** (0.10) (0.09) (0.10) (0.08) (0.10) (0.09) 6.009 6.440 0.00*** 6.028 6.539 0.00*** 6.090 6.281 0.00*** (0.71) (0.78) (0.76) (0.72) (0.60) (0.81) 0.304 0.350 0.00*** 0.299 0.369 0.00*** 0.215 0.349 0.00*** (0.20) (0.24) (0.21) (0.23) (0.21) (0.22) -0.014 -0.030 0.04** -0.011 -0.038 0.00*** -0.031 -0.022 0.35 (0.15) (0.14) (0.15) (0.14) (0.14) (0.14) 0.978 1.235 0.00*** 1.056 1.209 0.00*** 1.183 1.115 0.30 (0.72) (1.04) (0.87) (0.97) (0.87) (0.93) 0.274 0.258 0.06* 0.274 0.253 0.01** 0.282 0.262 0.11 (0.18) (0.16) (0.17) (0.15) (0.18) (0.16) 0.196 0.114 0.00*** 0.158 0.138 0.03** 0.113 0.155 0.00*** (0.19) (0.17) (0.18) (0.18) (0.17) (0.18) 677 894 874 697 221 1350 153 194 183 149 57 264

Note: FC, FM and FFM indicate family control, family management, and founding family management, respectively. Average (above) and standard deviation (bottom, in parenthesis). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 3 also shows that family control/management companies use more controlenhancing mechanisms (such as dual class) that create the wedge between control rights and cash flow. The fact that family companies posses greater wedge, fewer control rights, higher debt, higher cash holding, and lower dividend payments is consistent with the view that the family company adopts controlenhancing mechanisms to ensure company growth without assuming the risk of loss of control.

780


Table 4. Effect of family control on performance and cash dividends Q Selection Model Family Control Cash Flow Rights Wedge Control-Cash Flow Age Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk

0.7492* (1.68) 2.2323*** (3.79) -0.1426 (-0.54) 0.9465*** (3.02) 1.4973*** (2.83) -0.3389 (-1.44) -1.6342* (-1.70) 0.0932 (0.56) -2.9878*** (-2.03)

Tobin’s Q Cash Flow Size Total Debt Size2 Share Volatility Old Firm

Value

Constant Industry and Year Effect Chi-square Test Log pseudo likelihood Lambda Wald Test Indep. Eqs. (chi2) No. Observations

-0.1094 (-0.19) 13.497 (0.92) 0.6907* (1.79) -0.1243 (-1.05) -0.0041 (-0.26) 0.3212 (1.42) -44.959 (-0.99) Yes

ROA Outcome Selection Outcome Model Model Model -0.4639*** -0.0037 (-2.66) (-0.02) 0.3853** 0.8247 0.0257 (2.12) (1.62) (0.51) -0.3502* 2.1016*** 0.0138 (-1.80) (3.31) (0.12) -0.0274 -0.1242 0.0038 (-0.28) (-0.44) (0.30) -0.1733 0.7648** -0.0224 (-1.23) (2.52) (-0.49) 0.8544** 1.5876*** 0.0438 (2.36) (2.80) (0.46) 0.1342 -0.2935 -0.0056 (1.21) (-1.09) (-0.38) 2.0670*** (4.18) 0.1050 0.1323 0.0619*** (1.01) (0.79) (3.35) 0.5003 -2.9142* -0.1100 (1.50) (-1.78) (-0.63) -0.2252 0.0527*** (-1.81) (3.19) 2.5174*** (-4.70) -0.0927 1.1129*** 0.0215 (-1.35) (4.94) (1.22) 0.3994** 0.8194** -0.0475 (1.98) (1.97) (-1.11) -0.1088*** (-4.08) 0.0074 (0.24) 0.2530 (1.04) 1.6170*** -3.5051** -0.1130 (3.20) (-2.39) (-0.78) Yes Yes Yes

CASH DIVIDEND Selection Outcome Model Model 0.0876*** (6.30) 0.6773* 0.0088 (1.65) (-0.37) 1.6337*** -0.0454 (2.96) (-1.57) 0.1492 -0.0247* (0.60) (-1.75) 0.7638** 0.0594** (2.36) (2.40) 1.9162*** 0.0548 (3.64) (0.97) -0.2400 -0.0144 (-1.12) (-1.19)

0.0696 (0.41) -2.7656** (-2.40) -0.2661*** (-3.08) -0.8787 (-1.33) -0.4021 (0.40) 0.4438 (1.15) -0.0465 (-0.56) 0.0085 (0.64) 0.1474 (0.92) -15.284 (-0.48) Yes

-0.0377*** (-3.43) 0.1064** (2.46) 0.0290*** (4.89) 0.1344*** (2.81) 0.0208*** (2.64) -0.0699*** (-2.70)

-0.1216** (-2.31) Yes

437.68 -2447.86

416.65 672.17

173.20 1163.78

2.2654** 3.87**

0.0439 0.9649

-6.3913*** 32.03***

1568

1568

1568

Note: ***, **, * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

781


Table 5. Effect of family control on capital structure and cash holdings

Family Control Cash Flow Rights Wedge Control-Cash Flow Age Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk Tobin’s Q Cash Flow Size

TOTAL DEBT Selection Outcome Model Model -0.0881 (-1.27) 0.8836* -0.0309 (1.89) (-0.58) 2.0168*** -0.053 (3.28) (-0.70) -0.1641 -0.0045 (-0.67) (-0.19) 0.9594*** 0.0928** (3.18) (-2.50) 1.5025*** -0.0057 (2.83) (-0.08) -0.2714 -0.0064 (-1.14) -0.58 -2.0717** -0.0407 (-2.27) (-0.35) 0.1137 -0.0157 (0.63) (-0.76) -2.7299** -0.1359 (-2.06) (-1.12) -0.2148** 0.0064 (-2.27) (-0.48) 0.3501 -0.0911 (0.55) (-1.13) 0.8235 0.0341** (0.62) (2.16)

LONG-TERM DEBT Selection Outcome Model Model -0.0628* (-1.72) 0.8293* -0.0043 (1.82) (-0.11) 1.9985*** -0.0256 (3.24) (-0.49) -0.1347 -0.004 (-0.54) (-0.23) -0.9314*** 0.1090*** (3.06) (4.17) 1.5260*** 0.0747 (2.89) (1.36) -0.2773 0.0018 (-1.15) (0.10) -2.1367** 0.0348 (-2.26) (0.43) 0.1326 -0.0156 (0.75) (-1.11) -2.9100** -0.0777 (-2.18) (-0.84) -0.2035** -0.0023 (-2.27) (-0.32) 0.2778 0.0142 (0.48) (0.29) 1.2720 0.0511*** (0.89) (4.47)

Short-term Debt Tax Shield

-0.4031 (-1.54)

0.1508*** (4.34)

-0.4099 (-1.55)

Share Volatility Old Firm

Value

Constant Industry and Year Effect Chi-square Test Log pseudo likelihood Lambda Wald Test Index. Eqs. (chi2) No. Observations

-0.0850 (-0.80) 0.0123 (0.92) 0.2662 (1.30) -2.4894 (-0.60) Yes

0.0691 (0.62) Yes

-0.2426 (-1.03) -0.7664 (-0.82) 0.1164 (0.69) -2.9661** (-2.04) -0.172** (-2.29) 0.1075 (0.20) 2.1575 (1.43) 0.3913 (0.51)

0.0400*** (3.75) 0.2116** (2.34) 0.0300*** (3.03) -0.0191 (-0.29) 0.0166*** (2.62) 0.0020 (0.05) -0.0044 (-0.51) -0.1138** (-2.46)

0.1822 (0.43) 2.1575 (1.43) 0.0104 (0.80) 0.2843 (1.30) -6.4969 (-1.39) Yes

-0.1428*** (-4.11)

0.1037*** (3.98)

Other Liquid Assets Size2

CASH HOLDINGS Selection Outcome Model Model 0.0774*** (2.73) 0.8072* -0.0347 (1.80) (-1.34) 2.070*** -0.0659* (3.50) (-1.64) -0.2134 -0.0133 (-0.83) (-1.21) 0.8415*** -0.0845*** (2.71) (-3.53)

-0.1215 (-1.05) 0.0092 (0.74) 0.2473 (1.22) -3.8291 (-0.86) Yes

-0.1359* (-1.77) Yes

0.1380** (1.96) Yes

405.24 12.4575

353.44 454.56

223.29 725.62

1.8291* 3.09*

2.4729** 5.70**

-1.8214* 3.41*

1568

1568

1568

Note: ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

782


Table 6. Effect of family management on performance and cash dividends Q Selection Model Family Management Cash Flow Rights Wedge Control-Cash Flow Age Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk

-1.0165** (-2.25) -0.4637 (-0.85) 0.1089 (-0.36) 0.5766* (1.81) 1.3079** (2.38) -0.2756 (-1.16) -3.1841*** (-3.57) 0.0449 (0.25) -2.1288* (-1.95)

ROA Outcome Model -0.2430 (-0.56) 0.2108 (0.98) -0.6730*** (-3.51) -0.0377 (-0.38) -0.2701* (-1.90) 0.6911* (1.69) 0.1557 (1.28) 2.0819*** (4.06) 0.0990 (0.96) 0.7687** (1.96)

Tobin’s Q Cash Flow Size Total Financial Debt Size2 Share Volatility Old Firm

Value


-0.2624 (-0.44) -2.0697 (-1.60) 0.7740* (1.67) 0.1406 (1.41) -0.0066 (-0.25) -0.0073 (-0.03) 8.1698*** (1.98)

2.5889*** (4.46) -0.0911 (-1.24) 0.3899* (1.91)

Selection Model

-0.9195** (-2.10) -0.4333 (-0.82) 0.0690 (0.26) 0.4739 (1.41) 1.0075** (2.02) -0.1835 (-0.82)

0.0990 (0.54) -2.7719** (-2.47) -0.0895 (-1.26)

CASH DIVIDENDS Outcome Selection Outcome Model Model Model -0.0979*** 0.0907*** (-3.11) (8.20) -0.0014 -1.0876*** 0.0377 (-0.05) (-2.73) (1.49) -0.0002 -0.9174*** 0.0174 (-0.01) (-1.90) (0.58) 0.0080 0.3536 -0.0249* (0.67) (1.42) (-1.73) -0.0158 0.3208 0.0732*** (-0.75) (0.88) (2.72) 0.0672* 1.3788** 0.0768 (1.78) (2.44) (1.31) -0.0098 -0.2269 -0.0155 (-0.87) (-1.12) (-1.27)

0.0641*** (3.49) -0.1508** (-2.07) 0.0493*** (6.29)

0.0097 (0.06) -1.8566** (-2.27) -0.1786** (-2.17) -1.0001* (-1.72) -1.9792* (-2.09) 0.5340 (1.30) 0.1338* (1.88) 0.0096 (0.75) -0.0978 (-0.68) 7.7851** (2.49)

-0.0379*** (-3.37) 0.0815** (2.26) 0.0255*** (4.20) 0.1527*** (3.20) 0.0241*** (3.07) -0.0796*** (-2.98)

Yes

-3.6266** 0.0110 (-2.57) (1.16) 0.9587** -0.0203 (2.29) (-0.66) 0.2556** (2.37) -0.0075 (-0.58) -0.0256 (-0.12) 13.2560*** -0.0979*** (2.94) (-3.11) Yes

397.91 -2428.74

338.28 691.55

190.87 1206.13

0.4934 0.25

2.9830*** 7.06***

-8.2771*** 60.50***

1568

1568

1568

1.7002* (2.46)


783

-0.1956*** (-3.60) Yes


Table 7. Effect of family management on capital structure and cash holdings

Family Management Cash Flow Rights Wedge Control-Cash Flow Age Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk Tobin’s Q Cash Flow Size

TOTAL DEBT Selection Outcome Model Model 0.1126* (1.86) -1.1899*** -0.0217 (-2.57) (-0.43) -0.6721 -0.0848 (-1.24) (-1.57) 0.2414 -0.0103 (0.79) (-0.45) 0.6141* 0.0564* (1.86) (1.80) 1.2478** -0.0816 (2.12) (-1.14) -0.3154 0.0054 (-1.29) (0.24) -4.0280*** 0.0859 (-4.37) (0.72) 0.0144 -0.0205 (0.08) (-1.04) -1.9484** -0.0231 (-2.00) (-0.24) -0.0346 0.0103 (-0.47) (1.25) -0.2186 -0.0844 (-0.38) (1.15) -1.7678 0.0496*** (-1.37) (3.39)

LONG-TERM DEBT Selection Outcome Model Model 0.0495 (1.08) -1.1454** -0.0060 (-2.49) (-0.17) -0.6490 -0.0530 (-1.18) (-1.27) 0.1842 -0.0073 (0.61) (-0.45) 0.6228* 0.0872*** (1.90) (3.77) 1.2415** 0.0293 (2.17) (0.55) -0.2935 0.0087 (-1.20) (0.51) -3.9054*** 0.1024 (-4.26) (1.17) 0.0325 -0.0184 (0.18) (-1.37) -1.9730* -0.0107 (-1.86) (-0.12) -0.0293 0.0018 (-0.39) (0.28) -0.2027 0.0159 (-0.36) (0.34) -1.9817 0.0595*** (-1.55) (5.13)


-0.2470 (-1.03)

0.1654*** (5.14)

-0.2371 (-0.97)

Share Volatility Old Firm

Value


-0.2655 (-1.30) -2.6017** (-2.33) -0.0010 (-0.01) -1.8665** (-2.15) -0.0507 (-0.72) -0.1089 (-0.22) -2.9577*** (-3.49) -0.2971 (-0.44)

0.0296* (1.94) 0.1115 (1.03) 0.0336*** (3.00) -0.1410* (-1.83) 0.0110* (1.67) 0.0003 (0.01) -0.0211* (-1.95) -0.0834* (-1.68)

0.1153*** (4.68)


CASH HOLDINGS Selection Outcome Model Model -0.1327*** (-6.86) -1.1245*** -0.0507 (-2.78) (-1.61) -0.9473* -0.0383 (-1.77) (-0.90) 0.2078 -0.0088 (0.82) (-0.58) 0.5947** -0.0443* (1.95) (-1.66)

Yes

Yes

0.1697 -0.1297*** (0.42) (-3.48) 0.2164*** (3.32) 0.0042 (0.40) -0.2554* (-1.67) 10.9299*** 0.3650*** (3.99) (4.03) Yes

366.64 33.6779

419.61 473.89

202.91 765.96

-1.4354 1.99

-0.5523 0.30

7.3095*** 48.83***

1568

1568

1568

0.1196 (1.20) -0.0032 (-0.22) -0.1712 (-0.69) 7.3572* (1.80)

-0.1315 (1.02)

0.1361 (1.38) -0.0013 (-0.09) -0.1207 (-0.47) 8.0240** (1.97)

-0.2383** (-2.37)


784


Table 8. Effect of founder family management on performance and cash dividends Q Selection Model Founder Family Management Cash Flow Rights Control-Cash Wedge Age

Flow

Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk

ROA

-0.0804 (-0.16) 0.1722 (0.25) -0.4252 (-1.34) 0.0218 (0.06) 0.9668 (1.11) 0.3454 (1.19) -1.1589 (-1.11) 0.4509* (1.74) -6.9505

Outcome Model -0.0287 (-0.05) 0.274 -1.64 -0.6774 (-3.60) -0.0489 (-0.40) -0.3037 (-2.10) 0.6153* (1.67) 0.1727 (1.47) 2.2434*** (4.01) 0.0981 (0.95) 0.8714**

(-1.52)

(2.26)

-0.1611 (-0.20) 0.6162 (0.37) 0.6852 (1.27) -0.0875 (-0.67) 0.0343 (1.14) -0.2947 (-1.02) -0.7943 (-0.15)

2.5549*** (4.41) -0.0685 (-0.86) 0.3359 (1.39)

Tobin’s Q Cash Flow Size Total Financial Debt Size2 Share Value Volatility Old Firm Constant Industry and Year Effect Chi-square Test Log pseudo likelihood Lambda Wald Test Indep. Eqs. (chi2) No. Observations

1.3959** (2.17)

Selection Model

0.2492 (0.58) 0.2091 (0.35) -0.3307 (-1.39) 0.1577 (0.50) 0.9277 (1.43) 0.3322 (1.54)

CASH DIVIDENDS Outcome Selection Outcome Model Model Model -0.1392*** -0.0459* (-5.35) (-1.79) 0.0226 -0.0439 0.0127 (0.86) (-0.09) (0.61) 0.0111 0.2944 0.0065 (0.37) (0.42) (0.26) -0.0128 -0.5186* -0.0265* (-1.04) (-1.65) (-1.90) -0.0295 -0.1516 0.0771*** (-1.46) (-0.40) (3.45) 0.0528 0.3144 0.1036* (1.27) (0.36) (1.82) 0.0043 0.3597 -0.0166 (0.41) (1.29) (-1.60)

0.7369*** (3.35) 10.6313*** (-4.30) 0.1464* (1.65)

0.0778*** (4.10) -0.1413**

0.5015** (2.09) -6.1008**

-0.0293*** (-2.91) 0.0275

(-2.22) 0.0535*** (6.72)

-0.0205 (-0.02) 0.8427* (1.77) -0.0323 (-0.31) 0.0295** (2.14) -0.2076 (-1.04) 0.4765 (0.11)

0.0162* (1.77) -0.0266 (-0.78)

(-2.25) 0.0702 (0.70) 0.0013 (0.00) 0.6524 (0.43) 0.8888* (1.80) -0.0911 (-0.79) 0.0345* (1.89) -0.3355 (-1.23) -0.9864 (-0.20)

(1.10) 0.0233*** (4.61) 0.1346*** (3.31) 0.0126* (1.92) -0.0478** (-2.11)

-0.0341 (-0.49)

-0.0440 (-0.92)

Yes

Yes

Yes

418.82 -2133.74 0.2046 0.04

253.76 1016.13 5.3443*** 25.29***

228.66 1461.44 1.3731 1.84

1568

1568

1568


785


Table 9. Effect of founder family management on capital structure and cash holdings

Founder Family Management Cash Flow Rights Control-Cash Wedge Age

Flow

Capex Cash Holding Governance Quality Cash Dividends Growth Sales Operational Risk Tobin’s Q Cash Flow Size

TOTAL DEBT Selection Outcome Model Model 0.1942*** (2.71) -0.0985 -0.0492 (-0.21) (-1.01) 0.2409 -0.0987* (-0.39) (-1.72) -0.2582 0.0169 (-0.79) (0.71) 0.1018 0.0748** -0.31 (2.35) 0.4785 -0.0690 (0.69) (-0.91) 0.4784* -0.0138 (1.80) (-0.55) -2.4232** 0.0512 (-2.28) (0.47) 0.2932 -0.0385* (1.48) (-1.70) -7.4953*** -0.0232 (-3.55) (-0.23) 0.0074 0.0088 (0.08) (0.78) -0.1687 -0.0884 (-0.22) (-1.19) 0.9506 0.0454*** (0.68) (3.01)

LONG-TERM DEBT Selection Outcome Model Model 0.1318*** (3.45) -0.1218 -0.0173 (-0.26) (-0.49) 0.4011 -0.0582 (0.66) (-1.33) -0.3436 0.0104 (-1.07) (0.61) 0.0677 0.0958*** (0.20) (4.08) 0.4482 0.0306 (0.63) (0.56) 0.4075 -0.0027 (1.56) (-0.15) -2.0650* 0.0979 (-1.90) (1.23) 0.3849** -0.0316** (1.97) (-2.18) -7.9469*** 0.0008 (-3.67) (0.01) 0.0392 -0.0003 (0.50) (-0.06) -0.2812 0.0154 (-0.38) (0.33) 0.4913 0.0589*** (0.40) (5.31)


-0.2638 (-1.16)

0.1576*** (4.66)

-0.1989 (-0.92)

Share Value Volatility Old Firm Constant Industry and Year Effect Chi-square Test Log pseudo likelihood Lambda Wald Test Indep. Eqs. (chi2) No. Observations

-0.1107 (-1.03) 0.0263 (1.62) -0.3759 (-1.54) -1.9132 (-0.42)

-0.0871 (-0.77)

-0.0779 (-0.82) 0.0283* (1.79) -0.3528 (-1.40) -0.2231 (-0.06)

Yes

0.3593 (1.24) -1.7625 (-1.25) 0.3689 (1.61) -7.3427*** (-2.79) 0.0677 (0.72) -0.4109 (-0.54) 0.3256 (0.22) 0.5909 (0.68)

0.0364*** (3.27) 0.1964* (1.95) 0.0325*** (3.34) -0.0808 (-1.23) 0.0120* (1.82) 0.0095 (0.25) -0.0095 (-1.11) -0.1045** (-2.43)

1.2976** (2.43) -0.0587 (-0.50) 0.0415** (2.24) -0.2834 (1.02) -0.2273 (-0.05)

-0.1357*** (-4.05)

0.1086*** (4.22)


CASH HOLDINGS Selection Outcome Model Model -0.0037 (-0.14) 0.0803 -0.0153 (0.16) (-0.64) 0.2886 -0.0205 (0.43) (-0.62) -0.5265 -0.0121 (-1.57) (-0.98) 0.3749 -0.0635*** (1.06) (-2.87)

-0.2441*** (-3.17) Yes

0.1884*** (2.63) Yes

557.57 328.78 -2.571** 4.59**

371.59 772.54 -3.0833*** 7.03***

220.25 1048.78 1.1086 1.20

1568

1568

1568

Note: ***, **, * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

786


The selection models reported in Tables 4 and 5 indicate that family control is more likely in firms with greater cash flow rights. The use of enhancing control mechanisms, the level of capital expenditures, cash holdings, and total financial debt, however, tend to be lower in firms with higher cash dividends and increased operational risk. The lambda statistic indicates that performance (Tobin’s Q) and leverage are positively selected for family-controlled firms but negatively selected for the level of cash holdings and cash dividends. Tables 4 and 5 show the estimation results of the outcome models for the effect of family control on corporate performance and financial policy. Family control has a negative impact of approximately 46% for corporate performance as measured by Tobin’s Q but does not affect operating performance. Similarly, family control positively affects cash dividends and cash holdings by approximately 9% and 8%, respectively. The effect on leverage occurs only for long-term debt with a negative impact of approximately 6%. The selection models reported in Tables 6 and 7 show that family management is more likely in firms with less cash flow rights and a low level of capital expenditures, cash holdings, and total financial debt. However, family management tends to be lower in firms with higher cash dividends, greater size, and greater operational risk. The lambda statistic indicates that these observable characteristics positively select family management for performance (ROA), but negatively select family management for the level of cash holdings and cash dividends. Tables 6 and 7 show the estimation results for the outcome models of the family management effect on corporate performance and financial policy. Family management negatively impacts corporate operating performance by approximately 10% but does not affect the performance measured by Tobin’s Q. Similarly, family management positively affects cash dividends, total, and long-term debt by approximately 9%, 11%, and 5%, respectively. The impact on cash holdings is negative by approximately 13%. The selection models reported in Tables 8 and 9 show that founding family management is more likely in firms with a higher growth rate and greater share value volatility. However, founding family managementtends to be lower in firms with higher operational risk. , The lambda statistic indicates that these observable characteristics positively select founding family management for performance (ROA) but negatively select founding family management for leverage decisions. Tables 8 and 9 show the results of the estimations for the outcome models of founding family management effect on corporate performance and financial policies. Founding family management negatively impacts corporate operating performance by approximately 14% but does not affect the

performance measured by Tobin’s Q. Similarly, founding family management positively affects total and long-term debt by approximately 19% and 13%, respectively. Founding family management has no impact on cash holdings but negatively affects approximately 5% of cash dividends. The results of the selection models in Tables 4 to 9 suggest that the application of the Heckman model in this study is appropriate to treat selection bias. The estimates of the selection models for family ownership, family management, and founding family management are similar in that performance is positively selected, that is, the factors that increase the likelihood that the firm has family control or management are positively correlated with corporate performance. However, the selection bias regarding financial policy (leverage, cash holding, and cash dividends) differ depending on the family control conditions and the nature of the family management (family or founder). This suggests that financial policy differs in significance depending on the nature of the family control structure and management. In summary, the results of the outcome models reported in Tables 4 to 9 suggest that family control and/or management has a net negative effect on corporate performance but the effect is heterogeneous for performance measures. Family control negatively affects the forward-looking measure (Tobin’s Q), and family management (FM and FMM) negatively affects the backward-looking measure (ROA). The effect of family control and/or management on financial policy is sensitive to the nature and degree of family involvement. In short, the test results suggest that family control/management alone is not the source of under or over performance for Brazilian companies. Therefore, our empirical evidence corresponds with the argument (Demsetz and Lehn, 1985) that the control structure and family management are endogenously determined by corporate performance. While the net effect of family control on leverage is negative or negligible, the net effect of management (FM and FFM) is positive. Control (FC) and family management (FM) positively affect cash dividends. The effect of founding family management on cash dividends and cash holdings differs from the family control and family management effects. Family-controlled firms positively affect cash holdings and negatively affect long-term debt, which suggests that such firms are financially risk-averse to preserve control (risk-avoiding behavior). Similarly, family management firms negatively affect cash holdings and positively affect total and long-term debt, suggesting that aversion behavior is reduced when the family-controlled firm is also the management. Family-controlled and managed firms positively affect cash dividends, suggesting that the use of these financial policies along with leverage reduce the problems of free cash flow and discipline the insider but are not sufficient to prevent a negative

787


effect of family control and management on corporate performance. The net effects of founding family management on financial policies differ significantly from the other two control and management structures. Founding family management has a positive effect on long-term and total financial debt. Additionally, founding family management does not affect cash holdings and negatively affects cash dividends. This result reinforces previous evidence that founding family management exhibits risk-taking behavior in its financial policies. The contribution of this research is the observation of this behavior through multiple financial policies.

Acknowledgments The authors thank the support given by CAPES/PDSE (4139-2014-03), CAPES/PROSUP, Universidade do Vale do Rio dos Sinos, Instituto Federal de Educação, Ciência e Tecnologia do Rio Grande do Sul, and Fundo MackPesquisa. References 1.

2.

5. Conclusions We investigated the effect of family control/management on corporate performance and the financial policy capital structure, cash holdings, and cash dividends. Using a sample of Brazilian publicly-held companies and applying a treatment effect model to solve self-selection and endogeneity problems. The results show that family control/management has a negative net effect on corporate performance.The family effect on financial policy is sensitive to the nature of the family relationship and involvement (control, family, or founding family management). Family-controlled firms have a positive effect on the level of cash holdings and a negative effect on the level of longterm debt, suggesting aversive behavior to financial risk to preserve control. Similarly, family management has a positive effect on the level of cash dividend and total long-term debt. Family management positively (negatively) affect cash dividends (cash holding. Founding family management has a positive effect on long-term and total debt. Additionally, founding family management does not affect cash holdings and negatively affects cash dividends. This result suggests that family firm incurs in more risk-taking behavior in its financial policies when the firm have the control and management of firm. Additionally, the results indicate that the use of these financial policies along with leverage reduce the problems of free cash flow and discipline the firm. However, that such efforts are not sufficient to prevent the negative effect of family control and family management on corporate performance. We contribute to the literature in emerging market context, in wich the effect of family firm on performance and financial policies remains largely unexplored. This is the first study that takes into account the effect of family firm behavior through multiple financial policies.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

788

Almeida, H., Campello, M. & Weisbach, M.S., 2004. The Cash Flow Sensitivity of Cash. The Journal of Finance, LIX (4), pp.1777–1804. Anderson, R.C., Duru, A. & Reeb, D.M., 2012. Investment policy in family controlled firms. Journal of Banking & Finance, 36(6), pp.1744–1758. Available at: http://www.sciencedirect.com/science/article/pii/ S0378426612000337 [Accessed November 9, 2014]. Anderson, R.C., Mansi, S.A. & Reeb, D.M., 2003. Founding family ownership and the agency cost of debt. Journal of Financial Economics, 68(2), pp.263– 285. Available at: http://www.sciencedirect.com/ science/article/pii/S0304405X03000679 [Accessed November 12, 2014]. Anderson, R.C. & Reeb, D.M., 2003. Founding-Family Ownership and Firm Performance : Evidence from the S & P 500. The Journal of Finance, LVIII(3), pp.1301– 1328. Anderson, R.W. & Hamadi, M., 2009. Large powerful shareholders and cash holding, London. Available at: http://www.lse.ac.uk/fmg/workingPapers/discussionPa pers/fmgdps/dp631.pdf. Berle, A.A.J. & Means, G.C., 1932. The Modern Corporation and Private Property, New York: The Macmillian Company. Bianco, M. et al., 2013. Family firms’ investments, uncertainty and opacity. Small Business Economics, 40(4), pp.1035–1058. Available at: http://link.springer.com/10.1007/s11187-012-9414-3 [Accessed November 13, 2014]. Croci, E., Doukas, J. a. & Gonenc, H., 2011. Family Control and Financing Decisions. European Financial Management, 17(5), pp.860–897. Available at: http://doi.wiley.com/10.1111/j.1468036X.2011.00631.x [Accessed October 22, 2014]. Demsetz, H., 1983. The structure of ownership and the theory of the firm. Journal of Law and Economics, 26(2), pp.375–390. Demsetz, H. & Lehn, K., 1985. The Structure of Corporate Ownership : Causes and Consequences Kenneth Lehn. Journal of Political Economy, 93(6), pp.1155–1177. Demsetz, H. & Villalonga, B., 2001. Ownership structure and corporate performance. Journal of Corporate Finance, 7, pp.209–233. Faccio, M., Lang, L.H.P. & Young, L., 2001. Dividends and Expropriation. The American Economic Review, 91(1), pp.54–78. Fama, E.F. & Jensen, M.C., 1983. Separation of Ownership and Control. Journal of Law and Economics, 26(2), pp.301–325. Fazzari, S.M., Hubbard, R.G. & Petersen, B.C., 1988. Financing Corporate and Constraints Investment. Brookings Papers on Economic Activity, 1, pp.141– 206.


15. Fitzsimmons, J.R. & Douglas, E.J., 2006. Entrepreneurs and funding decisions : evidence from Australian SMEs. Int. J. Entrepreneurship and Small Business, 3(1), pp.76–91. 16. Gallo, M. a. & Vilaseca, A., 1996. Finance in Family Business. Family Business Review, 9(4), pp.387–401. Available at: http://fbr.sagepub.com/cgi/doi/10.1111/ j.1741-6248.1996.00387.x [Accessed November 7, 2014]. 17. Gama, A.P.M. & Galvão, G., 2012. Performance, valuation and capital structure : survey of family firms. Corporate Governance, 12(2), pp.199–214. 18. Gonzalez, M. et al., 2014. Family Involvement and Dividend Policy in Closely Held Firms. Family Business Review, 27(4), pp.365–385. Available at: http://fbr.sagepub.com/cgi/doi/10.1177/089448651453 8448 [Accessed November 12, 2014]. 19. Hiebl, M.R.W., 2013. Risk aversion in family firms : what do we really know ? The Journal of Risk Finance, 14(1), pp.49–70. 20. Himmelberg, C.P., Hubbard, R.G. & Palia, D., 1999. Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics, 53(3), pp.353–384. Available at: http://www.sciencedirect. com/science/article/pii/S0304405X99000252 [Accessed November 13, 2014]. 21. Holderness, C.G. & Sheehan, D.P., 1988. The role of majority shareholders in publicly held corporations. Journal of Financial Economics, 20, pp.317–346. Available at: http://www.sciencedirect.com/science/ article/pii/0304405X88900499 [Accessed November 13, 2014]. 22. Holmström, B. & Tirole, J., 1993. Market Liquidity and Performance Monitoring. Journal of Political Economy, 101(4), pp.678–709. 23. Jensen, B.M.C., 1986. Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. The American Economic Review, 76(2), pp.323–329. 24. Jensen, M.C. & Ruback, R.S., 1983. The market for corporate control. Journal of Financial Economics, 11(1-4), pp.5–50. Available at: http://www.sciencedirect.com/science/article/pii/03044 05X83900041 [Accessed October 20, 2014]. 25. King, M.R. & Santor, E., 2008. Family values : Ownership structure, performance and capital structure of Canadian firms. Journal of Banking and Finance, 32(11), pp.2423–2432. Available at: http://dx.doi.org/10.1016/j.jbankfin.2008.02.002. 26. Lease, R.C., Mcconnell, J.J. & Mikkelson, W.H., 1984. The Market Value of Differential Voting Rights in Closely Held Corporations. The Journal of Business, 57(4), pp.443–467. 27. Lee, J., 2006. Family Firm Performance: Further Evidence. Family Business Review, 19(2), pp.103– 114. Available at: http://fbr.sagepub.com/cgi/ doi/10.1111/j.1741-6248.2006.00060.x [Accessed November 13, 2014]. 28. McConaughy, D.L. et al., 1998. Founding family controlled firms: Efficiency and value. Review of Financial Economics, 7(1), pp.1–19. Available at: http://www.sciencedirect.com/science/article/pii/S1058 330099801426 [Accessed November 13, 2014]. 29. Miller, D. et al., 2007. Are family firms really superior performers ? Journal of Corporate Finance, 13, pp.829– 858.

30. Morck, R., Shleifer, A. & Vishny, R.W., 1988. Management ownership and market valuation: an empirical analysis. Journal of Financial Economics, 20, pp.293–315. 31. Morck, R., Stangeland, D. & Yeung, B., 2000. Inherited Wealth, Corporate Control, and Economic Growth The Canadian Disease ? In Concentrated Corporate Ownership. University of Chicago Press, pp. 319–372. Available at: http://www.nber.org/chapters/ c9014. 32. Pindado, J., Requejo, I. & Torre, C., 2012. Do Family Firms Use Dividend Policy as a Governance Mechanism? Evidence from the Euro zone. Corporate Governance: An International Review, 20(5), pp.413– 431. Available at: http://doi.wiley.com/10.1111/j.14678683.2012.00921.x [Accessed November 14, 2014]. 33. La Porta, R. et al., 2000. Investor protection and corporate governance. Journal of Financial Economics, 58(1-2), pp.3–27. Available at: http://www.sciencedirect.com/science/article/pii/S0304 405X00000659 [Accessed September 30, 2014]. 34. Schulze, W.S. et al., 2001. Agency Relationships in Family Firms : Theory and Evidence. Organization Science, 12(2), pp.99–116. 35. Setia‐Atmaja, L., 2010. Dividend and debt policies of family controlled firms. International Journal of Managerial Finance, 6(2), pp.128–142. Available at: http://www.emeraldinsight.com/doi/abs/10.1108/17439 131011032059 [Accessed November 14, 2014]. 36. Shleifer, A. & Vishny, R.W., 1986. Large Shareholders and Corporate Control. Journal of Pol, 94(3), pp.461– 488. 37. Steijvers, T. & Niskanen, M., 2013. The determinants of cash holdings in private family firms. Accounting and Finance, 53(December 2010), pp.537–560. 38. Stein, J., 1989. Efficient capital markets, innefficient firms: a model of myopic corporate behavior. The Quarterly Journal od Economics, 104(4), pp.655–669. 39. Stulz, R., 1988. Managerial control of voting rights. Journal of Financial Economics, 20, pp.25–54. Available at: http://www.sciencedirect.com/science/ article/pii/0304405X88900396 [Accessed November 1, 2014]. 40. Villalonga, B. & Amit, R., 2006. How do family ownership, control and management affect firm value? Journal of Financial Economics, 80(2), pp.385–417. Available at: http://www.sciencedirect.com/science/ article/pii/S0304405X05001157 [Accessed September 4, 2014]. 41. Wang, K. & Shailer, G., 2013. Ownership concentration and firm performance in emerging markets: a meta-analysis. Journal of Economic Survey, 00(0), pp.1–31. 42. Wei, Z. et al., 2011. Family control, institutional environment and cash dividend policy: Evidence from China. China Journal of Accounting Research, 4(1-2), pp.29–46. Available at: http://www.sciencedirect.com/ science/article/pii/S1755309111000025 [Accessed November 14, 2014]. 43. Zhou, H., 2012. Are Family Firms Better Performers during Financial Crisis? In EFMA Annual Meetings 2012. Barcelona, p. 45. Available at: http://www.efmaefm.org/0EFMAMEETINGS/EFMA ANNUALMEETINGS/2012-Barcelona/papers/ EFMA2012_0643_fullpaper.pdf.

789


U.S. COMMUNITY BANKING PERFORMANCE M. Kabir Hassan*, William J. Hippler**, Walter Lane* Abstract Community banking plays an important role in financial intermediation in the United States, especially in the context of providing financing in smaller, rural markets and for small businesses. However, recent trends in regulation, the economic environment, and industry practices have led to a significant decline in the amount of FDIC-chartered institutions that qualify as community banks. In addition, the share of community-bank-held assets in the United States is declining as well. The decline of the community banking industry has significant implications for the efficiency and growth of the real economy, as larger banks may not be able to serve the community banking demographic as efficiently. In this study, we develop a dataset that allows us to analyze banking data collected from all FDIC-charted institutions and published by the FDIC. We use this data to analyze the community banking industry in the U.S. We are able to report the trends, strengths, and weakness of the community banking industry for the past twenty years. In addition, we develop two sets of community banking indexes meant to assess the relative and nominal changes in the strength of the community banking industry. One set of indicators simply measure market share, while others are composite community banking indexes that represent a unique contribution to the analysis of the industry. Finally, we analyze developments in the community banking industry across the tenures of the past three FDIC Chairs, which can provide context and guidance with respect to the perspective of key regulatory officials on community banking issues. Analysis of the data shows that the community banking industry is declining in the United States. Our Community Bank Momentum Index (CMOM) shows that, on a nominal basis, the community banking industry has experienced some growth; however, our Community Banking Relative Growth Index (CRGI) shows that community banks have been weakening, relative to non-community banks.*** Keywords: Community Banks, Growth Index, Momentum Index, Political Regime JEL Classification: C14, C41, G21, G33 * Department of Economics and Finance, University of New Orleans, USA ** College of Business and Public Management, University of La Verne, USA *** The Authors acknowledge a research grant given by Gulf Coast Bank and Trust Company to help support this research.

1. Introduction Community banking is an important sector of the financial system in the United States. Smaller, community banks comprise a majority of banking institutions in the United States and are especially responsible for servicing the banking needs of small businesses and rural communities. FDIC Chairman Martin Gruenberg stated in a 2012 conference on community banking the fact that community banks “provide nearly 40 percent of all the small loans that insured financial institutions make to businesses and farms”. As such, the role that community banks play in providing capital to small and rural businesses is significant, and maintaining a healthy community banking system has important economic implications. Despite the important role that community banks play in the U.S. economy, the number of community banking institutions in the United States has been shrinking over the past several decades. Changes in the economic and regulatory environments within the banking industry have contributed to an increasing

amount of banking consolidation, which has shifted the landscape of financial intermediation in the United States away from traditional community banks towards larger banks and non-depository institutions. The relaxation of interstate banking regulations in the early 1990s helped pave the way for the merging of financial institutions across state lines. Additionally, economic conditions, such as low interest rates, as well as the development of new financial products, have changed the banking environment in a way that favors larger financial institutions. The effects of the changes that have been taking place in the banking industry have important economic implications. The goal of this study is to provide an examination of the status of the community banking industry in the United States. We develop a unique FDIC dataset on financial institutions to examine several key statistics regarding the size, composition, and efficiency of the community banking industry. Additionally, we create several community banking indexes that can be used by regulators, academics, and practitioners in order to

790


quickly assess the strength and trends in the community banking industry. The paper is divided into six sections. Section 2 gives a brief overview of pertinent academic literature on the U.S. community banking industry. Section 3 describes the data and examines the community banking industry the U.S. Section 4 describes our construction of community banking indices and how they measure relative and absolute changes in community banking conditions in the U.S. Section 5 examines the role played by FDIC regulators in promoting the community banking the in the U.S. Section 6 concludes. 2. Literature Review Avery and Samolyk (2004) study the role of consolidation activities in community banks in small business lending during two 3-year study periods. They argue that consolidation of big (community) banks lowers (increases) loan growth during 1994 – 1997 period. The effect somewhat decreases in the 1998 – 2000 period, however. Overall, community bank lending to small businesses increased on a nominal basis, net of organization reclassification, due to consolidation and asset growth. Critchfield, Samolyk, Davison, Hanc, Gratton, and Davis (2004) examine the role of community banks in small business lending and local deposit markets, the decline in the number of community banks from 1985 to 2003, and consolidation patterns in different areas. They also investigate changes in performance, characteristics, and the prospects of community banks. DeYoung and Duffy (2002) take an atypical approach to analyze the community banking sector in the U.S. Their analysis complements the usual dataintensive studies by using first-hand observation, collected in a Federal Reserve survey from August 2001. The survey provides insight on payment service needs and fundamental the missions of community banks. It also provides insight into the threats and opportunities posed by large banks, the perceptions that the playing field is not always level, and the growing tension between traditional high-touch relationship banking and potentially more efficient high-tech banking. The study finds bankers to be more optimistic about the future viability of community banks and less sanguine about changes in regulations and technology. Such changes provide opportunities (e.g. attracting relationship-based deposit customers who prefer bundled pricing), but pose threats (e.g. competition from non-bank financial firms, such as brokerage firms) for community banks to coexist with large, multi-state banks. This coexistence is strongly dependent on efficient operation, good management, and continuous innovation. DeYoung, Hunter, and Udell (2004) study the effects of deregulation, technology, and competitive

rivalry on the size and health of the U.S. community banking sector and on the quality and availability of banking products and services. They further devise a theoretical framework to analyze the effects of these changes on the competitiveness of community banks. Empirical evidence indicates that these changes have intensified overall competition, but created potentially exploitable strategic positions for well-managed community banks. The study also predicts changes in the number and distribution of community banks. DeYoung, Lang, and Nolle (2007) argue that the use of internet websites as an alternative distribution channel improves the profitability of community banks by increasing revenue from deposit service charges. Internet adoption further increases the movement of deposits from checking accounts to money market deposit accounts, the use of brokered deposits, and the average wage rates for bank employees. Internet adoption has little to no effect on loan mix, however. The evidence suggests that the internet is a complement to physical branches. Goddard, Liu and, Wilson (2014) investigate the entry, exit, and growth of commercial banks in the U.S. from 1984 – 2012 using hazard function estimation and cross-sectional growth regression. They find that exit via acquisition is negatively related to asset size and quality, profitability, managerial efficiency, and capitalization, but positively related to liquidity. Smaller banks face the risk of failure as loan share and credit risk increase. Growth is negatively related to size, and is persistent to some extent. Hakenes, Hasan, Molyneux, and Xie (2014) theoretically show that regional small banks, compared to big interregional banks, more effectively promote local economic growth, especially where initial endowment is low and credit rationing is severe. Empirical analysis using a sample of German banks and corresponding regional statistics confirm their theoretical hypothesis. Jagtiani, Kotliar, and Maingi (2014) examine the roles and characteristics of community banks to find their impact in small business lending (SBL) and relationship lending. The authors analyze risk characteristics of acquired community banks, compare the pre- and post-acquisition performances and stock market reactions to these acquisitions, and investigate how the acquisitions have affected small business lending. They find that a declining number of community banks does not affect SBL, as large acquiring banks tend to play a larger role in SBL. Mitts (2014) examines section 601(a)(2) of the Jumpstart Our Small Business (JOBS) Act of 2012 and finds positive effects of deregistration arising from the act using the quasi-experimental technique of regression discontinuity (comparative interrupted time series analysis to regression discontinuity). The study finds that net income (pre-tax income) increases (decreases) by $1.27 ($2.35) per $1 of average assets.

791


Peirce, Robinson, and Stratman (2014) analyze the Mercatus Center’s Small Bank Survey and find that the recent Dodd-Frank Act affects small banks and their customers significantly. Respondents find Dodd-Frank to be more burdensome than the Bank Secrecy Act, as it has increased compliance costs significantly. Thus, the banks are considering shrinking product and service offerings, so the DoddFrank Act will affect customers indirectly. Whalen (2007) argues that community banks show continued reliance on traditional intermediation activities to generate income, meaning lending strategy is a key determinant of survival for these banks. Analysis of lending trends indicates that the preference for commercial (non-commercial) lending is increasing (decreasing) in community banks. Typical community banks have changed lending strategies, leading to reduced returns and increased risk, all else being equal. Also, small banks are suffering a large performance disadvantage, regardless of lending strategy. Whalen (2013) empirically analyzes the ability of banks to switch among competing supervisors as an important factor contributing to increased bank failures from 2007 to 2011. A competing risk hazard model by Fine and Gray (1999) suggests that mergers and supervisory changes represent competing risks. The study also suggests that higher failure rates of supervisor-switching banks cannot be attributed to differences in agency funding sources or organizational responsibilities. Wright (2011) argues that commercial banks have a failure rate of about 1 percent on average each year. Before failing, or merging, community banks have provided intermediation services for decades. The author attributes such success to governance, where stockholders carefully choose and monitor bank officers, and charge the officers to produce steady dividends. 3. U.S. Community Banking Definition and Performance 3.1 Community Banking Definition and Data The first step in analyzing the community banking industry in the United States and how it has changed over time is defining a community banking institution and identifying a source of banking data that can be utilized to analyze the banking industry over time. To this end, we use previous community banking studies as a guide in developing an empirical definition of a community banking institution. In addition, we utilize banking data that is collected by the Federal Deposit Insurance Corporation (FDIC) and made available on a quarterly basis to the public. We collect the FDIC data and modify it in order to create a database that best suits the needs of our longitudinal study.

There are varying definitions that regulators and previous community banking studies have used to define a community bank. Typically, community banks are considered institutions that fall below a certain size as measured by total assets. Previous research has used total asset sizes ranging from $750 million to $5 billion as the threshold used to define a community bank. Additionally, community banks are often defined by a limited geographic reach and a focus on more traditional banking activities as their primary operations. Accordingly, many community banking studies expand the empirical definition beyond asset size to incorporate limitations on the geographic and operational scope of the institution. In this study, we define a community bank as an FDIC-chartered institution having total assets of less than $1 billion in 2013 constant-dollars. We adjust nominal total asset values for the entire sample period to 2013 levels using the CPI-U measure of inflation provided by the Bureau or Labor Statistics (BLS). For the purpose of defining a community bank, we consider the adjusted total asset values at the charter level. We apply our definition of community banks to a dataset comprised of all banking institutions in the U.S. The FDIC collects data on a quarterly basis for all U.S. banking institutions that are FDIC insured in the Statistics on Depository Institutions (SDI) database. The database collects data on an institutional level in the “Financial Data” database and on a branch level in the “Branch Office Deposits” database. The data are made available on a quarterly basis from the forth quarter of 1992 at the institutional level and on an annual basis from 1994 at the branch level. The data are available from the public website of the FDIC (www.fdic.gov). The Financial Data contain information on over 1,000 variables of interest for each FDIC-chartered institution. The data include important financial statement data, such as information regarding assets, liabilities, equity, income, and expenses. In addition, the data also contain more detailed information that are valuable in analyzing the community banking industry, such as letters of credit, derivative securities, and the types of loans held by financial institutions. Each data file contains the cross-sectional data for each quarter, and the data are presented in 63 files, based on several basic categories, such as assets and liabilities, changes in equity, derivatives, deposits, etc. Unfortunately, since the SDI data as provided on the FDIC website are presented cross-sectionally by category in this manner, they are of limited use for the purpose of broader analysis. In order to make the data more robust, we combine the data to create one database consisting of a panel dataset containing all the variables for every institution for each quarter. We first combine the 63 cross-sectional data sets for each quarter, which creates a single cross-sectional dataset for each quarter, consisting of every variable (>1,000 variables). Next, we combine the cross-

792


sectional data into one panel dataset that contains all the variables for the entire time series. The resulting dataset contains over 800,000 institution-level observations. The larger, combined dataset allows for the analysis of the relative performance of the community banking industry from both a crosssectional and time-series perspective. 3.2 Community Banking in the United States Table 1 illustrates how banking characteristics have changed over the past twenty years for both community banks and non-community banks in the United States. The Table reports the levels of several key characteristics for the average community and non-community bank in the sample. Included in the table are characteristics describing the size, profitability, efficiency, and capitalization of a representative bank in each industry.

3.2.1

Community Banking Industry Size

Community banks dominate non-community banks in terms of the number of institutions. In 2013:Q4 there are 6,149 FDIC charters that qualify as community banks, compared with only 671 non-community banks. Additionally, a considerable amount of consolidation has taken place in the commercial banking industry over this time period, which has occurred as a result of several key changes in the regulatory and operational environments of commercial banks. Both non-community and community banks saw a reduction in the number of FDIC charters from 1993 to 2013; however, the brunt of the consolation has been borne by the community banking industry. The number of non-community banks has decreased 19.1% from 1993 to 2014, while the number of community bank charters has decreased by 50.8%.

Table 1. U.S. FDIC-chartered Bank Statistics

No. of Institutions Total Assets Number of Employees Total Cash Balances Net Loans and Leases Loan Loss Allowance Total Liabilities Total Deposits Total Equity Capital Total Interest Income Net Interest Income Total Noninterest Income Net Operating Income Net Income Cash Dividends ROA ROE Efficiency Ratio Assets Per Employee Net Interest Margin Net Operating Income to Assets Equity Capital to Assets

Community Banks 1993:Q4 2013:Q4 % Change 12,495 6,149 -50.8% 94,027 222,266 136.4% 45 55 22.1% 4,993 18,492 270.4% 53,072 137,931 159.9% 880 3,821 334.4% 85,669 200,441 134.0% 80,995 186,206 129.9% 8,386 24,178 188.3% 6,391 8,527 33.4% 3,741 7,380 97.2% 953 2,476 159.8% 870 1,926 121.3% 977 1,974 102.0% 416 1,016 144.0% 1.09 0.93 -14.9% 12.12 7.01 -42.2% 73.14 108.62 48.5% 2.30 4.63 101.3% 4.55 3.66 -19.6% 0.97 0.90 -7.3% 9.58 11.37 18.7%

Both community and non-community banks have seen a reduction in the number of institutions that remain active in the U.S., but, as evidenced by Table 1, the impact on the community banking industry has been more pronounced, and the trend is continuing. Figure 1 depicts the trend in the number of FDIC chartered institutions over the twenty-year sample period. Until approximately 2001, both community and non-community banks experienced consistent declines in the number of chartered institutions. This trend is consistent with significant

Non-Community Banks 1993:Q4 2013:Q4 % Change 829 671 -19.1% 4,298,681 19,640,810 356.9% 1,492 2,571 72.3% 294,830 2,397,926 713.3% 2,502,002 10,137,891 305.2% 61,931 181,208 192.6% 3,979,296 17,449,481 338.5% 3,057,847 14,750,939 382.4% 320,934 2,214,431 590.0% 282,094 614,069 117.7% 150,555 545,823 262.5% 86,586 349,465 303.6% 37,961 208,282 448.7% 42,861 212,130 394.9% 23,200 121,852 425.2% 0.97 1.14 17.9% 14.73 9.70 -34.1% 67.60 66.46 -1.7% 5.34 23.66 342.9% 4.29 3.70 -13.6% 0.89 1.13 26.2% 7.65 11.44 49.5%

changes in the commercial banking environment over that time period. Regulatory changes reduced restrictions on interstate banking, making it easier for banks to conduct business across state lines. As a result, both community and non-community banks merged and adjusted operations as a response to these changes. As a result, the number of FDIC-chartered institutions declined. In addition, improvements in technology within the commercial banking industry created significant economies of scale that banks were

793


able to take advantage of by expanding and merging

across state lines.

Figure 1. Number of U. S. FDIC Charters

However, after 2001, the trend stabilizes for non-community banks, and the number of chartered non-community banks actually increases from 2001 to 2008, reflecting the growth experienced by large financial institutions leading up to the financial crisis. On the other hand, the community banking industry continued to decline during the pre-crisis period, and, in terms of the number of institutions, did not experience the growth seen by the non-community banking industry. The community banking industry saw a relative weakening in terms of the number of chartered institutions in the U.S., compared with noncommunity banks. In addition, the remaining community banks also exhibited less growth than their larger, non-community counterparts. In 1993, the average community bank had total assets of about $94 million. This grew to approximately $222.7 million by 2013, an increase of 136.4%. The average noncommunity bank, on the other hand, saw an increase of 356.9% from $4.3 billion in 1993 to $19.6 billion in 2013. Accordingly, growth in total liabilities and total deposits show similar growth patterns.

3.2.2

Community Bank Profitability

Larger, non-community banks have exhibited higher historical returns than their community bank counterparts. From 1993 to 2013, net income for noncommunity banks has increased by 394.9%, while net income for the average community banks has only increased by 102%. In 2013, the average community bank achieved a return on equity of 7.01% and the average noncommunity bank ROE was 9.70%. In addition, from 1993 to 2013, non-community bank ROE declined by 34.1%, while community bank ROE declined by 42.2%. In addition, the average ROA for a community bank in the fourth quarter of 2013 is 0.93%, while that of the average non-community bank is 1.14%. In fact, changes in the ability of commercial banks to generate returns on total assets illustrate the relative weakening of the community banking industry. In 1993, community banks generate higher average returns on assets (1.09%) than non-community banks (0.97%); however, from 1993 to 2013, average ROA for community banks declined 14.9%, while that of non-community banks actually increased 17.9%.

Figure 2. Return on Equity

794


Figure 2 shows the trend in the average community and non-community bank ROE for the sample period in question. The Figure shows that noncommunity banks have had a long established advantage in generating returns on equity. Except for a few exceptional quarters, the ROE of the average non-community bank is higher than that of the average community bank. In addition, aside for a period of instability surrounding the financial crisis of 2008, the ROEs generated by all FDIC-chartered institutions are stable over the past twenty years. Figure 3 shows the trend in the average return on assets for community and non-community banks from

1993 to 2013. We again document the relative weakening of the community banking industry. At the beginning of the sample period, community banks yielded returns on assets that were comparable to those of non-community banks. However, after 1995, aside from the period surrounding the 2008 financial crisis, non-community banks retain a consistent advantage in generating ROAs above those of their community bank counterparts. In addition, excluding the period surrounding the 2008 financial crisis, average ROA for FDIC-chartered institutions remains consistent over the time period analyzed in this study.

Figure 3. Return on Assets

3.2.3

Community Bank Efficiency

Banking Efficiency is also an important component of ensuring stable financial institutions, and this is another area in which non-community banks exhibit relative competitive advantages. The Efficiency Ratio is one measure commonly used by the FDIC as an indicator of banking Efficiency, and it is defined as noninterest expense divided by net operating revenue – a measure of fixed cost. A lower value of the efficiency ratio means that a smaller portion of revenues are spent on overhead, implying greater efficiency. The average efficiency ratio for community banks is 108.62% in 2013, compared with that of the average non-community bank of 66.46%. In addition, community banks have seen an increase in the average efficiency ratio of 48.5%, while noncommunity banks have experienced a decrease of 1.7%, indicating that non-community banks have become more efficient, while community banks have become slightly less efficient. Additionally, Figure 4 depicts the trends in the average efficiency ratios of community and non-community banks over time. Since the beginning of the sample period in 1993, non-community banks have held a consistent competitive advantage over community banks in the United States with regards to the efficiency ratio.

795


Figure 4. Efficiency Ratio

Another important measure of core-banking efficiency is the ability of depository institutions to profitably transform deposits into loans. Net interest margin essentially captures the degree to which banks are successful at maintaining sufficient margins between its inflows and outflows of cash flows from operations. Net interest margin is an area in which community banks have held a historical competitive advantage over their non-community bank counterparts; however, this advantage appears to be diminishing over time. Core banking activities – turning interest-bearing deposits into interest paying loans – has become increasingly difficult over the past several decades. Table 1 shows that the net interest margin for both community and non-community banking institutions has declined over the past twenty years. Community banking average net interest margin has declined 19.6%, from 4.55% in 1993 to 3.66% in 2013. Likewise, average net interest margin for noncommunity banks has declined by 13.6%, from 4.29% in 1993 to 3.70% in 2013. There are many contributing factors to this. Firstly, interest rates have become more stable over the past twenty years, compared with previous periods, such as the 1970s and 80s. A more stable interest rate environment reduces interest rate risks for banks, and, in a competitive banking environment, banks are able to lower the spread between interest bearing assets and liabilities, while still remaining profitable. Secondly, the period of extended low short-term interest rates induced by the Federal Reserve surrounding the recessions of 2001 and 2008 have put downward pressure on the net interest margins of commercial leading institutions. Thirdly, the development and increased implementation of hedging mechanisms, such as interest rate and default swaps, have reduced interest rate risk for financial intermediates, allowing for lower interest margins. Finally, developments in the capital markets, such as the growth of nondepository lending, has provided alternatives to the traditional banking systems and increased competition

for commercial banks, thus putting downward pressure on interest margins. Despite the decline in net interest margins seen in the industry as a whole, the community banking industry has been particularly affected and has lost one of its competitive advantages. Figure 5 depicts average net interest margin for community and noncommunity banks from 1993 to 2013. The figure depicts the general decline in net interest margin that has been seen industry-wide. In addition, it illustrates the fact that community banks have been particularly impacted by the decline in interest margin. Prior to 2009, community banks have a significant advantage over non-community banks with regards to net interest margins, as they are consistently higher. However, that advantage appears to have eroded in more recent periods.

796


Figure 5. Net Interest Margin

3.2.4

Community Banking Relative Capital

The amount of equity capital held by commercial banks is of particular concern, because equity capital contributes to the overall risk of the institution. Accordingly, capital ratios are closely regulated and monitored by the Federal Reserve, FDIC, and other regulators. Community banks have historically held a larger amount of capital, as measured by total equity capital-to-total assets. In 1993, community banks had a capital ratio of 9.58%, while non-community banks had a capital ratio of 7.65%. In addition, capital ratios have been increasing industry-wide over the past twenty years. However, the relative amount of capital held by community banks has been decreasing over

time. The average community bank’s capital ratio increased by 18.7%, from 9.58% to 11.37%, between 1993 and 2013. However, that of non-community banks increased by 49.5% to 11.44% over the same period. Figure 6 reflects the changes in the average capital ratio of community and non-community banks over the sample period. The figure illustrates both the increase in average capital ratios over time as well as a convergence between the capital ratios of community and non-community banking institutions. Interestingly, the capital ratio for all FDIC-chartered institutions peaked right before the financial crisis in 2007, yet many institutions still failed to absorb the financial strains posed by the crisis.

Figure 6. Capital Ratio

4. Community Banking Indexes Community banks serve an important economic role in the United States, yet the number of institutions qualifying as community banks in the U.S. has declined by over 50 percent since 1980. The effects of a declining community bank presence in the banking

market has the biggest impact on customers in rural communities and those seeking small business and personal loans, because community banks are historically the primary service providers for these customers. The net effect of a decline in community banking services on these customers and the U.S. economy as a whole is, therefore, of significant

797


interest to industry practitioners, policy makers, and academics. In order to evaluate the strength of the community banking system over time, it is important to develop key measures, or indicators, of the strength of the community banking system in the U.S. These measures can be used to both evaluate the status of the industry as well as gauge the effectiveness of policies aimed at promoting community banking activity. 4.1 Community Bank Index Development In this study, we develop several indicators of community banking strength in the U.S. The indicators focus on three major banking characteristics: Total assets; net loans and leases; and total deposits. These three banking characteristics are arguably the most important indicators of the size and growth of a traditional financial institution. Total assets measures the overall size of the bank or banking industry. Net loans and total deposits, on the other hand, focus more specifically on traditional banking activities – the transformation of deposits into loans. Thus, measures derived from loan and deposit characteristics can provide a good indication of trends in core banking operations. We generate two types of indicators. The first are market share indicators. These measures gauge the market power of community banks, relative to that of the entire banking system as a whole. In other words, market share indicators measure the size of the community banking industry. Changes in market share over time represent a relative expansion or contraction in community banking activities in the U.S. We generate three market share indicators from the bank characteristics previously mentioned: total

Asset Market Sharei,t =

asset market share; total loan market share; and total deposit market share. The second set of indicators developed in this study are composite indicators, which combine the asset, loan, and deposit activities of community banks into one index of community banking strength. We develop two composite measures of community banking strength: a relative growth measure, and a momentum measure. The relative growth indicator compares the growth rates in assets, loans, and deposits of community banks with those of larger, non-community banks. The momentum indicator examines the overall direction of growth in community banking assets, loans, and deposits. 4.2 Market Share Indexes Community bank market share indexes measure the relative size of community banks, compared to the banking system as a whole. We utilize data from the Statistics on Depository Institutions (SDI) database of the FDIC to construct three community bank market share measures. The data are sampled quarterly from 1992:Q3 to 2013:Q3. A community bank is defined as one with total assets less than $1 billion in constant 2013 dollars. We compute three market share measures: Asset Market Share; Loan Market Share; and Deposit Market Share. 4.2.1

Asset Market Share

Asset Market Share represents the percentage of total banking assets that are held by community banks. The market share indicator ranges from zero to 100 percent and is measured by:

Total Community Bank Assetsi,t , Total Bank Assetsi,t

where i represents the geographic area, and t represents quarter t.

4.2.2

(1)

Loan Market Share

Loan Market Share represents the percentage of total loans held in the banking system that are held by community banks. The market share indicator ranges from zero to 100 percent and is measured by:

Loan Market Sharei,t =

Total Community Bank Loansi,t , Total Bank Loansi,t


4.2.3

(2)

Deposit Market Share

Deposit Market Share represents the percentage of total deposits in the banking system that are held by community banks. The market share indicator ranges from zero to 100 percent and is measured by:

Deposit Market Sharei,t =

Total Community Bank Depositsi,t , Total Bank Depositsi,t

798

(3)



4.3.1

4.3 Composite Indexes The second set of indexes that we propose are two composite indexes that aggregate the asset, loan, and deposit characteristics of community banks into one index. We develop two such composite indexes: the relative growth index; and the momentum index.

CRGI i ,t

The Community Bank Relative Growth Index (CRGI) measures the relative growth disparity between the community banking industry and the non-community bank industry. The Index is defined as follows:

 Community Bank Noncommuni ty Bank       Asset Growthi ,t    Asset Growthi ,t    Community Bank Noncommuni ty Bank     100, =     Loan Growthi ,t Loan Growthi ,t        Community Bank Noncommuni ty Bank      Deposit Growthi ,t    Deposit Growthi ,t

where i represents the geographic area, and t represents quarter t. Growth is defined as the percentage change from period t-1 to t. In the calculation of CRGI, asset growth, loan growth, and deposit growth are all weighted equally. In other words, deviations in one area (say deposits) can be offset by changes in the opposite direction in another area (say loans). The value of CRGI is unbounded; however, positive values of CRGI indicate overall growth in the community banking industry, relative to that of larger banks. Conversely,

CMOM i ,t

Community Bank Relative Growth Index (CRGI)

(4)

negative values of CRGI represent a weakening of the community bank industry, relative to non-community banks. 4.3.2

Community (CMOM)

Bank

Momentum

The Community Bank Momentum Index (CMOM) measures the degree to which the community banking industry has expanded or contracted over the past quarter. The index is defined as follows:

 Community Bank Community Bank Community Bank    Assets i ,t Loansi ,t Deposits i ,t   =     3,  Community Bank Community Bank Community Bank   Assets i ,t -1 Loansi ,t -1 Deposits i ,t -1  

where i represents the geographic area, and t represents quarter t. In the calculation of CMOM, asset growth, loan growth, and deposit growth are all weighted equally, so large increases in one area can offset large declines in another. By construction, the CMOM Index has a lower bound of negative three and is unbounded above. However, realistic values of CMOM are around zero. A value of zero indicates that the community banking industry has remained roughly the same size from one quarter to the next. A positive value of CMOM means that the community banking industry is expanding, while a negative value means that it is shrinking.

Index

4.4.1

(5)

Market Share Measures

Several key statistics for the market share community bank indexes are presented in Table 3. The market share indexes provide indexes based on the asset, loan, and deposit market share of community banks for the U.S. Total community banking asset market share in the U.S. is 9.4%. The asset market share index measure of community banking strength reflects a decline in the role that community banks play in the national economy. Table 1 shows that the total number of community banking institutions has declined dramatically since 1993. Table 2 illustrates that this trend is reflected in the market share measures of community banking strength as well. Community bank asset share has declined since 1993

4.4 Community Banking Index Values

799


at the national level, with asset market share declining by 163.9%. Not surprisingly, the levels of community bank loan market share are similar to asset market share. The overall decline in the community bank loan share has been 118.5%, from 24.2% in 1993 to 11.1% in 2013.

The levels of community bank market share of deposits are similar to those of assets and loans. Deposit market share at the national level of 10.4% in 2013 represents a decline from that 28.5% in 1993.

Table 2. Market Share Community Banking Indexes Measure Asset Share Loan Share Deposit Share 4.4.2

1993:Q4 24.80% 24.20% 28.50%

Composite Community Banking Indexes

Several key statistics for the composite community bank indexes are presented in Table 4. The composite community banking indexes consist of the

2013:Q4 9.40% 11.10% 10.40%

% Change -163.90% -118.50% -175.20%

Community Bank Relative Growth Index (CGRI) and the Community Bank Momentum Index (CMOM). Table 3 presents statistics for the composite indexes for the U.S. community banking industry.

Table 3. Composite Community Banking Indexes Measure Community Bank Relative Growth Index (CRGI) Community Bank Momentum Index (CMOM) In terms of relative growth, Table 3 shows that the average value of the CRGI is negative, which reflects and confirms the fact that the community banking industry in the U.S. has been shrinking, relative to non-community banks. Figure 7 depicts the

Avg. -4.21 0.006

Std. Dev. 4.11 0.025

2013:Q4 -2.81 0.012

value of the CRGI from 1993 to 2013. It is clear that the vast majority of the values are negative, indicating a shrinking of the community banking industry, relative to non-community banks, for a majority of the quarters reported by the data.

Figure 7. Community Bank Relative Growth Index (CRGI)

In terms of momentum, the outlook on community banking is less grim, because no comparison is made to non-community banking growth. As shown in Figure 8, a majority of the CMOM are positive for the time period under study, indicating the industry had positive nominal growth over the past twenty years.

800


Figure 8. Community Bank Momentum (CMOM) Index

5. FDIC Leadership Banking

and

Community

The Federal Deposit Insurance Corporation is the main regulator and insurance provider for community banks in the United States. As such, the leadership of the FDIC can play a significant role in terms of transforming the community banking industry. The Chairperson of the FDIC is recognized as a transformative figure in this regard and must be appointed by the President and approved by Congress. In this section, we analyze the performance of the community banking industry under the past three FDIC Chairpersons, Donald Powell, Sheila Bair, and Martin Gruenberg, who have presided over the FDIC since 2001. Without a more thorough analysis, it is inappropriate to attribute the performance of the community banking industry with the actions of the FDIC or any individual therein; however, the viewpoints of key persons at the FDIC, namely the Chairperson, can have significant implications for the community banking industry and its outlook going forward. While all FDIC Chairs are concerned over the increasingly difficult competitive environment for community banks, the trends leading to consolidation in the industry have persisted nonetheless. Actions taken by the previous FDIC Chairpersons have led to the increased study and awareness of the trends affecting community banks in the United States; however, the ability and willingness to provide more fundamental support for the community banking industry is questionable, and it is clear that their policies have been unsuccessful at maintaining community bank market share. In fact, statements by recent FDIC chairs indicate that, while being concerned over recent pressures placed on the industry, they are satisfied with the resilience of the community banking industry and the competitive banking atmosphere, despite the continued loss of community banking institutions and market share.

Donald Powell became FDIC Chairman on August 29, 2001 after being nominated by George W. Bush. Powell has a background in community banking, having served as the President and CEO of The First Bank of Amarillo in Amarillo, Texas. He often acknowledged trends in the industry and the challenges faced by community banks, particularly consolidation. He noted before the Independent Community Bankers Association in 2004: “Consolidation has been relentless. The number of community banks declined by almost half between 1985 and 2001. Market share dropped significantly. The hardest hit were banks with assets of less than $100 million. But if you can pull back from this for a moment, and look at the longer-term trends, there is a positive story in the numbers. Community banks continue to maintain presence in all types of markets - urban, suburban, and rural. They remain profitable in both regions of population growth and population decline. Further, community bank performance is satisfactory when compared to that of the largest banks. From 1992, ROA, for example, has been at least 100 basis points - and this remains true even in those markets experiencing population declines. Finally, the community bank business model is still sought-after as you follow your customers into the suburbs and inner cities. We've seen more than 1,100 new banks formed since 1992, and their continued strength in small business and neighborhood lending has helped serve new customers, create jobs, bank the unbanked, and add to the economic vitality of their communities.” Powell left the FDIC in November of 2005 for a position aiding the Gulf Coast recovery in the aftermath of hurricanes Katrina and Rita. The chair position was vacant for seven months following Powell’s departure, but was filled by Sheila Bair in June of 2006. While FDIC Chair, Bair also acknowledged the importance of community banks and the challenges they face in the industry. To this end, under Bair’s tenure as Chairwoman, the FDIC

801


voted to approve the creation of the FDIC Advisory Committee on Community Banking in order to specifically address community banking issues. In establishing this Committee Bair said, “community banks are the lifeblood of our nation's financial system, supplying much-needed credit to countless individuals, small businesses, nonprofit organizations and other entities in large and small towns around the country”. Furthermore, in acknowledging the difficulties facing community banks, Bair stated in remarks to the Independent Community Bankers Association in 2010 that: “[It] is community banks that finance the startups where dreams are launched and new jobs are created. But, for too long, your business model has been at a critical disadvantage to larger financial companies with implicit government backing and insufficient regulatory oversight.” Additionally, Bair recognized the regulatory burdens that affect community banking institutions, In a recent 2014 Fortune Magazine Article, Bair commented on the costs faced by community banks due to the many new regulations established following the financial crisis: “The benefits [of the regulations] are less clear for regional and community banks. Many of the new regulations have hefty, fixed startup costs, which disproportionately impact smaller institutions.” Bair chaired the FDIC until July of 2011, when Martin Gruenberg acted as Chairman until confirmed by Congress in 2012. Gruenberg has long been active at the FDIC, as he has previously served as Acting Chairman, Vice Chairman, and a member of the Board of Directors throughout the past decade. Gruenberg, too, recognizes the importance of community banks. At the Conference on the Future of Community Banking in 2012, Gruenberg remarked: “Community banks play a crucial role in the financial system of the United States. Community banks with assets of less than $1 billion account for a little more than 10 percent of the banking assets in our country, but provide nearly 40 percent of all the small loans that insured financial institutions make to businesses and farms. Given the labor intensive, highly customized nature of many small business loans, it is not clear that large institutions would easily fill this critical credit need if community banks were not there. Community banks also play a crucial role in extending credit and providing financial services in rural communities, in small towns, and in inner-city neighborhoods. In many of those localities, if not for the community bank there would be no easy access to an insured financial institution. In my view there is a clear public interest in maintaining a strong community bank sector in the U.S. financial system.” Gruenberg went on to describe his views on the role of the FDIC in the assisting the community banking industry: “The FDIC is the lead federal regulator for the majority of community banks in the United States and

the insurer of all. In those capacities, it seems to me, the FDIC has a responsibility to use our resources to gain a better understanding of the challenges facing community banks and to share that understanding with the banks as well as the general public.” The FDIC is a regulator and insurance provider for the community banking industry, and the organizations Chairperson can have considerable impact on its focus. The past several FDIC chairs have all acknowledged the importance of the community banking industry and the challenges facing it. However, it is questionable whether the FDIC is able or wiling to enact policies that actively encourage the community banking industry. The FDIC has conducted a significant amount of support in terms of research surrounding the community banking industry. While the FDIC plays a role in identifying the impact of industry trends and regulations on community banks, its intention to affect these trends directly appears to be minimal. The FDIC’s main goal in promoting community banks appears to be through industry analysis that can be used by industry practitioners in order to adapt to industry trends that the leadership at the FDIC appear to hold as inevitable. Despite the FDIC’s concern over industry trends that have been negatively impacting certain aspect of the community banking business, these trends have continued under the supervision of the past three FDIC Chairpersons. Table 4 analyzes the changes in key banking statistics over the tenures of the past three FDIC chairs. 5.1 Community Banking Statistics and FDIC Leadership 5.1.1

Industry Size and FDIC Leadership

As many statements from the current and previous FDIC chairs illustrate, the number and prominence of FDIC-chartered community banks has been declining over the past several decades, and this trend has continued under the past three chairpersons. The total number of community banking institutions fell most dramatically under the tenure of Sheila Bair. During this time, the number of community banking institutions fell by 14.7%. This is not surprising since the time of her tenure coincides with the financial crisis of 2008. Donald Powell’s tenure also saw a dramatic reduction in the number of community banking institutions, as community banks fell by 9.3%, while non-community banks actually increased by 7.4%. This fact illustrates the consolidation and realignment in favor of non-community banks that occurred following the regulatory changes of the 1990s and early 2000s. Both the number of noncommunity and community banking institutions fell under the succeeding FDIC Chairs; however, as previously illustrated, community banks have been relatively more affected.

802


Average community bank size grew most rapidly under the tenure of FDIC Chairman Donald Powell, as total assets increased by 24.2% and total equity capital increased by 28.0%. This trend is consistent with the rapid growth that took place in the banking industry leading up to the financial crisis in 2008. Additionally, Powell’s tenure is currently the longest of the three, so more industry growth took place under his supervision. On the other hand, net income fell during the tenures of Donald Powell and

Sheila Bair, but increased by 24.2% under Martin Gruenberg. Martin Gruenberg’s tenure has also seen dramatic increases in cash dividends and net operating income, with increases of 106.5% and 134.4%, respectively. This fact echoes the sentiments of the FDIC regulators that community banks have remained resilient and competitive in spite of the 2008 financial crisis and continued declines in the number of institutions and their market share.

Table 4. U.S. FDIC Bank Statistics

No. of Institutions Total Assets Number of Employees Total Cash Balances Net Loans and Leases Loan Loss Allowance Total Liabilities Total Deposits Total Equity Capital Total Interest Income Net Interest Income Total Noninterest Income Net Operating Income Net Income Cash Dividends ROA ROE Efficiency Ratio Assets Per Employee Net Interest Margin Net Operating Income to Assets Equity Capital to Assets

5.1.2.

Powell Community Noncommunity % Change % Change -9.3% 7.4% 24.2% 33.0% 5.2% 5.5% 2.7% -3.7% 29.7% 36.5% 47.7% -11.7% 24.1% 30.9% 23.2% 33.8% 28.0% 55.2% -1.8% 3.8% 25.0% 20.5% 21.1% 24.2% 49.3% 51.6% -1.8% 3.8% 27.5% 13.3% 19.1% 8.6% 8.0% 4.8% -7.2% -1.5% 15.9% 58.6% 3.3% -5.1% 23.5% 12.7% 5.8% 13.0%

Community Bank Profitability and FDIC Leadership

Despite increased consolidation and eroding net interest margins, the remaining community banks remain relatively competitive in terms of returns on assets and equity. Both ROE and ROA increased most dramatically over the tenure of Gruenberg. Average community bank ROE grew by 222.1%, and ROA grew by 36.8%, respectfully, under his tenure as FDIC Chairman. This compares with increases in average non-community bank ROE and ROA of 37.4% and 30.0%, respectively. The fact that profitability measures have increased more rapidly for community banks in more recent years coincides with efficiencies gained from technology and consolidation. In addition, improved community bank performance may speak to the resilience of the community banking industry.

Bair Community Noncommunity % Change % Change -14.7% -7.6% 18.7% 26.5% 0.2% 6.1% 176.7% 242.2% 8.9% 819.8% 90.8% 209.6% 19.3% 25.0% 22.8% 43.0% 18.1% 40.2% -41.9% -42.4% -25.0% -6.6% -38.1% -32.0% -65.5% -41.5% -41.9% -42.4% -58.0% -33.1% -37.7% -38.7% -73.6% -48.6% 3.7% 20.4% 23.5% -36.8% -7.7% -1.8% -42.1% -35.9% -9.1% 4.1%

5.1.3.

Gruenberg Community Noncommunity % Change % Change -9.0% -2.5% 7.2% 9.8% 5.1% 0.9% 0.5% 37.2% 5.2% 8.7% 6.4% -30.9% 7.5% 10.1% 8.2% 15.9% 6.6% 8.0% 24.2% 26.8% 37.1% 35.2% 68.2% 46.2% 134.4% 70.4% 24.2% 26.8% 106.5% 59.9% 36.8% 30.0% 222.1% 37.4% 45.8% 1.8% 0.0% -35.8% -5.3% -3.5% 44.0% 39.0% -2.0% -1.3%

Community Bank Efficiency and FDIC Leadership

Changes in community banking efficiency across recent FDIC chairs reflects the increasingly difficult operating environment for both community banks and non-community banks. Average net interest margin shows the most improvement under Donald Powel, where net interest margins for community and noncommunity banks change by 3.3% and -5.3%, respectively. This corresponds with the rising interest rate environment that was in place during the period leading up to the financial crisis. For the remaining two tenures, however, net margins shrink for both community and non-community banks. Under Sheila Bair’s tenure, average community bank net interest margin experienced the largest decline of 7.7%, while those of non-community banks declined by 1.8%. Average net margins for community banks declined by 5.3%, compared with a decline of 3.5% for noncommunity banks, under Martin Gruenberg.

803


Efficiency Ratios have improved over time for both community and non-community banks, and this is reflected during the tenures of Sheila Bair and Martin Gruenberg. The average efficiency ratio for a community bank improved by 45.8% under Gruenberg and 3.7% under Bair. Average community bank efficiency ratios fell during the tenure of Donald Powell by 7.2%, while those of non-community banks fell by 1.5%. 5.1.4.

Community Banking Relative Capital and FDIC Leadership

Table 5 also shows that the average capital ratios for community banks increased the most under the tenure of Donald Powell, as capital increased by 5.8%. On the other hand, under the tenures of Sheila Bair and Martin Gruenberg, the average capital ratio declined by 9.1% and 2.0%, respectively. Average capital ratios for non-community banks, on the other hand, actually increased under the tenures of both Powell and Bair, at 13.0% and 4.1%, respectively, but it declined by 1.3% under Gruenberg.

5.2. Community Bank Indexes and FDIC Leadership The somewhat mixed performance of community banks over the tenures of the past three FDIC Chairs illustrate the need for more a more comprehensive measure of community banking strength. Table 5 illustrates how our community bank strength indicators have changed under the tenures of the past three FDIC Chairs. The Asset Market Share Index of community banking strength declined over each of the past three FDIC chair tenures, but it declined most significantly under the tenure of Donald Powell with asset share diminishing by 18.5%. Likewise, total community banking asset share under Sheila Bair and Martin Gruenberg fell by 12.0% and 8.1%, respectively. The decline in the asset market share index over the past three FDIC Chair tenures illustrates the decline in community bank asset market shares that has occurred over time due to changes in the banking industry. Therefore, the fact that the decline was most dramatic under Donald Powell’s term is not surprising, considering the fact that Powell has the longest tenure of the three, followed by Bair and Gruenberg.

Table 5. Community Banking Indexes Asset Share Chair Beg. Powell 15.4% Bair 11.8% Gruenberg 10.2% Community Bank Relative Growth Index (CRGI) Chair Avg. Powell -4.285 Bair -2.377 Gruenberg -3.990 Community Bank Momentum Index (CMOM) Chair Avg. Powell 0.024 Bair 0.001 Gruenberg -0.007 The two composite indexes give a better picture of community banking strength. Table 6 implies that the community banking industry performed poorly, when compared the non-community banking industry, under the terms of all three of the most recent FDIC chairs. The average value of Community Bank Relative Growth Index (CRGI) is highest (but still negative) under the term of Sheila Bair with an average value of -2.377. However, the variance is highest, which is likely due to the increased volatility caused by the onset of the financial crisis. On the other hand, the CGRI performed the worst under Donald Powell, with an average value of -4.285. While we have shown that the banking industry as a whole grew most dramatically during the tenure of Donald Powell, community banks lost the most

End 12.5% 10.4% 9.4%

% Chg. -18.5% -12.0% -8.1%

Std. Dev. 3.076 5.427 2.882 Std. Dev. 0.019 0.033 0.015 ground during this time in relation to their noncommunity bank counterparts. In addition, the improved relative performance of the community banking industry could in part be due to the increased focus of the FDIC on community-bank related issues, such as the establishment of the FDIC’s Advisory Committee on Community Banking. It is also likely that the relative performance of community banks also improved as the industry adjusted to the changing regulatory environment. Table 5 also shows the change in Community Banking Momentum Index (CMOM) for each of the recent FDIC Chairs. The average value of the CMOM Index is highest under the tenure of Donald Powell with an average value of 0.024, followed by that under Sheila Bair of 0.001. The CMOM is negative

804


under the tenure of Martin Gruenberg at -0.007, indicating the community banking industry contracted over his tenure. Table 5 indicates that the high average value of CMOM under the tenure of Donald Powell was caused by the overall growth of the banking industry during his appointment. The CMOM Index does not take into account community bank growth relative to that of non-community banks. When assessing the relative growth of the community banking industry by using the CRGI index, the composite indexes paint a clear picture of relative community banking strength. According to our composite community banking indexes, the community banking industry showed the most robust growth during the appointment of Sheila Bair as FDIC Chairperson. Community banks were relatively strong, despite the onset of the financial crisis. It is important to analyze the strength of community banking under different regulatory leaders. Although we cannot directly attribute changes in community banking performance to the direct actions of the FDIC or its agents, this analysis provides important context to the recent performance of the community banking industry. 6.

Conclusion

In this study, we show the dramatic reduction in the number of FDIC-chartered community banks in the United States. We utilize FDIC data to compare the relative size, efficiency, and performance of community banks, relative to their non-community counterparts. We show that, despite their declining numbers, community banks have historically held some operational advantages over larger banks. Larger banks, however, have higher returns on assets and equity, and efficiency, which has been a key driver of bank consolidation. Community banks serve an important economic function. Therefore, the decline in the number of community banking operations in the United States is of particular concern. Community banks serve an important role in servicing the needs of rural customers and small businesses. In this study, we provide some preliminary analysis regarding the strength of the U.S. community banking industry. We develop several unique community bank indexes, the Community Bank Relative Growth Index (CRGI) and the Community Bank Momentum Index (CMOM), that measure the relative and nominal growth of the community banking industry. These indexes provide quick snapshots of the strength of the community banking industry that can be used by regulators, researchers, and practitioners. In addition, we provide an analysis of how the community banking landscape is addressed by and has changed under the tenures of recent FDIC Chairs that can provide valuable context when analyzing the industry.

The development of the community banking indexes represents a unique and valuable contribution to the analysis of the community banking industry. Going forward, aside from being updated on a quarterly basis, these indexes will be updated to capture the most relevant trends in the community banking industry with regards to its growth and resilience. In addition, we develop a data set that contains detailed balance sheet information from all banking institutions. In future research, similar analyses can be expanded to analyze community banking activities in more detail, including analysis of different types of loans and other core banking activities, a comparison and analysis of different types of interest and non-interest income, or an analysis of off-balance sheet banking activities. References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

805

Avery, R. B., & Samolyk, K. A. (2004). Bank consolidation and small business lending: the role of community banks. Journal of financial services research, 25(2-3), 291-325. Critchfield, T., Samolyk, K., Davison, L., Hanc, G., Gratton, H., & Davis, T. (2004). The Future of Banking in America-Community Banks: Their Recent Past, Current Performance, and Future Prospects. FDIC Banking Review Series, 16(3). DeYoung, R., & Duffy, D. J. (2002). The challenges facing community banks: In their own words. Economic Perspectives, 4th Quarter. DeYoung, R., Hunter, W. C., & Udell, G. F. (2004). The past, present, and probable future for community banks. Journal of Financial Services Research, 25(2-3), 85-133. DeYoung, R., Lang, W. W., & Nolle, D. L. (2007). How the Internet affects output and performance at community banks. Journal of Banking & Finance, 31(4), 1033-1060. Goddard, J., Liu, H., & Wilson, J. O. (2014). Entry, Exit and Growth of US Commercial Banks. Available at SSRN 2398888. Hakenes, H., Hasan, I., Molyneux, P., & Xie, R. (2014). Small Banks and Local Economic Development*. Review of Finance, rfu003. Jagtiani, J., Kotliar, I., & Maingi, R. Q. (2014). The evolution of us Community banks and its impact on small business lending. FRB of Philadelphia Working Paper No. 14-16 Mitts, J. (2014). Did the JOBS Act benefit community banks? A regression discontinuity study. A Regression Discontinuity Study (February 19, 2014). Peirce, H., Robinson, I. C. & Stratman, T. (2014). How Are Small Banks Faring under Dodd-Frank?. GMU Working Paper in Economics No. 14-49. Whalen, G. (2007). Community Bank Strategic Lending Choices and Performance. Available at SSRN 981181. Whalen, G. (2013). Investigating the Relationship between Supervisory Change and Community Bank Failure in a Competing Risk Framework. Available at SSRN 2225418. Wright, R. E. (2011). Governance and the Success of US Community Banks, 1790-2010: Mutual Savings Banks, Local Commercial Banks, and the Merchants (National) Bank of New Bedford, Massachusetts. Local Commercial Banks, and the Merchants (National) Bank of New Bedford, Massachusetts (May 7, 2011).


DETERMINANTS OF CAPITAL ADEQUACY RATIO: AN EMPIRICAL STUDY ON EGYPTIAN BANKS Osama A. El-Ansary*, Hassan M. Hafez** Abstract Capital adequacy rules are safety valve for regulators and banks' clients/shareholders to reduce expected risks faced by commercial banks especially for cross border transactions as these rules are applied compulsory by all banks internationally. Applying these rules will achieve rational management and governance. This paper examines explanatory victors that influence capital adequacy ratio (CAR) in the Egyptian commercial banks. The study covers 36 banks during the period from 2004-2013. We examined the relationship between CAR as dependent variable and the following independent variables: earning assets ratio, profitability, and liquidity, Loan loss provision as measure of credit risk, net interest margin growth, size, loans assets ratio and deposits assets ratio. Furthermore, we investigate determinants of CAR before and after the 2007- 2008 international financial crises. Results vary according to the period understudy. For the whole period 2003 to 2013 results show that liquidity, size and management quality are the most significant variables. Before the period 2008 results show that asset quality, size and profitability are the most significant variables. After the period 2009 results show that asset quality, size, liquidity, management quality and credit risk are the most significant variable that explain the variance of Egyptian banks' CAR. Keywords: Capital Adequacy Ratio (CAR); Commercial Banks; Risk Based Capital; Basel (I) & (II); Egypt; Financial Crisis JEL classification: F15 

Business Administration Department, Faculty of Commerce, Cairo University, Egypt ** Business School at American University of the Middle East in Kuwait

1.

Introduction

empirical analysis. Finally, section six will illustrate results and future research.

The international financial community is very keen to apply recent regulations related to capital adequacy to reduce risk exposure for cross border and local transactions. The Egyptian Central Bank is applying capital adequacy rules according to Basel accord (II & III). These rules helped to reform the banking system and assured banks' ability to manage assets and liabilities against perceived risks. This paper aims to determine factors that explain the CAR variance in the Egyptian commercial banks in order to identify decisions that increase or decrease the quality of capital management. Research is divided into six sections. The introduction will followed by section two, theoretical and empirical background. Section three will display hypotheses development. Section four will highlight research methodology. Section five will demonstrate CAR =

2.

Theoretical Background

and

The literature related to capital adequacy ratios and regulations are very comprehensive and divided into two mean streams of studies. Stream one includes research focused on assessing the validity of applying capital adequacy and Basel accord regulations on banks' decisions to manage risks [Bailey (2005); Bank of International Settlement BIS (2009); Rose and Hudgins (2008) and Federal Reserve Bank (2003)]. According to Basel (II) CAR is calculated using two main items: core capital and supplementary capital. Both should be added together and divided by risk weighted assets (RWA) and contingent liabilities.

𝐶𝑜𝑟𝑒 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 (𝑡𝑖𝑟𝑒 1) + 𝑆𝑢𝑝𝑝𝑙𝑒𝑚𝑎𝑛𝑡𝑎𝑟𝑦 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 (𝑡𝑖𝑟𝑒 2) × 100 𝑅𝑊𝐴

Core capital [Tier (I)]: Is the core measure of a bank's financial strength and includes paid in capital (common shares and preferred stock), disclosed

Empirical

(1)

capital reserves, net income for the year and innovative capital instruments. Aspal et al., (2014) p. 33.

806


Supplementary capital [Tier (II)]: Measures banks' financial strength with regard to the second most reliable forms of financial capital from the regulator's point of view. It includes assets revaluation reserves, undisclosed reserves, general provisions, general loan loss reserves, long-term holding of equity securities, hybrid capital instruments and subordinated long term debts. Aspal et al., (2014) p. 34. The sum of core capital and supplementary capital should not be less than 8% as a percentage of total assets plus off balance sheet risk weighted assets. The Egyptian Central Bank increased this ratio to 10%. In addition, the supplementary capital should not exceed 10% of the core capital. It worth mentioning that CAR includes three important types of risks included in the RWA as expressed by Federal Reserve Bank (2003) and Hasan (2003), credit risk, market risk and operational risk as these are considered as the three pillars for CAR. A quite number of researches were conducted to examine the impact of applying Basel I & II regulations on bank' decision behaviour. Some authors support the applications of these rules such as Joosen (2002) who studied the impact of applying Basel (II) Accord on Netherland and European banks. The study stressed on the positive side of applying these rules to protect depositors and to avoid banks' solvency risk. Also, Girardone et al., (2004) examined the impact of capital adequacy rules on the economics of banks and Estrella et al., (2000) on predicating banks' failure. Samara and Nikaido (2007) analyzed expected challenges and issues which faced Indian banks when applying Basel (II) rules. One of the most important results of this study was that applying Basel (II) helped in enhancing CAR, on average, up to 12% compared to the international percentage (8%) and the Indian Federal Reserve Bank ratio (9%). On the other hand, Karacadug and Taylor (2000) focused on identifying restrictions imposed on banks by applying CAR. Meanwhile, Chami and Cosimano (2003) found that applying these rules may increase banks' cost of funds, decrease their profitability, and limit banks' ability of lending. The second stream of research was directed to develop models and to examine variables and factors that influence CAR in commercial banks. Also, this stream considered the link between the Central Bank's supervisory rules and commercial banks decision behaviour. We will focus on this stream on our paper. A group of academics Jackson et al. (1999) conducted a study through Basel committee which is affiliated with Bank for International Settlements (BIS) to test the impact of applying Basel regulations on the fixed capital percentage (8%) as minimum requirements by banks. They examined the impact of Basel rules on limiting banks' competitiveness or ability to provide credit. Results were not decisive to judge the impact of Basel (II) committee decisions on banks' ability to grand loans. Bertraned (2000)

focused on testing Swiss banks reaction to Basel rules and restrictions. He developed a model to analyze and adjust capital and financial policies in order to keep the minimum required capital. Results showed that regulations pressure encourage Swiss banks to increase capital and have no impact on the level of risks for banks' polices. The study examined the following factors: - Bank's total size: measured by total assets. - Bank's profitability: measured by return on assets. - Credit risk: measured by loans portfolio lose rate. - CAR: measured by the percentage of capital to risk weights assets, at least 8%. The study found that there is a positive statistical significant relationship between ROA and CAR. Swiss banks increased their capital to avoid any plenty for violating rules by the Central Bank and the impact of applying the CAR rules on Swiss banks were less compared to English and American banks. Another study conducted by Bouri and Ban Hamida (2006) to examine the impact of CAR and Basel (II) rules on Tunisian banks. Findings were very positive as banks were very keen to apply capital adequacy rules and to manage risks that threaten banks' decisions. Authors used a number of factors that affect CAR and tested by previous studies: bank's size, credit risk, equity/total loans, Loan Loss Provisions (LLP)/total deposits and bank's risks measurements. However, Rojas-Suarez (2002) examined the impact of applying of CAR and rules in six emerging markets using the following variables: - Change of banks' share market value. - Net income to total operating profits ratio. - Operating cost to total assets ratio. - Liquidity ratios. - Interest rate for bank's deposits. One of the major criticisms raised by the study was that in emerging markets owners/shareholders, most likely, can raise capital by issuing financial instruments with low quality and high degree of risk, relative to primary capital, to achieve the required adequacy in short time. The study recommended number of preventive measures to support Basel accord rules such as: to apply deposit insurance scheme, to limit banks' ability from having liquidity from Central banks, encourage banks to issue long term deposit certificates and the disclosure of information that affects banks' quality. Also, Diamond and Rajan (2000) stressed the importance of applying deposit insurance system to support capital adequacy control rules. Estrella et al., (2002) examined the relationship between CAR and the probability of banks' losses. They recommend simplifying the required standards to measure CAR in two ratios only: (a)leverage ratio measured by primary capital/total assets and (b)capital to operating revenue ratio=primary capital/interest & commissions received. Results

807


showed the possibility of using those two ratios to predict banks' loss with the same quality by using CAR rules. In Indonesia, Yudistira (2003) focused on examining the impact of CAR requirements on banks decision behaviour and proved a significant decrease in loans granted by banks to achieve Basel (II) requirements. In Spain, Barrios and Blanco (2003) focused to answer the following question: Is bank's capital adequacy affected by market conditions or by capital adequacy rules imposed by regulatory agencies? They developed two models one for banks that did not influenced by capital adequacy rules and one related to banks which influenced by these rules. A number of variables that affected banks' capital have been used such as: size, liquidity, operating cost variance, and return on assets, liquidity and credit risk. The study major contribution was that capital adequacy requirements are better influenced by market factors rather than capital adequacy rules imposed by regulatory authorities. As for the relationship between bank's capital and risk, Altunbas et al., (2007) examined this relationship in European banks. They found a positive relationship between risk and bank's efficiency. This result was explained as European banks are holding a large size of capital, exceed the minimum required, and in the same time invest their funds in less risky assets. The analysis showed a positive relationship between the degree of risk and the required level of capital and liquidity ratios. This result could interpret the reason for the attention of the regulators to increase bank's capital and liquidity to limit bank's ability to allocate funds in investments with high risks. In addition, it proved that the financial strengths for corporations which borrow loans from the banks, have a positive impact in reducing risks faced by banks, probability of default and a positive impact towards increasing CAR levels. The following variables used to develop a statistical model to test the relationship between bank's risks and capital adequacy. - Doubtful loans provisions. - Equity to total assets ratio. - Net Loans/ Total assets ratio. - Size: Log total assets. - Return on assets. - Liquidity ratio. A number of other studies focused on identifying determinants of CAR in Germany such as Kleff and Weber (2004); Stolz and Wedow (2005). Authors found a significant positive relationship between risky assets and the change of CAR especially for banks with high capital adequacy. Meanwhile, there was a negative correlation with banks' having law adequacy. In addition, there is a positive correlation between bank's profitability and CAR, a significant positive correlation between clients' deposits and CAR and a negative correlation between banks' size and CAR.

Evidence from Portuguese banks was introduced by Boucinha and Ribeiro (2007) found that larger banks hold less excess capital than small banks and banks with higher risk lend to hold higher capital reserves. Also, they found that the more banks were exposed to securities portfolio in capital markets the more they increase their capital to achieve the required adequacy. In addition, banks with high risk assets are holding high reserves in their capital components. The study confirmed that banks with large size keep law capital as there is a negative relationship with CAR as they can generate funds at law cost and less risk through external sources of finance, deposits, and in the time they preserve required CAR. If banks keep high ratio of primary capital relative to equity this contributes in reducing the need to increase capital reserves to maintain CAR at the required level. In an emerging market setting, recent study conducted by Bateni et al., (2014) examined the relationship between seven financial factors and CAR in Iranian private banks during the period 2006-2012. The study showed a negative relationship between size and CAR, a positive relationship with: loans assets ratio LAR, ROE, ROA and CAR. In addition, deposits assets ratio DAR and risk assets ratio were not having impact on CAR. In Nigeria, Olalekan and Adeyinko (2013) studied the impact of CAR on profitability. The results revealed that there is a significant positive relationship between capital adequacy and banks' profitability. Also Ogere et al., (2013) examined CAR variance in Nigeria money deposits banks. They found that changes in CAR were explained by a negative relationship between risk ratios, deposits to total assets ratio and CAR. There are many internal financial indicators have an influence on banks' capital adequacy rules introduced by central banks particularly with reference to size, earning assets, liquidity, loan loss provision loan loss provision LLP, ROA, net interest margin growth NIM. Our study will add evidence to the effect of Egyptian commercial banks risky financial decisions on CAR. In India, Aspal et al., (2014) reported that CAR is negatively related to loan assets ratio LAR, assets quality and management efficiency. Moreover, liquidity and sensitivity were positively related to CAR. The study showed that the Indian private sectors banks maintain a higher level of capital requirements than required by Reserve Bank of India. Recent study by Shingjergji and Hyseni (2015) showed that ROA and ROE are not correlated with CAR, meanwhile, size, nonperforming loans NPL, loans to deposits ratio LTD and equity multiplier EM have a negative and significant impact on CAR. An empirical study by Ozili (2015) investigated the relationship between bank profitability and Basel capital regulations. He found that bank capital adequacy is observed to be a significant determinant of bank profitability.

808


Polat and Al-Kalaf (2014) studied the determinants of CAR in the banking system of Kingdom of Saudi Arabia during the period from 2008 to 2012. They found that all the independent variable were significant with CAR except nonperforming loans. Loans to assets ratio was negatively significant but size and leverage were positively significant. Loans to deposits ratio was negatively significant and ROA was positively significant with CAR. A few numbers of researches were directed to study determinants of CAR in Islamic banks. Abusharaba, et al., (2013) examined the CAR in Indonesia and used the same independent variables used by other researchers. Results showed that ROA and Liquidity were positively related to CAR. Nonperforming loans was significant and negatively correlated to CAR. Indonesian Islamic banks are having excessive liquidity to meet their liabilities and to protect shareholders. On the other hand, Abdul Karim et al., (2013) conducted a research to examine capital adequacy, lending and deposit behaviours of conventional and Islamic banks. The study findings reported that there was a positive relationship between CAR and deposits and loans growth in both Islamic and conventional banks. According to Romdhane et al., (2012) the main determinants of CAR in developed countries are the same for developed ones. They analyzed determinants of CAR in Tunisia. The study found that the interest margin and risk affect strongly the capital ratio. On the other hand, Atici and Gursoy (2013) applied the same determinants of CAR on the Turkish banking system. Findings provided evidence that Turkish banks employ the capital buffering approach proposed under Basel (III) as an effective management tool. Capital buffer is mainly related to nonperforming loans, loans growth, loans to assets ratio and profitability. Also, Asarkaya and Ozcan (2007) analyzed determinants of CAR in Turkish banks. Findings pointed out that lagged capital, portfolio risk, economic growth, ROE, average capital level were positively correlated with CAR. On the other hand, deposits to assets ratio was negatively correlated with CAR. Al-Tamimi and Obeidat (2013) studied the important factors that determinate the CAR in Jordan listed commercial banks. Results showed a significant positive correlation between CAR and Liquidity risk, RAO and a negative, but not significant, relationship with credit risk. A study conducted by Rahari (2014) found that Indonesian state-owned banks CAR is affected by total assets growth, equity to total assets ratio, nonperforming loans, interest rate risk and operational cost and revenue ratios.

3. Hypothesis Development Prior literature shows that there are significant correlations between the independent variables and CAR. We preliminarily examine the significance and direction of the correlation between the independent variables and CAR in Egyptian commercial banks. Accordingly, the first hypothesis can be stated as: H1: “There is a significant correlation between CAR as a dependent variable and the study's independent variables”. As the expected impact for each individual independent variable on CAR will vary, bivariate correlation, all these variables jointly may have the same impact in explaining the capital adequacy variance. Thus, the second hypothesis can be stated as: H2: “All the independent variables, jointly, have equal relative impact on the banks' CAR”. The global banking system has been affected significantly by the financial crisis that hit banks in the late 2007. Kosak, M. et al., (2015) assessed the performance of banks during the financial crisis. They found that the existence of high quality funding strategy, tier 1 bank capital and retail deposits were very important for continuous bank lending during the financial crisis. The study suggested that in crisis period the high quality bank capital is a competitive strength. . Moreover, Nilsson, et al., (2014) examines the Swedish bank capital adequacy before and after 2007 financial crisis. They found that banks have been forced to made noticeable changes to their capital structure to achieve financial stability. Swedish banks set capital ratios above the regulations after the financial crisis. Berger and Bouwman (2013) examined the impact of banks' capital on performance across banking crises. They found that capital help small banks to increase their market share and profitability. In addition, capital enhances the performance of large banks during banking crises. Consequently, it is important to examine the determinants of CAR variance of Egyptian commercial banks before the financial boom and after the financial crisis period. We aim to evaluate the significant difference of CAR determinants before and after the financial crisis. Thus, the third hypothesis can be stated as: H3: There is no statistical significant difference between determinants of CAR before and after the financial crisis; year 2008. 4. Methodology This section will shed the light on the research methodology in addition; we will develop the research model that will be tested in the following section.

809

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 4.1 The Research Sample

Liquidity, asset management quality, loans to assets ratio, earning quality, credit risk and size. The following equation addresses the relationship between the independent variables and CAR.

We used a financial data gathered from annual financial statements through Bank-Scope database. We have a population of 40 commercial operating banks in Egypt. We conducted our empirical analysis on 33 banks represent 83% of operating commercial banks in Egypt during the interim period from 2003 till 2013. The sample includes all commercial banks operating in the Egypt and divided into three major groups: Group (A): Local Egyptian Banks count for 23 bank, Group (B): International Bank count for 6 banks. Group (C): Islamic Banks locally or regionally count for 4 banks.

CAR it = α + β1 (Earning assets/T. Assets it) + β2 (Securities/Assets it) + β3 (Loans/Assets it) + β4 (Loans Loss Reserves/Assets it)+ β5 (Provisions/Loans it)+ β6 (Loans/Deposits i, t) + β7 (ROA it) + β8 (ROE it) + β9 (Δ NII it) + β10 (lOG_Assets it) Ɛi+ i=1,…8 t=1,..9 Where, α is a constant, (β1: β10) are the parameters for the explanatory variables. The subscript (i) refers to the bank number and the subscript (t) denotes the time period. (Ɛi) is the unobservable individual heterogeneity, and vit is the remainder disturbance of the usual disturbance in the regression model that varies with individual units and time. Variables are used in the analysis are summarised in table (1) as follows:

4.2 Statistical Model The aim of this study is to investigate the influential factors of Egyptian commercial banks' capital adequacy ratio. We selected the independent variables that could explain the CAR variance according to literature review findings. Capital adequacy ratio of commercial banks operating in Egypt is a function of

Table 1. Variables Definitions Dependent Variable Capital Adequacy Ratio (CAR) Independent Variables Assets Management Quality

CAR =

𝑃𝑟𝑖𝑚𝑎𝑟𝑦 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 + 𝑆𝑢𝑏𝑙𝑚𝑖𝑛𝑡𝑎𝑟𝑦 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 × 100 𝑅𝑖𝑠𝑘 𝑊𝑒𝑖𝑔ℎ𝑡𝑒𝑑 𝐴𝑠𝑠𝑒𝑡𝑠 (Tier 1 + Tier 2) / Risky weighted assets

𝐸𝑎𝑟𝑛𝑖𝑛𝑔 𝑎𝑠𝑠𝑒𝑡𝑠 × 100 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 Earning assets= Balances with central bank and other banks + TB's + treasury bonds + securities portfolio +Loans portfolio 𝑁𝑒𝑡 𝐶𝑙𝑖𝑒𝑛𝑡 𝐿𝑜𝑎𝑛𝑠 𝐿1 = × 100 ′ 𝐶𝑙𝑖𝑒𝑛𝑡𝑠 𝐷𝑒𝑝𝑜𝑠𝑖𝑡𝑠 & 𝑆ℎ𝑜𝑟𝑡 𝑡𝑒𝑟𝑚 𝑓𝑢𝑛𝑑𝑠 AMQ =

Liquidity

𝑆𝑒𝑐𝑢𝑟𝑖𝑡𝑖𝑒𝑠 × 100 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 𝐿𝑜𝑎𝑛 𝑃𝑟𝑜𝑣𝑖𝑠𝑖𝑜𝑛 = × 100 𝑇𝑜𝑡𝑎𝑙 𝐿𝑜𝑎𝑛𝑠

𝐿2 =

Credit Risk (CR)

𝐿𝑜𝑎𝑛 𝐿𝑜𝑠𝑠 𝑅𝑒𝑠𝑒𝑟𝑣𝑒𝑠 × 100 𝑇𝑜𝑡𝑎𝑙 𝐿𝑜𝑎𝑛𝑠 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑎𝑓𝑡𝑒𝑟 𝑇𝑎𝑥 ROA = × 100 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 =

Profitability

ROE = Size (Total assets) Net Interest Income Growth Management Quality (MQ)

𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑎𝑓𝑡𝑒𝑟 𝑇𝑎𝑥 × 100 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 Log total assets

NIIG = Change (Interest Received – Interest Expenses) LAR =

810

𝑇𝑜𝑡𝑎𝑙 𝐿𝑜𝑎𝑛𝑠 × 100 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 5.

Empirical Analysis

5.1 Descriptive Statistics Table 2. Variables descriptive statistics Minimum Dependent Variable Capital Adequacy .0000 Independent Variables Earning assets/Total .0493 Assets Securities/ Assets .0029 Loans/ Assets .0181 Loans Loss .0059 Reserves/Assets Provisions/Loans .0083 Loans/Deposits .0203 Return on Assets .0000 Return on Equity .0000 Δ Net Interest .0000 Income LOG Assets 3.2759 Observations 363

Maximum

Mean

Median

Skewness

kurtosis

SD

JarqueBera

.6062

.16982

0.1433

1.3601

2.656

.10323

214,381

1.5130

.8432

0.8772

-1.5465

7.295

.1489

924.05

2.1107 1.2310

.2690 .4672

0.2472 0.4450

-1.5533.192

28.378 1.618

.1813 .1940

12443.03 81.91329

3.1111

.1605

0.0989

.858

68.207

.2400

71110.01

22.6111 17.2667 .0545 1.3942

.9256 .8872 .0110 .1238

0.610 0.5118 .0089 0.0963

6.707 8.690 5.858 -.987-

100.792 37.358 3.192 11.901

1.6158 1.8101 .01034 .1374

153949 22574.77 206.817 2333.773

26.2496

.41117

0.1480

-2.064-

148.395

1.5759

325085

5.5598 363

4.1880 363

4.171 363

9.439 363

.007 363

.4768 363

10.68 363

Table (2) illustrates the descriptive statistics of the study variables. The observed calculated descriptive statistics consists of minimum, maximum, standard deviation, skewness and kurtosis. As seen from the tables above all the variables are asymmetrical. Especially skewness is positive for Loans to assets, loans loss reserves to total assets, provisions to loans, loans to deposits, return on assets and log assets. While earning to total assets, securities

to assets, return on equity and change in net interest income have a negative skewness. Kurtosis value of all variables also indicates data is not normally distributed because values of kurtosis are deviated from 3. The measure of Jarque-Bera statistics and corresponding p-values are used to test for the normality assumption. Based on the JarqueBera statistics and p-value this assumption is rejected at 5% level of significance for variables.

Table 3. Pearson's Correlations Matrix between Capital Adequacy Ratio and Independent Variables CAR CAR EAR/TA Secc/TA Loans/TA Res/TA Pro/Loans Loa/Depo ROA ROE Δ NII Size

EAR /TA

1 .191** 1 -.063.127* .258** .475** -.067- -.308-** -.153-** -.377-** .363** .141** .257** .168** -.194-** -.100-.034- -.049-.489-** -.130-*

Secu./ Assets

Loans/ Assets

Res. / Assets

Prov/ Loans

Loans / Deposits

1 -.325-** -.003-.038-.286-** -.010-.027-.011.218**

1 -.224-** -.209-** .599** .147** -.127-* -.036-.251-**

1 .832** -.083-.232-** -.224-** .064 -.206-**

1 -.112-* -.191-** -.145-** .074 -.125-*

1 .357** -.055-.020-.188-**

ROA

ROE

Δ NII Size

1 .579** -.075-.021-

1 -.059.241**

1 .067

1

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

From the table above it is obvious that most of the independent variables have either a significant positive or negative relationship with the capital adequacy ratio except securities to total assets, reserves to total assets variable and the change in net interest income.

Earning to total assets, loans to total assets, loans to deposits and return on assets have a significant positive relationship with the capital adequacy at significant level .01 Provisions to total loans, return on equity and log assets variable have a significant negative

811


relationship with the capital adequacy ratio at a significant level 0.01. Securities to total assets, reserves to total assets and the change in net interest income have a negative relationship with the capital adequacy ratio but not significant. Therefore, according to the correlation analysis results we can partially accept the first hypothesis as most of the independent variables are significantly correlated with the leverage ratios except securities to total assets, reserves to total assets and the change in net interest income. Accordingly we will accept the hypothesis sating that: “There is a significant correlation between CAR as a dependent variable and the study's independent variables” Testing the second hypothesis H2: “All the independent variables, jointly, have equal relative impact on the banks' CAR”. We conducted a panel regression model to explore determinates of the capital adequacy ratio. Panel data involves the pooling of observations on a

cross-section of variables over several time periods (2003 till 2013). This approach is useful as the several data points, degree of freedom is increased and collinearity among the explanatory variables is reduced, thus improving the quality of results, Abor (2008; p.13). We used the statistical package for social science version (22) along with Eviews software version (8) to carry out the analysis. Table (4) reports the multiple regression analysis statistical results and the model's goodness of fit statistics. Three models are developed to test the second and third hypothesis. To test the second hypothesis we conducted the regression analysis for the whole period 2003 to 2013. Some variables were not normally distributed. Several trails of data transfer were performed, as explained by Field (2005; p.72), depending on the level of Skewness and Kurtosis using log, inverse or square root techniques in order to achieve normality. All the variables Kurtosis below the upper threshold of 3.29 will be accepted with large observations and samples. 5.2 Analysis Output

Table 4. Multiple regression analysis results (2003 to 2013) Independent Variables

Beta

t

Sig.

Constant Earning Assets / Total Assets Securities/ Total Assets Net Loans / Total Assets Loans Loss Reserves /Total Loans Provisions/ Total Loans Loans / Deposits Return on Assets Return on Equity Change in net interest income Log Assets (Asset Size) Goodness of Fit Statistics R2 Adjusted R2 F-equation Prob (F-statistics)

0.572 .002 .083 -.043.057 -.213.250 .201 -.114-.007-.454-

10.010 .037 1.679 -.638.687 -2.6184.076 2.601 -1.580-.173-9.499-

***.000 .970 .094* .524 .493 .009*** .000*** .010 .115 .863 .000***

Collinearity Statistics Tolerance VIF .552 .754 .394 .269 .276 .485 .306 .352 .985 .799

1.813 1.326 2.538 3.716 3.625 2.062 3.265 2.843 1.016 1.251

0.357 0.339 19.581 0.000****

*** Significant at 0.01, ** significant at 0.05, * significant at 0.10

As for multi collineariaty the variables coefficient did not face this problem as the variance inflation factor (VIF) was less than 5 for all variables and you can detect multi collineariaty if the largest VIF is greater than 10, then there is cause for concern. Hair, et al., (2006) highlighted that a maximum acceptable VIF value would be 10, anything higher would indicate a problem with multi collineariaty. The multiple regression analysis reports the following results:  Earning Assets to Total Assets ratio: is not significantly correlated with the capital adequacy



812

ratio, this means the portfolio mix of commercial Egyptian banks allocated between commercial loans, retail or even securities either traded or not traded is not affecting the capital adequacy ratio of Egyptian banks. We can interpret this conclusion to the fact that capital adequacy requirements for the Egyptian banks are better influenced by capital adequacy rules imposed by international regulatory authorities and Local regulatory authority rather than market factors. Securities to Total Assets is significantly correlated positively with the capital adequacy








ratio and this result due to the fact that a considerable portion of the Egyptian banks portfolios are invested in the stock market and since the stock market in Egypt is one of the emerging market characterized by the high volatility so we can expect that Egyptian banks are cautious and reflecting this in calculating the capital adequacy ratio and also previous research reported positive correlation between securities to total assets as a proxy to the measurement of liquidity and capital adequacy ratio. Net loans to Total Assets is not significantly correlated with the capital adequacy ratio. The net loans to total assets is a proxy to see if the banks is well operated and has a considerable market share in the credit market compared to other banks and has no impact on the capital adequacy ratio. Egyptian bank are concern to meet the capital adequacy ratio as per the Central Bank of Egypt regulation. And this justify why the relation between net loans to total assets ratio is not significant. A loans loss reserve to total assets is not significantly correlated to the capital adequacy ratio. Egyptian banks building reserves as a part of equity part in order to meet the nonperforming loans only and are not considering reserves for loans losses when they are justifying the capital adequacy ratio according to the regulations of Central Bank of Egypt. Provisions to Loans is significantly correlated positively with the capital adequacy ratio as long as Egyptian banks increasing provision to meet the unexpected percentage in nonperforming loans they have to consider provisions when calculating the capital adequacy ratio since the main target is of capital requirement is a cushion against unexpected loss. Of course this explains

why provision to loans has a significant relation with the capital adequacy ratio.  Loans to deposits ratio is significantly correlated positively with the capital adequacy ratio and this is supported by previous research as well (altunbas et al., 2007, Bateni et al., 2014) As long as Egyptian banks increase this ratio which into turn should be reflected in the increase of the capital adequacy ratio in order to secure banks against unexpected loss.  Return on assets is significantly correlated positively with the capital adequacy ratio. As the return on assets for Egyptian banks increase due to the increase in the portfolio of loans and assets banks have to increase the capital adequacy ratio to match the associated risk. Increase of return on assets is mainly due to the increase in credit portfolio.  Return on equity: the change in net interest income is not significantly but correlated negatively with the capital adequacy ratio.  Log assets as a proxy to measure the size of the bank is significantly correlated negatively with the capital adequacy ratio as long as the as large banks keep capital at low level as they have long term deposits to finance risky assets. Therefore, according to the correlation regression analysis results we can reject the second hypothesis as not all of the independent variables jointly have equal relative impact on the banks CAR. Accordingly we will reject the second hypothesis. Testing the third hypothesis: "There is no statistical significant difference between determinants of CAR before and after the financial crisis; year 2008". We conducted the multiple regression analysis for the period 2003 to 2007 and then for the period 2009 to 2013. We report the following results.

Table 5. Multiple regression analysis results (before and after financial crisis) Independent Variables Beta 0.584 -.204.094 .111 .206 -.392.141 .502 -.383.018 -.375-

Before Year 2008 t Sig. 6.673 .000*** -2.073.040** 1.249 .214 1.139 .256 1.458 .147 -2.908.004*** 1.585 .115 4.078 .000*** -3.276.001*** .302 .763 -5.388.000***

Constant Earning Assets / Total Assets Securities/ Total Assets Net Loans / Total Assets Loans Loss Reserves /Total Loans Provisions/ Total Loans Loans / Deposits Return on Assets Return on Equity Change in net interest income Log Assets (Asset Size) Goodness of Fit Statistics 0.482 R2 0.448 Adjusted R2 14.316 F-equation 0.000 Prob (F-statistics) *** Significant at 0.01, ** significant at 0.05, * significant at 0.10

813

Beta 0.577 .217 -.043-.292-.243.050 .233 -.070.172 -.040-.593-

After Year 2008 t Sig. 6.767 .000*** 2.924 .004*** -.507.613 -2.480.014** -2.908.004*** .604 .547 2.361 .019** -.640.523 1.637 .104 -.620.536 -8.577.000*** 0.386 0.346 9.661 0.000


The multiple regression analysis reports the following before and after financial crisis.  Earning Assets to Total Assets: the regression equation shows that it is significantly correlated negatively with the capital adequacy ratio before the crisis and is significantly correlated positively with the capital adequacy ratio after the crisis.  Securities to Total Assets: is not significantly correlated positively with the capital adequacy ratio before the crisis and is not significantly correlated negatively with the capital adequacy ratio after the crisis.  Net loans to Total Assets: is not significantly correlated positively with the capital adequacy ratio before and after the crisis.  A Loans loss reserves to Total assets: is not significantly correlated positively to the capital adequacy ratio before the crisis. but is significantly correlated negatively to the capital adequacy ratio after the crisis  Provisions to Loans: is a significantly correlated negatively with the capital adequacy ratio before the crisis and is not significantly correlated positively with the capital adequacy ratio after the crisis.  Loans to deposits: has is not significantly correlated positively with the capital adequacy ratio before and after the crisis.



Return on assets: is significantly correlated positively with the capital adequacy ratio before the crisis and is not significantly correlated positively with the capital adequacy ratio after the crisis. Return on equity: is significantly correlated negatively with the capital adequacy ratio before the crisis and is not significantly correlated negatively with the capital adequacy ratio after the crisis.  The change in net interest income: is not significantly correlated positively with the capital adequacy ratio before and after the crisis.  Log assets as a proxy to measure the size of the bank: is significantly correlated negatively with the capital adequacy ratio before and after the crisis. According the to the multiple regression analysis output we can reject the third hypothesis stating that "There is no statistical significant difference between determinants of CAR before and after the financial crisis; year 2008". Since the number of independent variable affecting the capital adequacy ratio before year 2008 -the year of financial crisis- are different than the number of independent variables affecting the capital adequacy ratio after year 2008. The following table reports only the main independent variables affecting the capital adequacy ratio of Egyptian banks for the three models as follows.

Table 9. Results summary Independent Variables Constant Earning Assets/ Total Assets Securities/ Total Assets Provisions/ Total Loans Loans / Deposits Return on Assets Log Assets (Asset Size) Net Loans to total Assets Loans Loss Reserves / Total Loans Return on Equity Goodness of Fit Statistics R2 Adjusted R2 F-equation Prob (F-statistics)

2003 to 2013 Beta Sig. 0.572 ***.000 Non Non .083 *.094 -.213***.009 .250 ***.000 .201 ***.010 -.454***.000 Non Non Non Non Non Non

Before 2008 Beta Sig. 0.584 .000 -.204.040 Non Non -.3920.004 Non Non .502 0.000 -.375***.000 Non Non Non Non -.3830.001

0.357 0.339 19.581 ****0.000

0.482 0.448 14.316 0.000

After 2008 Beta Sig. 0.577 .000 .217 .004 Non Non Non Non .233 0.019 Non Non -.593***.000 -.2920.014 -.2430.004 Non Non 0.386 0.346 9.661 0.000

*** Significant at 0.01, ** significant at 0.05, * significant at 0.10

6.

Conclusions

The aim of this paper was to investigate the determinants of capital adequacy ratio on commercial banks operate in Egypt such as: Asset quality, size, liquidity, profitability, and risk and management quality. To test such relation we used multiple

regression analysis. The results showed the following conclusions:  During the interim period from 2003 to 2013: profitability has no impact the capital adequacy ratio except return on assets is significantly correlated positively on the capital adequacy ratio. Assets quality represented in earning assets

814


to total assets measure is not correlated to the capital adequacy ratio. Liquidity represented in loans to deposits and securities to total assets is significantly correlated positively with the capital adequacy ratio. Management quality represented in total loans to total assets measure is significantly correlated positively with the capital adequacy ratio. Size of the bank represented in the log of total assets is significantly correlated negatively on the capital adequacy ratio. Risk represented in provisions and loans loss reserves to total loans ratios showed that provision to total loans is significantly correlated negatively to the capital adequacy ratio and loans loss reserves has no relation to the capital adequacy ratio.  Before year 2008 (which is the year of financial crisis). Size of the bank and Risk showed the same results as for the whole period under analysis from 2003 to 2013. Assets quality showed different results since it is significantly correlated negatively with the capital adequacy ratio. Management quality results showed no relation with the capital adequacy ratio. Liquidity has no impact on the capital adequacy ratio. In terms of profitability angle return on assets is significantly correlated positively with the capital adequacy ratio and return on equity is significantly correlated negatively with the capital adequacy ratio and the change in net interest income has no impact.  After year 2008 to 2013. Profitability showed no impact on the capital adequacy ratio. Liquidity represented only in loans to deposits is significantly correlated positively to the capital adequacy ratio. Asset quality is significantly correlated positively to the capital adequacy ratio. Size of the bank is significantly correlated negatively to the capital adequacy ratio. Risk represented only in loans loss reserves ratio is significantly correlated with the capital adequacy ratio. Management Quality represented in total loans to total assets is significantly correlated positively to the capital adequacy ratio. We can attribute these results to the fact that after the financial crisis and the failure of many banks worldwide Egyptian banks started to look carefully to loans portfolios and be more strict in providing loans to customer not only that but the portfolio mix for Egyptian banks of industrial loans, retail and securities have been change dramatically to see that retail loans and securities in some banks represent the high portions of Egyptian banks portfolios compared to the portfolio mix in the past years not only that but Egyptian banks started to be more cautious in building reserves and this justify the fact to see Loans Loss Reserves to total loans is significantly correlated negatively to the capital adequacy ratio. Also, earning assets to total assets appears to have an impact with the capital adequacy ratio. So after financial crisis

Egyptian banks are more concerned to the loans quality, credit risk. There is a high need to apply the same study on different banks classification operates in Egypt for example on International commercial banks, Islamic Banks and Egyptian local commercial banks. Also we need to consider other variables in the analysis and to consider the market and operation risk when measuring the capital adequacy ratio. References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

815

Abdul Karim, M., Hassan, M. K. and Hassan, T. (2013). Capital adequacy and lending and deposit behaviours of conventional and Islamic banks, PacificBasin Finance Journal, 28, 58-75. Abor, J. (2008), Determinants of Capital Structure of Ghanaian Firms, African Economic Research Consortium, AERC research paper 176. Abusharaba, M. T., Triyuwono, I., Ismail, M. and Rahman, A. F. (2013). Determinants of capital adequacy ratio (CAR) in Indonesian Islamic commercial banks, Global Review of Accounting and Finance, 4 (1), 139-170. Al-Tamimi, K. A. M. and Obeidat, S. F. (2013). Determinants of capital adequacy in commercial banks of Jordan an empirical study, International Journal of Academic Research in Economics and Management Science, 2 (4), 44-57. Altunbas,Y., Carbo, S., Gardener, E. P. M. and Molyneux, P. (2007). Examining the relationships between capital, risk and efficiency in European banking, European Financial Management, 13 (1), 49– 70 Asarkaya, Y., and Ozcan, S. (2007). Determinants of capital structure in financial institutions: the case of Turkey, Journal of BRSA and Financial Markets, 1 (1), 91-109. Aspal, P. K. and Nazneem, A. (2014). An empirical analysis of capital adequacy in the Indian private sector banks, American Journal of Research Communication, 2 (11), 28-42. Atici, G. and Gursoy, G. (2013). The determinants of capital buffer in the Turkish banking system, International Business research, 6 (1), 224-234. Bailey, R. (2005). Basel II and development countries: understanding the implications, London School of Economics, Working Paper, no. 05-71, 1-27. Bank of International Settlement (BIS) (2009) History of the Basel committee and its membership, Basel Committee on Banking Supervision. Available at http://www.bis.org. https://www.bis.org/publ/bcbsc101.pdf. [Accessed 26 November 2014]. Barrios, V. and Blanco, J. M. (2003). The effectiveness of capital adequacy requirements: a theoretical and empirical approach, Journal of Banking and Finance, 27, 1935-1958. Bateni, L., Vakiliford, H. and Asghari, F. (2014). The influential factors on capital adequacy ratio in Iranian banks, International Journal of Economics and Finance, 6 (11), 108-116. Berger, A. N. and Bouwman, H. S. (2013). How does capital affect bank performance during financial crises?, Journal of Financial Economics, 109 (1), 146176.


14. Bertrand, R. (2000). Capital requirements and bank behavior: empirical evidence for Switzerland, Swiss National Bank, Banking Studies Section, 1-23. 15. Boucinha, M. and Riberiro, N. (2007). The determinants of Portuguese banks' capital buffers, Financial Stability Reports, Banco Portugal, 153-164. 16. Bouri, A. and Ban Hamida, A. (2006). Capital and risk taking of banks under regulation: a simultaneous equations approach in the Tunisian context, Proposition for the sixth international congress of AFFI: Corporate finance and market finance, what complementarities?. 17. Chami, R. and Cosimano, T. F. (2003). The nature of capital adequacy constraints under the Basel Accord, September. Available at SSRN: http://ssrn.com/abstract=456981 or DOI: 10.2139/ssrn.456981. [Accessed 12 December 2014]. 18. Diamond, D. W. and Rajan, G. R. (2000). A theory of bank capital, Journal of Finance, 55 (6), 2431-2465. 19. Estrella, A. , Park, S. , and Peristiani, S. (2000). Capital ratios as predictors of bank failure, FRBNY Economic Policy Review, 6 (2), 33-52. 20. Federal Reserve Bank (2003). Capital standards for banks: the evolving of Basel Accord, Federal Reserve Bulletin, September, 395-405. 21. Field, A. (2005), Discovering Statistics Using SPSS, 2nd edition, London, SAGE Publications. 22. Girardone, C., Molyneux, P., and Gardener, E. P. M. (2004). Analyzing the determinants of bank efficiency: the case of Italian banks, Applied Economics, 36 (3), 215-227. 23. Grier, W. A. (2012). Credit analysis of financial institutions, 3rd edition, London, Euromoney Books. 24. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. and Tatham, R. L. (2006). Multivariate Data Analysis, 6th edition, Person Prentice Hall. 25. Hasan, M. (2003). The significance of Basel 1 and Basel 2 for the future of the banking industry with special emphasis on credit information, a paper presented at the Credit Information Alliance Regional Meeting, Amman, 3-4 April. 26. Jackson, P. D. , Fufine, C., Groeneveld, H., Hancock, D., Jones, D., Perraudin, W., Radeci, L. and Yoneyama, M. (1999). Capital requirements and bank behavior: the impact of the Basle Accord, Bank of International Settlements, Basel Committee on Banking Supervision, Working Paper, no. 1, April. 27. Joosen, B. P. M. (2002), Solvency and capital adequacy regulations for banks in the Netherlands, International Financial Law Review, 21 (4), 17-20. 28. Karacadug, C. and Taylor, M. W. (2000). The new capital adequacy framework: institutional constraints and incentive structures, International Monetary Fund, Working Paper, no.39. 29. Kleff, V. and Weber, M. (2008), How do banks determine capital? Evidence from Germany, German Economic Review, 9 (3), 354-372. 30. Kosak, M., Li, S., Loncarski, I. and Marinc, M. (2015). Quality of bank capital and bank lending behavior

31.

32.

33.

34.

35.

36.

37.

38.

39. 40.

41.

42.

43.

816

during the global financial crisis, International Review of Financial Analysis, 37, 168-183. Nilssona, V., Nordströmb, J. and Bredmar, K. (2014). Need for liquidity and the capital structure of Swedish banks following the financial crisis, International Journal of Finance and Banking Studies, 3 (2), 10-27. Ogere, A. G., Peter, Z. and Inyang, E. E. (2013). Capital adequacy ratio and banking risks in the Nigeria money deposits banks, Research Journal of Finance and Accounting, 4 (17), 17-25. Olalekan, A. and Adeyinka, S. (2013). Capital adequacy and banks profitability: an empirical evidence from Nigeria, American International Journal of Contemporary Research, 3 (10), 87-93. Ozili, P. (2015). Determinants of bank profitability and Basel capital regulation: empirical evidence from Nigeria, Munich Personal RePEc Archive, Paper no. 61069. Available at http://mpra.ub.unimuenchen.de/6109/ [Accessed 20/1/2015]. Polat, A. and Al-Khalaf, H. (2014). What determines capital adequacy in the banking system of Kingdom of Saudi Arabia? a panel data analysis of Tadawal Banks, Journal of Applied Finance and Banking, 4 (5), 27-43. Raharjo, P. G., Hakim, D. B., Manurung, A. H. and Maulana, T. N. A. (2014). Determinants of capital ratio: a panel data analysis on state-owned banks in Indonesia, Buletin Ekonomi Moneter Jan Perbankan, 16 (4), 395-414. Rojas-Suarez, L. (2002), Can international capital standards strengthen banks in emerging markets?, Institute for International Economics, Working Paper. Available at http://www.iie.com/publications/wp/0110.pdf. [Accessed 12 December 2014]. Romdhane, M. (2012). The determinants of bank' capital ratio in developing countries: empirical evidence from Tunisia, Research Journal of Finance and Accounting, 3 (1), 35-46. Rose, P. and Hudgins, S. C. (2008). Bank management and financial services, 7 edition, McGraw-Hill. Sarma, M. and Nikaido, Y. (2007). Capital adequacy regime in India, an overview, Indian Council for Research on International Economics Relations, Working Paper, no 196. Shingjergii, A. and Hyseni, M. (2015). The determinants of capital adequacy ratio in Albanian banking system during 2007-2014, International Journal of Economics, Commerce and Management, 3(1), 1-10. Stolz, S. and Wedow, M. (2005), Banks’ regulatory capital buffer and the business cycle: evidence for German savings and cooperative banks, Discussion Paper Series 2: Banking and Financial Studies, No. 7/2005, Deutsche BundesBank Research Center, 1-42. Yudistira, D. (2003), The impact of bank capital requirements in Indonesia, May, Paper provided by EconWPA in its series Finance, number 0212002. Available at URL: http://128.118.178.162/eps/fin/ papers/0212/0212002.pdf. [Accessed 12 December 2014].


INVESTIGATING THE RELATIONSHIP BETWEEN CORPORATE GOVERNANCE AND BANKS’ FINANCIAL PERFORMANCE: EGYPT CASE Amr Youssef*, Mohamed Bayoumi** Abstract In the last few decades, policy makers around the world have focused on corporate governance reform since the Asian financial crisis and scandals in the United States such as the Enron debacle. In addition, there is no doubt that banks have significant position in the welfare of any economy. Corporate governance involves in how banks’ businesses and affairs are governed by its board of directors that raises a fundamental question of how this could affect banks’ financial performance. The focus of this research is to investigate the relationships between some of the corporate governance variables that are related to the board of directors on the financial performance of these banks working in the Egyptian market. Thirteen banks that are listed in the Egyptian Stock Exchange were selected with data collected for the period from 2011 till 2013 which is the post Egyptian revolution era. Research analyses adopted in this study are descriptive, correlation and regression analyses to test the research hypotheses. Findings of this research provide evidence that some of these variables such as board independence, foreign board members ratio, women board members ratio and board educational ratio have significant effect on the financial performance of these banks; however, board size and CEO qualities do not have any significant effect on banks’ performance. The research reaches some implications that are important to different stakeholders on practical and academic levels. Key Words: Corporate Governance, Performance Measurement, Agency Theory, Conventional Banks, Developing Countries, Egypt Jel Classification: G3, G2 * Head of Finance and Accounting Department, College of Management and Technology, Arab Academy for Science, Technology, and Maritime Transport, Egypt ** Relationship Manager at National Bank of Egypt and MSc of Investment Student, Graduate School of Business, Arab Academy for Science, Technology, and Maritime Transport, Egypt

1.

Introduction

In the 20th century the management was the focus, however; corporate governance is the focal point for the 21st century (Tricker, 2012). The term corporate governance drives from an analogy between the governance of corporations and the government of nations (Becht et al., 2002). The word governance is ancient and it comes from the Greek word steering (Carrol and Bucholtz, 2009). However, the phrase corporate governance is young. Recent corporate governance scandals make the corporate governance field receives a lot of attention from all interest groups and an increase in media coverage which turned terms like transparency, governance failure and weak board of directors as house hold phrases (Tirole, 2006). From the banking industry perspective, corporate governance involves the manner in which how the banks’ businesses and affairs are governed by its board of directors and its senior management. The board of directors is elected by the shareholders as the

decision making body of the bank which has as one of its responsibilities to formulate bank loan strategy (Sumner and Webb, 2005). And since the higher cost of capital will hurt the overall economic development, so the governance of banks is different from unregulated non-financial companies for several reasons, for one is that the number of parties with a stake complicate the governance of banks, in addition to investors, depositors and regulators have a direct interest in banks performance. One more reason is that regulators are more concerned with the effect that governance has on the performance of banks because the health of the overall economy depends upon banks performance (Adams and Mehran, 2003). The international financial landscape is changing rapidly; acquisitions and mergers wave has changed the banking industry shape. All things changed in that new global banking industry except for one thing remains unchanged which is the need to have a strong banking system with good corporate governance practices which will help any bank to survive in that

817


increasingly open environment (Kaheeru, 2001). Banking supervision functions are affected if the corporate governance system reliability has been affected (Heidi and Marleen, 2003). Between 1990 and 2000 the rapid changes in the banks ownership globally affected the governance systems of banks (Mayes et al., 2001). These changes in the corporate governance of banks have raised a very important question and fundamentally that question is how these changes in banks governance systems will affect banks’ performance? Hence, there is no globally accepted set of corporate governance principles, which could be applied to the board structures as they depend on business, legal and political environment which varies from one country to another. However, board structure is considered as an important corporate governance mechanism which would result in an improved performance, also the board composition, CEO qualities and board size all addressed due to their importance as components of the board structure and in affecting banks performance (Cadbury, 1992). 2.

Literature Review

Better corporate governance should lead to higher stock prices or better long-term performance, because when managers are better supervised, agency costs are decreased (Albanese et al., 1997). However, it’s suggested that, the evidence of a positive association between corporate governance and firm performance may be traced to the agency explanation, so in connection with the relationship between corporate governance and firm performance, the most studied governance practices include board composition and board size as elements of the board structure (Gompers et al., 2003). 2.1 Board Composition Performance

Effect

on

The composition of board members has been proposed to help reduce the agency problem (Weisbach, 1988). Empirical studies on the effect of board membership and structure on performance generally show results either mixed or opposite to reduce agency problem. While some studies find better performance for firms with boards of directors dominated by outsiders (Pfeffer and Salancik, 1978; Ogus, 1994; Pearce and Zahra, 1992; Vafeas, 1999), others find no such relationship in terms of accounting profits or firm’s value (Weisbach, 1988; Daily and Dalton, 1992; Mehran 1995; Daily and Ellstrand, 1996; Rosenstein and Wyatt, 1997; Klein, 1998; Weir et al., 2001; Bhagat and Bolton, 2005). The analysis of 54 empirical studies of board composition and 31 empirical studies of board leadership structure and their relationships to firm financial performance resulted in a little evidence of a relationship between board composition or leadership

and firm financial performance (Daily and Dalton, 1992). This is also evident in the studies of Hermalin and Weisbach (1999) and Bhagat and Black (1999). There is also a study showed that the cost of debt, as determined by bond yield spreads, is inversely related to board independence (Anderson et al., 2004). However, Hermalin and Weisbach (1999) observed no association between the proportion of outside directors and Tobin’s Q; and Bhagat and Black (1999) find no linkage between the proportion of outside directors and Tobin’s Q, return on assets, asset turnover and stock returns. Attiya and Robina (2007) in Pakistan analyzed the relationship between firm value using Tobin’s Q and governance subindices “board ownership and shareholdings”. The result indicates that corporate governance does matter in Pakistan and that board composition has significant effects on firm performance. Thus, the relationship between the proportions of outside directors, as a substitute for board independence, and firm performance is mixed. Studies using financial statement data and Tobin’s Q have found no link between board independence and firm performance, while those using stock returns data or bond yield data find a positive link. 2.2 Board Size Effect on Performance Unlike in board composition, a fairly clear negative relationship appears to exist between board size and firm performance (Yermack, 1996). A similar pattern has been documented for a sample of small and midsize firms. The study also revealed that board size and firm value are negatively correlated (Eisenberg et al., 1998). Other studies also confirmed that; limiting board size is believed to improve firm performance because the benefits by larger boards of increased monitoring are outweighed by the poorer communication and decision-making of larger groups (Lipton and Lorsch, 1992; Jensen, 1993). A large board is likely to be less effective in substantive discussion of major issues and to suffer from freerider problems among directors in their supervision of management (Hermalin and Weisbach, 2002). Harris and Raviv (2005) and Bennedsen et al. (2006) quoted the study of Yermack (1996) as a first ever-empirical study conducted on board size effect. Yermack has conducted his study on 452 US firms between 1984 and 1991. He took Tobin’s Q as an approximation of market valuation. He documented an inverse association between board size and firm value. He further asserted that the fraction of lost value occurs more when size of firm is increasing from small to medium for example from 6-12 as compare to the firm whose board size is increasing from medium to big from 12-24. In Ghana, it has been identified that small board sizes enhances the performance of firms (Kyereboah-Coleman and Nicholas-Biekpe, 2006).

818


While in a study conducted in Nigeria, Sanda et al. (2005) found that firm performance is positively related with small size as opposed to large boards. In their studies, Klein (1998); Booth and Deli (1999) and Anderson et al. (2004) tried to find out the relation between board size and ratio of debt to assets. They presented a different result that firms with bigger boards have lower cost of debt. On contrary to the theory that larger boards are ineffective monitors, they stated that board plays an important advisory role that enables firms to gain access to low-cost debt. They observed that the board will be larger in firms with high leverage. 2.3 Theoretical Framework Corporate Governance

for

Corporate governance importance is growing, particularly with regards to the monitoring role of the board of directors. As a result, the theoretical perspective that’s relevant to this study is based on the governance structures that affect the value of the firms (Sanda et al., 2005). This section reviews agency theory as the relevant theoretical perspective of a board’s accountability to this study. Agency theory has its roots in economic theory and it dominates the corporate governance literature. It’s been pointed to two factors that influence the prominence of agency theory. Firstly, the theory is a conceptually simple one that reduces the corporation to two participants, managers and shareholders. Secondly, the notion of human beings as selfinterested is a generally accepted idea (Daily et al., 2003). In its simplest form, agency theory explains the agency problem arising from the separation of ownership and control. It provides a useful way of explaining relationships where the parties’ interests are at odds and can be brought more into alignment through proper monitoring and a well-planned compensation system (Davis et al., 1997). Eisenhardt (1989) explains that the agency problem arises when the desires or goals of the principal and agent conflict and it is difficult or expensive for the principal to verify what the agent is actually doing. The problem is that the principal is unable to verify that the agent is behaving inappropriately. Shleifer and Vishny (1997) further explained that the agency problem in this context refers to the difficulties financiers have in assuring that managers do not expropriate funds or waste them on unattractive projects. Agency theory set that the control function of an organization is primarily exercised by the board of directors. With regard to the board as a governance mechanism, the issues that appear most prominent in the literature is board structure, more specifically board size, inside versus outside directors, board

composition and CEO characteristics, and the role and responsibilities of the board (Biserka, 2007). In relation to the research objectives, this study will adopt the agency theory because it focuses on the board of directors as a mechanism, which dominates the corporate governance literature. 3.

Research Methodology

This study made use of the secondary data of the annual reports of the 13 listed banks in the Egyptian Stock exchange to find out the relationship that exists between corporate governance variables and financial performance of banks. Therefore, this study is based on a positivist paradigm used deductive reasoning and quantitative techniques. Besides, this study adopted a positivist approach, because a positivist approach seeks facts or causes of social phenomena. The reasoning is deductive because the hypotheses were derived first and the data were collected later to confirm or negate the propositions. 3.1 Research Problem It’s been stated before that the financial crises made it necessary to measure the role of corporate governance on the banks’ performance (Ermina and Patsi, 2010). And since a lot of researches have analyzed the bank stability, accounting performance and the structure of ownership, but there is a few that examine the relationship between corporate governance and banks’ performance. Even most of the previous empirical literatures that analyses the connection between governance and firm performance using board size, board composition and CEO qualities are mostly focused on financial firms in general or the industrial firms in particular. This study will then address the questions emerging within the domain of study problems as follows:  To what extent (if any) does board size affects the financial performance of the listed banks in Egypt?  Is there any significant relationship between board composition and the financial performance of the listed banks in Egypt?  Is the relationship significant between the CEO qualities and the financial performance of the listed banks in Egypt? 3.2 Variables Development The research is going to define the variables that will be used to investigate the relationship between governance and banks’ financial performance as seen in table (1).

819


Table 1. Definition of Variables Variable

Definition Previous Studies Measures of Bank Performance (Dependent Variables) Return on Assets (ROA) Net Income After Taxes + Interest Finkelstein and D'Aveni, 1994; Kiel and Expense /Total Assets. Nicholson, 2003; Weir et al., 2001; Bonn et al., 2004; Epps and Cereola, 2008; Baysinger and Butler, 1985; Return on Equity (ROE) Net Income After Taxes/Total Dehaene et al., 2001. Equity. Measures of Board Structure (Independent Variables) Hanrahan et al., 2001; Tomasic et al., Board size: 2003; Bhagat and Jefferis, 2002; Defond Board Size (BSIZE) A natural logarithm of the total and Hung, 2004; Morin and Jarrell, number of board members. 2001; Holmstrom and Milgrom, 1994; Monks and Minow, 2001; Yermack, 1996; Heinrich, 2002; Zahra and Pearce, Board composition: A proportion of independent non- 1989; Bhagat and Bolton, 2005; Weir et Board Independence executive members in the board. al., 2001; Sang-Woo and Lum, 2004; (BIND) Bhagat and Black, 2002; Lipton and A proportion of foreign members in Lorsch, 1992; Jensen, 1993; Hermalin the board. and Weisbach, 2002; Mak and Li, 2001; Healey, 2003; Beiner et al. 2003; Mak Foreign members of Board and Yuanto, 2003; Bennedsen et al. Ratio A proportion of women members in 2006; Harris and Raviv, 2005; (FBRATIO) the board. Kyereboah-Coleman and NicholasBiekpe, 2006; Coles et al. 2004; Women members of Board Anderson and Reeb, 2003; De Andres et Ratio (WBRATIO) A proportion of members in the al., 2005; Jackling and Johl, 2009. Supervisory board holding PhD. Board Educational Ratio (EDURATIO) (1): if the CEO is foreign citizen; (0): if otherwise. CEO qualities: Dummy for CEO (CEO)

Dummy for CEO Power (CEOPOWER) Control Variables Bank’s age (BAGE)

Dummy for Bank’s nature (BNATURE)

(1): if the CEO serves longer than one-term “3 years”; (0): if otherwise.

A natural logarithm of the difference between the principle year of analysis and the year of bank’s foundation. (1): if a bank is subsidiary of a multinational bank; (0): if otherwise.

3.3 Research Hypotheses A business exists for the profit of shareholders, and the board of directors should focus on that objective (Ferrel et al., 2013). Board serves as a bridge between shareholders and managers, playing a major governing role in the corporate governance framework (Cadbury, 2002). The study of corporate governance is complicated by the fact that the structure, role and impact of boards have been studied

from a variety of theoretical and practical perspectives. Numerous studies are dedicated on detection a link between corporate governance and bank performance (Jensen and Meckling, 1976; Hovey et al., 2003). By including the board of directors’ characteristics such as director’s shareholding, gender, director size, director’s race and directors’ independence, it brings the new avenue for the researcher and regulators of the importance of

820

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 board of directors’ characteristics on the performance (Shukeri et al., 2012).

 H2e: There is significant relationship between

3.3.1 Board Size



The following are the hypotheses that will be tested empirically with regard to the impact of the Board size:  H1a: There is significant relationship between board size and banks financial performance measured by ROA.  H1b: There is significant relationship between board size and banks financial performance measured by ROE. 3.3.2 Board Composition The hypotheses to test the significance of the impact of Board composition are defined with the following statements:  H2a: There is significant relationship between board independence and banks financial performance measured by ROA.  H2b: There is significant relationship between board independence and banks financial performance measured by ROE.  H2c: There is significant relationship between foreign board members ratio and banks financial performance measured by ROA.  H2d: There is significant relationship between foreign board members ratio and banks financial performance measured by ROE.

 

women board members ratio and banks financial performance measured by ROA. H2f: There is significant relationship between women board members ratio and banks financial performance measured by ROE. H2g: There is significant relationship between educational board ratio and banks financial performance measured by ROA. H2h: There is significant relationship between educational board ratio and banks financial performance measured by ROE.

3.3.3 Chief Executive Officer Qualities The significance of the impact of CEO qualities will be tested though the following hypotheses:  H3a: There is significant relationship between CEO nature and banks financial performance measured by ROA.  H3b: There is significant relationship between CEO nature and banks financial performance measured by ROE.  H3c: There is significant relationship between CEO power and banks financial performance measured by ROA.  H3d: There is significant relationship between CEO power and banks financial performance measured by ROE.

Figure 1. Research Framework of the Relationships between Corporate Governance and Banks’ Financial Performance Corporate Governance (independent variables):

Bank’s financial performance (dependent variable):

Board Size Board Independence Foreign Board Members Ratio Women Board Member Ratio Board Education Ratio CEO Nature CEO Power

ROA ROE

Control variables: Bank’s age Bank’s nature

821

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 3.4 Research Model

3.5 Data Sampling

As shown in figure (1), the large number of independent variables used in the study forced researchers to develop two multiple regression analyses with limited variables in order to assess the relationships between corporate governance variables and banks’ financial performance in Egypt as follows: Equation 1 defined as:

The population for this study consists of 41 registered Commercial banks at the Central Bank of Egypt (CBE) in 2013. The time frame considered for this study is from 2011 to 2013. This 3-year period, although shorter than most studies of this nature, it will help sharpen a deep understanding of the relation between governance and banks’ financial performance in the post Egypt 25th of January revolution era. Thus, it’s noted that the banking sector performance hasn’t been affected during this mentioned period. The research sample size is the 13 listed banks in the Egyptian Stock Exchange (EGX); which represents about 31.7% of the total population. Therefore, this sampling technique enabled us to have easy accessibility to their annual reports which is the major source of our secondary data. The data used for this study were secondary data derived from the audited financial statements of the banks listed in the Egyptian Stock Exchange (EGX) for the three years period from 2011 till 2013.

Financial Performance Governance)

=

f

(Corporate

(1)

With replacing the measures of the financial performance within the equation, the resulted equations will be as follows: ROE= f (Board Size + Board Composition + CEO Qualities + Control Variables) ROA= f (Board Size + Board Composition + CEO Qualities + Control Variables)

(2a) (2b)

Then the research reaches to the final regression equations:

4. ROE = β0+ β1 BSIZE it + β2 BIND it+ β3 FBRATIO it + β4 WBRATIO it + β5 EDURATIO it+ β6 CEO it + β7 CEOPOWER it + β8 BAGE it + β9 BNATURE it + ε ROA = β0+ β1 BSIZE it + β2 BIND it + β3 FBRATIO it + β4 WBRATIO it + β5 EDURATIO it+ β6 CEO it + β7 CEOPOWER it + β8 BAGE it + β9 BNATURE it + ε

(3a)

(3b)

Where: (β0): constant. (i): represents the cross sectional dimensions of the data (it refers to a specific bank). (t): represents time. (β): represents the coefficients of the independent and control variables. (ε): represents the error term (the difference between expected and observed performance of the sample of banks used in this study).

Statistical Results and Analysis

The analysis uses descriptive statistics and t-tests to report the significance of the change. Spearman’s correlation analysis assesses the association between variables, and an analysis of variance assesses the suggested relationships of the research hypotheses. The results from the statistical analysis discuss the integrated results to find out if the hypotheses are supported. The study is presenting the results of the analysis performed on the data collected to test the propositions made in the study and answer the research questions. Analyses were carried out with the aid of SPSS software package. The following table (2) provides descriptive statistics of the mean and standard deviation of the dependent, independent variables and control variables regarding the sample used in this study:

Table 2. Descriptive statistics Variable ROA ROE BSIZE BIND FBRATIO WBRATIO EDURATIO CEO CEOPOWER BAGE BNATURE

N 39 39 39 39 39 39 39 39 39 39 39

Minimum -0.0587 -1.3684 6 .0769 0 0 0 0 0 5 0

Maximum 0.0264 0.2504 15 1 0.7143 0.2500 0.4444 1 1 38 1

822

Mean 0.0088 0.0558 10.72 0.8005 0.3189 0.0540 0.1060 0.33 0.82 29 0.62

Std. Deviation 0.0157 0.2859 2.470 0.1557 0.2501 0.0719 0.1163 0.478 0.389 10.180 0.493


Generally, from the 39 observations as seen in table (2), the table revealed that on average, the banks included in the research sample generates Return on Equity (ROE) of about 5.58% and a standard deviation of 28.59%. The average board size from the 39 observations is about 11 suggesting that banks in Egypt have relatively moderate board sizes as suggested by Kyereboah-Coleman and Biekpe (2006) with a maximum board size of fifteen (15) and deviation of 247%. The implication is clear that banks in Egypt have relatively similar board sizes. In addition, the average of the CEO Nature from the 39 observations shows that about 33% of the sample banks’ CEOs are not Egyptian citizens, and about 82% of them spend more than one term as the bank

CEO. Also, the study shows on average that about 62% of the banks within the sample are subsidiaries of foreign multinational banks. The research measured the degree of association between the governance variables and profitability variables i.e. if the governance proxies (board size, board composition and CEO qualities) will increase profitability. From the prior, a positive relationship is expected between the variables of corporate governance and profitability measures (ROE and ROA). The next table (3) presents the Pearson's correlations matrix between the dependent and independent variables; the correlation analysis supports some expectation.

Table 3. Pearson’s Correlation Coefficients Matrix Pearson’s Correlation ROA ROE BSIZE BIND FBRATIO WBRATIO EDURATIO CEO CEOPOWER

ROA 1 0.959 0.302 -0.079 -0.225 -0.258 0.317 0.124 0.159

ROE BSIZE BIND FBRATIO WB RATIO EDU RATIO CEO CEO POWER 1 0.332 -0.007 -0.305 -0.312 0.286 0.205 0.093

1 0.171 -0.402 -0.331 0.047 0.081 0.064

1 -0.653 -0.098 0.201 -0.157 -0.348

1 0.013 -0.217 0.038 0.246

From the correlation results, board size (BSIZE) has a positive correlation of (0.332) with return on equity (ROE) which is significant. Similar trend was observed from the correlation result that board size also have a positive correlation of (0.302) with return on asset (ROA). The outcome for board size is consistent with the earlier study of Arslan et al. (2010) as they argued that large board size improves corporate performance through enhancing the ability of the company to establish external connection with the environment. But however, this result is not consistent with Bennedsen et al. (2006) as they argued that larger board is ineffective as compared to smaller boards. The proportion of outside independent directors (BIND) is another governance variable that recorded a negative but weak correlation of (-0.079) and (0.007) with both (ROA) and (ROE) respectively. This is consistent with Yermack (1996) and Bhagat and Black (1999) in their studies, where they found a negative correlation between the proportion of outside directors and corporate performance. Furthermore, another study conducted in UK, Weir and Laing (1999) did not find any correlation between the proportion of non-executive directors and corporate performance. As for the foreign board members ratio (FBRATIO), there is a negative correlation of (0.225) and (-0.305) with both (ROA) and (ROE) respectively. Similar trend was observed from the

1 0.052 -0.395 -0.314

1 0.359 0.198

1 0.172 1

correlation result with respect to women board members ratio (WBRATIO) that had a negative correlation of (-0.258) and (-0.312) with both (ROA) and (ROE) respectively, this is consistent with Forbes and Milliken (1999) as they argued that board diversity also generates various costs associated with coordination problems and decision making times. Further, board diversity may lead to a less cooperative and conflicts within the board (Lau and Murnighan, 2005). However, in contrary it’s been observed that there is positive correlation of (0.317) and (0.286) between board education ratio (EDURATIO) and both (ROA) and (ROE) respectively. Thus, it’s been analyzed that there is positive correlation of (0.124) and (0.205) between CEO nature (CEO) and both (ROA) and (ROE) respectively. Similar trend was observed from the correlation of (0.159) and (0.93) between CEO power (CEOPOWER) and both (ROA) and (ROE) respectively. However, Berger et al. (2012) argued that the effect of a powerful CEO can be counterbalanced by other executives. The research used the panel data regression analysis to investigate the impact of corporate governance on banks’ financial performance measured by both return on equity and return on assets. In doing this, two simple definitional models were developed to guide the analyses as shown in table (4).

823


Table 4. Model Summary

Model 1 2

R 0.636 0.630

R Square 0.405 0.397

Adjusted R Square 0.220 0.210

 Model (1): ROA as Dependent Variable.  Model (2): ROE as Dependent Variable.  Predictors: (Constant), BSIZE, BIND, FBRATIO, WBRATIO, EDURATIO, CEO, CEOPOWER, BAGE and BNATURE. The results show the explanatory power of the model as measured by the R Square and adjusted R Square. The adjusted R Square provides a better estimation of the true population value, especially with a small sample (Tabachnick and Fidell, 2001). From model (1), the coefficient of determination (R Square) indicates that about 40.5% of change in ROA is accounted for by the explanatory variables while the adjusted R-squared of 22% which is

Standard Error of the Estimate 0.0138872 0.2541442

powerful. Also for model (2), 39.7% of change in ROE is accounted for by the independent variables while the adjusted R-squared of 21% which is further justified as having powerful effect of the independent variables over the financial performance of the sampled banks. Std. Error of the Estimate, also called the root mean square error, is the standard deviation of the error term, and is the square root of the Mean Square Residual (or Error). The value of Std. Error of the Estimate in the current study is (0.0139) for model (1) and (0.2541) for model (2); this result is relatively small and indicates to the quality of the model.

Table 5. Regression Results of the Variables (Coefficients)

Model (1,2) (Constant) BSIZE BIND FBRATIO WBRATIO EDURATIO CEO CEOPOWER BAGE BNATURE

Unstandardized Coefficients B Std. Error 1 2 1 2 0.095 1.511 0.038 0.705 0.001 0.010 0.001 0.021 -0.067 -1.034 0.024 0.438 -0.032 -0.639 0.016 0.290 -0.111 -1.967 0.045 0.831 0.079 1.131 0.028 0.515 -0.011 -0.109 0.007 0.120 -0.008 -0.157 0.009 0.156 -0.001 -0.015 0 0.007 0.007 0.084 0.007 0.124

Standardized Coefficients Beta 1 2 0.088 -0.668 -0.513 -0.508 0.583 -0.335 -0.189 -0.551 0.221

 Model (1): ROA as Dependent Variable.  Model (2): ROE as Dependent Variable.  95.0% Confidence Interval for B. Table (5) shows that P-values (Sig.) whenever they’re lower than or equal to (0.05) are significant. Table (5) also shows that board size, board independence, foreign board ratio, women board ratio, educational ratio and bank age all have significant effect with financial performance of banks. Meanwhile, board size, CEO nature, CEO power and bank nature all have non-significant effect with financial performance of banks in Egypt.

0.083 -0.563 -0.559 -0.495 0.460 -0.182 -0.213 -0.523 0.145

t 1 2.469 0.495 -2.820 -2.036 -2.447 2.804 -1.678 -0.900 -2.380 1.040

Sig. 2 2.144 0.461 -2.362 -2.204 -2.366 2.197 -0.906 -1.008 -2.246 0.675

1 0.020 0.624 0.009 0.049 0.021 0.009 0.104 0.375 0.024 0.307

2 0.041 0.648 0.025 0.036 0.025 0.036 0.372 0.322 0.032 0.505

ROE= β0+ β1 BSIZE it + β2 BIND it+ β3 FBRATIO it + β4 WBRATIO it + β5 EDURATIO it+ β6 CEO it + β7 CEOPOWER it + β8 BAGE it + β9 BNATURE it+ ε

(2)

After the substitution with (B) values for both models (1) and (2) from table (5), the regression models will be read as follows: ROA= 0.095 + 0.001 BSIZE - 0.067 BIND 0.032 FBRATIO - 0.111 WBRATIO + 0.079 EDURATIO - 0.011 CEO - 0.008 CEOPOWER - 0.001 BAGE + 0.007 BNATURE

(1)

ROE= 1.511 + 0.010 BSIZE - 1.034 BIND 0.639 FBRATIO - 1.967 WBRATIO + 1.131 EDURATIO - 0.109 CEO - 0.157 CEOPOWER - 0.015 BAGE + 0.084 BNATURE

(2)

4.1 Hypotheses Testing As mentioned earlier the final regression models were as followed: ROA= β0+ β1 BSIZE it + β2 BIND it + β3 FBRATIO it + β4 WBRATIO it + β5 EDURATIO it+ β6 CEO it + β7 CEOPOWER it + β8 BAGE it + β9 BNATURE it+ ε

(1)

824


4.1.1 Hypothesis (1) From the analysis, the correlation between board size and both ROE and ROA have coefficients of (0.332, 0.302) respectively, indicating the positive correlation between the two variables. Also, the regression coefficients of the model are positive (0.001, 0.010) for both ROA and ROE respectively, with a p-value of (0.624, 0.648) respectively (significant when p≤0.05). This indicates a non-significant effect of board size on the financial performance of the listed banks. On the premise of these results, since the effect is non-significant, the research therefore reject the null hypothesis and accept the alternate hypothesis which states that there is no significant relationship between board size and the financial performance of the listed banks in Egypt. However, this result is different from other results reported by KyereboahColeman and Biekpe (2006) which concluded that a positive relationship between a firms’ value and board size exists. The result of the hypothesis also differs from Pathan et al. (2007) as it’s been stated that the board members tend to become involved in dysfunctional conflicts where the board is not cohesive which results in deteriorating the value of a firm. 4.1.2 Hypothesis (2) From the hypotheses (H2a, H2b), the research assume that there is a significant relationship between the proportion of outside independent directors sitting on a board and the financial performance of banks. The correlation result shows a negative correlation with both ROA and ROE of (-0.079, -0.007) respectively which entails that the more the number of outside directors, the lower the financial performance of banks in Egypt. However, the regression coefficients of the model are negative (-0.067, -1.034) for both ROA and ROE respectively. The regression result shows that the negative association observed between the variables is significant only when p≤0.05 with a pvalue of (0.009, 0.025). This also confirms that outside directors does have significant but negative impact on bank performance as measured by both ROA and ROE. Based on the fact that the association is significant, the research therefore accepts the null hypothesis. The negative effect noticed is likely to be because non-executive independent directors are likely not to have a hands-on approach or are not necessarily well versed in the business; hence do not necessarily make the best decisions. This is in tune with the studies of Pi and Timme (1993); Bosch (1995); Belkhir (2006); Staikouras et al. (2007) and Adams and Mehran (2008) which found a negative but significant relation between the tested variables. However, the research findings disagree with

Bebchuk et al. (2009) and Pathan et al. (2007) who found a positive relationship between these variables. From hypotheses (H2c, H2d), a negative correlation of (-0.225, -0.305) is observed between the foreign board members ratio and both ROA and ROE respectively; the regression coefficients of the model are negative (-0.032, -0.639) for both ROA and ROE respectively. The regression result further reveals that a significant negative relationship with a p-value of (0.049, 0.036) significant when p≤0.05. However, based on these findings, the research therefore accepts the null hypothesis and rejects the alternate hypothesis. This result is consistent with Forbes and Milliken (1999) as they argued that board diversity generates various costs associated with coordination problems and decision making times. From hypotheses (H2e, H2f), a negative correlation of (-0.258, -0.312) is observed between the women board members ratio and both ROA and ROE respectively; the regression coefficients of the model are negative (-0.111, -1.967) for both ROA and ROE respectively. The regression result further reveals that a significant negative relationship with a p-value of (0.021, 0.025) significant when p≤0.05. However, based on these findings, the research therefore accepts the null hypothesis and rejects the alternate hypothesis. This result is consistent with Lau and Murnighan (2005) as they argued that board diversity may lead to a less cooperative and conflicts within the board. From hypotheses (H2g, H2h), a positive correlation of (0.317, 0.286) is observed between the educational board ratio and both ROA and ROE respectively; the regression coefficients of the model are positive (0.079, 1.131) for both ROA and ROE respectively. The regression result further reveals that a significant positive relationship with a p-value of (0.009, 0.036) significant when p≤0.05. However, based on these findings, the research therefore accepts the null hypothesis and rejects the alternate hypothesis. This result may support the research point of view to this particular relation as it’s been stated before that the PhD holders of the board put their hands on the proper knowledge which enables them to guide the rest of the board members through markets uncertainty. 4.1.3 Hypothesis (3) From hypotheses (H3a, H3b), assumed that there is a relationship between CEO nature and financial performance of banks in Egypt. From the analysis, the correlation between CEO nature and both ROE and ROA have correlation of (0.124, 0.205) respectively, indicating the positive correlation between the two variables. Also, the regression coefficients of the model are negative (-0.011, -0.109) for both ROA and ROE respectively, with a p-value of (0.104, 0.372) respectively significant when p≤0.05. Therefore, the research rejects the null hypothesis which states that

825


the profitability of the banks with foreign directors is significantly different from the profitability of banks without foreign directors and accepts the alternate hypothesis. This is in line with Hoschi et al. (1991) and Fich (2005) but however not in agreement with Chibber and Majumdar (1999) and Djankov and Hoekman (2000) in their studies in which they stated that firms with foreign directors tend to perform better than those without foreign directors. Finally, hypotheses (H3c, H3d), assumed that there is significant relationship between CEO power and financial performance of banks in Egypt. From the analysis, the correlation between CEO power and both ROE and ROA have correlation of (0.159, 0.093) respectively, indicating the positive correlation

between the two variables. Also, the regression coefficients of the model are negative (-0.008, -0.157) for both ROA and ROE respectively, with a p- value of (0.375, 0.322) respectively significant when p≤0.05. Therefore, the research rejects the null hypothesis which states that the profitability of the banks with powerful directors is significantly different from the profitability of banks without powerful directors and accepts the alternate hypothesis. This result is consistent with Berger et al. (2012) as they stated that the effect of a powerful CEO can be counterbalanced by other executives. A summary of these results is provided in the following table (6):

Table 6. Summary of Hypotheses Results H H1

Hypothesis There is significant relationship between board size and banks financial performance measured by both ROA and ROE. There is significant relationship between board independence and banks financial performance measured by both ROA and ROE.

Results Non-Significant and Positive.

There is significant relationship between foreign board members ratio and banks financial performance measured by both ROA and ROE.

Significant and Negative.

There is significant relationship between women board members ratio and banks financial performance measured by both ROA and ROE.


There is significant relationship between board educational ratio and banks financial performance measured by both ROA and ROE. There is significant relationship between CEO nature and banks financial performance measured by ROA and ROE.

Significant and Positive.


H2

Non-Significant and Positive.

H3 There is significant relationship between CEO power and banks financial performance measured by ROA and ROE. 5.

Summary, Conclusion Recommendations

Non-Significant and Positive.

and

5.1 Recommendations and Implications of the Study

The study concludes that negative and significant relationships exist between bank performance from a side and board independence, foreign board members ratio and women board members ratio from the other side. Also, a positive and significant relationship exists between board educational ratio and financial performance. While there are non-significant and positive relationships between bank performance and board size, CEO nature and CEO power.

Based on the findings of this research, the following recommendations are presented which should be useful to different stakeholders: 1. Efforts to improve corporate governance shouldn’t focus on the board size as a mechanism of corporate governance of banks operating in Egypt, since it isn’t significantly related to the financial performance of banks in Egypt. 2. Proponents of board independence should note with caution the negative relationship between board independence and future operating performance. Hence, if the purpose of board independence is to improve performance, then

826


3.

4.

5.

6.

such efforts might be misguided. However, if the purpose of board independence is to discipline management of the Egyptian banks or otherwise monitor, then board independence has merit. In other words, to have proper monitoring by independent directors, bank regulatory bodies should require additional disclosure of financial or personal ties between directors (or the organizations they work for as independent board member) and the bank or its CEO. By so doing, they will be more completely independent. Also, banks should be allowed to experiment with modest departures from the current norm of a “supermajority independent” board with only one or two inside directors. Steps should also be taken for monitoring and ensuring that the Egyptian banks have already taken steps in compliance with the code of corporate governance set by the Central Bank of Egypt (CBE). As this code will ensure the rights and obligations of a bank, its directors, shareholders, specific disclosure requirements and provide for effective enforcement of the law. Thus, efforts to improve corporate governance in the Egyptian banking sector has to put into consideration to analyze the proportion of both foreign and women members within banks boards, since both are significant and negatively related to the financial performance of banks in Egypt. Also, the significant and positive relationship between the proportion of board members who hold PhD degree and the financial performance of banks in Egypt, should be analyzed further, as to encourage current board members and the future leaders in the Egyptian banks to pursue further graduate academic studies and not to pay all their attention towards obtaining professional internationally recognized certificates only. Finally, there is a need to set up a unified corporate body which would be responsible for collecting corporate governance related data and constructing the relevant indices to facilitate corporate governance research in general and in the Egyptian banking sector in particular.

5.2 Limitations of the Research The scope of the study was limited to only the 13 listed banks in the Egyptian Stock Exchange (EGX), because these listed banks were more likely to have the resources and motivation to take the opportunity to adopt good corporate governance practices as to increase the investors trust in the value of their listed stocks. Although the sample was small, it represented different local and multinational subsidiary banks. The small size of the sample may have affected the relationships between the variables. Therefore, the findings may have been different if a larger sample was included and the study period was extended,

however unfortunately it couldn’t be extended due to different political and socioeconomic circumstances that were existing after year 2011, besides the financial statements and the annual reports of the banks represented in the sample were not yet ready to be published for the year ended 2014, till the date when the research had been conducted. Although, most research in the area of corporate governance has been conducted in the developed economies, there is a very limited research on corporate governance practices and performance of companies in developing countries like Egypt, which operate in different, difficult and turbulent political and economic environment. Furthermore, institutional legal frameworks in emerging economies are not well developed compared to developed countries, which limits the benefits of their corporate governance efforts. Besides these emerging economies show significant differences in terms of economic growth, business environments, income levels and management practices. Moreover, a lot of researches have analyzed the bank stability, accounting performance or the structure of ownership, but there is a few that examine the relationship between corporate governance and banks performance. Even most of the previous empirical literatures that analyze the connection between governance and firms’ performance are mostly focused on financial firms in general or the industrial firms in particular. Thus, the research had used only the board of directors’ attributes as mechanisms of evaluating corporate governance in the Egyptian banking sector. 5.3 Suggestions Studies

for

Further

Future

The limitations of the study have prompted suggestions for further research as listed below: 1. This research has gone some way in exploring corporate governance variables and corporate performance of banks in a broader context. Further research could explore the relationship in more specific categories for example, in notfor-profit organizations, in government organizations, and in family businesses. Since this study focused on the Egyptian banking sector, it would be beneficial to have a clearer understanding of corporate governance roles in other types of organizations. Such research could address the similarities and differences of the roles in different organizations and consider also the legal requirements for different organizations. 2. The period of study for this research is only three years i.e. (2011-2013), which the post Egypt’s uprising period. This limitation was imposed by the non-availability of data pertaining to the reviewed banks. However, further research can consider more time frames

827


3.

based on the availability of the annual reports and even can make a comparison between the era before the Egyptian revolution and the era after it. Further research is also required on the behavioral aspects of boards. Researchers in developed countries have recently started examining board processes by attending actual board meetings; however this also needs to be expanded by researchers in developing economies. There is therefore the need to go beyond the quantitative research, which is yielding a mixture of results, to perhaps a more qualitative approach as to how boards work. Expanding this current research into a wider study of board dynamics and decision making would be a start in developing a better understanding of corporate governance.

References Adams, R. and Mehran, H. (2008), "Corporate Performance, Board Structure and their Determinants in the Banking Industry” Federal Reserve Bank of NY Staff Report No. 330. 2. Adams, R., and Mehran, H. (2003), "Is corporate governance different for banking holding companies?", Economic Policy Review, Vol. 9., pp. 123-142. 3. Albanese, R, Dacin, M. T. and Harris, I. C. (1997), "Agents as Stewards", The Academy of Management Review, Vol. 22, No.3, pp. 609-611. 4. Anderson, R. C., and Reeb, D. M. (2003), "Foundingfamily ownership and firm performance: Evidence from the S&P 500", The Journal of Finance, Vol. 58. No.3, pp. 1301-1328. 5. Anderson, R., Mansi, S. and Reeb. D. (2004), "Board Characteristics, Accounting Report Integrity, and the Cost of Debt", Journal of Accounting and Economics, Vol. 37 (Sept.), pp. 315-342. 6. Arslan, O., Karan, M. B., and Eksi, C. (2010), "Board Structure and Corporate Performance". Managing Global Transitions, Vol. 8, No.1, pp. 3-22. 7. Attiya Y. and Robina, I. (2007), "Relationship between Corporate Governance Indicators and Firm Value: A Case Study of Karachi Stock Exchange". MPRA Paper No. 2225. 8. Baysinger, B. and Butler, H. (1985), "Corporate Governance and Board of Directors: Performance Effects of Changes in Board Composition", Journal of Law Economics and Organization, Vol. 1, pp. 101-24. 9. Bebchuk, L., Cohen, A. and Ferrell, A. (2009), "What Matters in Corporate Governance?", The Review of Financial Studies, Vol. 22, No. 2, pp.783-807. 10. Becht, M., Bolton, P., and Röell, A. (2002), "Corporate Governance and Control". The National Bureau of Economic, Research Working Paper. No. 9371, pp. 1-168. 11. Beiner, S., Drobetz, W., Schmid, M. and Zimmermann, H. (2003), "An Integrated Framework of Corporate Governance and Firm Valuation - Evidence from Switzerland", University of Switzerland, Finance Working Paper, No. 34. 12. Belkhir, M. (2006), "Board Structure, Ownership Structure and Firm Performance: Evidence from Banking. Retrieved from

13.

14.

15.

16.

17.

18.

1.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

828

http://papers.ssrn.com/so13/papers.cfm. On 24th of November, 2009. Bennedsen, M., Kongsted, H. And Nielsen, K., (2006), "Board Size Effects in Closely Held Corporations, Working Papers 09-2004, Copenhagen Business School, Department of Economics. Berger, A. N., Imbierowicz, B., and Rauch, C. (2012), "The role of corporate governance in bank failures during the recent crisis". Available at SSRN: http://ssrn.com/abstract=2021799 or http://dx.doi.org/10.2139/ssrn.2021799. On 13th September, 2014. Bhagat, S. and Black, (2002), “The Non-Correlation Between Board Independence and Long-Term Firm Performance”, Journal of Corporation Law, Vol. 27, pp. 231-254. Bhagat, S. and Bolton, B. (2005), "Corporate Governance and Firm Performance”, Working Paper No.17-2005, University of Colorado. Bhagat, S. and Jefferis, R., (2002), "The Econometrics of Corporate Governance Studies, Cambridge, MIT Press. Bhagat, S., and Black, B. (1999), "The Uncertain Relationship between Board Composition and Performance", Journal of Global Finance, Vol. 17, No.1, pp. 515-530. Biserka, S., (2007), "The Role of Non-Executive Directors in Corporate Governance: An Evaluation", PhD Thesis submitted to the Department of Business in the Faculty of Business and Enterprise, Swinburne University of Technology. Bonn, I., Yoshikawa, T., and Phan, P. H. (2004), "Effects of Board Structure on Firm Performance: A Comparison between Japan and Australia, Asian Business & Management, Vol. 3, pp. 105-25. Booth, J. and Deli, D., (1999), "On Executives of Financial Institutions as Outside Directors”, Journal of Corporate Finance, Vol. 5, pp. 227-250. Bosch, H., (1995), “The Director at Risk: Accountability in the Boardroom”, The Business Lawyer, Vol. 54, No. 3, pp. 921-953. Cadbury, A. (1992), "The Committee on the Financial Aspects of Corporate Governance". London: Gee and Company. Cadbury, A. (2002), "Overview of Corporate Governance: A Framework for Implementation". The World Bank Group, Washington, D.C: V-VI. Carrol, A. B., and Buchholts, A. K. (2009), "Business and Society: Ethics and Stakeholder Management. Seventh edition, Ohio: Cengage Learning. Chibber, P. K., and Majumdar, S. K. (1999), "Foreign Ownership and Profitability: Property Rights, Control and the Performance of Firms in Indian industry", Journal of Law and Economics, Vol. 42, pp. 209-238. Coles, J., Daniel, L. and Naveen, L. (2004), "Boards: Does One Size Fit All?" Retrieved from papers.ssrn.com/sol3/papers.cfm_id. Daily, C. and Ellstrand, A. (1996), "Boards of Directors: A Review and Research Agenda", Journal of Management, Vol. 22, No. 3, pp. 409 - 418. Daily, C. M. and Dalton, D. R., (1992), "The Relationship Between Governance Structures and Corporate Performance in Entrepreneurial Firms", Journal of Business Venturing, Vol. 7, No. 5. Daily, C. M., Dalton, D. R. and Canella, A. A. (2003), “Corporate Governance: Decades of Dialogue and


31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Data”, Academy of Management Review, Vol. 28, No. 3, pp. 371-382. Davis, J. H., Schoorman, F. D. and Donaldson, L. (1997), "Toward a Stewardship Theory of Management", Academy of Management Review, Vol. 22, pp. 20-37. De Andres, P. A., Azofra, V., and Lopez, F. (2005), "Corporate Boards in some OECD Countries: Size, Composition, Functioning and Effectiveness", Corporate Governance: An International Review, Vol. 13, No. 2, pp. 197-210. Defond, M. and Hung, M. (2004), "Investor Protection and Corporate Governance: Evidence from Worldwide CEO Turnover", Journal of Accounting Research, Vol. 42, No. 2, pp. 86- 94. Dehaene, A., De Vuyst, V., and Ooghe, H. (2001), "Corporate Performance and Board Structure in Belgian Companies", Long Range Planning, Vol. 34, No. 3, pp. 383-98. Djankov, S., and Hoekman, B. (2000), "Foreign investment and Productivity Growth in Czech Enterprises", World Bank Economic Review, Vol. 37, pp. 49-64. Eisenberg, T., Sundgren, S., and Martin T. (1998), "Larger Board Size and Decreasing Firm Value in Small Firms", Journal of Financial Economics, Vol. 48, pp. 35-54. Eisenhardt, K. M. (1989), "Agency Theory: An Assessment and Review", International Journal of Management, Vol.5, pp. 341 – 353. Epps, R. W. and Cereola, S. J. (2008), "Do Institutional Shareholder Services (ISS) Corporate Governance Ratings Reflect a Company's Operating Performance?", Critical Perspectives on Accounting, Vol. 19, pp. 1138-48. Ermina, P. and Patsi, M. (2010). “Corporate Governance and Bank Performance", International Hellenic University, M.Sc. In Banking and Finance. Ferrell, O. C., Fraedrich, J., and Ferrell, L. (2013), "Business Ethics: Ethical Decision Making and Cases", Ninth edition. Ohio: South Western Cengage Learning. Fich, E., (2005), "Are Some Outside Directors Better than Others? Evidence from Directors", Appointments by Fortune 1000 Firms, Journal of Business, Vol. 78, No. 5, pp. 1943-1971. Finkelstein, S. and D'Aveni, R. A., (1994), "CEO Duality as a Double-edged Sword: How Board of Directors Balance Entrenchment Avoidance and Unity of Command", Academy of Management Journal, Vol. 37, No. 5, pp. 1079-108. Forbes, D. P., and Milliken, F. J. (1999), "Cognition and Corporate Governance: Understanding Boards of Directors as Strategic Decision-Making Groups", The Academy of Management Review, Vol. 24, pp. 489505. Gompers, P., Ishii, L and Metick, A. (2003), "Corporate Governance and Equity Prices", Quarterly Journal of Economics, Vol. 118, pp.107-125. Hanrahan, P., Ramsay, I. and Stapledon, G. (2001), "Commercial Applications of Company Law, 2nd ed.", Sydney, CCH Australia. Harris, M. and Raviv, A. (2005), "Allocation of Decision-Making Authority". Paper presented at the Twelfth Annual Utah Winter Finance Conference; on 11th March.

47. Healey, J. (2003), "Corporate Governance and Wealth Creation in New Zealand", Palmerston North, Dunmore Press Ltd, New Zealand. 48. Heidi, V. B., and Marleen, W. (2003), "Voluntary Disclosure on Corporate Governance in the European Union”. Dept. of Applied Economics, University of London. 49. Heinrich, R. (2002), "Complementarities in Corporate Governance", Berlin, Springer. 50. Hermalin, B. and Michael, W. (2002), "Boards of Directors as Endogenously Determined Institutions: A Survey of the Economy Literature", Economy Policy Review, March, pp. 48- 62. 51. Hermalin, B. E., and Michael, S. W., (1999), "The Effects of Board Composition and Direct Incentives on Firm Performance", Financial Management, Vol. 20, pp. 101-112. 52. Holmstrom, B. and Milgrom, P. (1994), “The Firm as an Incentive System”, American Economic Review, Vol. 84, No. 4, pp. 972-991. 53. Hoschi, T., Kashyap, A., and Scharfstein, D. (1991), "Corporate Structure Liquidity and Investment: Evidence from Japanese Industrial Group's", Quarterly Journal of Economics, Vol. 106, pp. 33-60. 54. Hovey, M., Li, L., and Naughton, T. (2003), "The Relationship between Valuation and Ownership of Listed Firms in China", Corporate Governance, Vol. 11, No. 2, pp. 112-122. 55. Jackling, B., and Johl, S. (2009), "Board Structure and Firm Performance: Evidence from India’s Top Companies". Corporate Governance: An International Review, Vol. 17, No. 4, pp. 492-509. 56. Jensen, M., (1993), "The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems", Journal of Finance, Vol. 48, pp. 831-840. 57. Jensen, M., and Meckling, W. (1976), "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure", Journal of Financial Economics, Vol. 3, Issue 4, pp. 305-360. 58. Kaheeru, V. (2001), “Institute of Corporate Governance of Uganda Manual”, Institute of Corporate Governance of Uganda, Lipton, M. And J. W. Working Paper. 59. Kiel, G. C. and Nicholson, G. J. (2003), "Board Composition and Corporate Performance: How the Australian Experience Informs Contrasting Theories of Corporate Governance", Corporate Governance, Vol. 11, No. 3, pp. 189-205. 60. Klein, E. (1998), “Firm Performance and Board Committee Structure”, Journal of Law and Economics, Vol. 41, No. 2, pp. 275- 293. 61. Kyereboah-Coleman, A., and Nicholas- Biekpe, N. (2006), "Corporate Governance and the Performance of Microfinance Institutions (mfis) in Ghana”, Working paper 4330-05, UGBS, Legon. 62. Lau, D.C., and Murnighan, J. K. (2005), "Interactions within Groups and Subgroups: The Effects of Demographic Fault lines", Academy of Management Journal, Vol. 48, pp. 645-659. 63. Lipton, M., and Lorsch, J. (1992), “A Modest Proposal for Improved Corporate Governance", Business Lawyer, Vol. 48, No. 1, pp.59-67. 64. Mak, Y. T. and Li., Y. (2001), "Determinants of Corporate Ownership and Board Structure: Evidence from Singapore", Journal of Corporate Finance, Vol. 7, pp. 235-256.

829


65. Mak, Y. T. and Yuanto, K. (2003), "Board Size Really Matters: Further Evidence on the Negative Relationship between Board Size and Firm Value", Journal of Corporate Finance, Vol. 5, pp. 145-166. 66. Mayes G. D., Halme, L. and Aarno, L. (2001), "Improving Banking Supervision". New York, Palgrave, Macmillan. 67. Mehran, H., (1995), "Executive Compensation Structure, Ownership, and Firm Performance", Journal of Financial Economics, Vol. 38, pp. 163-84. 68. Monks, R. and Minow, N. (2001), "Corporate Governance", 2nd ed., Blackwell Publishers, Oxford. 69. Morin, R. and Jarrell, S. (2001), "Driving Shareholders Value: Value-Building Techniques for Creating Shareholder Wealth", Sydney, McGraw-Hill Publishers. 70. Ogus, A. I. (1994), "Regulation, Legal Form and Economic Theory", Clarendon Law Series, Oxford, University Press. 71. Pathan, S., Skully, M. and Wickramanayake, J. (2007), "Board Size, Independence and Performance: An Analysis of Thai Banks", Asia-Pacific Financial Markets, Vol. 14, No. 2, pp. 211-227. 72. Pearce, J. A. and Zahra, S. A. (1992), "Board Composition from a Strategic Contingency Perspective", Journal of Management Studies, Vol. 29, No. 4, pp. 414-438. 73. Pfeffer, J. and Salancik, G. R. (1978), "The External Control of Organization: A Resource Dependency Perspective", New York: Harper and Roe. 74. Pi, L. and Timme S. G. (1993), "Corporate Control and Bank Efficiency", Journal Banking Finance, Vol. 21, No.7, pp. 515-530. 75. Rosenstein, S and Wyatt, J. C. (1997), "Outside Directors, Board Effectiveness and Shareholders Wealth", Journal of Financial Economics, Vol. 26, pp. 175-191. 76. Sanda, A. U, Mukaila, A. S. and Garba, T. (2005), "Corporate Governance Mechanisms and Firm Financial Performance in Nigeria", Final Report Presented to the Biannual Research Workshop of the AERC, Nairobi, Kenya, pp. 24-29. 77. Sang-Woo, N. and Lum, A. (2004), "Relationship Banking and Its Role in Corporate Governance", ADBI Research Paper, ADB Institute.

78. Shleifer, A., and Vishny, R. (1997), "A survey of corporate governance", Journal of Finance, Vol. 52, pp. 737-783. 79. Shukeri, S. N., Shin, O.W., and Shaari, M. S. (2012), "Does Board of Director’s Characteristics Affect Firm Performance? Evidence from Malaysian Public Listed Companies", International Business Research, Vol. 5, No. 9, pp. 120-127. 80. Staikouras, C., Maria-Eleni, K., Agoraki, A., Manthos, D. and Panagiotis, K. (2007), "The effect of board size and composition on bank efficiency". Retrieved from http://www.efmaefm.org/0EFMAMEETINGS/EFMA on 13th of October 2014 81. Sumner, S.W., and Webb, E. (2005), "Does Corporate Governance Determine Bank Loan Portfolio Choice?", Journal of the Academy of Business and Economics. Vol. 5, No.2, pp. 1-15. 82. Tabachnick, B. G., and Fidell, L. S. (2001),"Using multivariate statistics (4th ed.)". MA: Allyn and Bacon. 83. Tirole, J. (2006), "The Theory of Corporate Finance". New Jersey: Princeton University Press. 84. Tomasic, R., Pentony, B. and Bottomley, S. (2003), "Fiduciary Duties of Directors", Interview Schedule, Personal Communication, Melbourne. 85. Tricker, B. (2012), "Corporate Governance: Principles, Policies and Practices". Second edition. Oxford: Oxford University Press. 86. Vafeas, N. (1999), "Board Meeting Frequency and Firm Performance", Journal of Financial Economics, Vol. 53, pp. 113-132. 87. Weir, C., Laing, D., and mcknight, P. J. (2001), "An Empirical Analysis of the Impact of Corporate Governance Mechanisms on the Performance of UK Firm", Working Paper, Retrieved from http://www.ssrn/ awwfc.com. 88. Weisbach, M. S (1988), "Outside Directors and CEO Turnover", Journal of Financial Economics, Vol. 20, pp. 431-450. 89. Yermack, D. (1996), "Higher Market Valuation of Companies with a Small Board of Directors". Journal of Financial Economics, Vol. 40, pp.185–211. 90. Zahra, S. A. and Pearce, J. A. (1989), "Board of Directors and Financial Performance: A Review and Integrative Model", Journal of Management, Vol. 15, pp. 291-334.

830


THE IMPACT OF CAPITAL STRUCTURE AND CERTAIN FIRM SPECIFIC VARIABLES ON THE VALUE OF THE FIRM: EMPIRICAL EVIDENCE FROM KUWAIT Ahmad Mohammad Obeid Gharaibeh*, Adel Mohammed Sarea**

Abstract The main objective of this study is to empirically examine the impact of leverage and certain firmcharacteristics that are believed to have significant effects on the decision to use debt and on the value of the firm. The sample is composed of 48 companies listed in the Kuwait Stock Exchange (KSE) representing four different sectors. The study uses actual and historical panel data set obtained from the published annual reports of individual firms in addition to the publications of KSE. The study was accomplished using 8 years of data with a total of 239 observations representing the study period 2006-2013. The study uses descriptive statistics, correlation, and multiple-regression analyses to examine the impact of explanatory variables on the value of the firm. The study findings lead to the conclusion that capital structure (leveraging) is the most influential factor on firm’s value. Business risk, previous year’s value (one-year lagged ROA), dividends payout ratio, size, growth opportunities and liquidity of the firm are found to have significant influence on the firm’s value in Model 1 (where ROA is used as a proxy for the value of the firm). In model 2 (i.e., where ROE is used as a proxy of the firm’s value), the findings reveal that capital structure (leveraging); firm’s size, growth opportunities and liquidity of the firm are significant influential of the firm’s value. The study is valuable to academicians, finance managers, policy makers and other stakeholders as it fills the gap of literature by providing up-to-date evidence of the impact of capital structure and other firm specific variables on the value of the firm in Kuwait. Keywords: Capital Structure, Firm Value, Kuwait Stock Exchange (KSE) * Assistant Professor, Department of Banking and Finance, College of Business and Finance, Ahlia University, Manama, P.O. Box: 10878 Kingdom of Bahrain ** Director of MBA program & Assistant Professor of Accounting, Department of Accounting & Economics, College of Business and Finance, Ahlia University, Manama, P.O. Box: 10878 Kingdom of Bahrain

1. Introduction The relationship between capital structure and firm value has been a discussed in the previous studies regionally and globally. In theory, the relationship between both predict either positively, negatively as stated in the previous studies conducted in this regards. For instance, Modigliani and Miller (1963), (Ross 1977, and Leland and Pyle 1977) Rajan and Zingales (1995), Imad Ramadan (2015) Chowdhury and Chowdhury (2010) ElKelish and Andrew (2007), Booth et al. (2001). However, in the same context for other theories such as the trade –off theory (Myers,1984), pecking order theory (Myers and Majluf,1984) and agency cost theory (Jensen and Meckling, 1976) argue that if capital structure decision is irrelevant in a perfect market, then, imperfection which exist in the real world may be adduce for its relevance (Maxwell & Kehinde, 2012). The objective of this paper is to investigate the relationship between the capital structure and firm values in Kuwait exchange Market, the market value of the firms may be affected by the capital structure

decision as discussed in the previous studies. Therefore, the research question is, Does the capital structure of listed firms in Kuwait Exchange Market affect its market value? The structure of the paper is as follows. In next section, we discuss what others have done in this regards, reviewing briefly some of the previous studies conducted regionally and globally as well as the main underline theory which is MM theory. And then discussing Methodological framework, The Study Hypotheses, The study model, Research results and discussion including descriptive statistics, correlation Analysis, regression Analysis, and Conclusion of the work research. 2. Literature review Prior studies on the capital market made tremendous efforts to ensure practical and theoretical aspects of the capital structure. Capital structure is a term used in corporate finance to describe the mix of a company's long-term debt, some short-term debt, common and preferred equity. The capital structure refers to how a

831


company finances its operations and its growth by using various accessible sources of funds. When people refer to capital structure they are most likely referring to a firm's debt-to-equity ratio, which provides insight into how risky a company is. Decisions concerning the right hand side of the balance sheet of the firm (liabilities and stockholders' equity) result in a given capital structure of the firm. Sub optimal financing decisions, mostly, could lead to corporate failure. The objective of all financing decisions is wealth maximization and the immediate way of measuring the quality of any financing decision is to examine the effect of such a decision on the firm’s performance (Mwangi L. et al, 2014). The term financial performance is a subjective measure of how well a firm can use assets from its primary mode of business and generate revenues. It is also used as an overall measure of a company's financial health over a particular period of time. To evaluate the financial performance of a company, one should use financial analysis to assessing the firm’s profitability, leverage, solvency, and operational efficiency. The challenge is to know which ratios to choose and how to interpret the results. Operating income, cash flow from operations, and total unit sales can also be used to measure the financial performance of any firm. Financial leverage is a term used to refer to the usage of debt to finance activities and acquire additional assets in order to increase the expected return on equity. It is measured by dividing total debt by total assets or total debt by total debt and equity. Highly leveraged firms are those using more debt than equity. However, the presence of fixed cost of fund pertinent to financial leverage may add to the volatility of cash flows and, thus, net income especially when operating income is falling. Hence, leverage, increases the company's risk of bankruptcy. Capital structure decision is a crucial decision in corporate finance for almost all enterprises in the world. It is mainly consisted of debt and equities with proportions differ between firms based on many factors and variables. Both types of financing carry costs though they have their own benefits. Advantages to using debt vary; it provides tax shield as the interest rate paid as a cost of debt is normally tax deductible; it is not dilutive from shareholders standpoint; the cost of debt is generally less than that of equity to the firm. However, using debt financing increases the companies risk level (financing risk). Also, the borrowing firm has to meet loan covenants. In addition, assets may be taken as collateral and agency cost between creditors and shareholders may increases. The risk to shareholders is generally more than that to lenders as payment of debt is required by law irrespective of a company's profitability. Financial managers as well as some other stakeholders (investors and policy makers) of all firms around the globe conceivably will want to know the proper mix of debt and equity (capital structure) that

maximizes the firms’ performance. They may need to know the factors that influence the capital structure of their firms. They need to measure the influence of changing the capital structure of the firm on the profitability or the financial performance of their firm. In particular, they need to identify the relationships between financing decisions and the company performance. This may vary by country, by business environment, by sector, by company, or even by time. The risk to shareholders is greater than to lenders, since payment on debt is required by law regardless of a company's profit margins (Mwangi, L. et al, 2014). Other factors also may influence the company’s performance given the unique characteristics of certain economies. The findings of this study may make a contribution to the body of knowledge in this regard. This paper examines the effect of capital structure on the market value in Kuwait. The review of the literature shows that only a few researchers have been conducted to examine the impact of capital structure on the market value in GCC, but no studies have been conducted to examine the impact of capital structure on the market value in Kuwaiti market. In the international context for instance, study conducted by Sunder and Myers (1999) one of the most prominent theory is the static tradeoff theory which proposed that, there is a target level of debt-to equity ratio, in which the present value of tax benefits would equal the financial distress cost (bankruptcy risk). Furthermore, does the capital structure affect its market value of the firms? In theory, no according to Modigliani-Miller, assuming no transactions costs and taxes, etc. but in practices could be quite different. However, changes in capital structure can lead to change in market price. The change in the market can depend on a wide range of factors such as market mood, expectations of the market, and market needs. 2.1 Research questions Based on the above discussions, the followings are the research questions that this study seeks to answer: 1. Does the capital structure of listed firms in Kuwait affect its market value? 2. What is the type of the relationship between the capital structures and market value (i.e., positive or negative)? 3. What is the strength and intensity of the relationship between the capital structure and market value (i.e., significant or insignificant and at what level of significance)? 2.2

Research objectives

1. The aim of this study is to answer the research question; does the capital structure of listed firms in Kuwait affect its market value?

832


2. To examine the type of the relationship between the capital structures and market value (i.e., positive or negative)? 3. To examine how the capital structure negatively or positively influences the firm’s market value in the financial & nonfinancial in Kuwait as well as the relationship between the capital structure and market value (i.e., significant or insignificant and at what level of significance)? 3. Methodological framework To study the influence of capital structure (leverage) on the firm’s value, this study examined 48 companies from multiple sectors of Kuwait Stock Exchange. The study sample was selected from four different sectors including manufacturing (industrial), services, basic materials, and oil and gas. Certain sectors like financial, real estate, and communications were excluded from the analysis as they are considered as either having special characteristics or having high leverage. The nonmanufacturing firms were selected randomly and based on the availability of data, whereas all industrial (manufacturing) firms were included in the sample. This is, of course, to avoid the sampling error resulted from mixing all the listed firms and also to increase the reliability of the study. The panel data used for analysis was mainly collected from the published annual reports of 48 firms listed in the Kuwait stock exchange (KSE). Most of the selected firms were listed Kuwaiti while the remaining were non Kuwaiti companies. Some other financial data were obtained from information published by KSE. The actual financial data obtained embraces financial ratios including dividends to net profit ratio representing dividends policy of the firm, market price of the stock to its book value representing growth opportunities of the firm, total liabilities to total assets to represent leverage, total liabilities to total equities to represent liquidity position of the firm, natural logarithm of total assets to represent size of the firm, fixed assets to total assets ratio to represent tangibility of the firm. In addition the study uses firms age as a factor influencing the decision to use financing (capital structure) proxied by the number of years the firm is in business, and type of business which is used as a dummy variable where 0 denotes industrial and 1 denotes otherwise. The study also uses dummy variables to represent ownership structure where 1 signifies closely held companies and 0 signifies publicly held companies. These financial ratios and parameters were designed to aid the empirical model of the study and all were considered as independent variables and used as proxies for the capital structure decision of the firm. On the other hand, the study uses return on assets (ROA) and return on equity (ROE) as proxies of firm value. The 48 firms represent the sample of the study chosen from a population of 215 firms including nonKuwaiti listed companies. A total of 239 after

adjustments observations were obtained for the investigation covering the period of the study i.e., 2008 to 2013. 3.1 The Study Hypotheses Yu-Shu Cheng et al. (2010) and Gill et al. (2011) use return on equity (ROE) to measure firm value, when studying the relationship between capital structure and firm value. Joshua Abor (2005) uses return on equity (ROE) to measure firm value when investigating the impacts of capital structure on profitability of US companies. ROE and EPS were used by Chien-Chung et al. (2008) to identify firm value. Imad Ramadan (2015) used ROA as a proxy for firm value when studying the association between leverage and the Jordanian Firms’ Value. Some other studies such as those of Ben Naceur and Goaied (2002), Feng-Li and Chang (2008), use market-to-book-value ratio as a proxy for firm value. Earnings per share (EPS) and price earning ratio (P/E) were used as proxies for firm value by Mathanika et al (2015) in their study of the impact of capital structure on firm value in Srilanka. In addition, Chung and Pruitt (1994) and Feng-Li Lin (2010) use Tobin’s q to measure firm value. This study uses both ROE and ROA as proxies for firm’s value. The independent variables used in this study include the debt ratio measured by total liabilities to total assets and used as a proxy for capital structure (leveraging or gearing). Other variables were used in this study include those that may have influence on the capital structure and thus on the value of the include dividend policy, firm’s age, asset tangibility, firm’s size, ownership structure, growth opportunities, business risk, liquidity, and type of industry. Based on the above discussions and in order to explore the relationship between firms’ value and of leveraging (capital structure) the following 10 null hypotheses are formulated and used for testing: H1: There is no statistically significant relationship between firm’s value and its leverage (capital structure). H2: There is no statistically significant relationship between firm’s value and its dividend policy. H3: There is no statistically significant relationship between firm’s value and its age. H4: There is no statistically significant relationship between firm’s value and its assets tangibility. H5: There is no statistically significant relationship between firm’s value and its size. H6: There is no statistically significant relationship between firm’s value and its ownership structure. H7: There is no statistically significant relationship between firm’s value and its growth opportunities.

833


H8: There is no statistically significant relationship between firm’s value and its business risk. H9: There is no statistically significant relationship between firm’s value and its liquidity. H10: There is no statistically significant relationship between firm’s value and its type of industry. 3.2 The study model This study uses multiple-regression model to test the association between firm’s value and capital structure. Other control variables were also tested by the regression model comprising those that are believed to have influence on the decision to use debt financing or leveraging. The variables used in this study were determined based on the results reached by previous researches (prior studies) in addition to the availability of data. The econometrics model used expresses the firm’s value as a function of capital structure, dividend policy, age of the firm, tangibility of assets, size of the firm, ownership structure, growth opportunity, business risk of the firm, liquidity of the firm, and type of industry of the firm. The actual panel data obtained is analyzed through OLS regression. Since the efficiency of the estimates can be improved and thus the collinearity of the explanatory factors can be reduced, a panel data is used. A panel data approach is according to Joshua

Abor (2008) is more useful than either cross-section or time-series data alone. A multiple regression model is employed in this study as the study has more than one independent variable. The hypothesized independent variables include total liabilities to total assets ratio as a proxy for capital structure (leverage), dividends to net profit (DTNP) ratio as a proxy of dividend policy, number of years in business as a proxy of firm’s age (FAGE), fixed assets to total assets (FATTA) ratio as a proxy of tangibility of assets, natural logarithm of total assets (LNTA) as a proxy for size of the firm, ownership structure (OWNS) measured by a dummy variables where 1 denotes closely held companies and 0 denotes publicly held companies, price per share to book value per share (PTBV)ratio as a proxy of growth opportunities of the firm, the standard deviations of ROE of the firm (SDROE) as a proxy of business risk of the firm, total liability to total equity ratio (TLTE) as a proxy for liquidity, and type of industry (TYPE) denoted by dummy variables where 0 signifies industrial (manufacturing) and 1 signifies otherwise. The dependent variables used are return on Assets (ROA) for Model 1, and return on equity (ROE) for Model 2 measured by net income to total assets and net income to total stock holder’s equity, respectively. Following are the econometric regression models estimated to test the above-mentioned study hypotheses:

Firm’s value=f (TLTA, DTNP, FAGE, FATTA, LNTA, OWNS, PTBV, SDROE, TLTE, TYPE) ROA i, t = β0 + β1 TLTA i, t + β2 DTNP i, t + β3 FAGE i, t + β4 FATTA i, t + β5 LNTA i, t + β6 OWNS i, t +β7 PTBV i, t + β8 SDROE i, t + β9 TLTE i, t + β10 TYPE i, t + ε (1) ROE i, t = β0 + β1 TLTA i, t + β2 DTNP i, t + β3 FAGE i, t + β4 FATTA i, t + β5 LNTA i, t + β6 OWNS i, t +β7 PTBV i, t + β8 SDROE i, t + β9 TLTE i, t + β10 TYPE i, t + ε (2) Where: TLTA i, t = Total liability to total assets ratio representing leverage or debt ratio of firm i in time t β0: The intercept or constant amount β1 - β10 = Coefficients of the explanatory variables DTNP i, t = Dividends to net profit ratio to represent dividend policy of firm i in time t FAGE i, t = Number of years since the firm is in business to represent Age of firm i in time t FATTA i, t = Fixed assets to total assets ratio to represent tangibility of assets for firm i in time t LNTA i, t = Natural logarithm of total assets to represent the size of firm i in time t OWNS i, t = Dummy variables (0, 1) to represent ownership structure for firm i in time t PTBV i, t = price per share to book value per share to measure growth opportunities of firm i in time t SDROE i, t = Standard deviations of ROE to represent business risk of firm i in time t TLTE i, t = Total liquidity to total equity ratio to represent liquidity of firm i in time t TYPE i, t = Dummy variables (0, 1) to represent industry type of firm i in time t ε: the error term 4. Research results and discussion

4.1 Descriptive statistics

The following sections represent the study findings. In addition to the descriptive statistics, the findings include the correlation and regression analyses.

Table (1) shows the descriptive statistics for the study variables. It shows Mean, Maximum, Minimum, Standard deviation, Skewness statistics, kurtosis,

834


Jarque-Bera and probability for each of the dependent and independent variables. The average (Mean) leverage (TLTA) ratio, as can be seen in the Table, equals 36.6% which implies that firms in Kuwait in general and in the sample of the study in particular are not highly leveraged. The maximums and minimums show the ultimate highest and lowest values of the study variables. The low standard deviation values for most variables indicate that most of the firms are in the same range of value, leverage, dividends payout, growth opportunities, ownership structure, riskiness, tangibility and type, but not the firms’ age and liquidity as the standard deviations values of these two is quite high (11.528 and 9.604) respectively.

Table (1) shows that Skewness is positive for 8 out of 12 series indicating that fat tails on the right hand side of the distribution. Positive and negative signs of skewness values indicate that the results of this study are generally not normally distributed. Since values of kurtosis are deviated from 3, Kurtosis values show also that data is not normally distributed. To test for the normality of data the study uses Jarque-Bera statistics and its corresponding probability (probability). Based on these values the normality assumption is rejected at significance level of 1% (probability is less than or equal to 0.01) for all the variables except the size variable (LNTA) which shows a probability of (0.56).

Table 1. Descriptive Statistics

Mean Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Probability Observations

ROA 0.026 0.301 -0.397 0.086 -1.286 7.33 252. 0.00 239

ROE TLTA DTNP 0.067 0.366 0.370 7.8081 0.990 2.460 -0.758 0.010 -6.098 0.525 0.224 0.636 13.45 0.418 -3.687 199.7 2.210 47.28 39243 13.17 20066 0.000 0.001 0.000 239 239 239

FAGE FATTA LNTA 27.00 0.280 11.12 53.00 1.012 14.39 5.000 0.000 8.52 11.528 0.245 1.28 -0.131 0.815 0.16 1.911 2.832 2.87 12.50 26.76 1.13 0.0019 0.000 0.56 239 239 239

OWNS PTBV SDROE TLTE TYPE 0.31 1.1335 0.0927 1.576 0.393 1.00 4.900 5.847 148.2 1.000 0.00 -9.400 0.000 0.009 0.000 0.463 0.977 0.386 9.604 0.490 0.824 -4.555 14.01 14.98 0.437 1.678 59.01 209.2 229.0 1.191 44.41 32069. 43118 5177 40.20 0.00 0.000 0.000 0.000 0.000 239 239 239 239 239

4.2 Correlation Analysis Utilizing E-views analysis tool, the study uses correlation to test for multicollinearity of the variables. Table 2 displays the correlating analysis of study variables. Besides the degrees of correlation (association) between each pair of variables, the table shows the sign or the direction of association (positive or negative).

The Table reveals that none of the study variables have multicollinearity problem with each other as they are all shown to be low correlated. This indicates that none of the variables will be excluded from further analysis.

Table 2. Correlation Matrix

ROA ROE TLTA DTNP FAGE FATTA LNTA OWNS PTBV SDROE TLTE TYPE

ROA 1.000 0.167 -0.229 0.324 0.044 0.112 0.053 0.160 0.307 -0.183 -0.111 -0.137

ROE

TLTA

DTNP

FAGE FATTA LNTA

OWNS PTBV SDROE TLTE

TYPE

1.000 0.133 0.042 0.123 0.110 0.070 -0.009 -0.590 0.899 0.942 -0.062

1.000 -0.191 0.1930 0.042 0.4295 -0.082 -0.032 0.2031 0.284 0.447

1.000 -0.0118 0.081 0.0128 0.034 0.109 -0.085 -0.058 -0.098

1.000 0.127 0.211 0.334 0.025 0.098 0.144 -0.108

1.000 0.131 -0.055 -0.047 -0.095

1.000

1.000 -0.084 0.041 0.002 0.052 0.076 -0.023

The correlation matrix shows that DTNP, FAGE, FATTA, LNTA, OWNS, PTPV have positive relationships with ROA, which implies that the value of the firm represented by ROA increases as the values of dividend payout ratio, firms age, tangibility,

1.000 -0.088 -0.083 0.0299 0.100 0.166

1.000 -0.696 -0.676 0.101

1.000 0.966 -0.026

1.000 -0.017

size, ownership structure, and growth opportunities increase. The Table shows negative association between the value of the firm represented by ROA and each of TLTA, SDROE, TLTE, and TYPE, which indicates that the value of the firm decreases with an

835


increase in leverage, business risk, liquidity. It is also influenced negatively by the type of industry. The Table also shows positive relationships exist between ROE and each of TLTA, DTNP, FAGE, FATTA, LNTA, SDROE, and TLTE. This suggest that the value of the firm represented by ROE is positively influenced by using leverage, dividend payout ratio, age, size, business risk and liquidity of the firm. On the other hand the correlation matrix shows negative associations exist between ROE and each of OWNS, PTBV, and TYPE, which implies that, the value of the firm represented by ROE is negatively influenced by ownership structure, growth opportunities and type of industry of the firm. 4.3 Regression Analysis The study uses multiple regression analysis to examine the effect of the independent variables on the value of the firm. The study uses return on assets (ROA) and return on equity (ROE) as proxies for firm value. ROA is used as a proxy for firm’s value for Model 1 and ROE is used as a proxy of firm’s value for Model 2. Durbin-Watson statistics, p-value and adjusted R-squared were used by both models for decisionmaking criteria. To decide whether accept or reject the hypotheses, the study uses P-values (Prob.). The alternative hypothesis is accepted and the null hypothesis is rejected at 1% level of significance if the Prob. value is less than or equal to 0.01. Similarly, the alternative hypothesis is accepted at 5% and 10% level

of significance if the P-value is less than or equal to 0.05 and 0.10 respectively. The Adjusted R squared is used to measure goodness-of-fit that penalizes additional explanatory variables. The adjusted R squared value of 0.350524 s indicates that 35% of the variability of the value of the firm in Model 1 is explained by the independent variables. To test for first order serial correlation in the errors of a regression model, the study uses DurbinWatson Statistic method. Durbin-Watson helps in specifying the right combination of explanatory variables (Gujarati, 2004). It is also used to test the presence of autocorrelation in the residuals. The D-W statistic value of 1.865085 (very close to 2.0) indicates an absence of autocorrelation in model 1 and confirms that serial correlation is not existed. For such a number of observations, this indicates neither underestimation nor overestimation of the level of significance. The calculated F-statistic of 12.67721 at probability (F-value) of 0.0000 for the data regression (Model 1) indicates the null that all coefficients are simultaneously zero is rejected. This implies that the regression is generally significant. Table 3 displays the regression results of the independent and the independent (predictors) variables. It shows the relationships (degrees of association) between ROA and each of the independent variables (i.e., ROA (-1), TLTA, DTNP, FAGE, FATTA, LNTA, OWNS, PTBV, SDORE, TLTE and TYPE).

Table 3. Regression results between ROA and the independent variables using Least Square Method (MODEL 1) Variable C ROA(-1) TLTA DTNP FAGE FATTA LNTA OWNS PTBV SDROE TLTE TYPE R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

Coefficient -0.099978 0.169761 -0.124265 0.019692 -0.000423 0.023038 0.011423 0.016330 0.033112 -0.152349 0.008084 -0.004140 0.380542 0.350524 0.069523 1.097179 304.2284 12.67721 0.000000

Std. Error t-Statistic 0.045879 -2.179167 0.062137 2.732070 0.029963 -4.147275 0.007566 2.602494 0.000452 -0.935767 0.019116 1.205181 0.004181 2.732342 0.010636 1.535370 0.007064 4.687709 0.057076 -2.669253 0.002383 3.391561 0.010747 -0.385204 Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter. Durbin-Watson stat

Prob. 0.0303 0.0068*** 0.0000**** 0.0099*** 0.3504 0.2294 0.0068*** 0.1261 0.0000*** 0.0082*** 0.0008*** 0.7004 0.026130 0.086267 -2.445426 -2.270876 -2.375087 1.865085

***, **, and *, signify 1%, 5% and 10% respectively.

The Table shows ROA (-1) coefficient of 0.169761 is positive and statistically significant at 1%

level with a p-value of 0.0068. This suggests that oneyear-lagged ROA has a positive significant impact on

836


the value of the firm (represented by ROA). The Table also shows total liability to total equity (TLTA) coefficient of -0.124265 is negative and statistically significant at 1% level with a p-value of 0.0000. Therefore, the First hypothesis that there is no statistically significant relationship between firm’s value and its leverage (capital structure) is rejected, and thus, the alternative hypothesis is accepted. The negative sign of the relationship (coefficient) suggests that an increase in debt ratio (leverage) decreases the value of the firm. This can be explained by the fact that high leverage implies higher bankruptcy cost (and risk) for low quality firms (Ross 1977, and Leland and Pyle 1977) and). This result is consistent with the research results of Kinsman and Newman (1998) who pointed out that firm can maximize its value by choosing low debt or zero debt. It is also consistent with the results of Rajan and Zingales (1995) and Imad Ramadan (2015) who found leverage to be significantly inversely correlated with the firm value. This result can also be explained by the views of Modigliani and Miller (1963) who proven that a firms cost of equity increases as debt increases. However, it is not consistent with the results of Chowdhury and Chowdhury (2010) who found leverage measured by long term debt to total assets ratio to have positive coefficient with firm’s value. It is also not consistent with the results of ElKelish and Andrew (2007) who investigates the impact of financial structure on firm value in the United Arab Emirates and concluded that debt to equity ratio has no impact on the value of the firm. This result also contradicts with the irrelevance theory of Modigliani and Miller (1958) who postulate that there is no relationship between capital structure and firm’s value in the perfect world. Table 3 reveals that there is a significant positive relationship exists between ROA and the dividends payout ratio with level of significance of 1% and a pvalue of 0.0099. Therefore, the Second hypothesis that there is no statistically significant relationship between firm’s value and its assets tangibility is rejected and thus, the alternative hypothesis is accepted. The positive sign of the coefficient value of the firm and its payout ratio suggests that as dividends payout ratio increases, the value of the firm increases. This suggests that firms may increase their values through paying more dividends to their shareholders. This result comports with the research results of Chowdhury and Chowdhury, (2010) who found dividend payout ratio to have positive coefficient with the value of the firm in Bangladesh. The empirical results reveal an insignificant negative association exists between firms age (FAGE) and ROA with a coefficient of -0.000423 and p-value of 0.3504. This implies that the Third hypothesis that there is no statistically significant relationship between firm’s value and its age is accepted. This indicates that firm’s age is not a significant factor in explaining firm’s value when measured by ROA. This result is consistent with the research results of Bender

and Ward (1993) who maintained that the capital structure could be affected by the firm’s life stage, as financing needs could vary once firm’s circumstances do. They also maintained that business risk decreases with the progress of the firm’s age, allowing financial risk to increase, and thus the value of the firm to decrease. This result is also consistent with the research results of Frielinghaus et al. (2005) who concluded that mature companies have more debt in their capital structure. The empirical results an insignificant positive association exists between tangibility of the firm proxied by fixed assets to total assets (FATTA) and the value of the firm as measured by ROA. This indicates that the Fourth hypothesis that there is no statistically significant relationship between firm’s value and its assets tangibility accepted. This result is consistent with the research results of Imad Ramadan (2015) who find asset structure (measured by fixed assets to total assets) to be significantly positively correlated with the firm value expressed as ROA. The Table also displays positive and significant relationship exists between ROA and Size of the firm measured by the natural logarithm of total assets (LNTA) at 1% level of confidence with a coefficient of 0.011423 and p-value of 0.0068. This means that the Fifth hypothesis that there is no statistically significant relationship between firm’s value and its size is rejected and, thus, the null hypothesis is accepted. This result implies that the value of the firm increases as its size increases. This result is comports with the results of Imad Ramadan (2015) who find firm size to be significantly positively correlated with the firm value expressed as ROA. However, it is not consistent with the results of Booth et al. (2001) who concluded that profitability has an inverse relationship with debt level and size of the firm. Ownership structure (OWNS) is revealed by the results to have positive and statistically insignificant association with ROA with a coefficient of 0.016330 and p-value of 0.1261. This means that the Sixth hypothesis that there is no statistically significant relationship between firm’s value and its ownership structure is accepted. The positive sign of the relationship in this result indicates that closely held companies have more value than publicly held companies. However, this finding does not comport with the results of Chowdhury and Chowdhury, (2010) who found public shareholding to have negative impact on the value of the firm. The empirical results show statistically significant positive relationship at 1% level exists between ROA and growth opportunities of the firm with p-value of (0.0086). Therefore, the Seventh hypothesis that there is no statistically significant relationship between firm’s value and its growth opportunities is rejected and, thus, the alternative hypothesis is accepted. This suggests that firms with more growth opportunities have higher values than those with less growth opportunities. This finding is

837


consistent with the research results Imad Ramadan (2015) who find sales growth to be significantly positively correlated with the firm value expressed as ROA. However, it is does not comport with the results of Chowdhury and Chowdhury (2010) who found sales growth to have negative coefficient with the value of the firm. Table 3, shows the coefficient of business risks, measured by the standard deviations of return on equity (SDROE), of -0.152349 is statistically insignificant at 10% level with p-value of 0.0082. Therefore the Eighth hypothesis that there is no statistically significant relationship between firm’s value and its business risk is rejected and thus the alternative hypothesis is accepted. The negative sign of the coefficient of this result indicates that companies with high business risk have less value than those having less risk.

The results show that liquidity of the firm proxied by total liquidity to total assets (TLTA) is positive and statistically significant at 1% level with P-value of 0.0008. The positive association between these two variables implies that firms with higher liquidity have higher value. Therefore the Ninth hypothesis that there is no statistically significant relationship between firm’s value and its liquidity is rejected and, thus, the null hypothesis is accepted. The results also show type of industry of the firm (TYPE) has a negative and statistically insignificant relationship with ROA at 10% level and a p-value of 0.7004. This implies that type of industry is not a significant factor in influencing the value of the firm. Therefore the Tenth hypothesis that there is no statistically significant relationship between firm’s value and its type of industry is accepted.

Table 4. Regression results between ROE and explanatory variables using Least Square Method (MODEL 2) Variable C ROE(-1) TLTA DTNP FAGE FATTA LNTA OWNS PTBV SDROE TLTE TYPE R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

Coefficient -0.106094 0.376295 -0.377463 0.023899 -0.000844 0.040573 0.014444 0.016930 0.037401 -0.015631 0.058531 0.023057 0.935548 0.932425 0.136447 4.226240 143.0742 299.5451 0.000000

Std. Error t-Statistic 0.090078 -1.177799 0.069234 5.435107 0.059263 -6.369301 0.014778 1.617172 0.000888 -0.950038 0.037587 1.079455 0.008200 1.761546 0.020832 0.812700 0.014106 2.651445 0.115181 -0.135711 0.004741 12.34478 0.021205 1.087316 Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter. Durbin-Watson stat

Prob. 0.2401 0.0000*** 0.0000*** 0.1072 0.3431 0.2815 0.0795* 0.4172 0.0086*** 0.8922 0.0000*** 0.2781 0.066962 0.524892 -1.096855 -0.922305 -1.026516 1.817459

***, **, and *, signify 1%, 5% and 10% respectively.

On the other hand, the study uses ROE as another model (Model 2) to investigate the influence of capital structure (leveraging) and the other control variables on the value of the firm. Table 4 below shows the multiple regression results of Model 2 of the study. It shows the regression analysis between return on equity (ROE) and each of the independent variables (i.e., ROE (-1), TLTA, DTNP, FAGE, FATTA, LNTA, OWNS, PTBV, SDORE, TLTE and TYPE). The Table shows statistically significant association exists between ROE and leverage ratio (a proxy for capital structure) of the firm and represented by total liability to total assets (TLTA) ratio. This indicates that capital structure is a significant factor in influencing the value of the firm when measured by ROE. Leverage ratio has the highest coefficient of -

0.377463 and this designates that this variable is the most influential. The negative sign of the coefficient implies that the value of the firm is inversely affected by the use of debt (leveraging). In other words, the study determines that the higher the leverage ratio, the lower the value of the firm. A significant positive relationship is found between ROE and the one-year-lagged return on equity (ROE (-1)), which implies that previous year’s value explains the current year’s value of the firm. The Table also reveals positive and statistically significant associations between value of the firm (measured by ROE) and each of LNTA, PTBV, and TLTE. This implies that size of the firm, growth opportunities, and liquidity of the firm are major determinants of the firm’s value. The remaining variables are found to be statistically insignificant in

838


determining the value of the firm when proxied by ROE. Conclusions This study has empirically examined the impact of capital structure and some other firm-characteristics variables on the value of the firm. Data were obtained from 48 companies listed in the Kuwait stock exchange. The sample was selected from multiple sectors including manufacturing, basic materials, oil and gas, and services. Some sectors were considered as having special characteristics or considered as highly leveraged were excluded from the analysis. In Model 1 (i.e., where ROA is used as a proxy of the firm’s value), the study findings reveal that capital structure (leverage) is a significant determinant of the value of the firm. The results also reveal that business risk, previous year value (one-year lagged ROA), dividends payout ratio, size, growth opportunities and liquidity of the firm are significant determinants of the firm’s value. On the other hand the study found insignificant association between firm’s value measured by ROA and firm’s age, tangibility (asset structure), ownership structure, and type of industry of the firm. In model 2 (i.e., where ROE is used as a proxy of the firm’s value), the findings revealed that capital structure (captured by debt ratio) as the most influential factor in explaining the value of the firm. The empirical results also reveal that previous year’s value explains the current year’s value of the firm. In addition, the study reveals that size, growth opportunities, and liquidity of the firm are major determinants of the firm’s value when proxied by ROE. The variation in the results of model 1 of the study (when ROA is taken as the dependent variable) and model 2 (when ROE is taken as the dependent variable) is perhaps due to the differences in the sizes of the selected firms and their market shares. Most of our results are consistent with those in the literature. However, the inconsistencies between this study’s findings and some of those in the previous literature is, probably, due to the dissimilarities in the country or countries used as a home or subject of study and because of the variation(s) in the timing used as a period for data collecting. The study is valuable to managers, academicians, policy makers and other stakeholders as it fills the gap of literature by providing evidence of the impact of capital structure and other firm specific variables on the value of the firm in Kuwait. References: 1. Ben Naceur, Samy and Goaied, Mohamed (2002), “The relationship between dividend policy, financial structure, profitability and firm value”, Applied Financial Economics, 2002, 12, pp. 843-849. 2. Bender, Ruth and Ward, Keith (1993), Corporate Financial Strategy, Oxford: Butterworth-Heinemann.

3. Booth, L., Aivazian V, Demirguc-Kunt A, Maksimovic V. (2001), “Capital structure in developing countries”, The Journal of Finance, Vol. LVI, issue: 1, pp. 87-130. 4. Chien-Chung Nieh, Hwey-Yun Yau, and Wen-Chien Liu (2008), “Investigation of target capital structure for Electronic listed firms in Taiwan”, Emerging Markets Finance & Trade, Vol. 44, issue: 4, pp: 75–87. 5. Chowdhury, Anup and Chowdhury, Suman P. (2010) “Impact of capital structure on firm’s value: Evidence from Bangladesh”, vol. 3, Issue: 3, pp: 111-122. 6. Chung, K. H. and S. W. Pruitt (1994), “A simple approximation of Tobin’s q”, Financial Management, vol. 23, issue: 3, pp: 70-74. 7. ElKelish Walaa W. and Marshal Andrew (2007), "Financial structure and firm value: empirical evidence from the United Arab Emirates", International Journal of Business Research, vol. VII, issue: 1, pp: 69-76. 8. Feng-Li Lin (2010), “A panel threshold model of institutional ownership and firm value in Taiwan”, International Research Journal of Finance and Economics, pp. 54-62. 9. Feng-Li Lin and Tsangyao Chang (2008), “Does ownership concentration affect firm value in Taiwan? A panel threshold regression analysis”, The Empirical Economics Letters, vol. 7, issue: 7, pp: 673-680. 10. Frielinghaus, A., Mostert, B. and Firer, C. (2005), “Capital Structure and the firm’s life stage”, South African Journal of Business Management, vol. 36, issue: 4, pp: 9-18. 11. Gill, Amarjit, Nahum Biger, and Neil Mathur (2011), “The effect of capital structure on profitability: Evidence from the United States”, International Journal of Management, Vol. 28, issue: 4, pp: 3-15. 12. Imad Zeyad Ramadan (2015), “Leverage and the Jordanian Firms’ Value: Empirical Evidence”, International Journal of Economics and Finance, vol. 7, issue: 4, pp: 75-81. 13. Joshua Abor (2005), “The effect of capital structure on profitability: an empirical analysis of listed firms in Ghana”. Journal of Risk Finance, vol. 6, issue: 5, pp: 438-445. 14. Kinsman, Michael and Newman, Joseph (1998) “Debt tied to lower firm performance: Finding calls for review of rise in debt use”, Graziadio Business Review, vol. 1, issue 3. 15. Leland, Hayne E. and Pyle, David H. (1977) “Informational asymmetries, financial structure, and financial intermediation”, Journal of Finance, vol.32, issue: 2, pp.371-387. 16. Maxwell, O and Kehinde, E (2012), Capital Structure and Firm Value: Empirical Evidence from Nigeria, International Journal of Business and Social Science, Vol. 3 No. 19; pp, 252-261 17. Mathanika .T, Virginia Vinothini. A.G and Paviththira R. (2015), “Impact of Capital Structure on Firm Value: Evidence from Listed Manufacturing Companies on Colombo Stock Exchange (CSE) In Srilanka”, Proceeding of International Conference on Contemporary Management - (ICCM-2015), pp: 24-35. 18. Mwangi, L, Makau, M and Kosimbei, G (2014), “Relationship between Capital Structure and Performance of Nonfinancial Companies Listed In the Nairobi Securities Exchange, Kenya. Global Journal of Contemporary Research in Accounting, Auditing and Business Ethics (GJCRA). Vol: 1 Issue 2. pp 72-90 19. Modigliani, F. and Miller, M. H. (1963), “Corporate Income Taxes and the Cost of Capital: A Correction”,

839


American Economic Review, vol. 53, issue: 3, pp: 433443. 20. Modiglinai, Franco and Merton, H. Miller (1958), “the Cost of Capital, Corporate Finance and the Theory of Investment”, American Economics Review, volume XLVIII, issue: 3, pp. 261 – 297. 21. Rajan, R. G. and Zingales, L. (1995), “What do we know about capital structure? Some evidence from international data”, The Journal of Finance, vol. 50, issue: 5, pp: 1421-1460.

22. Ross, Stephen A. (1977) “The determination of financial structure: The incentive signaling approach,” Bell Journal of Economics, vol. 8, issue: 1, pp. 23-40. 23. Yu-Shu Cheng, Yi-Pei Liu and Chu-Yang Chien (2010), “Capital structure and firm value in China: A panel threshold regression analysis”, African Journal of Business Management Vol. 4, issue: 12, pp: 2500-2507. 24. Sunder, L., & Myers, S. (1999). Testing static tradeoff against pecking order models of capital structure. Journal of Financial Economics, 51 (2), 219-244.

840


CORPORATE GOVERNANCE ON FINANCIAL PERFORMANCE OF INSURANCE INDUSTRY Gardachew Worku Fekadu*

Abstract The role of corporate governance in financial institutions differs from that of nonfinancial institutions for the discretionary power of the board of directors would be limited especially in regulated financial systems where financial institutions are obliged to function through legislative and prescriptive procedures, policies, rules and regulations. This study, therefore, was aimed at examining the impact of corporate governance on the performance of closely regulated Ethiopian insurance Industry. The study employed explanatory research design with an econometric panel data of 10 Insurance companies that covers the period 2007 to 2014. Board size, board independence and board diversity have negative and insignificant effect on the performance of insurance companies while size and independence of audit committee and frequency of board meetings have positive but insignificant effect on the performance of insurance companies in Ethiopia. Thus it could be concluded that all corporate governance mechanisms have insignificant effect on the performance of insurance companies measured by return on asset. This vividly affirms that the role of board of directors in closely regulated financial sector is dismal and insignificant for they have limited discretionary power to exercise as board of directors. Thus it would be recommendable if the regulatory body could relax its prescriptive and stringent policies and devolve its power to board of directors without endangering the viability of insurance companies. Keywords: Corporate Governance, Insurance Companies, Board of Directors, Ethiopia * University of South Africa, Pretoria, South Africa Acknowledgements I greatly appreciate my mentor Prof. Daniel Makina, Department of Finance, University of South Africa, for constructive comments on the paper.

1. Introduction Corporate governance can be defined as the relationship among shareholders, board of directors, top management, employees, regulators, any other stakeholders and the community in determining the direction and performance of the corporation (Ruin, 2001). The boards of directors have an important role in alleviating the agency costs that arise from the separation of ownership and decision control in corporations (Fama and Jensen, 1983). Short et al. (1999) also took this view and argued that the boards of directors are the central corporate governance control mechanism responsible for monitoring the activities of managers, whilst Jensen (1993) describes the board of directors as the apex of the internal control mechanism in an organization. Therefore, the sole existence of board of directors is to protect the interests of shareholders from where it receives its authority for internal control (Jensen, 1993). Firms with better corporate governance mechanisms continue to attain organizational objectives and goals than those that do not have (Bradley (2004). Adams and Mehran (2003), argues

that organizations with better systems and procedures are important for firms’ performance. Better policies and procedures have been recognized as a significant factor in improving financial performance of organizations. Many authors argues that if an organization pays attention in having and following systems, then it will be in position to generate better returns to its shareholders (Matama, 2005; Gompers et al. (2003). Corporate Governance is aimed at ensuring proper governance of business as well as complying with all the governance norms prescribed by regulatory body for the benefit of all interested parties including society. The board of directors has an important role in alleviating the agency costs that arise from the separation of ownership and decision control in corporations (Cheung and Chan, 2004). Studies also show that corporate governance in financial institutions differs from that in the nonfinancial institutions because of the broader risk that financial firms pose to the economy. As a result the regulator plays a more active role in establishing standards and rules to make management practices in financial institutions more accountable and efficient and hence financial sector regulators place additional

841

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 responsibilities on board of directors’ that often result in detailed regulations regarding their decisionmaking practices and strategic aims. These additional regulatory responsibilities for management have led some experts to observe that banking regulation is a substitute for corporate governance (Renee and Hamid, 2003). The apex regulatory body is more active than the board of directors of financial institutions in devising governance standards, promoting the interests of shareholders, depositors and other stakeholders. Thus the role played by directors of financial institutions would not be at par with non-financial institutions. Corporate governance matters are governed and driven by countries’ company’s codes, securities and exchange commissions, the stock exchange listing requirements, regulations and rules and other country specific regulatory agencies (Anthony, 2007). In Ethiopia where the financial sector is closed and subject to stringent regulatory system, exploring impact of corporate governance mechanisms on the performance of insurance industries is quite interesting. The extant literatures explain only the effect of corporate governance on the performance of firms (Matama, 2005; Joan, et.al, 2010; Rashid, 2011; David and Tobias 2013). These studies attempted to explain the effect of corporate governance on the performance firms in relatively liberalized economy where the board of directors has more discretionary power to exercise their power and to make decisions that they felt are more worthy to their firms. This paper argues that board of directors have limited role to play in Ethiopian insurance companies, where they are obliged to implement legalistic and prescriptive policies, procedures, rules and regulations set by the regulatory body, National Bank of Ethiopia. The regulatory body prohibits foreign entry into the industry, regulates capital requirements, imposes ownership limitations, restricts types of business and investment insurance companies could engage and the amount of investment they could make, remuneration and appointment of executives and inter alia. This study aims to explain the role of board of directors on stringently regulated insurance industry in Ethiopia and would contribute its part in explaining the role of corporate governance mechanisms in the absence of capital markets in closed Ethiopian Financial sector. The rest of the paper is organized as follows. Section 2 presents review of related literature. Section 3 describes the research deigns and methodology. Section 4 deals with results and discussion. Section 5 presents the conclusion and policy implications of the findings.

2. Literature Review 2.1 Review of Theoretical Literature There are various theories that can be used to explain corporate governance conventions and also the issues that arise as a result of these conventions (Rashid, 2011). These theories include the agency theory, stakeholder theory, stewardship theory and resource dependency theory (Sanda and Garba, 2005; David and Tobias 2013). These four theories as the main and most significant theories of corporate governance are explained further respectively below. 2.1.1 Agency Theory The essence of this theory is based on the existence of separation between ownership and management of corporations. In such corporations, the managers (agents) are hired to work and make decision on behalf of the owners (principals) in order to maximize return to the shareholders. However, the managers (agent) who are put in control of the affairs of the organization may not always consider the best interest of the owners and may pursue their self-activities to the detriment of the welfare of the principals (David and Tobias 2013). As a result of these agency problems, the principal might end up incurring costs known as Agency costs. This Agency cost is a value loss to the shareholders and usually involves the cost of monitoring the activities of managers so that goal congruence can be achieved between shareholders and managers. The effect of this agency theory is that one can only try to mitigate against this agency problem when the board is composed largely by non-executive directors (independent and dependent) who will be able to control the activities of managers and thereby maximize shareholders‘wealth. The governance structures suggested by the agency theory involve size of the board, composition of the board, remuneration to CEO, directors ‘shareholding and shareholder right (Luan & Tang, 2007; Rashid, 2011). The theory also suggests that the role of the chairman and the role of the CEO should not be occupied by the same person as this can limit the monitory role bestowed on the board of directors and can also have a negative impact on the performance of the firm. It was suggested that the reason for limit in the monitory role by the board will be loss of board independence as a result of CEO duality (Elsayed, 2007 and Kang &Zardkoohi, 2005). This theory is based on the belief that there is a basic conflict of interest between the owners and managers of the company (Kiel & Nicholson, 2003). 2.1.2 Stewardship theory A steward is defined by Davis, Schoorman & Donaldson (1997) as one who protects and maximizes

842


shareholders wealth through firm performance, because by so doing, the steward’s utility functions are maximized. In this perspective, stewards are company executives and managers working for the shareholders, protects and make profits for the shareholder (cited in David and Tobias 2013). This theory is a contrast or a direct opposite to the agency theory and this theory adopts a more idealistic view of humans. This theory is based on a belief that the agent is not self-opportunist but a steward that perceives greater utility in the interest of the principal and the organization as a whole. The theory assumes that a significant correlation exist between the firm‘s success and the manager‘s satisfaction. This trade-off is achieved by the steward admitting that working towards achieving company‘s and collective goals will lead to self actualization. The theory argues for the post of Chief Executive Officer and Chairman to be held by the same person. Therefore, control lowers the motivation of steward and weakens motivational attitude (Davis et al., 1997). Stewardship theory poses that stewards are likely to ignore selfish interests in order to pursue the best interest of the firm. Davis et al, (Ibid) observed that when a steward has been in a company for so long, the steward and the firm becomes one entity. Instead of using the firm for their own selfish interest, the stewards seems to be more in ensuring the continuous existence and long term success of the firm because they now see the firm as an extension of themselves (David and Tobias 2013). 2.1.3 Stakeholder theory The other popular theory of corporate governance is the Stakeholder theory. The stakeholder theory originated from the management discipline and gradually developed to include corporate accountability to a broad range of stakeholders (Abdullah and Valentine, 2009). Unlike the agency theory, whereby managers are predominantly responsible for satisfying the interests of shareholders, stakeholder theory maintains that managers in organizations are not only responsible for the interests of shareholders but also for a network of relationships to serve which includes the suppliers employees and business partners (Ibid). According to stakeholder theory decisions made regarding the company affect and affected by different parties in addition to stockholders of the company. Hence, the managers should on the one hand manage the company to benefit its stakeholders in order to ensure their rights and their participation in decision making and on the other hand the management must act as the stockholder’s agent to ensure the survival of the firm to safeguard the long term stakes of each group (Fontain et al., 2006).

2.1.4 Resource Dependency Theory Resource dependency theory concentrates on the role of board directors in providing access to resources needed by the firm while the stakeholder theory focuses on relationships with many groups for individual benefits. Hillman et al. (2000) contend that resource dependency theory focuses on the role that directors play in providing or securing essential resources to an organization through their linkages to the external environment. Indeed, Johnson et al, (1996) concurs that resource dependency theorists provide focus on the appointment of representatives of independent organizations as a means for gaining access in resources critical to firm success. It has been argued that the provision of resources enhances organizational functioning, firm’s performance and its survival (Daily et al, 2003). 2.2 Empirical literature There are scanty and inconclusive studies and findings on the effect of corporate governance mechanism on the financial performance of Insurance companies in developing countries (Joan, et.al, 2010). In many of empirical studies corporate governance mechanisms had been dealt and categorized as endogenous and exogenous governance mechanisms. Endogenous corporate governance mechanisms are otherwise known as internal corporate governance. Internal corporate governance is about mechanisms for the accountability, monitoring, and control of a firm’s management with respect to the use of resources and risk taking this starts with the board of directors which is the supreme governing body of insurance company. Exogenous corporate governance mechanisms are external governance mechanisms related external force and regulation with the power to discipline the agent (Joan, et.al, 2010; Sapovadia, 2009). Many researchers argued that the board of directors is the central corporate governance control mechanism responsible for monitoring the activities of managers and improving the performance of firms and board of directors have been described as the apex of the internal control mechanism in an organization (Jensen (1993; Hillman et al. 2000; Joan, et.al, 2010). It is constantly debated what the right mix of governance structure (size of the board, composition of the board and frequency of board meetings should be and how a company performs is dependent on these governance structures (Das and Gosh, 2004). AlHawary, S (2011) Investigated the effect of governance mechanisms such as board size, CEO duality, percentage of non-executive directors, capital adequacy, the ownership percentage of large shareholders, and the ownership percentage of the largest shareholders of Jordanian commercial banks as measured by Tobin’s Q and found that CEO duality, and percentage of nonexecutive directors had

843


statistically significant positive effect on performance; whereas leverage had statistically significant negative effect on performance. With regard to CEO-Chairman duality, National bank of Ethiopia has already prohibited the duality of the CEO in contrary to what is common in the boards of some firms in some countries that allow the CEO to be a board chairman. This CEO-Chairman split in the Ethiopian case is a positive move towards more independent boards to discharge their oversight and monitoring role and this variable could not be an issue for consideration. The empirical review of the literature has focused on more relevant explanatory variables that are deemed to have effect on the performance of Ethiopian Insurance companies. These endogenous explanatory variables considered in this paper include boar size, board independence, size and independence of audit committee, board diversity and frequent of board meeting. Board size is defined as the number of directors on the board. There is a view that larger boards are better for firm value because they have a range of expertise to help make better decisions, and are harder for a powerful CEO to dominate. Dallas, G (2004), states that the size of the board has positive effect and is an important governance consideration. However, some authors have advocated for smaller boards. Fama & Jensen (1983) argue that large boards are less effective and are easier for the CEO to control. When a board gets too big, it becomes difficult to coordinate, encourages free riding and poses problems. Smaller boards however reduce the possibility of free riding, and increase the accountability of individual directors. Hence there will be a positive or negative relationship between board size and firm value. Adetunji and Olawoye (2009) argue that board size determines the number of directors in a board and the board should be of reasonable size, and the terms of its directors should be fixed and advocates for optimal size of board of directors for good corporate governance as well as performance in the firm. Of course the National bank of Ethiopia (2014) has stated the minimum number of board size of an insurer to be nine. The question “what would be the optimum board size remains debatable and inconclusive (Houssem and Ines, 2011; Ishaya, Francis and Solomon, 2013; Adeusi et al., 2013; Musa et al., 2013; Turku, 2014; Anthony, 2007). The aforementioned empirical review of the literature leads to develop the hypothesis that board size has positive and significant effect on the performance of insurance company. Board independence: sometimes called board composition is measured as the ratio of independent (external) board members to the total number of board members. There are empirical evidences supporting that the higher proportion of outsiders on a board can better monitor and control the opportunistic behavior of the incumbent management, thus, minimizing the agency problem and maximizing shareholders' wealth

(Martin and Sebastian, 2011; Anthony, 2007; Cassandra et al., 2009; Lorne and Jun , 2012; Adeusi et al.,2013; Musa et al., 2013). Of course negative association between board composition and firm performance was reported by Agrawal and Knoeber (1996), who find that more outsiders on the board negatively affect the performance and conclude that outsiders are added on boards for political reasons and they reduce performance directly or by proxy for the underlying political constraints that led to their board memberships. With such inconclusive findings in this study it has been hypothesized as the board independence has positive and significant effect on financial performance of insurance company Audited Committee size and independence: Review of the literature has revealed that existence of independent and competent audit committees has positive effect on firm performance (Anthony, 2007; Cassandra et al., 2009). Audit committee help to ensure that accounting policies are sound and financial statements are properly prepared and audited. Moreover, the existence of audit committee composed of external board members in the firm will create a transparent and credible environment between management, external auditors and the board members. The evidence suggests that existence of audit committee improves governance quality and financial performance of firms (Defond, et al., (2005; Green, 2005). Thus it has been the hypothesized as there is positive relationship between independent audit committee and financial performance of insurance company. Board diversity: In recent years, there has been an increasing interest in investigating the impact of gender diversity on the firm’s performance, which is whether the addition of women to the board affects performance, and a number of research projects have attempted to provide evidence for this argument. The empirical study by Smith et al., 2005; Huse, 2007 and Mersland and Strom, 2007) have found that the presence of women in the board positions have a positive effect on the firm’s performance. Thus it could be hypothesized as board diversity has positive effect on the financial performance of insurance company. Frequency of board meetings: frequency of boar meeting as corporate governance are considered as important proxies for the time directors spend monitoring managerial performance and also as an important resource in improving the effectiveness of a board (Funmi,2014). When boards hold regular meetings, they are more likely to remain informed and knowledgeable about relevant performance of the company leading them to take or influence and direct the appropriate action to address the issue (Adams, 2000; Abbott et al., 2003; Funmi, 2014). Indeed Jensen (1993) found negative relationship and suggests that board meetings were a reactive response and not a proactive measure.

844


The negative association between frequency of board meeting and performance was further confirmed by Vafeas (1999) and Karamanou and Vafeas (2005). National bank of Ethiopia (2014) stated the board shall set up and put in use rules for the manner of conducting board meetings and Board meetings shall be held at least once in a month. Keeping in view of these empirical literatures and regulatory requirement, frequency of board meeting has positive and significant effect on financial performance of insurance company. 3. Research specification

Methodology

and

Model

3.1 Sample Selection According to the data gathered from the National Bank of Ethiopian website the numbers of insurance companies operating in the market at the end of June 2014 were 16 in numbers. Of these 6 insurance companies were established since 2010 and are in operation for few years. Thus purposively, 10 of the insurance companies were included in the sample for analysis. 3.2 Source of Data Data for this study is secondary data obtained from audited annual financial statements of the respective insurance companies and from the website of National bank of Ethiopia. The study included 10 Insurance companies. The study used panel data of 10 Insurance companies that covers the period 2007 to 2014 (10 years) (80 observations). 3.3 Selection of the variables Corporate governance mechanisms such as boar size, board independence, Audit committee size and independence, frequent of boar meeting, board diversity were selected as endogenous independent variables in the study. Age and size of the insurance companies were also incorporated as control variables. ROA was considered as dependent variables. 3.4 Econometric Model specification A quantitative method of data analysis was employed which involved descriptive and inferential statistical analysis. The descriptive statistics were used to analyze the means and standard deviations of regression variables. The assumptions and tests of Classical Linear Regression Model (CLRM) were tested before conducting regression analysis. The following regression model was used to explain the effect of corporate governance mechanisms on financial performance of insurance company:

ROAit = α0 + a1BSIZE + a2BIND + a3ACSIZE + a4ACIN+ a5BDIV + a6FBM + a7FSIZE + a8FAGE+ E Where: α0= Intercept ROA=Dependent variable, Return on Asset BSIZE= Board size representing the number of directors sitting in the board BIND= the percentage of external board members to the total number of board members ACSIZE=Number of Audit committee members ACIN: percentage of independent audit committee members to total audit committee members BDIV: Board diversity, Percentage/proportion of women in the board FBM= Frequency of board meetings (number of board meetings per year) FSIZ= Firm Size as natural logarithm of total assets of an insurance company FAGE= Age of the insurance company in years 3.5 Diagnostic Test for the Regression Assumptions Diagnostic test were conducted by using STATA version 12. The goodness of fitness of the model was tested through ANOVA and F-statistic and was proven that the explanatory variables used in the model actually explain the variations in the dependent variable (ROA). The correlation matrix, Variance Inflation Factor (VIF) and Tolerance values shows that there is not multicolinearity among the explanatory variables. Normality and heteroscedasticity tests also portrayed that the normality, homoscedasticity assumptions of the regression model were satisfied to run the regression analysis. Finally, to select a best fitted model between the alternatives of random effect model and fixed effect model, Hausman test was conducted. The p-value of Hausman test is 0.0358 which is less than the level of significance (0.05). Therefore, the null hypothesis which claims the unique errors (Ui) are not correlated with the repressors was rejected and fixed effects become more appropriate than Random effect. Further test was also conducted to choose between fixed effects versus pooled OLS regression model by using Breush and pagan Lagrangian multiplier test and the result shows that Fixed effect is fitted for the study since the P-value is 0.0004 which is less than the significant level (0.05). Therefore, our suitable model could be fixed effect model. 4 Analyses and Interpretation of Results Descriptive, correlation and regression analysis and interpretation of the results were made hereunder.

845

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 4.1 Descriptive statistics Table 1 reports the descriptive statistics of the dependent and independent variables for the panel period from 2007 to 2014. The table presents the

mean, minimum, maximum and standard deviation for the panel data variables for the period from 2007 – 2014.

Table 1. Descriptive Statistics Variables ROA BSIZE BIND ACSIZE ACIN BDIV FBM FSIZE FAGE

Obs 80 80 80 80 80 80 80 80 80

Mean 0.077 9.6 0.74 3.4 0.65 0.13 15.6 19.11 15.6

Std. Dev. 0.04 1.59 0.14 0.49 0.19 0.23 2.7 1.04 8.0

Min -0.01 7 .42 0 0 0 11 16.9 1

Max 0.18 16 0.89 4 0.75 0.20 21 21.2 39

Source: Author’s computation

For the ten insurance companies included in this study, the average return on asset (ROA) was 7.7% while the minimum and maximum return being negative 1% and 18%, respectively with standard deviation of 4%. The average size of the board was nearly 10 with minimum size of 7 and maximum of 16 board members with significant standard deviation of 159%. With regard to board independence on the average 74% of the bard members are external with minimum and maximum percentage of 42% and 89%, respectively, implying that board of directors are relatively independent as they are mostly dominated by non executive directors. The average size of audit committee members was nearly 3 with minimum size of 0 and maximum of 4 audit committee members. The average percentage of independent audit committee members were 65% with minimum and maximum of 0 and 75%, respectively. The minimum audit committee size of zero and Audit committee independence of zero implies that in some of the insurance companies audit committee was not established during that specific observation period. With regard to board diversity, on the average only 13% of the board members are composed of

female directors with minimum of 0 and 20%, implying that there are insurance companies whose boards of directors are 100% composed of males. The average number of board meetings was nearly 16 times per year while the minimum and maximum numbers of bear meetings were held for 11 and 21 times per year. The size of the sampled insurance companies, taken as the logarithmic of total asset, indicated a mean value of 19.11, and a minimum and a maximum value of nearly 17 and 21, respectively. The average age of the insurance companies were nearly 16 years with minimum and maximum of 1 and 39 years. This shows that except the state owned Ethiopian insurance company, all private insurance companies were established following deregulation of the financial sector since 1994. 4.2 Correlation matrix Table-2 shows the summary of correlation coefficient between dependent variables (ROA) and explanatory variables. From the table it was observed that multicollinearity was not a threat to the model variables.

Table 2. Correlation matrix --ROA BSIZE ROA 1.0000 BSIZE 0.1188 1.0000 BIND 0.0327 -0.6005 ACSIZE -0.1447 0.1487 ACIN -0.0105 0.2135 BDIV -0.1129 -0.1074 FBM 0.4120 -0.1864 FSIZE 0.4643 0.1362 FAGE 0.4201 -0.0131 Source: Author’s computation

BIND

ACSIZE

ACIN

BDIV

FBM

FSIZE

FAGE

1.0000 0.1097 0.1217 0.2026 0.0739 0.0747 0.1575

1.0000 0.4234 0.1585 -0.0249 -0.0573 -0.0199

1.0000 0.1409 -0.0768 -0.0108 -0.0341

1.0000 0.0222 0.1239 0.0182

1.0000 0.4756 0.5115

1.0000 0.7342

1.0000

846


As portrayed in table-2, board size, frequency of board meetings, firm size and age have positive correlation with the performances of the insurance companies. However independence of the board, size of audit committee and its independence and board diversity have negative relationship with the profitability of the insurance companies. The correlation matrix also revealed the relationship between explanatory variables. The correlation matrix has also indicated that the multicolinearity is not a threat to the model variables as all correlation coefficients are below the threshold level of 0.8 (Gujarati, 2004).

4.3 The Regression Interpretation

Analysis

and

The regression analysis in Table-3 shows the impact of corporate governance mechanisms on the performance of Ethiopian insurance companies. As it has been already discussed in the research methodology fixed effect model was found fit for panel data analysis and therefore, our empirical analysis were based on fixed effect model. The pvalue of” F” was 0.0053, which is less than5% and confirms that the model is appropriate to explain the panel data.

Table 3. Regression Analysis Fixed-effects (within) regression Number of obs = 80 Group variable: FIRM Number of groups = 10 R-sq: within = 0.2853 Obs per group: min = 8 between = 0.2905 avg = 8.0 overall = 0.1986 max = 8 corr(u_i, Xb) = -0.8782 F(8,62) = 3.09 Prob > F = 0.0053 ROA Coef. Std. Err. T P>|t| [95% Conf. Interval] BSIZE -.0042949 .006302 -0.68 0.498 -.0168924 .0083025 BIND -.0816083 .0804827 -1.01 0.315 -.0168924 .0792744 ACSIZE .0035088 .0101678 0.35 0.731 -.0168164 .023834 ACIN .0055516 .0265773 -0.21 0.835 -.0586788 .0475757 BDIV -.0159456 .0175904 -0.91 0.368 -.0511083 .0192172 FBM .0032748 .0028332 1.16 0.252 -.0023886 .0089382 FSIZE .0009031 .0117126 0.08 0.939 -.02251 .0243163 FAGE .0082839 .0031139 2.66 0.010 .0020594 .0145085 _cons -.0256501 .206894 -0.12 0.902 -.4392254 .3879251 sigma_u .06127165 sigma_e .03153539 Rho .79057815 (fraction of variance due to u_i) F- test that all u_i=0: F(9, 62) = 4.85 Prob > F = 0.0001 Source: Author’s computation

The empirical result of the study under fixed effect estimation technique shows that board size has negative and insignificant effect on the performance of the insurance companies leading to the rejection of research hypothesis. Though insignificant this finding was consistent with findings of (Fama & Jensen, 1983; Adetunji and Olawoye, 2009) who advocate that large board size has adverse effect on the performance of firms but contradict with findings of Dallas, G (2004) who states that the size of the board has positive effect on performance of firms. Unless the type and nature of the industry and the discretionary power of board of directors is considered the optimum size board size remains debatable and inconclusive as claimed by Houssem and Ines, 2011; Adeusi et al., 2013; Musa et al., 2013 and Turku, 2014). The empirical study would dare to prove that in repressed and regulated financial sector the boards of directors have limited power and its size doesn’t matter as decisions are made within prescribed polices and rules of regulatory body. Board independence which is measured as the ratio of external board

members to the total number of board members has negative and insignificant effect on the performance of insurance companies proving for rejection of research hypothesis. This finding is supported with the previous empirical finding of Agrawal and Knoeber (1996) who found negative association between board independence and firm performance. This find ,of course, contradict the findings of Anthony, 2007; Cassandra et al., 2009; Lorne and Jun, 2012; Adeusi et al., 2013; Musa et al., 2013) who reported positive association between board independence and firm performance. Despite the fact that on average 75% of board members are composed of independent board members (Table-1), the negative association between their independence and firm performance shows that board of directors have limited power to exercise as board of directors. Interestingly size of audit committee and audit committee independence has positive but insignificant effect on the performances of insurance companies. This finding compliments with (Green, 2005); Anthony, 2007 and Cassandra et al., 2009), who

847


revealed the positive effect of independent and competent audit committees on the performances of firms. Board diversity measured by the presence of women in the board has negative effect on the firm’s performance. This finding contradicts with (Smith et al., 2005; Huse, 2007 and Mersland and Strom, 2007) who have found that the presence of women in the board positions have a positive effect on the firm’s performance. This would not come as surprise as board size itself has negative effect on the performances of insurance companies. Frequency of board meetings which is considered as an important proxies for the time directors spend monitoring managerial performance has positive but insignificant effect on the performances of insurance companies. This is consistent with studies of (Adams, 2000; Abbott et al., 2003; Funmi, 2014). Indeed it contradicts with (Jensen (1993 Karamanou and Vafeas, 2005) who reported negative relationship between number of meetings and firm performances. The control variables firm size and firm age have got positive relationship with the performances of the insurance companies. In a nutshell, board size, board independence and board diversity have negative and insignificant effect on the performance of insurance companies in Ethiopia. However, Size and independence of audit committee and frequency of board meetings have positive but insignificant effect on the performance of insurance companies in Ethiopia. The study in general found that all corporate governance mechanisms considered for investigation have insignificant effect on the performances of insurance companies implying that under stringently regulated financial industry board of directors have limited dictionary power and regulatory body has much more influential power than the board of directors. Conclusions and Recommendations The effect of corporate governance on firm performance has been such a buzz research topic that has attracted the attention of both academicians and researchers. Though there are extant empirical literatures, the outcome of such empirical studies has remained inconclusive. Few studies have shown that corporate governance in financial institutions differs from that of nonfinancial institutions and hence the discretionary power of the directors would be limited especially in regulated financial systems, where financial institutions are obliged to function through legislative and prescriptive procedures, policies, rules and regulations. This study, therefore, aims at examining the impact of corporate governance on the performance of closely regulated Ethiopian insurance Industry. Eight years panel data from ten insurance companies were employed and analyzed though fixed effect model. The return on assets of the sampled

insurance companies on average was 7.7% with average board size of nearly 10 board members. Of the total board members, 75% of them were non executive board members that in turn have paved the way for establishing an audit committee that is composed of independent board members. Almost 90% of the board members were composed of males and on average board of directors have held 16 meetings per year. Board size, board independence and board diversity have negative and insignificant effect on the performance of insurance companies while size and independence of audit committee and frequency of board meetings have positive but insignificant effect on the performance of insurance companies in Ethiopia. Thus it could be concluded that all corporate governance mechanisms have insignificant effect on the performance of insurance companies measured by return on asset. This vividly affirms that the role of board of directors in closely regulated financial sector is dismal and insignificant for they have limited discretionary power to exercise as board of directors. Thus it would be recommendable if the regulatory body could relax its prescriptive and stringent policies and devolve its power to board of directors without endangering the viability of insurance companies. References: 1. Abdullah, H & Valentine, B (2009). ‘Fundamental and ethics theories of corporate governance’, Middle Eastern Finance and Economics Journal, Issue 71 (2009). 2. Adams, R.B. and Mehran, H. (2003). Board structure, banking performance and the bank holding Company organizational form. Federal Reserve Bank of Chicago Proceedings (May), 408-422. 3. Adetunji Babatunde and Olawoye (2009). The effects of internal and external mechanism on governance in Nigeria. Corporate Ownership & Control / Volume 7, Issue 2. 4. Al-Hawary, S 2011, ‘The Effect of banks governance on banking performance of the Jordanian commercial banks 5. Anthony Kyereboah Coleman (2007). Corporate governance and firm performance in Africa: a dynamic panel data analysis, sabanci university, Istanbul, Turkey. 6. Bradley, N., (2004). Corporate governance scoring and the link between corporate governance and performance indicators: Corporate Governance Int. Rev., 12(1): 8-10. 7. Cheung syl and Chan, (2004). “Corporate Governance in Asia”, Asia Pac. Dev. J., 11(2): 1-31. 8. Cull, R., A. DemigÄuc-Kunt, and J. Morduch (2007). A global analysis of banks. Economic Journal, 117 (517), 107-133. 9. Daily, C.M., et al. (2003) “Corporate Governance: Decades of Dialogue and Data”. Academy of Management Review, 28(3), 371-382 10. Das, A. and S. Ghosh, (2004, March 20): Corporate Governance in Banking System: An Empirical Investigation. Economic and Political Weekly, pp. 12631266. 11. Dallas, G 2004, Governance and Risk: An Analytical Handbook for Investors, Managers, Directors, & Stakeholders, McGraw-Hill Inc., U.S.

848


12. David Wanyonyi Wanyama1* Tobias Olwen (2013). Public Policy and Administration Research ISSN 22245731(Paper) ISSN 2225-0972(Online) Vol.3, No.4. 13. Davis, J.H., Schoorman, F.D. and Donaldson, L. (1997). Toward a Stewardship Theory of Management. Academy of Management Review, Vol.22, pp. 20-37. 14. Defond, M. L., Hann, R. N., & Hu, X. (2005). Does the Market Value Financial Expertise on Audit Committees of Boards of Directors?. Journal of Accouting Research, 43(2). 15. Fama, E. & Jensen, M. (1983) Separation of Ownership and Control. Journal of Law and Economics. 26:301325. 16. Fontaine, C, Haarman, A and Schmid, S (2006) ‘The Stakeholder theory: stakeholder theory of The MNC’ (No volume & number). 17. Funmi Babington. (2014).Code of Corporate Governance for the Insurance Industry, Insurance and you (no.4). 18. Gompers, P.A., Ishii J.L. and Metrick, A. (2003). Corporate governance and equity prices. Q. J. Econ., 118(1): 107-157. 19. Green, S. (2005). Sarbanes-Oxley and the Board of Direcors: Techniques and Best Practices for Corporate Governance. New Jersey: John Wiley and Sons, Inc. 20. Gujarat D.N. (2004). Basic Econometrics. 4th edition. The McGraw−Hill Companies 21. Hillman, A.J., et al. (2000): The Resource Dependency Role Of Corporate Directors: Journal of Management Studies, 37(2), 235-25 22. Johnson, J.L., et al. (2006): Boards of Directors. A Review of Research Agenda. Journal of Management, 22(3), 409-438 23. Kang, E. &Zardkoohi, A. (2005). Board Leadership Structure and Firm Performance. Corporate submitted to

24.

25.

26.

27. 28.

29.

30.

31. 32. 33.

34.

the Centre for Strategic Economic Studies, Victoria University. Kaymark, T. & Bektas, E. (2008). East Meets West? Board Characteristics in an Emerging Market: Evidence from Turkish Banks. An International Review, 16(6), pp. 550 - 561. Kiel, G.C. & Nicholson, G. J. (2003). Board Composition and Corporate Performance: Corporate Governance: An International Review, 11(3), pp. 189 205. Lorne N. Switzer and Jun Wang (2012). Default Risk and Corporate Governance in Financial vs. NonFinancial Firms, Finance Department, Concordia University, Montreal, CANADA. Matama, R., (2005). Corporate governance and financial performance of selected commercial banks in Uganda. Rashid, A. (2011). Board Composition Board Leadership Structure and Firm Performance: Evidence from Bangladesh. Renee Adams & Hamid Mehran (April, 2003) ‘Is Corporate Governance Different for Bank Holding Companies?’ Federal Reserve Bank of New York Economic Policy Review 123 Ruin, Josef Eby. (2001). Essentials of Corporate Management. Kuala Lumpur: Malaysian Institute of Corporate Governance (MICG). Sanda M. & Garba, T. (2005). Corporate Governance Mechanisms and Firm Financial Performance in Nigeria. Sapovadia, V. (2009) Good Corporate Governance: An Instrument for Wealth Maximization. Smith, N., Smith, v., & Verner, M. (2005). Do Women in Top Management Affect Firm Performance? A Panel Study of 2500 Danish Firms. IZA. Wheelen, T.L., & Hunger, J. (2006) Strategic Management and Business Policy, Pearson Prentice Hall. (http://en.wikipedia.org:15th December, 2012).

849


IMPACT OF SUPERVISORY BOARD MEMBERS’ PROFESSIONAL BACKGROUND ON BANKS’ RISK-TAKING Dennis Froneberg, Florian Kiesel*, Dirk Schiereck

Abstract This paper examines the impact of financial expertise of supervisory board members on the risk-return profile of 200 German regional cooperative banks during the period 2004–2009. The results show that with more financial expertise the bank performance does not improve, but bank risk increases. These findings induce concerns that mandating financial expertise on boards is not necessarily beneficial for the risk-return profile of regional banks. We suggest that overconfidence of entrepreneurs in the supervisory boards leads to this unfavorable development since they represent the largest fraction of professionals within the sample. Keywords: Regional Banks, Governance, Supervisory Board, Financial Expertise, Risk-Return Profiles, Germany Jel classification: C23, G21, G32, G38 * Department of Business Administration, Economics and Law, Technische Universität Darmstadt, 64289 Darmstadt, Germany

1. Introduction European banking markets became increasingly integrated and more competitive subsequent the implications of deregulation and innovation in the financial markets over the two decades prior to the credit crisis that started in late 2007. Consequently banks have expanded, changed, and discarded various business models in recent years due to shifts between regulation and deregulation, macroeconomic and political trends, industry competition, changing customer demands as well as domestic, foreign, and state ownership (e.g., Dermine, 2003, Goddard et al., 2007, DeYoung et al., 2004, Berger et al., 2005). On the back of this evidence bank managers have constantly adjusted their banks’ risk-return profiles through policies with respect to financing, investment, organization and merger decisions (Malmendier and Tate, 2005, Bertrand and Schoar, 2003, Adams et al., 2005, Manove and Padilla, 1999). So far, theoretical evidence has not provided a clear picture concerning the ultimately optimal business model, and the started in 2007 banking crisis has proved that not all business models and their risk-return profiles are equally feasible and sustainable (Rajan, 1992, Diamond, 1991, Stein, 2002, Demirgüç-Kunt and Huizinga, 2010, Altunbas et al., 2011, Santos, 2001, Giammarino et al., 1993). Healthy risk-return profiles are not only important for the stability and profitability of individual banks, but also for the whole financial system. Failures of single firms in the banking industry have a systemic dimension in contrast to other industries. Based on that both regulatory bodies

and academic research came to the conclusion to differentiate between the pure size of a financial institution and the systemic importance it has. New terms are focusing on the potential systemic impact if a particular institution fails. This systemic importance of financial institutions is the key issue in both financial stability assessments (Zhou, 2010). Previous theoretical considerations state that large banks, which are perceived “too big to fail,” require rescue measures by governments as their failures are likely to result in the collapse of the whole financial system (Kaufman, 2002, Hoggarth et al., 2004). On the other hand in the case of the simultaneous failure of several weak, and not necessarily large banks, regulators might face a “too many to fail” problem, which could end in a systemic collapse as well (Brown and Dinç, 2011, Acharya and Yorulmazer, 2007). Given the economic relevance and interaction banks can hardly be fully crisis resilient. Their business models are associated with uncertainty about liquidity needs and a maturity mismatch between assets and liabilities which is seen as substantial systematic-risk exposure yielding towards the fragility of institutions (Diamond and Rajan, 2001, Bhattacharya et al., 1998, Farhi and Tirole, 2012). Albeit risk-taking has significant ramification for the economy it is desirable that banks take some risks to bolster the economy (Levine, 2006, Rajan and Zingales, 1998). Financial intermediaries have fundamental channels which are linked to wealthincreasing projects of the private sector, banks facilitate the capital accumulation, trading, hedging, and pooling of risks (Levine, 1997, Kroszner et al.,

850


2007, Pang and Wu, 2009, Beck et al., 2000, Dell'Ariccia et al., 2008). Supervisory entities and government bodies reform financial regulations frequently to empower bank governance structures with the intention to ensure balanced risk-return profiles and to keep the banking system stable. Consequently a bank manager is calibrating a banks’ risk-return profile within the limitations set by the regulation such as bank activities as well as capital and funding requirements (Barth et al., 2008). During and post financial crisis, the lack of sufficient industry expertise and effective framework have been recognized as an important shortcoming of internal bank governance mechanisms (Choundhry, 2011, Ard and Berg, 2010, Kirkpatrick, 2009, Hau and Thum, 2009, Peni and Vähämaa, 2012, Spong and Sullivan, 2010). Furthermore, the business model itself has become more complex and more opaque since the banks grew significantly and started expanding into new business areas creating unregulated exposure (Mehran et al., 2011). On the back of this evidence, superior financial expertise of board members is considered as a potential framework to assess and manage risks better as well as to create more stable banks with balanced risk-return profiles from the inside. Regulators imposed comprehensive risk management frameworks in combination with the recommended governance structures focusing on the overall balance of the board in relation to the risk strategy of the business, taking into account the experience, behavioral and other qualities of individual directors (Walker, 2009, Aebi et al., 2012). The supervision of bank management and the risk-return profile of the bank are being performed through mandating independent directors in a one-tier system and supervisory board members in a two-tier system (Hopt and Leyens, 2004, John and Senbet, 1998, Jungmann, 2006). Especially the mandate holds for the latter one since this system separates management board and explicitly assigned management monitoring. The question, what the impact of greater financial expertise in internal governance on a banks’ risk-return profile is, has not been analyzed in-depth yet. But without waiting for respective empirical evidence, German bank regulation was sharpened with the target of increasing the required expertise for bank supervisory board members to foster bank stability and limit risks. Since 2009, provisions on vetting members of administrative and supervisory bodies were inserted in the Banking Act (Kreditwesengesetz – KWG) and the Insurance Supervision Act (Versicherungsaufsichtsgesetz – VAG) for the first time by virtue of the Act to Strengthen Financial Market and Insurance Supervision (Gesetz zur Stärkung der Finanzmarktund der Versicherungsaufsicht). The Federal Financial Supervisory Authority (BaFin) shall be authorized to dismiss members of the supervisory board of institutions and insurance companies if they are not

adequately qualified or not trustworthy, or if they act negligently in the exercise of their control functions. New elected members of supervisory boards of financial institutions such as banks and insurance companies in Germany have to meet new legal requirements in terms of financial expertise. In the future prospective members’ competence will be checked by the BaFin in order to make sure that the supervisory board can exercise its tasks (BaFin and Deutsche Bundesbank, 2010). This paper examines how financial expertise in internal governance mechanisms influences a bank’s risk-return profile. It is challenging to divulge the actual influence of financial experts on the risk-return profile of a bank. Factors such as regulatory environment, business models, and competition in the banking industry have direct influence on the performance and the risk of banks. Furthermore, banks are characterized by different governance structures with either a one- or two-tier system across borders. These two factors are of high relevance since they result in a distortion of the analysis with regards to who mainly affects a banks’ risk-return profile. For the purpose of the intended analysis the German cooperative banks as an important pillar of the regional banking sector display a unique field of data. Since all banks within the sector operate in the same regulatory environment, and have similar business models, comparable regional reach, strategies, and organizational structures this sector provides a homogeneous object of investigation. The competitive advantage of the relatively small regional banks results from the concentration on selected market segments with two main roles, lending to small and midsized companies as well as to private households (Mercieca et al., 2007). The typical German two-tier board system that separates management and supervision also applies to regional banks including the general regulations on supervisory board members and labor codetermination (Hopt and Leyens, 2004). With regards to the governance perspective, the managers of these credit institutions have a particular characteristic compared to exchange-listed commercial banks since they are not confronted with a market for corporate control (Manne, 1965). Consequently, the absence of one governance instrument increases the importance of all other governance mechanisms. To analyze the consequences of different levels of financial expertise of supervisory board members this paper focuses on the period before 2009 where banks were completely free in the selection of board members. In response to the financial crisis, modifications of the German legislation have been implemented in 2009 to increase the quality of internal control and stability of the banking system. For our examination on the influence of the financial expertise of supervisory board members on the risk-return profile of German regional banks the present study relies on a unique, manually collected

851


sample with more than 200 institutions comprising only entities being cooperative banks in its legal form. The size of the data set, the focus on a two-tier board structure as well as the proprietary compilation of data including the granularity is unrivaled given access to non-public data provided by the Deutsche Bundesbank. The paper identifies two important areas to discuss for the German banking sector and its regulator which we want to highlight at this point. Is financial expertise beneficial for the development of a German regional bank and its risk-return profile? It is evident that in a sector where risk taking is at the core of each company’s business model, financial expertise should display a particularly important factor for the bank’s risk-taking behavior. Based on the results showing that financial expertise does not lead to a favorable development of a risk-return profile of a bank it remains open what the optimal composition of a supervisory board of a regional bank is. In particular, the large fraction of entrepreneurs within a supervisory board has presumably a significant contribution to the deteriorated risk-return profile through their overconfidence. The remainder of this paper is structured as follows: Section 2 provides an overview of the related literature with a particular focus on the influence of the financial expertise of board members, board composition and performance on regional banks’ operations. Section 3 introduces the data, their sources and the empirical model. Section 4 shows the summary statistics and the results of the regression analysis. This section also exhibits the findings based on different measures and outlier treatments for robustness checks. Section 5 concludes the paper. 2. Literature Review Studies with regards to the impact of financial expertise on risk-taking and performance can be classified into three segments: (i) board competence and expertise, (ii) board size and composition, and (iii) the performance, efficiency and stability of cooperative banks. Board competence and expertise are mainly characterized by the professional experience of the board members. Positive relationships between financial expertise and performance measures has been identified for example by Cunningham (2008), Davidson et al. (2004), DeFond et al. (2005), Dhaliwal et al. (2006), Fernandes and Fich (2009), Lee et al. (1999), and Swan and Forsberg (2014). Davidson et al. (2004), DeFond et al. (2005), Fernandes and Fich (2009), and Lee et al. (1999) apply financial market information using stock price movement and abnormal returns as key metrics. They conclude that capital markets reward competence on the boards following the positive reaction of capital markets after the announcement of appointing financial experts. A different approach for performance measures is applied by Dhaliwal et al. (2006) who find a positive

correlation between accounting expertise and accruals quality based on the definition of three types of expertise: accounting, finance, and supervisory. This finding is underpinned by Cunningham (2008) who points out that particular accounting expertise is more important than any other kind of financial expertise. Swan and Forsberg (2014) show that former executives in the board, such as now retired CEOs and those who retain links with management, make better acquisition decisions, increase the proportion of incentives in CEO pay, and raise dividend payouts. They show that replacing such board members with independent directors declines the firm’s performance. Mixed results are provided by Carcello et al. (2008), Güner et al. (2008), Rosenstein and Wyatt (1990), and Minton et al. (2010). Güner et al. (2008) focusing on bankers as board members conclude that there is no significant impact on appointment decisions as far as there are no conflicts of interest between the bankers and the concerned company. Bankers can be detrimental to shareholder wealth in such incidents. According to Rosenstein and Wyatt (1990), all occupations are equally valuable to shareholders, with regard to share price reactions. Based on the findings of Carcello et al. (2008), expertise is not beneficial for real earnings management since other governance mechanisms are as good as financial expertise and have a positive impact on the quality of financial reporting. Focusing on the recent banking crisis Minton et al. (2010) analyze a sample of more than 650 unique firms consisting of over 300,000 board members over the 2000 to 2007 period the impact of financial expertise on bank risk and conclude that bank risk-taking is positively associated with greater financial expertise. Focused on the peculiarities of the German twotier system, only three studies investigate the relationship between expertise and effectiveness of supervisory boards to our knowledge. Kaplan (1994) shows that supervisory board monitoring works efficiently, as indicated by the management turnover, resulting from poor firm performance. The quality of supervision in banking is addressed in Hau and Thum (2009) who comprise a sample of the 29 largest German banks. They define 14 biographical criteria to assume competence by the members of the supervisory board. The results suggest that the financial fragility of the banks in the sample correlates with the monitoring ability of the supervisory board members. Delegating supervisory mandates to individuals with financial expertise is considered an option to stabilize the banking system. Schmielewski and Wein (2012) find that the risk-taking attitudes are closely related to the ownership structure. They analyze the ownership and the risk-taking behavior of bank managers of 397 German banks between 2000 and 2010. They conclude that risk-taking of bank managers depend on the ability to control them. The lower the monitoring capabilities of bank owners are,

852


the greater the probabilities of failures in choosing the optimal portfolio. The second strand of related literature addresses the impact of board composition and board size on performance. Lipton and Lorsch (1992) and Jensen (1993) suggest that larger boards tend be less effective due to co-ordination problems and director free-riding which might have adverse effects on performance and risk-taking. Empirical studies support this view (Yermack, 1996, Eisenberg et al., 1998). Another study portrays a U-shaped relation between the size of Monetary Policy Committees and inflation, implying that lowest level of inflation will be reached at an optimum level of committee members. Research with focus on the banking industry does not necessarily support the theory that larger boards are less effective. A more recent study shows that strong bank boards[1] particularly small and less restrictive boards positively affect bank risk-taking (Pathan, 2009, Berger and Nitsch, 2011). This supports other views that larger boards lead to an increase in supervision and in advising management (Adams and Mehran, 2008). The analysis of the majority of studies related to board composition is focused on the proportion of outside directors - not having the same scope given the peculiarity of the German two-tier board structure. Most studies do not find a causal relation between board composition and firm performance (Hermalin and Weisbach, 2003, Wintoki et al., 2012). Whereas Harris and Raviv (2008) develop theoretical models of board structure finding that optimal boards will employ large numbers of outside directors, and can be larger in overall size, Bermig and Frick (2010) confirm the missing consistent link between board composition and valuation and performance based on a sample of listed German firms in the period between 1998 and 2007 in contrast to Berger et al. (2012) who find a clear relationship between the socio-economical compositions of a board structure and bank’s risktaking. They analyze a data set for the entire board composition of German banks for the timeframe 19942010 and find that boards represented by younger executive teams and female executives tend to take on more risk, contrary to boards with a higher representation of Ph.D. degree holders where risktaking declines. The third related category of literature focuses on risk-taking, performance, and stability of regional banks with a focus on cooperative banks. The studies analyze these factors on a national as well as an international level (Beck et al., 2009, Cihák and Hesse, 2007, Westman, 2011, Altunbas et al., 2001, Iannotta et al., 2007, Ayadi et al., 2010, Hasan et al., 2012, Lang and Welzel, 1996). Overall, the evidence suggests that regional banks tend to be more stable than commercial banks. This finding can be confirmed for the German regional bank sector. Beck et al. (2009) show, using the z-score as metric for financial soundness, that German regional banks are more stable than German commercial banks due to the

lower earnings volatility. International comparisons confirm that regional banks are more stable than commercial banks, underlined by the better loan quality and lower asset risk of cooperative banks (Cihák and Hesse, 2007, Westman, 2011, Iannotta et al., 2007). In terms of efficiency and performance the results are mixed. Iannotta et al. (2007) exhibit that regional banks are slightly more efficient than other banks, however exhibiting lower profits than commercial banks. In this respect Altunbas et al. (2001) can confirm the results only partially, since they demonstrate that cooperative have slight cost but also profit advantages over privately owned banks. Ayadi et al. (2010), conducting a European wide study, can confirm these mixed picture, with respect to Germany they find that regional banks are more profitable in terms of ROE and ROA, while being less efficient due to a higher cost-to-income ratio than other banks. It can be concluded that existing findings exhibit that financial expertise of board members is one of the factors influencing the risk-return profile of firms. Although, previous studies mainly focus on issues of board composition and board size in jurisdictions where a one-tier board system is in place. Further, the majority of studies, investigating the relationship between financial expertise and firm performance, do not deal with banks in particular. Finally, the quality of supervision in terms of financial expertise in twotier board systems was largely neglected. The study of Hau and Thum (2009) can be considered as the most similar to our analysis, but their focus is slightly different. It deals with a relatively small sample of 29 large-scale German banks and bank losses only during the banking crisis years. The results suggest that the financial fragility of the banks in the sample correlates with the monitoring ability of the supervisory board members. Delegating supervisory mandates to individuals with financial expertise is considered an option to stabilize the banking system. 3. Data and Methodology 3.1 Sample and data There is no public data base available for the supervisory board composition of German cooperative banks. The supervisory board composition and the individual’s profession are collected manually from each bank’s annual report. Balance sheet and profit and loss statement data are obtained from Deutsche Bundesbank’s prudential database Bankaufsichtliches Informationssystem (BAKIS). The data access to the prudential information system BAKIS contains bankspecific data and is subject to special restrictions. The Hoppenstedt Banken database is used to access the merger history of each cooperative bank. Data for macroeconomic and structural control variables have been provided by the Regionaldatenbank Deutschland of the German Federal Statistical Office. The data

853


used in this study are based on the 200 largest regional banks in the legal form of cooperative banks during the period 2004–2009. Bank size is defined as total assets measured at the end of year 2009. Cooperative banks that are not of the legal form of a cooperative firm are excluded from the dataset. In order to increase the explanatory power of the sample, the data set has to comply with several requirements. Financial expertise is approximated by the profession of supervisory board members. This requires concrete information about this item for each person. However, this is not always clearly mentioned. Given the category “other” where no assessment regarding financial expertise of an individual can be made, a maximum threshold of 10% for this category per bank has been defined. Consequently a maximum of 10% of the overall supervisory board, or one individual for supervisory boards of up to nine individuals, is allowed to be classified as “other”. If a cooperative has not been included in the analysis due to lack of data or a too high representation of individuals who could not clearly be labeled, the institution ranking subsequent in size is included, instead. Our sample covers 57% of the total assets of the German cooperative bank sector as of the end of 2009. The average cooperative bank in the sample has total assets of EUR 1.9 billion. The largest cooperative bank has total assets of EUR 41.4 billion, while the smallest bank has EUR 0.7 billion. In addition, the present paper analyzes a mergeradjusted sample as robustness test. We run our analysis on the 150 largest cooperative banks which have not been involved in any mergers activities over the panel horizon and the two years preceding this horizon in order to control for potential distortions for merger activities. 3.2 Model and hypotheses In the present study we address with our empirical model the following research question: “How are the risk, stability, and performance of regional banks impacted by (outside) financial expertise?” In the course of this research question, we test the following two hypotheses: Hypothesis 1: (Risk-adjusted) performance of a regional bank is unrelated to the (outside) financial expertise of the supervisory board members.

Hypothesis 2: The stability and risk of a regional bank are not driven by the (outside) financial expertise of the supervisory board members. The empirical examination is based on a panel analysis via a random effects regression model of the following general form:

DVit     FEit   Bit   M it   it

(1)

where DVit is the dependent variable of cooperative bank i at time t. Dependent variables for stability and risk are z-score and non-performing loans (NPL) ratio, whereas the dependent variables for performance are return on risk-weighted assets (RORWA) and return on equity (ROE), FEit is the financial expertise, Bit is a vector of bank-specific variables, Mit is a vector of macroeconomic and of cooperative bank i at time t. Vector B comprises (a) bank size, (b) bank size growth, (c) bank efficiency, (d) bank loan volume growth, (e) bank claims on monetary financial institutions (MFI), (f) bank claims on non-monetary financial institutions (non-MFI), and (g) supervisory board size. Vector M consists of the macroeconomic and structural control variables (h) area, (i) federal state, and (j) local GDP per capita. 3.3 Financial expertise We define financial expertise as the ratio of members with financial expertise in the supervisory board to the total members in the supervisory board. The special access to micro data of German banks by BAKIS allows us to obtain the supervisory boards of German regional banks. A total of approximately 14,800 data points at the individual level are included in this study[2]. Financial expertise on the individual level is gauged based on their professional backgrounds. This follows previous studies (e.g., Davidson et al., 2004, Dhaliwal et al., 2006, Minton et al., 2010). The following professional backgrounds with assumed financial expertise have been constituted according to occupations with the help of publicly available data in the annual reports of the banks. This allows assumptions to be made regarding the financial expertise of the supervisory board members which is defined as outlined in Table 1.

Table 1. Professional backgrounds with assumed financial expertise Level 1 Employment Level 2 Occupation

Self-employed persons with assumed financial expertise Entrepreneur, Merchant, Owner

Employed persons with assumed financial expertise Non-bank employee with assumed financial focus (e.g. members of management board)

Managing Partner

Bank employee

Partner

Public officer with assumed financial focus (e.g. treasurer) Federal and State Minister, State Secretary Local and regional politician

854

Retired persons with assumed financial expertise Retired self-employed person with assumed financial expertise Retired employed person with assumed financial expertise


The main idea behind this methodology is the obvious relation to financial issues. For instance, financial expertise is expected in the case of management board members. This approach follows other studies that were conducted in relation to the US-American Sarbanes-Oxley Act (SOX), which was released in 2002. SOX comprises requirements for the Audit Committees of companies whose securities are traded in the United States. Related studies analyze the impact of financial expertise in general and knowledge in accounting and auditing in particular on the risk and performance of companies (Carcello et al., 2008, Dhaliwal et al., 2006, Güner et al., 2008, Hau and Thum, 2009, Hermalin and Weisbach, 2003). These studies mainly use individuals’ professional backgrounds as indications of qualifications. Only Hau and Thum (2009) and the recent study of Körner et al. (2014) consider additionally the educational background. Further, the definition of financial expertise is mainly on the basis of the German regulatory bodies (BaFin and Deutsche Bundesbank, 2010). For instance, following the regulators’ joint bulletin, financial expertise is assumed for potential mandatory members, such as mayors and district administrators. Both regulatory bodies do not explicitly specify the required competence levels, but provide indications concerning occupations and experiences that allow assumptions to be made regarding financial expertise. Professional experience gained in other areas might be also sufficient to justify financial expertise. Additionally, supervisory board members can gain relevant expertise through similar supervisory mandates and professional activities in the banking sector. Furthermore, regulatory bodies consider potential mandatory members, such as mayors and district administrators, whose board presence is legally required, to be financially literate (BaFin and Deutsche Bundesbank, 2010). In the present study, we assume that selfemployed persons generally have financial expertise. Self-employed persons are merchants within the meaning of the German Commercial Code. The definition of a merchant is very important in the German law, because merchants according the HGB law have certain obligations and duties. Especially §238 HGB commits to legal obligation to keep records. Given the fact that details in the annual reports differ between banks, several assumptions concerning tenure, employment classification, and politicians have been applied. To control for the portion of employee representatives on the board the regression analysis is conducted twice for each sample. First, the financial expertise of the total supervisory board and its impact on risk and performance is tested. Second, employee representatives who are employed at the respective bank are excluded in order to analyze the impact on performance and risk of the outside

expertise of the free eligible supervisory board members. 3.4 Measurement of risk and performance We apply two different measures for the analysis of bank risk. The first risk measure is the z-score that has widely been used for the analysis of bank risk (Boyd and Runkle, 1993, Laeven and Levine, 2009). We calculate z-scores for each cooperative bank to measure the individual bank’s insolvency risk and follow Laeven and Levine (2009) for the calculation. Based on the idea that insolvency is the state in which a bank’s capital does not suffice to absorb losses the probability of insolvency is defined as prob(-ROA < CAR) where ROA is the return on assets and CAR is the capital assets ratio (Roy, 1952). Given that profits are normally distributed the probability of insolvency becomes (ROA+CAR)/σ(ROA), where σ(ROA) is the 10 years standard deviation of ROA (Laeven and Levine, 2009). A higher z-value indicates a more stable bank. The ratio of loan loss provisions to loans serves as second risk measure. The variable NPL ratio indicates the loan loss provision. As we outlined at the beginning lending to small and mid-sized enterprises as well as to private households is one of the main pillars of regional banks’ business and therefore represents an important measure for regional banks. We measure performance with ROE and RORWA. ROE is defined as income before tax divided by average equity. RORWA is defined as income before tax divided by average risk-weighted assets. This allows reflecting the bank specific risk profile. Operating result is used for both performance measures in order to avoid distortions due to undisclosed reserves. These reserves are typically used when banks are in trouble (Beck et al., 2009). 3.5 Control variables Besides bank-specific issues, we also control for macroeconomic as well as structural aspects. To factor in different bank sizes and growth rates, we apply the log of total assets and the annual change in total assets; for different efficiency levels we use the cost income ratio (CIR), which is defined as the ratio of general administrative expenses to operating result. To account for different levels of lending engagements we use the growth rate of the total loan volume. Additionally, the ratios of claims on MFI to total assets and claims on non-MFI to total assets are factored in. Further we control for the size of supervisory boards as larger boards tend to incentivize free-riding and this might lead to a negative effect of the supervisory board size on performance and risk (Jensen, 1993). As regional banks have legally specified business areas, it is important to control for the region where the bank is located. We capture these structural differences by an index distinguishing between urban

855


and rural areas. Further a binary variable identifying federal states is applied. Regional banks which used to be in the former German Democratic Republic are assigned a value of one. The latter variable is important to control for the long-term business relationships. Regional banks in Eastern German federal states worked in a different structure and business environment under the East German regime until 1990. We measure the difference between urban and rural areas based on population. The differentiation between regional, district, and county centers is captured by an index with five clusters

which have been defined. For the discrimination between district and regional centers, areas with up to 1,000,000 inhabitants are divided into four clusters of 250,000 inhabitants each. The fifth cluster is comprised by areas with more than 1,000,000 inhabitants. This allows separating sparsely populated areas from densely populated and metropolitan areas. In order to differentiate the economic development, we considered the local GDP per capita. Table 2 gives an overview of the definitions and sources of all control variables.

Table 2. Definition of control variables and sources Bank level control variables Description Bank size Ln of total assets Bank size growth Year-to-year change in total assets Bank efficiency (costGeneral administrative expenses to operating income ratio) result Bank loan volume growth Year-to-year change in total loans Bank claims on MFI Claims on MFI divided by total assets Bank claims on non-MFI Claims on non-MFI divided by total assets Bank supervisory board size Number of supervisory board members Macroeconomic and structural control variables Area Index based on population within business area Federal state Binary variable separating Western and Eastern federal state GDP per capita GDP of business area divided by population within business area

4. Results 4.1 Summary statistics Table 3 shows the supervisory board composition of the 200 largest cooperative banks and their financial expertise. There is no obvious strong stakeholder group dominating solely the composition of cooperative banks supervisory boards. Rather cooperative banks supervisory boards are characterized by a large fraction of self-employed and employed people, they account for almost the same

Source BAKIS (Deutsche Bundesbank) BAKIS (Deutsche Bundesbank) BAKIS (Deutsche Bundesbank) BAKIS (Deutsche Bundesbank) BAKIS (Deutsche Bundesbank) BAKIS (Deutsche Bundesbank) Annual reports Regionaldatenbank Deutschland Annual reports Regionaldatenbank Deutschland

proportion of the board, comprising 40% and 50% respectively of the supervisory board. Notable is that only 5% of the members of cooperative banks’ are employed at the respective bank, suggesting that insiders play a minor role. Further, our analysis reveals that politicians account for approximately 3% of the board members implying that cooperative banks are impacted by few political interest and influence. Other notable fractions are retired individuals which make up almost 10% of the supervisory board.

Table 3. Supervisory board composition Variables Insider (%) Outsider (%)

Mean 5.20 94.80

Std. Dev. 11.88 11.88

Coeff. of variation 228.34 12.53

Min 0.00 47.06

Max 52.94 100.00

Employment Employed individuals (%) Self-employed individuals (%) Retired individuals (%) Others (%)

50.22 39.55 9.20 1.03

22.94 23.49 11.86 2.76

45.68 59.40 128.99 268.43

0.00 0.00 0.00 0.00

100.00 91.67 77.78 14.29

Politicians Federal politicians (%) Federal state politicians (%) Local politicians (%) Current and retired politicians (%)

0.09 0.24 1.83 2.80

0.90 1.53 4.24 5.53

1,020.84 629.23 231.21 197.83

0.00 0.00 0.00 0.00

11.11 20.00 25.00 35.29

856


The summary statistics of the variables is provided in Table 4. Despite the same focus and core business, regional banks display a relatively wide range of performance, with coefficients of variation of 65.70% for RORWA and 62.04% for ROE. The distribution range for RORWA is from -1.82% to 6.27% and for ROE from -17.62% to 55.21%. The average performance measured by RORWA is 1.21%,

whereas ROW is 12.19%. Risk and stability metrics show a similar pattern and variation, with coefficients of 65.24% for z-score and 74.15% for the NPL ratio. Z-score shows an average of 22.71, ranging in the distribution pattern from 3.61 to 103.23. The minimum NPL ratio in the sample reaches a peak of 13.47%, whereas the average NPL ratio is 2.66%.

Table 4. Summary statistics Dependent variables RORWA (%) ROE (%) z-score NPL ratio (%) Bank level variables Financial expertise (%) Outside financial expertise (%) Bank size (ln ‘000s of EUR) Bank efficiency (%) Bank size growth (%) Board size Loan growth (%) Claims on MFI (%) Claims on non-MFI (%) Macroeconomic and structural variables Federal state GDP per capita ('000s of EUR) Population

Mean 1.21 12.19 22.71 2.66

Std. Dev. 0.80 7.56 14.81 1.97

Coeff. of variation 65.70 62.04 65.24 74.15

Min -1.82 -17.62 3.61 0.00

Max 6.27 55.21 103.23 13.47

47.76 45.25 20.90 66.48 2.61 12.42 4.98 11.89 58.67

22.87 23.23 0.69 9.55 4.61 4.44 7.44 6.59 11.39

47.88 51.34 3.28 14.37 176.76 35.76 149.52 55.47 19.41

0.00 0.00 19.60 15.98 -14.80 5.00 -13.19 0.06 11.17

100.00 100.00 24.45 112.92 31.47 37.00 51.54 45.65 83.57

0.02 28.04 2.67

0.14 7.32 1.39

700.29 26.12 52.07

0.00 9.39 1.00

1.00 62.04 5.00

The analysis reveals that financial expertise at regional banks has a high variation, given the variation coefficient of 47.88. This is underlined by the fact that the distribution of financial expertise ranges from boards with zero percent financial expertise to boards composed by individuals assigned only with financial expertise. On average a board of a regional bank has a fraction of financial experts of 47.76%. The data sample adjusted for insiders, exclusion of bank employees, shows similar results and not much variation. Besides the board composition also the board size indicates high variation with a coefficient of variation of 35.76. The board size ranges from 5 individuals at the smallest board to 37 individuals at the largest board. On average a supervisory board of a regional bank has 12.42 members. With regards to other bank-level variables, bank size growth and loan volume growth show the largest variation, whereas bank size has the lowest coefficient of variation within the sample. Our results suggest that, over the years, regional banks shift their focus on the more profitable lending business by allocating their assets without growing in total bank size by the same factor. Whereas average bank growth is 2.61%, the loan volume grows at an average rate of 4.98%. Bank size shows the lowest coefficient of 3.28 indicating that bank size of our sample is nearly comparable. Bank efficiency is relatively homogeneous among all banks as indicated by the

relatively low coefficient of variation of 14.37%. The non-MFI engagement by regional banks does not diverge much, as indicated by the low coefficient of variation, regardless of a broad range between minimum and maximum values. In contrast regional banks are rather different in their inter-bank activities as the coefficient of variation for claims on MFI shows. Table 5 reports pair wise correlation coefficients. The matrix explaining the correlation between RORWA and ROE is positive and statistically significant, while the correlation between RORWA and financial expertise shows a negative relationship. These correlations indicate that cooperative banks with higher financial expertise have a lower level of RORWA and the performance of a bank decreases with a higher financial expertise ratio in the supervisory board. A significantly positive association is shown for the correlation between financial expertise and the NPL ratio, indicating that banks with higher financial expertise in the supervisory board are likely to enlarge their non-performing loans. This risk measure is highly important for cooperative banks due to the large engagement in the lending business. However, the opposite is true for the relation between financial expertise and the z-score. Overall, these correlation coefficients suggest that financial expertise has a negative influence on the performance and at least partial negative influence on the risk of cooperative banks.

857


Table 5. Correlation coefficients among variables of interest

RORWA ROE Z-score NPL ratio Financial expertise Outside financial expertise

RORWA ROE

Z-score

1.0000 0.9023* 0.1428*

1.0000

1.0000 0.0842* -0.0610* 0.0595* -0.1008* 0.0728* -0.0800* 0.0627*

-0.0025 0.1527* 0.1431* 0.1088*

NPL ratio

Outside Financial Bank financial expertise size expertise

Bank Bank size efficiency growth

0.2384* 0.9575*

1.0000

0.2010* 0.2792*

0.0085

1.0000

0.1230*

1.0000 0.1898*

Bank efficiency

-0.3240*

-0.0240 0.3421*

0.1415* 0.1005*

Bank size growth

0.0362

0.0379

-0.0928* -0.1109* 0.1061* -0.1050* 0.2707*

Bank loan growth

0.2621*

Claims on non-MFI

-0.0031 -0.0553 0.0741* -0.0589* -0.0354 0.0311 0.1285* 0.1304* -0.1589* 0.2899* 0.2025* -0.0059 0.0633*

Supervisory board size

-0.0259

GDP per capita

-0.0807*

Population

-0.0271

Federal state

0.0280

0.1778*

0.0051

0.1771* 0.0602* -0.0101

0.1111* 0.0671* -0.0359 0.1609* 0.0236 0.0815*

Claims on Non MFI

Board size

GDP per capita

Population

Federal state

0.1755* 1.0000

-0.0785* -0.0486

Claims on MFI

Claims on MFI

1.0000

Bank size

-0.0560

Bank loan growth

0.0863* 0.1339* -0.0115 0.3973* -0.0093

0.0549

1.0000

-0.0732* 0.1463* -0.0624*

0.6166* 1.0000

0.0387

0.0615* 0.1001*

0.0970* 0.0052

-0.0025

0.0887* 0.1987*

1.0000 0.1558* 0.1568* 0.4040* -0.0230 -0.0446 0.0917* 0.1281* 0.0866* 0.0784* 0.1931* 0.1029* 0.1223* 0.0956* 0.1393* 0.2820* -0.0267 -0.0249 0.0601* 0.2834*

-0.0793* 0.2213* -0.0203 0.0455

0.2971* 0.0071

-0.1354* 0.6413* -0.1898* 0.0695*

0.0849* 0.1280*

* significant at the 5% level

858

1.0000

1.0000 1.0000 0.1170* 0.0000

0.3317* 1.0000

-0.0362

-0.0517 0.0908*

1.0000


In a next step, all cooperative banks are split into four equal clusters by financial expertise consisting of 50 banks each in order to analyze the linkage between the different levels of financial expertise and the performance, risk, and stability of a bank. Table 6 shows the bank-level variables clustered by financial expertise. Cluster 1 comprises the banks with the highest levels of financial expertise, and Cluster 4 contains those with the lowest levels. The analysis indicates that financial expertise of the supervisory board tends to reduce stability. Another interpretation is that in situations of high risk cooperative banks might specifically appoint new supervisory board members with high expertise. Regional banks in Cluster 1 and Cluster 2 have a meaningful lower zscore and higher NPL ratio than those in Cluster 3 and 4. The increased risk-taking within Cluster 1 and 2 does not get paid off by higher performance. The performance of regional banks tends to deteriorate

with increasing financial expertise. This trend is consistent across both metrics RORWA and ROE from Cluster 1 to 3 whereas Cluster 4 shows a slightly lower performance than Cluster 3 and 2, but still higher than Cluster 1. Other variables worth looking at are bank size and efficiency. The analysis demonstrates that larger banks have a higher level of financial expertise. This could be traced back to the hypothesis that the larger the bank, the more complex its business model and strategy as well as the range of activities, and the higher the requirements for financial expertise are. In contrast to the size effect the variable bank efficiency shows that regional banks with high expertise reveal significantly higher CIRs than low-expertise regional banks which leads to the hypothesis that larger banks are less efficient.

Table 6. Bank-level variables, clustered by financial expertise Dependent variables

Cluster 1

RORWA (%) 1.05 ROE (%) 10.86 z-score 19.05 NPL ratio (%) 3.14 Bank level variables Financial expertise (%) 76.53 Outside financial expertise 72.95 (%) Bank size (ln ‘000s of 21.09 EUR) Bank efficiency (%) 67.92 Bank size growth (%) 1.82 Board size 12.65 Loan growth (%) 4.08 Claims on MFI (%) 12.06 Claims on non-MFI (%) 59.73

Cluster 2

Cluster 3

Cluster 4

1.24 12.37 22.18 2.70

1.34 13.54 24.52 2.59

1.22 11.99 25.06 2.22

Delta 1-4 -0.17 -1.13 -6.01 0.92

54.89

39.85

19.78

56.75

53.04

36.87

18.13

54.82

20.89

20.88

20.74

0.35

66.54 2.96 11.66 5.26 11.97 57.06

65.61 2.62 12.67 5.03 13.06 58.46

65.88 3.01 12.69 5.54 10.47 59.44

2.04 -1.19 -0.04 -1.46 1.59 0.29

The cluster analysis shows that the loan volume of Cluster 4, 3 and 2 have higher growth rates than Cluster 1. This suggests that loan growth could lead to stability. This finding is, however, inconsistent with the results of the correlation matrix. The correlation coefficient between loan growth and z-score is negative and significant on the 5% level. Therefore, we cannot conclude that loan growth leads to more stability. Table 7 provides evidence of the significance between the clusters. The performance and risk

Cluster

variables of Cluster 1 are significantly different to the other clusters. The z-score does not significantly change for the analysis between Cluster 3 and 4, whereas the NPL ratio is not significantly different between Cluster 2 and 3, respectively. We do not find any significance between Cluster 2 and 4 for the performance measures. However, we find in general that our performance and risk measures differ significantly between the clusters. Therefore, we conclude that financial expertise on supervisory boards have an impact on the risk-return profile.

859


Table 7. Pairwise test of the significance between expertise clusters Dependent variables RORWA ROE z-score NPL ratio

Cluster 1 vs. 2 t (p-value) -2.9746 *** 0.0032 -2.4392 ** 0.0153 -2.9416 *** 0.0035 2.9874 *** 0.0031

Cluster 1 vs. 3 t (p-value) -5.1122 *** 0.0000 -4.8278 *** 0.0000 -5.1116 *** 0.0000 4.0449 *** 0.0001

Cluster 1 vs. 4 t (p-value) -2.8960 *** 0.0041 -2.0750 ** 0.0388 -5.3941 *** 0.0000 6.0319 *** 0.0000

Cluster 2 vs. 3 t (p-value) -1.7083 * 0.0886 -2.0154 ** 0.0448 -2.1057 ** 0.0361 0.5127 0.6085

Cluster 2 vs. 4 t (p-value) 0.2822 0.7780 0.7044 0.4817 -2.2044 ** 0.0283 2.9269 *** 0.0037

Cluster 3 vs. 4 t (p-value) 2.2395 ** 0.0259 3.0085 *** 0.0028 -0.3978 0.6910 2.2893 ** 0.0228

Bank level variables 36.4796 *** 62.4921 *** 107.0168 *** 31.6913 *** 0.0000 0.0000 0.0000 0.0000 25.2058 *** 37.7623 *** 79.3250 *** 20.1509 *** Outside financial expertise 0.0000 0.0000 0.0000 0.0000 4.0484 *** 3.5593 *** 6.9548 *** 0.2905 Bank size 0.0001 0.0004 0.0000 0.7716 1.8552 * 3.4015 *** 3.2641 *** 1.2181 Bank efficiency 0.0645 0.0008 0.0012 0.2241 -3.3467 *** -1.9672 * -3.2731 *** 1.0335 Bank size growth 0.0009 0.0502 0.0012 0.3023 2.7820 *** -0.0637 -0.0953 -3.5862 *** Board size 0.0057 0.9492 0.9242 0.0004 -2.6742 *** -2.0712 ** -3.1177 *** 0.4078 Loan growth 0.0080 0.0393 0.0020 0.6837 0.1639 -1.9450 * 2.9122 *** -1.9462 * Claims on MFI 0.8699 0.0527 0.0039 0.0526 2.8703 *** 1.7566 * 0.3068 -1.4140 Claims on non-MFI 0.0044 0.0800 0.7592 0.1584 * indicates significance at the 10% level, ** at the 5% level, and *** at the 1% level, respectively Financial expertise

4.2 Regression Results The results of the random-effect regressions with the dependent variables RORWA and ROE are shown in Table 8. As indicated by the Wald-Chi square value, all regressions with RORWA and ROE as dependent variable are significant at the 1% level. The regression results document that financial expertise does not have a positive impact on the performance of regional banks. On the contrary, financial expertise tends to influence negatively the bank’s performance. However, the coefficients of total board financial expertise and outside financial expertise are not significant for RORWA and ROE. This finding is at odds with the corporate finance theory postulating appropriate compensation for risk according to the risk-return trade-off. Loan volume growth is next to board size the only variable which is positively associated with the performance whereas all other variables have a negative impact. The variable loan volume growth is significant on the 1% level, whereas board size shows only a positive tendency without significance. This finding suggests that the lending

58.9953 0.0000 47.0515 0.0000 3.3427 0.0009 0.8768 0.3813 -0.3325 0.7397 -2.8178 0.0052 -0.4854 0.6278 2.8777 0.0043 -2.4490 0.0149

*** 35.3432 0.0000 *** 21.5942 0.0000 *** 2.2812 0.0232 -0.4186 0.6758 -0.8804 0.3795 *** -0.0343 0.9726 -0.7861 0.4325 *** 4.6932 0.0000 ** -1.1065 0.2694

*** *** **

***

activities as core business and focus of a regional bank are still a meaningful profitability driver. The result of bank size, which is negative and highly significant, is in contrast to the hypothesis that larger boards might lead to members’ perceived lower personal responsibility for performance and, therefore, ease free-riding (Harris and Raviv, 2008). There is further evidence as shown by the considerations of Jensen (1993) as well as of Lipton and Lorsch (1992) that larger boards are detrimental to performance. With regards to the negative impact of bank size profitable growth of a regional bank is restricted due to the limitations in terms of highly profitable projects. Consequently large regional banks can only underwrite projects at some point with lower profitability leading to shrinking total returns. This might explain the effect that bank size growth reduces profitability. Remarkably, economic strength, defined as per capita GDP, is the only of the structural variables having a significant impact on performance, neither population density nor the regional aspects have a significant impact on performance.

860


Table 8. Random-effect regressions with dependent variables RORWA and ROE RORWA (RE) RORWA (RE) ROE (RE) ROE (RE) Financial expertise Out. financial expertise Financial expertise Out. financial expertise SE SE SE SE Coeff. Coeff. Coeff. Coeff. (robust) (robust) (robust) (robust) Financial expertise -0.0010 0.0013 -0.0033 0.0139 Financial Outside financial expertise -0.0012 0.0013 -0.0067 0.0136 expertise Bank size -0.2680 0.0598 *** -0.2720 0.0575 *** -1.9734 0.6668 *** -1.9690 0.6355 *** Bank efficiency -0.0262 0.0033 *** -0.0262 0.0034 *** -0.2707 0.0345 *** -0.2702 0.0345 *** Bank size growth -0.0408 0.0073 *** -0.0409 0.0073 *** -0.2505 0.0700 *** -0.2515 0.0699 *** Board size 0.0074 0.0068 0.0071 0.0069 0.0835 0.0661 0.0809 0.0667 Bank level Loan growth 0.0433 0.0041 *** 0.0434 0.0041 *** 0.2693 0.0334 *** 0.2692 0.0333 *** Claims on MFI -0.0088 0.0048 * -0.0088 0.0048 * -0.1319 0.0459 *** -0.1318 0.0459 *** Claims on Non-0.0144 0.0029 *** -0.0145 0.0029 *** -0.0634 0.0312 ** -0.0639 0.0311 ** MFI Macroeconomic Federal state -0.1279 0.0869 -0.1282 0.0850 0.5330 1.4624 0.5577 1.4346 and structural GDP per capita -0.0062 0.0042 -0.0061 0.0042 -0.1055 0.0433 ** -0.1047 0.0430 ** environment Population 0.0054 0.0334 0.0036 0.0332 0.0882 0.3083 0.0700 0.3079 constant 9.5131 1.1456 *** 9.6152 1.1022 *** 77.9055 13.5145 *** 78.0171 12.8751 *** Variable

Wald Chi2 279.63 *** 282.37 *** 179.88 *** 181.73 *** R-sq 0.2548 0.2553 0.1858 0.186 No of observations 1,116 1,116 1,116 1,116 * indicates significance at the 10% level, ** at the 5% level, and *** at the 1% level, respectively. Standard errors control for clustering at the bank level. Multicollinearity is controlled with variance inflation factors lower 2.

Table 9 shows the results of the random-effect regressions with the dependent variables z-score and NPL ratio. All regression models with z-score and NPL ratio as dependent variables are significant at the 1% level as well - as indicated by the Wald-Chi square value. In contrast to the performance regressions, the financial expertise coefficients are significant in the regressions of the risk variables. We find that financial expertise has a negative sign and the coefficient of outside financial expertise is significant at the 5% level, whereas the variable financial expertise is not significant for the z-score. The results suggest that financial expertise leads to greater risk-taking rather than stabilizing banks. This is supported by the observation that financial expertise tends to increase the loan portfolio risk. These results are remarkable as financial expertise reduces profitability and leads to greater risk taking. Boards of regional banks, in particular cooperative banks, are characterized by a relatively high fraction of entrepreneurs[4]. This group of professionals is supposed to have a solid understanding of business risk and selection. Our results show that they tend to overestimate their abilities though and advocate for higher risk taking. The results also assume that the members of cooperative banks might specifically appoint new supervisory board members with high expertise in situations of high risk. In this case the risk taking induces appointing board members with financial expertise. However, most of the individuals of the supervisory board have been member of the board before the financial crisis started. Moreover, we studied the annual reports of banks with high and low supervisory board expertise for divergences in reported risk taking strategies but we could not detect any indications in this direction. Nevertheless, we

cannot finally exclude the possibility of such a reverse causality. Compared to z-score, most of the bank-level control variables are significant in the regression with the NPL ratio as dependent variable. In this connection the variables show different impacts on loan portfolio risk and bank stability. We find that while loan growth increases the portfolio risk, bank size growth contributes to a lower NPL ratio. Interestingly this effect is conversely on the bank stability. Given the negative effect of loan volume growth on stability, we hypothesize that regional banks can only primarily grow by underwriting business which jeopardize the risk profile of the bank and subsequently its loan portfolio. Concerning the bank size and bank size growth, our analysis shows that these variables have a consistent positive effect on loan portfolio risk given the reduction of the NPL ratio, however both variables are not significant for the bank stability, measured by the z-score. Whereas regional banks can improve their loan portfolio and stability through economies of scale, our results indicate that there are not any further diversification effects on the loan portfolio side which help reducing the loan portfolio risk and increasing the bank stability. The negative impact of board size on the loan portfolio underlines the free-riding hypothesis from the performance regressions. The perceived lower personal responsibility in large boards might result in a less strict lending policy, which explains the increasing effect on the NPL ratio. The results also show a tendency that board size has a negative impact on the z-score, however the results are not significant.

861


Table 9. Random-effect regressions with dependent variables z-score and NPL ratio z-score (RE) z-score (RE) NPL Ratio (RE) NPL Ratio (RE) Financial expertise Out. financial expertise Financial expertise Out. financial expertise SE SE SE SE Coeff. Coeff. Coeff. Coeff. (robust) (robust) (robust) (robust) Financial expertise -0.0719 0.0453 0.0118 0.0041 *** Financial Outside financial expertise -0.0960 0.0408 ** 0.0122 0.0039 *** expertise Bank size -1.7478 2.4256 -1.9359 2.4345 -0.5093 0.1680 *** -0.4444 0.1633 *** Bank efficiency -0.0640 0.0254 ** -0.0627 0.0257 ** -0.0012 0.0059 -0.0011 0.0059 Bank size growth 0.0492 0.0585 0.0468 0.0578 -0.1578 0.0144 *** -0.1576 0.0144 *** Board size -0.1938 0.2619 -0.2124 0.2641 0.0442 0.0195 ** 0.0460 0.0196 ** Bank level Loan growth -0.0926 0.0385 ** -0.0911 0.0380 ** 0.0945 0.0106 *** 0.0941 0.0106 *** Claims on MFI -0.0067 0.0716 -0.0104 0.0724 0.0338 0.0116 *** -0.1576 0.0144 *** Claims on Non0.2081 0.1205 * 0.2021 0.1201 * 0.0292 0.0083 *** 0.0941 0.0106 *** MFI -4.6956 4.8127 -4.5279 4.9777 -0.2500 0.8592 -0.2207 0.8572 Macroeconomic Federal state and structural GDP per capita -0.0359 0.1503 -0.0249 0.1497 -0.0454 0.0139 -0.0459 0.0140 environment Population -0.8731 1.1510 -1.0190 1.1407 -0.2536 0.0815 -0.2458 0.0816 Constant 60.838 56.2943 66.290 57.1181 12.010 3.4530 *** 10.580 3.4139 *** Variable

Wald Chi2 65.45 R-sq 0.0959 No of observations 1,116

*** 64.05 0.1004 1,116

*** 268.99 0.2979 1,103

*** 269.11 0.3005 1,103

***

* indicates significance at the 10% level, ** at the 5% level, and *** at the 1% level, respectively. Standard errors control for clustering at the bank level. Multicollinearity is controlled with variance inflation factors lower 2

Greater financial expertise does not necessarily lead to greater controlling and monitoring of bank risk as shown by our analysis (e.g., Chuang and Lee, 2006, Fellner et al., 2004). Further evidence documents Odean (1998) who shows that the greater expertise individuals have, the more overconfident they tend to be. This reasearch field is also investigated by Dittrich et al. (2005) who demonstrate that overconfidence tends to increase with task complexity. The regression analyses show that financial expertise on supervisory boards of regional banks in the form of cooperative banks cannot contribute to more bank stability and less loan portfolio risk, rather lead to greater risk taking. These findings lend support to the evidence from Minton et al. (2010), who show that higher bank risk levels are associated with higher financial expertise. The detrimental impact on risk and stability suggests that financial experts explicitly advocate more risk-taking, which Acharya et al. (2011) and Wallison (2009) describe as a “race to the bottom” in relation to the mortage quality and ability to increase support for affordable housing. Overall, financial experts tend to deteriorate the risk-return profile. Since the negative impact on risk cannot be set off by a higher performance. Our results document that financial expertise does not improve bank profitability and suspends the risk-return trade-off.

4.3 Merger-controlled sample Besides our sample which consists of the largest 200 cooperative banks, we run our analysis on the 150 largest cooperative banks which have not been involved in any mergers activities over the panel horizon and the two years preceding this horizon in order to control for potential distortions for merger activities. Merger activities are one major confounding event for risk taking. We find 107 banks with merger activities in our 200 largest cooperative banks sample. Therefore, we extend the data set with 57 banks which have no merger activities in our investigation period, but are smaller than the 200 largest cooperative banks. In total 257 cooperative banks have been considered in this study. The merger-controlled banks account for 35% of the cooperative bank sector’s total assets. The average of total assets in the merger-controlled sample is EUR 1.6 billion. The findings of the mergercontrolled sample remain qualitatively identical. Table 10 shows the results of the random-effect regression for the merger-controlled sample with the dependent variables RORWA and ROE. We find that financial expertise has no significant impact on RORWA or ROE. This is consistent with the findings of the 200 largest cooperative banks without controlling for merger activities. The results also indicate that the findings are robust for the control variables. This suggests that merger activities do not change the results so far.

862


Table 10. Random-effect regression with dependent variable RORWA and ROE analyzing the 150 largest merger-controlled cooperative banks RORWA (RE) RORWA (RE) ROE (RE) ROE (RE) Financial expertise Out. financial expertise Financial expertise Out. financial expertise SE SE SE SE Coeff. Coeff. Coeff. Coeff. (robust) (robust) (robust) (robust) -0.0110 0.0170 -0.0008 0.0015

Variable

Financial expertise Outside financial expertise Bank size -2.0404 Bank efficiency -0.2993 Bank size growth -0.2776 Bank level Board size 0.1622 Loan growth 0.2688 Claims on MFI -0.1873 Claims on Non-MFI -0.0540 3.7231 Macroeconomic Federal state and structural GDP per capita 0.0259 environment Population 0.5925 constant 75.4903 Financial expertise

-0.0134 0.0166 0.6844 0.0389 0.0736 0.1056 0.0399 0.0530 0.0354 1.3706 0.0300 0.3859 13.2886

*** -2.0849 0.6603 *** -0.2991 0.0389 *** -0.2790 0.0737 0.1582 0.1065 *** 0.2690 0.0399 ** -0.1875 0.0530 *** -0.0556 0.0356 3.7074 1.3594 0.0260 0.0299 ** 0.5757 0.3876 *** 76.6627 12.9471

-0.0011 0.0015 *** -0.2937 0.0630 *** -0.0283 0.0037 *** -0.0435 0.0078 0.0094 0.0096 *** 0.0414 0.0045 ** -0.0132 0.0053 *** -0.0137 0.0034 0.2776 0.2216 0.0009 0.0028 ** 0.0813 0.0356 *** 9.6942 1.2090

*** -0.2965 0.0609 *** -0.0282 0.0037 *** -0.0436 0.0078 0.0090 0.0097 *** 0.0415 0.0045 *** -0.0132 0.0053 -0.0138 0.0034 *** 0.2768 0.2214 0.0010 0.0028 0.0796 0.0355 *** 9.7778 1.1848

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

***

Wald Chi2 164.84 *** 166.49 *** 239.51 *** 238.69 *** R-sq 0.2055 0.2056 0.2663 0.2664 No of observations 896 896 896 896 * indicates significance at the 10% level, ** at the 5% level, and *** at the 1% level, respectively. Standard errors control for clustering at the bank level. Multicollinearity is controlled with variance inflation factors lower 2.

The results of the random-effect regression for the merger-controlled sample with the risk variables zscore and NPL ratio are shown in Table 11. Financial expertise is still negative for the z-score, but financial expertise is not statistically significant. In the regression model of the non-controlled merger sample, outside financial expertise was significant at

the 5% level. This indicates that merger activities may influence the results of the z-score. However, we find that the NPL ratio for the merger-controlled sample is highly significant. Therefore, we conclude that our results are overall robust for merger activities. The results are not driven by merger activities.

Table 11. Random-effect regression with dependent variables z-score and NPL ratio analyzing the 150 largest merger-controlled cooperative banks z-score Financial expertise SE Coeff. (robust) Financial expertise -0.0213 0.0232 Financial Outside financial expertise expertise Bank size -3.4713 1.3903 Bank efficiency -0.0148 0.0079 Bank size growth -0.0571 0.0171 Bank level Board size 0.1439 0.1722 Loan growth 0.0022 0.0123 Claims on MFI -0.0337 0.0197 Claims on Non-MFI 0.0484 0.0261 -1.2522 3.1739 Macroeconomic Federal state and structuralGDP per capita 0.4232 0.4853 environment Population 0.0137 0.0136 constant 85.9926 27.8159 Variable

Wald Chi2 197.97 R-sq 0.0835 No of observations 896

(RE)z-score (RE)NPL Ratio (RE)NPL Ratio (RE) Out. financial expertise Financial expertise Out. financial expertise SE SE SE Coeff. Coeff. Coeff. (robust) (robust) (robust) 0.0115 0.0044 *** -0.0258 0.0229

0.0116 0.0043

** -3.4892 1.4183 * -0.0145 0.0080 *** -0.0575 0.0170 0.1368 0.1712 0.0025 0.0123 * -0.0347 0.0197 * 0.0479 0.0259 -1.2137 3.1599 0.3898 0.4798 0.0137 0.0138 *** 86.7101 28.3948

** -0.4423 0.1557 * -0.0025 0.0065 *** -0.1634 0.0161 0.0676 0.0301 0.0947 0.0112 * 0.0180 0.0124 * 0.0370 0.0091 1.0533 0.7308 -0.3939 0.0882 -0.0122 0.0042 *** 9.5360 3.3429

*** -0.3892 -0.0025 *** -0.1631 ** 0.0696 *** 0.0946 0.0183 *** 0.0377 1.0720 *** -0.3882 *** -0.0123 *** 8.3848

*** 196.09 0.0876 896

*** 292.94 0.347 885

*** 295.91 0.348 885

0.1580 0.0064 0.0161 0.0302 0.0112 0.0124 0.0092 0.7272 0.0887 0.0042 3.4395

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

* indicates significance at the 10% level, ** at the 5% level, and *** at the 1% level, respectively. Standard errors control for clustering at the bank level. Multicollinearity is controlled with variance inflation factors lower 2.

4.4 Robustness Checks In order to run the robustness checks, the panel of the largest 200 cooperative banks as well as the merger-

controlled sample is adjusted in two ways. All variables are truncated, except for the binary variable (federal state) and the index variable on population (area). First, all observations smaller than the 1st

863


percentile of the distribution are set to the value of the 1st percentile. Observations larger than the 99th percentile are set to the value of the 99th percentile. Second, the same procedures has been applied to all observations smaller than the 5th percentile. Overall, the outlier-corrected regression results can confirm the same results as per our original data set providing robustness to our findings. The models with outlier-corrected datasets have higher explanatory power with regards to the financial stability and risk than the models without outlier correction, as indicated by the higher R-squared. The significance of the financial expertise coefficients of regional banks increases as the risk regressions indicate. With regard to performance, all coefficients on financial expertise and a significant part on bank level retain their signs from the original regressions. Despite the outlier correction, the coefficients in the ROE and RORWA regression outcomes remain insignificant. A comparison with the coefficients of the outliercorrected data set underlines our initial findings. There is neither a change in the signs of the expertise coefficients nor a material difference between the original panel and the merger-adjusted panel due to the outlier corrections compared to the original regressions. Conclusion The present study examines the influence of financial expertise of supervisory board members on a bank’s risk-return profile based on German cooperative banks during the period 2004–2009. Our analysis demonstrates that financial experts in supervisory boards impact the risk-return profile. The risk-return trade off does not work for regional banks in which an increase in performance is associated with an increase in risk. The results are remarkably since financial expertise leads to greater risk-taking, measured as NPL ratio. However, we do not find evidence that financial expertise increases bank performance. The results indicate that cooperative banks are not able to benefit from financial experts. These findings are confirmed with a merger-controlled data set of the largest 150 banks without merger activities and outlier-corrected robustness checks. Boards of cooperative banks are characterized by a large fraction of self-employed people, making up 40% of the total supervisory board population. In contrast, self-employed individuals represent less than 13% of the supervisory board in savings banks. Interestingly, entrepreneurs do not exert a positive influence on the bank performance. Based on their profession and background this group of individuals is supposed to have a solid knowledge and distinctive awarenes of business selection and risktaking. This phenomen can be explained by a portion of overconfidence with a self-attribution bias where

people tend to ascribe any success they in some activity to their own abilities and talents (Barberis and Thaler, 2003). Odean (1998) demonstrates that the greater the expertise individuals have, the more overconfident they tend to be. Further support is exhibited by the analysis of Dittrich et al. (2005) showing that overconfidence is positively related to the task complexity. Employed people with an average value of 50% of the total supervisory board represent a large fraction and notable reservoir of financial expertise. Insiders and politicians play a minor role representing 5% and 3% respectively of the board members. Therefore, there is hardly any change between the results for insider and outsider financial expertise. From a regulator’s perspective the results provide evidence that the prescription of universal financial expertise in bank’s internal governance mechanisms does not lead to the desired effect of increasing the stability of the banking system. Our findings suggest that the Act to Strengthen Financial Market and Insurance Supervision is not necessarily contributing to an enhancement of banking system stability as it was intended to. However, we do not compare the composition of financial expertise in supervisory boards before and since the introduction of the new German regulation. This unresolved question could provide a promising avenue for future research. In addition, our findings leave the research question what the optimal threshold of the board composition with regards to financial expertise and board size is in order to increase the stability of the banking system. Endnotes: [1] Strong bank boards are characterized as boards representing more of bank shareholders interest. [2] The full dataset comprises 39,365 data points of 257 cooperative banks and 209 savings banks at the individual level of the supervisory board members. [3] We compare the supervisory board composition of the largest 200 cooperative banks with the 200 largest savings banks as of the end of 2009. The ratio of employed individuals in savings banks is more than 73%, whereas self-employed individuals represent less than 13% of the supervisory board. [4] We find 40% self-employed individuals in supervisory boards of cooperative boards, whereas in savings banks the ratio of self-employed individuals to all board members is less than 13%. References: 1. Acharya, V., Richardson, M., van Nieuwerburgh, S. and White, L. J. (2011), Guaranteed To Fail: Fannie Mae, Freddie Mac and the Debacle of Mortgage Finance, Princeton University Press., Princeton. 2. Acharya, V. and Yorulmazer, T. (2007), "Too many to fail - An analysis of time-inconsistency in bank closure

864


3.

4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

policies", Journal of Financial Intermediation, Vol. 16 No. 1, pp. 1-31. Adams, R., Almeida, H. and Ferreira, D. (2005), "Powerful CEOs and their impact on corporate performance", Review of Financial Studies, Vol. 18, pp. 1403-1432. Adams, R. B. and Mehran, H. (2008), Corporate performance, board structure, and their determinants in the banking industry, Staff Reports, No. 330. Federal Reserve Bank of New York, New York. Aebi, V., Sabato, G. and Schmid, M. (2012), "Risk management, corporate governance, and bank performance in the financial crisis", Journal of Banking & Finance, Vol. 36 No. 12, pp. 3213-3226. Altunbas, Y., Evans, L. and Molyneux, P. (2001), "Bank Ownership and Efficiency", Journal of Money, Credit and Banking, Vol. 33 No. 4, pp. 926-954. Altunbas, Y., Manganelli, S. and Marques-Ibanez, D. (2011), “Bank risk during the great recession: Do business models matter?”, ECB Working Papers Series. Ard, L. and Berg, A. (2010), “Bank Governance Lessons from the Financial Crisis, World Bank, Policy Briefs, Note No. 13”, The World Bank, Washington D.C. Ayadi, R., Llewellyn, D. T., Schmidt, R. H., Arbak, E. and Groen, W. P. D. (2010), “Investigating Diversity in the Banking Sector in Europe - Key Developments, Performance and Role of Cooperative Banks”, Center for European Policy Studies, Brussels. BaFin and Deutsche Bundesbank (2010), "Merkblatt zur Kontrolle von Mitgliedern von Verwaltungs- und Aufsichtsorganen gemäß KWG und VAG" [Online], available at: http://www.bafin.de/cln_152/nn_721290/SharedDocs/V eroeffentlichungen/DE/Service/Merkblaetter/mb__1002 22__Verwaltungs__und__Aufsichtsorgane.html?__nnn= true [Accessed 04 July 2015]. Barberis, N. and Thaler, R. (2003), A survey of behavioral finance, Elsevier, Amsterdam. Barth, J. R., Caprio, G. J. and Levine, R. (2008), Bank Regulations Are Changing: For Better or Worse?, World Bank Policy Research, Working Paper Series, no. 4646. The World Bank, Washington D.C. Beck, T., Hesse, H., Kick, T. and Westernhagen, N. v. (2009), "Bank Ownership and Stability: Evidence from Germany", FDIC - Center for Financial Research. Beck, T., Levine, R. and Loayza, N. (2000), "Finance and the sources of growth", Journal of Financial Economics, Vol. 58 No. 1, pp. 261-300. Berger, A. N., Clarke, G. R. G., Cull, R., Klapper, L. and Udell, G. F. (2005), "Corporate governance and bank performance: A joint analysis of the static, selection, and dynamic effects of domestic, foreign, and state ownership", Journal of Banking & Finance, Vol. 29 No. 8-9, pp. 2179-2221. Berger, A. N., Kick, T. and Schaeck, K. (2012), "Executive board composition and bank risk taking", Journal of Corporate Finance, Vol. 28, pp. 48-65. Berger, H. and Nitsch, V. (2011), "Too Many Cooks? Committees in Monetary Policy", Southern Economic Journal, Vol. 78 No. 2, pp. 452-475. Bermig, A. and Frick, B. (2010), "Board Size, Board Composition, and Firm Performance: Empirical Evidence from Germany", Working Paper, University of Paderborn, Paderborn. Bertrand, M. and Schoar, A. (2003), "Managing with style: The effect of managers on firm financial policies",

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

865

Quarterly Journal of Economics, Vol. 118, pp. 11691208. Bhattacharya, S., Boot, A. W. A. and Thakor, A. V. (1998), "The economics of bank regulation", Journal of Money, Credit, and Banking, Vol. 30 No. 4, pp. 745770. Boyd, J. H. and Runkle, D. E. (1993), "Size and performance of banking firms: Testing the predictions of theory", Journal of Monetary Economics, Vol. 31 No. 1, pp. 47-67. Brown, C. O. and Dinç, I. S. (2011), "Too Many to Fail? Evidence of Regulatory Forbearance When the Banking Sector Is Weak", Review of Financial Studies, Vol. 24 No. 4, pp. 1378-1405. Carcello, J. V., Hollingsworth, C. W., Klein, A. and Neal, T. L. (2008), "Audit Committee Financial Expertise, Competing Corporate Governance Mechanisms, and Earnings Management in a Post-SOX World". University of Illinois 18th Symposium on Audit Research, Illinois. Choundhry, M. (2011), "Effective bank corporate governance: observations from the market crash and recommendations for policy", Journal of Applied Finance & Banking, Vol. 1 No. 1, pp. 179-211. Chuang, W.-I. and Lee, B.-S. (2006), "An empirical evaluation of the overconfidence hypothesis", Journal of Banking & Finance, Vol. 30 No. 9, pp. 2489-2515. Cihák, M. and Hesse, H. (2007), "Cooperative Banks and Financial Stability". Working Paper Series, International Monetary Fund, Washington D.C. Cunningham, L. A. (2008), "Rediscovering Board Expertise: Legal Implications of the Empirical Literature", University of Cincinnati Law Review, Vol. 77, p. 465. Davidson, W. N., Xie, B. and Xu, W. (2004), "Market reaction to voluntary announcements of audit committee appointments: The effect of financial expertise", Journal of Accounting and Public Policy, Vol. 23 No. 4, pp. 279-293. DeFond, M. L., Hann, R. N. and Hu, X. (2005), "Does the Market Value Financial Expertise on Audit Committees of Boards of Directors?", Journal of Accounting Research, Vol. 43 No. 2, pp. 153-193. Dell'Ariccia, G., Detragiache, E. and Rajan, R. (2008), "The real effect of banking crises", Journal of Financial Intermediation, Vol. 17 No. 1, pp. 89–112. Demirgüç-Kunt, A. and Huizinga, H. (2010), "Bank activity and funding strategies: The impact on risk and returns", Journal of Financial Economics, Vol. 98 No. 3, pp. 626-650. Dermine, J. (2003), "Banking in Europe: Past, present and future", Gaspar, V., Hartmann, P., Sleijpen, O. (Ed.), Book, Banking in Europe: Past, present and future, ECB, Frankfurt am Main. DeYoung, R., Hunter, W. C. and Udell, G. F. (2004), "The past, present, and probable future for community banks", Journal of Financial Services Research, Vol. 25 No. 2/3, p. 85. Dhaliwal, D. S., Naiker, V. and Navissi, F. (2006), "Audit Committee Financial Expertise, Corporate Governance and Accruals Quality: An Empirical Analysis". Working Paper, University of Arizona, Tucson. Diamond, D. W. (1991), "Monitoring and Reputation: The Choice between Bank Loans and Directly Placed Debt", Journal of Political Economy, Vol. 99 No. 4, pp. 689-721.


36. Diamond, D. W. and Rajan, R. G. (2001), "Liquidity risk, liquidity creation, and financial fragility: A theory of banking", Journal of Political Economy, Vol. 109 No. 2, pp. 287-327. 37. Dittrich, D. A. V., Güth, W. and Maciejovsky, B. (2005), "Overconfidence in investment decisions: An experimental approach", The European Journal of Finance, Vol. 11 No. 6, pp. 471-491. 38. Eisenberg, T., Sundgren, S. and Wells, M. T. (1998), "Larger board size and decreasing firm value in small firms", Journal of Financial Economics, Vol. 48 No. 1, pp. 35-54. 39. Farhi, E. and Tirole, J. (2012), "Collective Moral Hazard, Maturity Mismatch and Systemic Bailouts", American Economic Review, Vol. 102 No., pp. 60-93. 40. Fellner, G., Güth, W. and Maciejovsky, B. (2004), "Illusion of expertise in portfolio decisions: an experimental approach", Journal of Economic Behavior & Organization, Vol. 55 No. 3, pp. 355-376. 41. Fernandes, N. and Fich, E. M. (2009), "Does financial experience help banks during credit crises?", IMD, Lausanne. 42. Giammarino, R. M., Lewis, T. R. and Sappington, D. E. M. (1993), "An incentive approach to banking regulation", The Journal of Finance, Vol. 48 No. 4, pp. 1523-1542. 43. Goddard, J., Molyneux, P., Wilson, J. O. S. and Tavakoli, M. (2007), "European banking: An overview", Journal of Banking & Finance, Vol. 31 No. 7, pp. 19111935. 44. Güner, A. B., Malmendier, U. and Tate, G. (2008), "Financial expertise of directors", Journal of Financial Economics, Vol. 88 No. 2, pp. 323-354. 45. Harris, M. and Raviv, A. (2008), "A theory of board control and size", Review of Financial Studies, Vol. 21 No. 4, pp. 1797-1832. 46. Hasan, I., Schmiedel, H. and Song, L. (2012), "Returns to retail banking and payments", Journal of Financial Services Research, Vol. 41 No. 3, pp. 163-195. 47. Hau, H. and Thum, M. (2009), "Subprime crisis and board (in-)competence: private versus public banks in Germany", Economic Policy, Vol. 24 No. 60, pp. 701752. 48. Hermalin, B. E. and Weisbach, M. S. (2003), "Boards of directors as an endogenously determined institution: a survey of the economic literature", Economic Policy Review, Apr, pp. 7-26. 49. Hoggarth, G., Reidhill, J. and Sinclair, P. (2004), "On the resolution of banking crises: theory and evidence", Working Paper No. 229, Bank of England, London. 50. Hopt, K. J. and Leyens, P. C. (2004), "Board Models in Europe - Recent Developments of Internal Corporate Governance Structures in Germany, the United Kingdom, France, and Italy", European Company and Financial Law Review, Vol. 1 No. 2, pp. 135-168. 51. Iannotta, G., Nocera, G. and Sironi, A. (2007), "Ownership structure, risk and performance in the European banking industry", Journal of Banking & Finance, Vol. 31 No. 7, pp. 2127-2149. 52. Jensen, M. C. (1993), "The modern industrial revolution, exit, and the failure of internal control systems", The Journal of Finance, Vol. 48 No. 3, pp. 831-880. 53. John, K. and Senbet, L. W. (1998), "Corporate governance and board effectiveness", Journal of Banking & Finance, Vol. 22 No. 4, pp. 371-403. 54. Jungmann, C. (2006), "The effectiveness of corporate governance in one -tier and two-tier board systems –

55.

56.

57.

58.

59.

60.

61.

62.

63.

64. 65.

66.

67.

68.

69.

70.

71.

72.

73.

866

Evidence from the UK and Germany", European Company and Financial Law Review, Vol. 3 No. 4, pp. 426-474. Kaplan, S. N. (1994), "Top executives, turnover, and firm performance in Germany", Journal of Law, Economics, & Organization, Vol. 10 No. 1, pp. 142-159. Kaufman, G. G. (2002), "Too big to fail in banking: What remains?", The Quarterly Review of Economics and Finance, Vol. 42 No. 3, pp. 423-436. Kirkpatrick, G. (2009), "The corporate governance lessons from the financial crisis", OECD Journal: Financial Market Trends, Vol. 2009 No. 1, pp. 61-87. Körner, T., Müller, O., Paul, S. and Schmidt, C. M. (2014), “Glas halb voll oder halb leer? Eine Analyse der Qualifikation von Kontrollorganmitgliedern deutscher Banken”, RWI Materialien. Kroszner, R. S., Laeven, L. and Klingebiel, D. (2007), "Banking crises, financial dependence, and growth", Journal of Financial Economics, Vol. 84 No. 1, pp. 187228. Laeven, L. and Levine, R. (2009), "Bank governance, regulation and risk taking", Journal of Financial Economics, Vol. 93 No. 2, pp. 259-275. Lang, G. and Welzel, P. (1996), "Efficiency and technical progress in banking Empirical results for a panel of German cooperative banks", Journal of Banking & Finance, Vol. 20 No. 6, pp. 1003-1023. Lee, Y. S., Rosenstein, S. and Wyatt, J. G. (1999), "The value of financial outside directors on corporate boards", International Review of Economics & Finance, Vol. 8 No. 4, pp. 421-431. Levine, R. (1997), "Financial Development and Economic Growth: Views and Agenda", Journal of Economic Literature, Vol. 35 No. 2, pp. 688-726. Levine, R. (2006), Finance and Growth: Theory and Evidence, Elsevier, Amsterdam. Lipton, M. and Lorsch, J. W. (1992), "A modest proposal for improved corporate governance", Business Lawyer, Vol. 48 No. 1, pp. 59-77. Malmendier, U. and Tate, G. (2005), "CEO overconfidence and corporate investment", Journal of Finance, Vol. 60, pp. 2661-2700. Manne, H. G. (1965), "Mergers and the market for corporate control", Journal of Political Economy, Vol. 73 No. 2, pp. 110-120. Manove, M. and Padilla, A. J. (1999), "Banking (conservatively) with optimists", RAND Journal of Economics, Vol. 30 No. 2, pp. 324-350. Mehran, H., Morrison, A. and Shapiro, J. (2011), "Corporate governance and banks: What have we learned from the financial crisis?". Staff Reports, No. 502, Federal Reserve Bank of New York, New York. Mercieca, S., Schaeck, K. and Wolfe, S. (2007), "Small European banks: Benefits from diversification?", Journal of Banking & Finance, Vol. 31 No. 7, pp. 1975-1998. Minton, B. A., Taillard, J. P. A. and Williamson, R. (2010), "Do independence and financial expertise of the board matter for risk taking and performance?". Fisher College of Business Working Paper Series, 2010-03014, Ohio State University, Columbus. Odean, T. (1998), "Volume, volatility, price, and profit when all traders are above average", The Journal of Finance, Vol. 53 No. 6, pp. 1887-1934. Pang, J. and Wu, H. (2009), "Financial markets, financial dependence, and the allocation of capital", Journal of Banking & Finance, Vol. 33 No. 5, pp. 810818.


74. Pathan, S. (2009), "Strong boards, CEO power and bank risk-taking", Journal of Banking & Finance, Vol. 33 No. 7, pp. 1340-1350. 75. Peni, E. and Vähämaa, S. (2012), "Did good corporate governance improve bank performance during the financial crisis?", Journal of Financial Services Research, Vol. 41 No. 1-2, pp. 19-35. 76. Rajan, R. G. (1992), "Insiders and outsiders: The choice between Informed and Arm's-Length Debt", The Journal of Finance, Vol. 47 No. 4, pp. 1367-1400. 77. Rajan, R. G. and Zingales, L. (1998), "Financial dependence and growth", The American Economic Review, Vol. 88 No. 3, pp. 559-586. 78. Rosenstein, S. and Wyatt, J. G. (1990), "Outside directors, board independence, and shareholder wealth", Journal of Financial Economics, Vol. 26 No. 2, pp. 175191. 79. Roy, A. D. (1952), "Safety first and the holding of assets", Econometrica, Vol. 20 No. 3, pp. 431-449. 80. Santos, J. A. C. (2001), "Bank capital regulation in contemporary banking theory: A review of the literature", Financial Markets, Institutions & Instruments, Vol. 10 No. 2, pp. 41-84. 81. Schmielewski, F. and Wein, T. (2012), "Are private banks the better banks? An insight into the principal– agent structure and risk-taking behavior of German banks", Journal of Economics and Finance, Vol. 39 No. 3, 518-540. 82. Spong, K. R. and Sullivan, R. J. (2010), "Bank ownership and risk taking: Improving corporate

83.

84.

85.

86.

87.

88.

89.

90.

867

governance in banking after the crisis", Federal Reserve Bank of Kansas City. Stein, J. C. (2002), "Information production and capital allocation: Decentralized versus hierarchical firms", The Journal of Finance, Vol. 57 No. 5, pp. 1891-1921. Swan, P. L. and Forsberg, D. (2014), "Does board “independence” destroy corporate value", Working Paper. Walker, D. (2009), "A review of corporate governance in UK banks and other financial industry entities: Final recommendations.", The Walker Review Secretariat, London. Wallison, P. J. (2009), "Cause and effect: Government policies and the financial crisis", Critical Review, Vol. 21 No. 2-3, pp. 365-376. Westman, H. (2011), "The impact of management and board ownership on profitability in banks with different strategies", Journal of Banking & Finance, Vol. 35 No. 12, pp. 3300-3318. Wintoki, M. B., Linck, J. S. and Netter, J. M. (2012), "Endogeneity and the dynamics of internal corporate governance", Journal of Financial Economics, Vol. 105 No. 3, pp. 581-606. Yermack, D. (1996), "Higher market valuation of companies with a small board of directors", Journal of Financial Economics, Vol. 40 No. 2, pp. 185-211. Zhou, C. (2010), "Are banks too big to fail? Measuring systemic importance of financial institutions", International Journal of Central Banking, Vol. 34 No. 6, pp. 205-250.


OWNERSHIP CONCENTRATION AND INVESTMENT SENSITIVITY TO MARKET VALUATION Ghada Tayea*

Abstract During the past decade, Jordan has undertaken substantial reforms aiming at restructuring its stock market in order to strengthen its role in promoting investment and allocating capital efficiently. This paper empirically investigates the impact of stock market development on capital investment at the firm level by assessing the investment-q sensitivity. In addition, this paper examines the impact of concentrated ownership, a salient institutional feature of listed Jordanian companies, on the investment-q sensitivity. The findings of this study indicate that investments by Jordanian firms respond significantly and positively to market signals. Furthermore, the results show that a company responds more efficiently to market signals as ownership concentration increases, which suggests that large ownership stakes align the interests of large shareholders with those of the firm. Keywords: Jordan, Investment Efficiency, Tobin Q, Concentrated Ownership, Largest Shareholder Jel Classification: G31, G34 * Assistant Professor in Finance. The University of Jordan, School of Business, Amman 11942, Jordan

1.

Introduction

The allocation of capital to its most productive investment uses is a fundamental question in corporate finance (Stein, 2003). However, the role of the stock market in this allocation process has been long debated.[1] According to Tobin’s (1969) q theory of investment and its extension by Hayashi (1982), a firm’s investment is directly related to stock market valuation of that firm. Firms with high market valuation to replacement cost, a proxy of marginal q, respond by increasing their investment. This link between investment and stock market valuation implies that insiders learn new information from the stock market about a firm’s growth opportunities. Theoretical literature suggests that the stock market plays an important role in aggregating information not known to insiders from outside investors and hence managers can learn new information about their own firms from the stock price (Dow and Gorton, 1995, Subrahmanyam and Titman, 1999). For example, external investors are likely to have access to critical information not available to insiders about customer demand for firm’s products and competition with other firms (Chen et al., 2007). Therefore, under this theoretical framework, stock prices can serve as a useful signal that aids insiders in mobilizing capital towards the most value adding investment opportunities. Recent empirical evidence supports this notion, as for example, Chen et al. (2007) shows that

1

This point is discussed in more details in Section 2.

investment-q sensitivity improves as stock price informativeness increases. The above argument, nonetheless, assumes that ownership structure is irrelevant in the process of learning from the stock price. However, Jiang et al. (2011) argue that the pyramid ownership structure of East Asian companies can affect the investment-q sensitivity. According to the authors, controlownership wedge allows for sizeable divergence between voting and cash flow rights and hence gives controlling shareholders incentives to extract private benefits at the expense of minority shareholders. Jiang et al. (2011) find that companies with larger controlownership wedge are more likely to ignore signals from the stock price, which in turn weakens the investment-q sensitivity. In addition, Andres (2011) studies family ownership of listed German companies and report that the investment-q sensitivity is significant only for family firms. This study complements this literature by examining the influence of other aspects of ownership structure, namely the degree of concentrated ownership and the ownership of the largest shareholder, on strengthening/attenuating the link between market valuation and a firm’s capital investments. Jordanian companies listed in the Amman Stock Exchange (ASE) are mostly characterized by highly concentrated ownership structures. In this context, it is difficult to gauge the propensity of companies with highly concentrated ownership to listen to the market. On one hand, concentrated ownership allows shareholders with large ownership stake to exercise control over the firm. This is especially true for

868


Jordanian firms where the largest shareholder typically assumes the responsibilities of the Chairman of the Board and/or the CEO (Abdel-Halim and Bino, 2013). Hence, in the case of Jordan the largest shareholder becomes effectively an insider. Therefore, large shareholders have incentives and discretionary power to extract private benefits in the form of suboptimal investment which weakens their propensity to learn from the market and attenuates the investment-q sensitivity.[2] Conversely, and similar to Jensen and Meckling (1976), as ownership stake of shareholders increases, private benefits of sub-optimal investment may not exceed benefits of efficient investment and hence the investment-q sensitivity is strengthened as ownership increases. In this study, therefore, we resort to empirical findings to resolve the differential nature of investment sensitivity to stock price that arises from concentrated ownership. Examining the investment behavior in Jordan is interesting for several reasons. First: there are few studies that link the analysis of investment-q sensitivity to aspects of corporate governance.[3] For example, Jaing et al. (2011) examine the case of six East Asian countries characterized by pyramid ownership structures that give rise to divergence between voting and cash flow rights and hence agency problems between insiders and minority shareholders. However, Jordanian companies are characterized by concentrated ownership with largest shareholders exerting control over the firm, a structure that can align the interests of large shareholders with those of the firm. Andres (2011) examines the case of family businesses in Germany, however, his study focuses mainly on the investment cash flow sensitivity and pays only little attention to the investment-q sensitivity. Second: there is little known about the behavior of investment in the context of small emerging economies, such as Jordan. Most studies focus on the US and other well-established markets. However, there is large disparity of capital expenditures among economies. For example, the mean (median) for the change in net fixed assets scaled by capital and book assets for Jordanian listed industrial companies are 0.009 (-0.007) and 0.004 (.024) respectively. Conversely, the mean (median) for the change in net fixed assets scaled by total assets reported in Jaing et al. (2011) for six East Asian 2

In Jensen (1986), managers with free cash flows have incentives and abilities to overinvest as it increases their personal utility. 3 It is important to note, though, that the link between investment-cash flow sensitivity and corporate governance has been extensively examined. For example, Wei and Zhang (2008) find that for listed companies in eight East Asian countries investment cash flow sensitivity increases as the degree of the divergence between control and cash-flow rights increases. Andres (2011) finds that family businesses in Germany are less sensitive to internal cash flows. Pawlina and Renneboog (2005) find that outside blockholders in UK companies reduce the cash flow sensitivity of investment via effective monitoring. Other contributions include Goergen and Renneboog (2001) for the UK, Gugler (2003) for Austria and Haid and Weigand (2001) for Germany.

markets is 0.038 (0.012) and the mean for the ratio of capital expenditures scaled by net fixed assets reported in Andres (2011) for Germany is 0.255.[4] The disparity of capital expenditures calls for greater attention to the investment behavior of small emerging economies. Third: the study of the investment behavior and the role played by the stock market in enhancing investment efficiency has far reaching implications for emerging markets. The last two decades have witnessed unprecedented growth in equity markets globally as many countries progressed to marketbased economy, with Jordan being no exception. Jordan officially launched its economic reforms in 1989 in response to a severe economic crisis (Harrigan et al., 2006). However, it was not until the end of the 1990s that Jordan took serious steps to reform and liberalize its economy (Harrigan et al., 2006). The economic reform consists of plans aimed at strengthening the private sector and limiting the role of the public sector in economic activities with the Jordanian stock market being in the heart of these reforms. Securities Law of 1997 was enacted under which the stock market was restructured to include three distinct institutions: Amman Stock Exchange (ASE); Securities Depository Centre (SDC); and the supervisory body Jordan Securities Commission (JSC). In addition, electronic trading, settlement and clearing systems were introduced and the Securities Law of 1997 was supplemented with numerous bylaws and regulations adhering to corporate governance best practice (Zeitun, 2006, Kanaan and Kardoosh, 2005). Therefore, it is important to evaluate the success of equity markets of emerging economies in enhancing investment efficiency. Using a q theory framework, this study documents a positive and significant impact of stock market valuation on a firm’s investment decisions. The results show that capital expenditures of industrial companies listed in the ASE are positively related to stock market signals as captured in average Q. In addition, the results show that concentrated ownership strengthens the impact of investment-q sensitivity. Using alternatives measures of concentrated ownership, including the sum of the largest three shareholders and the percentage and existence of a large shareholder, I find that the interaction term between concentrated ownership and stock market valuation is positive and significant. This result indicates that greater ownership stakes increase the propensity of the firm to listen to the market. On a closer examination, the results show that there is a positive and significant influence of ownership on a firm’s investment-q sensitivity in the ownership range beyond 20% and insignificant influence below that bound. These results imply that, when largest 4

The choice to report the fixed assets ratios from Jiang et al. (2011) and Andres (2011) is because these two studies are the closest to this paper. Similar conclusion can be drawn by from other studies that examine the investment behavior.

869


shareholders own a small stake, and in the absence of other effective corporate governance and market discipline mechanisms, there is little incentive for the firm to follow market signals.[5] However, when ownership stakes increase, largest shareholders are subject to a larger share of the costs of squandering corporate wealth (Jensen and Meckling, 1976) and therefore their propensity to listen to the market increases. This explanation is consistent with arguments from studies on the impact of insider ownership on investment behavior where they find that for some range of ownership there is an alignment effect between the interests of insiders and shareholders (Morck et al., 1988, Cho, 1998, Pawlina and Renneboog, 2005). The remainder of this article is organized as follows. The next section contains a short review of related literature. Section 3 reviews the q theory of investment and the econometric specifications. Section 4 provides information on the construction of the dataset, variable definitions and descriptive statistics. Section 5 presents the regression analysis and section 6 concludes the paper. 2. Stock Markets, Ownership

Investment,

and

The stock market is important to an economy in insuring investors against idiosyncratic risk, disseminating more information about investment projects, exerting corporate governance, and hence raising the rate of return to economy wide investment (Holmstrom and Tirole, 1993, Atje and Jovanovic, 1993, Levine and Zervos, 1996). Indeed, empirical studies find a positive impact of stock market development on economic growth using pooled crosscountry analysis (Levine and Zervos, 1996, Levine and Zervos, 1998, Arestis et al., 2001, Beck and Levine, 2004, Ndikumana, 2005). One important area of this literature is concerned with investigating the role of the stock market in allocating investment efficiently. The seminal work by Barro (1990) examines the relation between market valuation using index returns on private domestic investment in the US over the period 1891-1987 and documents a significant impact of market returns on investment. Similarly, Henry (2000) finds that the growth rate of private domestic investment for eleven developing countries that liberalized their stock market has increased significantly following stock market liberalization. However, as Samuel (1998) points out investigating investment at the firm level is more appropriate given that investment decisions are made 5

It is well documented that emerging markets in general have weak legal protection rights (La Porta et al., 1999, La Porta et al., 2000) and higher levels of insider trading, price and market manipulation and false disclosure (Cumming et al., 2011) which makes it reasonable to conclude that effective corporate governance and market discipline mechanisms in Jordan are weak.

at the firm level. Moreover, Morck et al. (1990) note that the stock market is likely to play a role in allocating investments across sectors and firms more than over time, which renders the study of investment at a firm level more relevant. At the heart of the link between market valuation and investment at the firm level is the question on whether insiders learn new information from external investors through signals contained in the stock price. This question has been long-debated as external investors have smaller information sets about firm’s investment opportunities than managers and therefore managers can safely ignore stock market movements (Morck et al., 1990). However, theoretical models in Dow and Gorton (1995) and Subrahmanyam and Titman (1999) show that the stock price contains new information that is aggregated from external investors and hence managers can improve their investment decisions by observing stock-price movements. The empirical evidence documented in Morck et al. (1990) shows that for US non-financial firms there is a weak influence of stock return on firm level investment which suggests that the stock market is not central for firm-level investments. This finding is reinforced in Samuel (1998) where he finds that a q model performs relatively poor to other investment models in explaining US capital expenditures. However, Chen, Goldstein, and Jiang (2006) examine the connection between the sensitivity of investment to Tobin’s q and find that the investment-q sensitivity increases when measures of external information embedded in the stock price improve. This result, according to the authors, indicates that managers learn new information from stock prices when they make investment decisions. Luo (2005) arrives at a similar conclusion by looking at merger deal completion where the author finds that returns on merger announcements successfully predict deal completion. The evidence from Arab countries on investment-q sensitivity is rare. In Bolbol and Omran (2005) the authors examine firms from five Arab countries and find no evidence that stock return is related to investment. Their evidence, however, is based on a small number of firms (83 firms from five countries). Therefore, this study aims to complement the international evidence by thoroughly examining the investment-q sensitivity in the context of Jordan, a small emerging market. 3. Methodology: Economic Model and Estimation Specifications To answer the questions of this study I utilize Tobin’s q theory of investment (Tobin, 1969). The advantage of this theory is that it incorporates future expected profits and hence links market valuation of the firm to its investment decision. The q theory is derived from a profit maximization function that includes real rental price of capital (that is the price of output relative to the price of investment goods) and costs of adjustment

870


(such as costs of installation and training). The first order condition of this function yields the following equation: (𝐼𝑡 /𝐾𝑡 ) = 𝑓(𝑞𝑡 )

(1)

where I is firm’s investment in period t; K is the capital stock; and q is the marginal q defined as the increase in market value due to having an additional unit of capital. According to the definition of marginal q, it is optimal for a firm to invest when q exceeds 1. One main problem with estimating Equation 1 is that q is not observed since it is a marginal amount. However, in his seminal work, Hayashi (1982) shows that under certain assumptions, average Q can be used as a proxy of marginal q, where Q is the current market value of the firm divided by the replacement cost of the firm’s capital. Hence, Equation 1 is modified as a linear function of average Q: (𝐼𝑡 /𝐾𝑡 ) = 𝑎 + 𝑏𝑄𝑡 + 𝑒𝑡

(2)

This study defines Q as the ratio between the sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. However, many empirical studies document that Q alone fails to fully explain the behavior of a firm’s investment due to the existence of capital market frictions that cause firms in the real world to deviate from the optimal investment behavior (see Stein, 2003 for a comprehensive review). Early studies have focused on the impact of liquidity constraints and financing frictions on a firm’s investment behavior (Fazzari et al., 1988, Hubbard, 1997, Kaplan and Zingales, 1997). Financial constraints relates to the differential cost between internal and external financing, with a premium required on external financing. This has the effect of a constraining a firm’s ability to utilize all of its investment opportunities. In addition, recent empirical studies document evidence that a firm’s financial leverage determines in part its investment (Aivazian et al., 2005). The number of employees is included as a proxy for a firm’s size and is expected to be negatively correlated with the firm’s investment (Shanmugam and Bhaduri, 2002). Finally, I include the current ratio to capture a firm’s liquidity position (Xiao, 2009). Therefore, equation 2 is modified to include a vector of control variables as follows:

the specification in equation 3 includes firm effects to control for unobserved firm heterogeneity (Bond and Meghir, 1994) and time fixed effects in order to account for macroeconomic conditions (Xiao, 2009). Finally, the specification uses one period lag for all control variables since an investment decision at time t is likely to be influenced by information available at the beginning of the period (Xiao, 2009). Using lagged explanatory variable has the added benefit of alleviating endogeneity. This study main question relates to the impact of ownership structure on a firm’s investment decision. Following Claessens et al. (2000) I include Large3 which is the total ownership of the largest three shareholders owning 5% or more of a company (I vary the 5% cut-off in the robustness checks). In addition, I examine the impact of the largest shareholder Largest, defined as the percentage ownership of the largest shareholder, on a firm’s investment decision. I also consider a third ownership measure, LargestDum, which is an indicator variable that takes the value of one if a firm has a largest shareholder with an ownership of 20% or more and zero otherwise. Furthermore, the variables Large3, Largest, and LargestDum are interacted, respectively, with the variable Q in order to examine the impact of a firm’s ownership structure on investment efficiency (if any). To accounts for the influence of ownership structure on a firm’s decision to invest, I modify Equation 3 to include a vector of the previously discussed variables: 1

(𝐼𝑖𝑡 /𝐾𝑖𝑡 ) = 𝜆𝑄𝑖𝑡−1 + ∑ 𝛿𝑗 Χ𝑗𝑖𝑡−1 1

+ ∑ 𝛾𝑘 OWN𝑘𝑖𝑡−1 + 𝜈𝑖 + 𝜏𝑡 𝑘

+ 𝑢𝑖𝑡

(4)

Lastly, I make a final adjustment to equation 4. I include a lagged term of I/K in Equation 4 to account for a firm’s dynamic adjustment towards an optimal capital level, that is a firm’s adjustment of investment spending until it reaches an optimal level of capital (Devereux and Schiantarelli, 1990, Bond and Meghir, 1994). Therefore equation 4 is adjusted to include a lagged dependent variable: 1

(𝐼𝑖𝑡 /𝐾𝑖𝑡 ) = 𝛽(𝐼𝑖𝑡−1 /𝐾𝑖𝑡−1 ) + 𝜆𝑄𝑖𝑡−1 + ∑ 𝛿𝑗 Χ𝑗𝑖𝑡−1 𝑗

1

1

+ ∑ 𝛾𝑘 OWN𝑘𝑖𝑡−1 𝜈𝑖 + 𝜏𝑡

(𝐼𝑖𝑡 /𝐾𝑖𝑡 ) = 𝜆𝑄𝑖𝑡−1 + ∑ 𝛿𝑗 Χ𝑗𝑖𝑡−1 + 𝜈𝑖 + 𝜏𝑡

𝑘

𝑗

+ 𝑢𝑖𝑡 (3) where Χ𝑗𝑖𝑡 is a vector of control variables defined as follows. CashFlow is the ratio between EBIT to its capital. Employees is the logarithm of the number of total employees. Firms included in the analysis are the ones with 10 employees or more. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. Also,

𝑗

+ 𝑢𝑖𝑡

(5)

However, when a lagged dependent variable is included as a control variable, the lagged dependent variable becomes correlated with the error term, 𝑢𝑖𝑡 . Therefore, estimating equations 5 in static form using standard panel data techniques leads to biased and inconsistent estimators. In order to overcome this

871


problem, equation 5 is first differenced to eliminate the fixed effects. Then, an instrumental variable approach is used to estimate the dynamic panel in equation 5 (Anderson and Hsiao, 1981, Arellano and Bond, 1991). 4. Data and Summary Statistics This study utilizes firm-level data on companies listed in the ASE during the period 2002-2011. Financial firms are excluded from the sample because their financial data and nature of investment is very different from nonfinancial firms. Furthermore, service companies are excluded because they rely heavily on labour and they have small incremental changes in their fixed assets base. For each year during the study period I collected data on listed companies for that year. Major databases provide data only on a small number of large listed Jordanian companies, therefore, the financial data used in this study is hand-collected from the annual Corporate Guide published by the ASE. The ownership data is collected from the annual Corporate Guides for the period 2002-2007 and from the companies’ financial statements for the period 2008-2011. Jordanian

Securities Commission requires listed companies to disclose equity holdings of 5% and more. Data is disregarded for reasons of consistency in case a company reports shareholdings less than 5%. Table 1 shows summary statistics for the sample used in the study. The number of observations for all variables is 610 representing 84 companies. However, the number of observations on ownership data is 517 representing 78 companies. The loss of ownership data is due to the unavailability of recorded data in the Corporate Guides or the unavailability of a company’s annual report(s). The primary variables of interest in this study are Q and the ownership variables: Largest, Largest3 and Largest5. On average Q is above 1 with a value of 1.28 and its median is also above 1 with a value of 1.123. The mean (median) of equity holdings of the largest shareholder is 28% (22%), while the sum of equity holdings of the largest three and the largest five shareholders are 48% (46%) and 54% (57%) respectively. These figures indicate that the average equity holding stakes of Jordanian investors are large enough to induce an interest in the investment activities of the company.

Table 1. Summary Statistics Table 1 reports descriptive statistics for a sample of industrial Jordanian firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. Concentrated Ownership is the total ownership percentage of shareholders owning 5% and more. Large5 is the total ownership percentage of the largest five shareholders owning 5% and more. Large3 is the total ownership percentage of the largest three shareholders owning 5% and more. Largest is the total ownership percentage of the largest shareholder owning 5% and more. CashFlow is the ratio between EBIT to its capital. Employees is the logarithm of the number of total employees. Firms included in the analysis are the ones with 10 employees or more. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets.

Fixed Assets (Million JDs) (ln) Fixed Assets Change in Fixed Assets to Total Assets Change in Fixed Assets to Capital Market to Book Ratio (Q) Concentrated Ownership Total Ownership of Largest 5 (Large5) Total Ownership of Largest 3 (Large3) Largest Owner (Largest) Cash Flow (CashFlow) Book Assets (Million JDs) (ln) Book Assets Number of Employees (ln) Number of Employees (Employees) Debt Ratio (DebtRatio) Current Ratio (Liquidity)

Mean 14.800 15.401 0.009 0.044 1.281 54.396 53.679 47.891 28.308 0.309 48.100 16.544 380.2443 5.067 0.345 2.834

Median 4.538 15.328 -0.007 -0.024 1.128 57.850 57.140 46.220 22.600 0.135 12.700 16.355 150 5.011 0.315 1.941

Table 2 reports the correlation matrix between the key variables used in the study. As predicted, Q is positively and significantly correlated with the investment measure at the 5% significance level. The

SD 1.431 0.089 0.254 0.581 22.400 22.132 21.188 18.500 1.083 1.353 745.430 1.237 0.218 2.538

Min 0.113 11.632 -0.442 -0.536 0.300 0 0 0 0 -1.342 0.626 13.347 10 2.302 0.009 0.120

Max 252.000 19.345 0.631 1.651 3.904 98.5 98.5 98.5 98.5 10.647 873.000 20.588 4786 8.473 0.945 18.05

Skewness 4.029 0.226 2.364 2.835 1.618 -0.322 -0.279 0.148 1.543 6.979 3.877 0.715 3.733 0.240 0.657 2.342

Kurtosis 21.376 3.300 18.779 14.963 6.109 2.372 2.393 2.419 5.725 58.619 19.821 3.499 17.217 3.976 2.789 9.921

ownership variables (Large3 and Largest) are not significantly correlated with the investment measure. This study does not make predictions on the relation between a firm’s investment spending and its

872


ownership structure, since investment spending does not reflect spending efficiency. The variable CashFlow is positively and significantly related to the investment measure, a result that is also confirmed in the regression analysis. Other notable observation is the positive and significant correlation between

ownership variables (Large3 and Largest) and Q. This indicates that firms with larger ownership concentration have larger investment opportunities.

Table 2. Correlation Matrix Table 2 presents the correlation between the variables used in the study. The sample consists of industrial Jordanian firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the ratio between sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. Large3 is the total ownership percentage of the largest three shareholders owning 5% and more. Largest is the total ownership percentage of the largest shareholder owning 5% and more. LargestDum is an indicator variable that takes the value of one if the firm’s largest shareholder owns 20% and more and zero otherwise. CashFlow is the ratio between EBIT to its capital. Employees is the logarithm of the number of total employees. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. p-values are in parentheses. I/K I/K Large3 Largest LargestDum Q CashFlow Employees DebtRatio Lquidity

Large3

Largest

Largest -Dum

Q

CashFlow

Employees

DebtRatio

Liquidity

1 0.058 (0.189) 0.005 (0.912) 0.048 (0.271) 0.101 (0.012) 0.115 (0.004) 0.064 (0.112) 0.085 (0.036) -0.042 (0.298)

1 0.865 (0.000) 0.761 (0.000) 0.239 (0.000) -0.035 (0.422) 0.090 (0.040) 0.019 (0.658) -0.039 (0.370)

1 0.659 (0.000) 0.172 (0.000) -0.051 (0.246) 0.096 (0.029) -0.008 (0.862) -0.087 (0.046)

1 0.166 (0.000) 0.030 (0.497) 0.015 (0.735) 0.070 (0.113) -0.100 (0.022)

5. Data Analysis and Results The analysis starts by estimating equations 4 and 5 without including ownership variables. The results are reported in Table 3. Panel I reports the estimation results of regressing Q on investment to capital ratio in order to capture the stand alone impact of Q on investment, while Panel II reports the estimation results that include Q and other control variables. Columns 1, 2 and 3 report the estimation results using pooled OLS, firm fixed effects, and Arellano-Bond first differenced dynamic panel respectively. All estimations use robust standard errors and include time fixed effects. The results show that the impact of Q on a firm’s capital expenditures is positive and statistically significant in all estimations. This result

1 0.016 (0.691) 0.127 (0.002) -0.089 (0.026) 0.035 (0.384)

1 0.002 (0.960) -0.160 (0.000) 0.199 (0.000)

1 0.315 (0.000) -0.192 (0.000)

1 -0.579 (0.000)

1

indicates that investment spending of listed Jordanian companies responds positively and significantly to stock market valuation. In terms of the magnitude of the coefficient, and based on the static fixed effects estimator, an increase in Q by one standard deviation is associated with 4.7% increase in investment to capital ratio. Other variables do not have a statistically significant impact on the investment to capital ratio except for cash flow. The sign on the coefficient of cash flow is positive and is statistically significant in all estimations, indicating that industrial firms listed in the ASE can be financially constrained. However, the coefficients on Employees, DebtRatio and Liquidity are not significant.

873

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Table 3. Investment Efficiency – Base Model Table 3 reports estimation results for testing equations 4 and 5. Panel I reports the estimation results of regressing Q on investment to capital ratio, while Panel II reports the estimation results that include Q and other control variables. The sample consists of industrial Jordanian firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the ratio between sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. CashFlow is the ratio between EBIT to its capital. Employees is the logarithm of the number of total employees. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. t-statistics (alternatively z-statistics) are in parentheses. ***, **, * indicate significance at the 1%, 5%, and 10% respectively. Column 1 OLS

Column 2 Fixed Effects

Column 3 Arellano-Bond

Panel I -

I/Kt-1

-

0.049 (1.65) *

Q

-0.015 (-0.23)

0.091 (2.45)**

0.187(1.72)*

Yes

Yes

Yes

610

610

358

Sargan Test (Chi )

-

-

68.784

Arellano-Bond test (z)

-

-

-0.914

Time Effects Observations 2

Panel II -

I/Kt-1

-

-0.067 (-1.02)

Q

0.051 (1.78) *

0.081 (2.36) *

0.186 (1.76)*

CashFlow

0.020 (3.11) ***

0.047 (6.81) ***

0.043 (3.43)***

Employees

0.006 (0.63)

-0.005 (-0.17)

-0.067 (-1.14)

DebtRatio

0.016 (0.83)

0.019 (0.50)

0.053 (0.91)

Liquidity

-0.021 (-0.88)

0.028 (0.86)

0.052 (1.10)

Yes

Yes

Yes

610

610

355

Sargan Test (Chi )

-

-

68.673


-

-

-0.398


In order to account for any possible adjustment effect, the model includes the lagged dependent variable as a predictor. To estimate the dynamic panel, this study applies the Arellano-Bond first difference model. The results are presented in column 3. The sign on the coefficient of the lagged dependent variable has the negative predicted sign but is statistically insignificant. This result indicates that there is no adjustment effect on investment among industrial firms listed on the ASE. In addition, the signs and significance of the base-model variables are stable. For example, the coefficients on Q and cash flow have their expected positive sign and are statistically significant, while the coefficients on Employees, DebtRatio and Liquidity are not significant. Because the results of the Arellano-Bond first difference model suggests that there is no significant adjustment effect,

further estimations will not include the lagged I/K as a predictor. Therefore, further analysis will report the estimation results using the fixed effects model.[6] The main purpose of this study is to explore the effect of concentrated ownership on the propensity of a firm to respond positively to market valuation, Tobin’s q. This study applies two operational definitions of ownership concentration. The first definition focuses on the sum of ownership of the largest three investors owning above 5% of equity capital and the second measures ownership of the largest shareholder (Demsetz and Lehn, 1985,

6

The results of estimations using the lagged I/K as a predictor are qualitatively similar to the ones reported in Tables 4 and 5.

874

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Demsetz and Villalonga, 2001).[7] In order to test the impact of ownership concentration on a firm’s propensity to listen to the market, the estimation includes the interaction between ownership measures with Q. A negative sign on the interaction term supports an expropriation effect. Largest shareholders extract private benefits in the form of sub-optimal investment which weakens their propensity to learn from the market and attenuates the investment-q sensitivity. A positive sign, however, supports an alignment effect. As ownership stake of shareholders increases, private benefits of sub-optimal investment may not exceed benefits of efficient investment and hence the investment-q sensitivity is strengthened as ownership increases. Table 4 reports the results of estimating equation 3 including Large3 and Largest as regressors and their respective interactions with Q, Large3*Q and Largest*Q. The signs and significance of the basemodel variables are stable compared to the estimation results reported in Table 3. The coefficient on CashFlow has its expected positive sign and is statistically significant, while the coefficients on Employees, DebtRatio and Liquidity are not significant. The coefficients on Q and the interaction terms Large3*Q and Largest*Q are positive and significant. The results show that the impact of Q on investment to capital ratio is given by .077 + .0026*Largest3 in case of using the ownership of the largest three shareholders and by .061 + .005*Largest in case of using the ownership of the largest shareholder. Taking the case of the ownership of the largest shareholder, a company with a second quartile largest shareholder ownership has a total Q effect over investment of .143 (.061+.005*16.47), while a company with a median largest shareholder ownership has a total Q effect of .183 (.061+.005*24.47). This result indicates that the propensity of an industrial company listed in the ASE to listen to the market increases with the percentage of ownership of the largest shareholder.

7

The results are robust when the sum of the largest five shareholders owning 5% and more is used. It is also robust to the use of the 10% cut-off point instead of the 5%.

875


Table 4. Ownership Concentration and Investment Efficiency Table 4 reports estimation results for testing the differential impacts of ownership concentration on investment efficiency using two alternative definitions of ownership concentration: Large3 and Largest. The sample consists of industrial firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the ratio between sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. Large3 is the total ownership percentage of the largest three shareholders owning 5% and more. Largest is the total ownership percentage of the largest shareholder owning 5% and more. CashFlow is the ratio between EBIT to its capital. Employees are the logarithm of the number of total employees. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. T-statistics are in parentheses. ***, **, * indicate significance at the 1%, 5%, and 10% respectively. Model 1

Model 2

Q

0.077 (1.98)*

0.061 (1.79)*

Large3

0.002 (1.44)

-

Q*Large3

0.003 (1.74)*

-

Largest

-

0.000 (-0.25)

Q*Largest

-

0.005 (1.76)*

CashFlow

0.044 (6.92)***

0.046 (7.22) ***

Employees

-0.005 (-0.15)

-0.001 (-0.04)

DebtRatio

-0.002 (-0.05)

0.001 (0.03)

Liquidity

0.038 (1.17)

0.038 (1.12)

Time Effects

Yes

Yes

Observations

517

517

The positive sign on the interaction terms Large3*Q and Largest*Q lend support to the alignment effect. Largest shareholder with high ownership stakes are subject to a larger share of the costs of sub-optimal investment and hence are more likely to invest efficiently. To further analyze the impact of large shareholding on investment-q sensitivity, I split the sample into two subsamples based on two cut-offs 10% and 20% ownership. These two cut-offs are well-accepted in the literature as representing control interest in the company (see for

example (La Porta et al., 1999). I report results based on the ownership of the largest shareholder in Table 5. The results show that for firms with a large shareholder owning 10% (alternatively 20%) or less Q do not have a significant impact on ownership, while firms with a large shareholder owning above 10% (alternatively 20%) Q have positive and significant impact on investment. These results indicate that the positive incremental impact of ownership on investment-q sensitivity is present only in firms with large ownership.

876


Table 5. Ownership Concentration and Investment Efficiency by Ownership Percentage Table 5 reports estimation results for estimating equation 4 sub-samples divided based on ownership percentages. The sample consists of industrial Jordanian firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the ratio between sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. CashFlow is the ratio between EBIT to its capital. Employees is the logarithm of the number of total employees. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. T-statistics are in parentheses. ***, **, * indicate significance at the 1%, 5%, and 10% respectively. 10% Ownership Below 10%

20% Ownership

Above 10%

Below 20%

Above 20%

Q

-0.109 (-0.92)

0.124 (2.77)***

-0.044 (-0.72)

0.206 (2.94)***

CashFlow

0.024 (0.39)

0.045 (6.97)***

-0.027 (-0.49)

0.032 (3.06)***

Employees

0.011 (0.74)

-0.024 (-0.51)

0.021 (0.84)

-0.130 (-3.27)***

DebtRatio

0.139 (1.95) *

-0.023 (-0.51)

0.020 (0.25)

0.091 (1.38)

Liquidity

0.114 (1.33)

0.039 (1.24)

0.090 (1.27)

0.048 (1.34)

Time Effects

Yes

Yes

Yes

Yes

Observations

77

440

208

309

To test the significance of the differential impact of ownership level on Q, this study creates an indicator variable, LargestDum, that takes the value of zero if the firm’s largest shareholder owns less than 20% and one otherwise. This variable is interacted with Q to create the interaction term LargestDum*Q. The results are reported in Table 6. Column 1 reports the estimation results of the main variable of interest and shows that the impact of Q in firms with an ownership

stake less than 20% for the large owner is positive but statistically insignificant. However, the impact of Q in firms with an ownership stake more than 20% for the large owner is positive and statistically significant. Columns 2 and 3 reports the estimation results with the set of control variables and using fixed effects and Arellano-Bond model respectively. The results are consistent with the ones reported in Column 1.

877

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Table 6. Ownership Concentration and Investment Efficiency – Extended Model Table 6 reports estimation results for testing the differential impacts of ownership concentration on investment efficiency using two alternative estimation methods: Fixed Effects and Arellano-Bond dynamic panel. The sample consists of industrial firms listed in the ASE over the period 2002-2011. Industries are based on the ASE classification. Q is the ratio between sum of the market value of a firm’s equity and the book value of its liabilities divided by the book value of a firm’s assets. LargestDum is an indicator variable that takes the value of one if the firm’s largest shareholder owns 20% and more and zero otherwise. CashFlow is the ratio between EBIT to its capital. Employees are the logarithm of the number of total employees. DebtRatio is the ratio between total liabilities to total assets. Liquidity is the ratio between current assets to total assets. T-statistics (alternatively z-statistics) are in parentheses. ***, **, * indicate significance at the 1%, 5%, and 10% respectively. Model 1 Fixed Effects I/Kt-1

Model 2 Arellano-Bond

-

-0.094 (-1.48)

Q

0.008 (0.20)

-0.087 (-0.84)

LargestDum

0.044 (1.46)

-0.131 (-1.82)*

Q*LargestDum

0.138 (2.42)**

0.298 (2.47)**

CashFlow

0.044 (6.63)***

0.047 (3.94)***

Employees

-0.009 (-0.31)

-0.031 (-0.65)

DebtRatio

-0.002 (-0.05)

0.083 (1.57)

Liquidity

0.040 (1.23)

0.081 (1.83)*


Yes

Yes

517

335

Sargan Test (Chi )

-

136.131


-

-0.066

Conclusion This article investigates the investment behavior of Jordanian companies listed in the ASE with a focus on the impact of ownership concentration on attenuating/strengthening the investment-q sensitivity. Concentrated ownership along with the possibility of assuming management responsibilities allow large shareholders to exercise control over the firm. Therefore, large shareholders have incentives and abilities to extract private benefits in the form of suboptimal investment which weakens their propensity to listen to the market and hence attenuates the investment-q sensitivity. In contrast, the costs relative to the benefits of maximizing private benefits and wasting cash flows on sub-optimal investment are increasing in percentage of equity holding. Therefore, the investment-q sensitivity is strengthened as ownership increases In this study, I address these competing predictions by using a q theory framework applied on

a sample of listed Jordanian companies for the period from 2002 until 2011. Base model results indicate that listed firms respond to signals from the market regarding their investment opportunities. The propensity to listen to the market increases with ownership concentration. In addition, using subsamples the results show that there is a positive and significant influence of ownership on a firm’s investment-q sensitivity in the ownership range beyond 20% and insignificant influence below that bound. These results imply that, when largest shareholders own a small stake, and in the absence of other effective corporate governance and market discipline mechanisms, there is little incentive for the largest shareholder to follow market signals. However, when ownership stakes increase, largest shareholders are subject to a larger share of the costs of squandering corporate wealth and therefore their propensity to listen to the market increases.

878

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 Refеrences: 1. Abdel-Halim, M. & Bino, A. (2013), "Corporate Governance and Dividend Policy when Investor Protection is Weak", Social Science Research Network (SSRN). 2. Aivazian, V. A., Ge, Y. & Qiu, J. (2005), "The Impact of Leverage on Firm Investment: Canadian Evidence", Journal of Corporate Finance, Vol. 11 No. 1, pp. 277291. 3. Anderson, T. W. & Hsiao, C. (1981), "Estimation of Dynamic Models with Error Components", Journal of the American Statistical Association, Vol. 76 No. 375, pp. 598-606. 4. Andres, C. (2011), "Family Ownership, Financing Constraints and Investment Decisions", Applied Financial Economics, Vol. 21 No. 22, pp. 1641-1659. 5. Arellano, M. & Bond, S. (1991), "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations", The Review of Economic Studies, Vol. 58 No. 2, pp. 277-297. 6. Arestis, P., Demetriades, P. O. & Luintel, K. B. (2001), "Financial Development and Economic Growth: The Role of Stock Markets", Journal of Money, Credit and Banking, Vol. 33 No. 1, pp. 16-41. 7. Ase 2001-2012. Various Issues of the Corporate Guides. Jordan. 8. Atje, R. & Jovanovic, B. (1993), "Stock Markets and Development", European Economic Review, Vol. 37 No. 2, pp. 632-640. 9. Barro, R. J. (1990), "The Stock Market and Investment", Review of Financial Studies, Vol. 3 No. 1, pp. 115-131. 10. Beck, T. & Levine, R. (2004), "Stock Markets, Banks, and Growth: Panel Evidence", Journal of Banking and Finance, Vol. 28 No. 3, pp. 423-442. 11. Bolbol, A. A. & Omran, M. M. (2005), "Investment and the Stock Market: Evidence from Arab Firm-Level Panel Data", Emerging Markets Review, Vol. 6 No. 1, pp. 85-106. 12. Bond, S. & Meghir, C. (1994), "Dynamic Investment Models and the Firm's Financial Policy", The Review of Economic Studies, Vol. 61 No. 2, pp. 197-222. 13. Chen, Q., Goldstein, I. & Jiang, W. (2007), "Price Informativeness and Investment Sensitivity to Stock Price", Review of Financial Studies, Vol. 20 No. 3, pp. 619-650. 14. Cho, M. H. (1998), "Ownership Structure, Investment, and the Corporate Value: An Empirical Analysis", Journal of Financial Economics, Vol. 47 No. 1, pp. 103121. 15. Claessens, S., Djankov, S. & Lang, L. H. (2000), "The Separation of Ownership and Control in East Asian Corporations", Journal of Financial Economics, Vol. 58 No. 1–2, pp. 81-112. 16. Cumming, D., Johan, S. & Li, D. (2011), "Exchange Trading Rules and Stock Market Liquidity", Journal of Financial Economics, Vol. 99 No. 3, pp. 651-671. 17. Demsetz, H. & Lehn, K. (1985), "The Structure of Corporate Ownership: Causes and Consequences", The Journal of Political Economy, Vol. 93 No. 6, pp. 11551177. 18. Demsetz, H. & Villalonga, B. (2001), "Ownership Structure and Corporate Performance", Journal of Corporate Finance, Vol. 7 No. 3, pp. 209-233. 19. Devereux, M. & Schiantarelli, F. (1990). "Investment, Financial Factors, and Cash Flow: Evidence from U.K. Panel Data". In: Hubbard, G. (ed.) Asymmetric

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

879

Information, Corporate Finance, and Investment. University of Chicago Press. Dow, J. & Gorton, G. (1995), "Stock Market Efficiency and Economic Efficiency: Is there a Connection?", National Bureau of Economic Research (NBER). Fazzari, S., Hubbard, R. G. & Petersen, B. C. (1988), "Financing Constraints and Corporate Investment", Brookings Papers on Economic Activity, Vol. 1988 No. 1, pp. 141-206. Goergen, M. & Renneboog, L. (2001), "Investment Policy, Internal Financing and Ownership Concentration in the UK", Journal of Corporate Finance, Vol. 7 No. 3, pp. 257-284. Gugler, K. (2003), "Corporate Governance and Investment", International Journal of the Economics of Business, Vol. 10 No. 3, pp. 261-289. Haid, A. & Weigand, J. (2001), "R&D, Liquidity Constraints, and Corporate Governance", Journal of Economics and Statistics, Vol. 221 No. 2, pp. 145-167. Harrigan, J. R., El-Said, H. & Wang, C. (2006), "The IMF and the World Bank in Jordan: A Case of Over Optimism and Elusive Growth", Review of International Organizations, Vol. 1 No. 3, pp. 263–292. Hayashi, F. (1982), "Tobin's Marginal q and Average q: A Neoclassical Interpretation", Econometrica: Journal of the Econometric Society, Vol. 50 No. 1, pp. 213-224. Henry, P. B. (2000), "Do Stock Market Liberalizations Cause Investment Booms?", Journal of Financial Economics, Vol. 58 No. 1–2, pp. 301-334. Holmstrom, B. & Tirole, J. (1993), "Market Liquidity and Performance Monitoring", Journal of Political Economy, Vol. 101 No. 4, pp. 678-709. Hubbard, R. G. (1997), "Capital-Market Imperfections and Investment", Journal of Economic Literature, Vol. 36 No. 1, pp. 193-225. Jensen, M. C. (1986), "Agency cost of free cash flow, corporate finance, and takeovers", American Economic Review, Vol. 76 No. 2, pp. 323-329. Jensen, M. C. & Meckling, W. H. (1976), "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure", Journal of Financial Economics, Vol. 3 No. 4, pp. 305-360. Jiang, L., Kim, J.-B. & Pang, L. (2011), "ControlOwnership Wedge and Investment Sensitivity to Stock Price", Journal of Banking and Finance, Vol. 35 No. 11, pp. 2856-2867. Kanaan, T. & Kardoosh, M. (2005), "Law-Making for Trade Liberalization and Investment Promotion in Jordan", Center for Development Research: University of Bonn. Kaplan, S. N. & Zingales, L. (1997), "Do InvestmentCash Flow Sensitivities Provide Useful Measures of Financing Constraints?", The Quarterly Journal of Economics, Vol. 112 No. 1, pp. 169-215. La Porta, R., Lopez-De-Silanes, F., Shleifer, A. & Vishny, R. (2000), "Investor Protection and Corporate Governance", Journal of Financial Economics, Vol. 58 No. 1, pp. 3-27. La Porta, R., Lopez‐De‐Silanes, F. & Shleifer, A. (1999), "Corporate Ownership Around the World", The Journal of Finance, Vol. 54 No. 2, pp. 471-517. Levine, R. & Zervos, S. (1996), "Stock Market Development and Long-Run Growth", The World Bank Economic Review, Vol. 10 No. 2, pp. 323-339. Levine, R. & Zervos, S. (1998), "Stock Markets, Banks, and Economic Growth", American Economic Review, Vol. 88 No. 3, pp. 537-558.


39. Luo, Y. (2005), "Do Insiders Learn from Outsiders? Evidence from Mergers and Acquisitions", The Journal of Finance, Vol. 60 No. 4, pp. 1951-1982. 40. Morck, R., Shleifer, A. & Vishny, R. W. (1988), "Management Ownership and Market Valuation: An Empirical Analysis", Journal of Financial Economics, Vol. 20 No. January–March, pp. 293-315. 41. Morck, R., Shleifer, A., Vishny, R. W., Shapiro, M. & Poterba, J. M. (1990), "The Stock Market and Investment: Is the Market a Sideshow?", Brookings Papers on Economic Activity, Vol. 1990 No. 2, pp. 157215. 42. Ndikumana, L. (2005), "Financial Development, Financial Structure, and Domestic Investment: International Evidence", Journal of International Money and Finance, Vol. 24 No. 4, pp. 651-673. 43. Pawlina, G. & Renneboog, L. (2005), "Is Investment‐ Cash Flow Sensitivity Caused by Agency Costs or Asymmetric Information? Evidence from the UK", European Financial Management, Vol. 11 No. 4, pp. 483-513. 44. Samuel, C. (1998), "The Investment Decision: a Reexamination of Competing Theories Using Panel Data", Applied Economics, Vol. 30 No. 1, pp. 95-104.

45. Shanmugam, K. & Bhaduri, S. N. (2002), "Size, Age and Firm Growth in the Indian Manufacturing Sector", Applied Economics Letters, Vol. 9 No. 9, pp. 607-613. 46. Stein, J. C. (2003). "Agency, Information and Corporate Investment". In: Constantinides, G. M., Harris, M. & Stulz, R. M. (eds.) Handbook of the Economics of Finance. Elsevier. 47. Subrahmanyam, A. & Titman, S. (1999), "The GoingPublic Decision and the Development of Financial Markets", The Journal of Finance, Vol. 54 No. 3, pp. 1045-1082. 48. Tobin, J. (1969), "A General Equilibrium Approach to Monetary Theory", Journal of Money, Credit and Banking, Vol. 1 No. 1, pp. 15-29. 49. Wei, K. J. & Zhang, Y. (2008), "Ownership Structure, Cash Flow, and Capital Investment: Evidence from East Asian Economies before the Financial Crisis", Journal of Corporate Finance, Vol. 14 No. 2, pp. 118-132. 50. Xiao, F. (2009), "Does the Stock Market Affect Investment by Chinese Firms? Some New Evidence", International Review of Applied Economics, Vol. 23 No. 2, pp. 197-213. 51. Zeitun, R. (2006), "Firm Performance and Default Risk for Publicly Listed Companies in Emerging Markets: A Case Study of Jordan", University of Western Sydney.

880


GUIDING CRITERIA FOR OPERATIONAL RISK REPORTING IN A CORPORATE ENVIRONMENT J. Young* Abstract Risk reporting is most probably one of the most important components of a risk management process. Operational risk reporting, in many organisations, is not developed to such a degree that it will add value to the organisation and is mostly based on regulatory requirements. This means that risk reports mostly aim to comply with regulations rather than add value in terms of providing useful information to ensure effective decision-making. Within this context, this research aims to develop guidelines for operational risk reporting which will be based on a comprehensive literature review of operational risk to determine criteria which can serve as guidelines for effective risk reporting. The criteria will be subject to an empirical analysis by means of an anonymous questionnaire completed by experienced managers in a corporate environment. The data will be analysed in terms of descriptive statistical analysis in order to confirm the applicability of the criteria in terms of operational risk reporting. The information will be used to compile a prioritised list of criteria which could serve as a guideline to corporate organisations during operational risk reporting. Keywords: Integrated Risk Reporting, Risk Reporting, Risk Control, Risk Communication, Operational Risk Management, Key Risk Indicators, Risk Incidents, Risk Identification, Risk Assessment, Risk Information, Residual Risk, Inherent Risk, Risk Owner *University of South Africa, PO Box 52185, Wierda Park, Centurion, Pretoria, South Africa, 0149 Tel: +27 12 429 3010, +27 8307 6265

1 Introduction The management of operational risk, as an independent risk type and management discipline should be in an advanced phase of implementation in most organisations. There are a number of reasons to support this statement, seeing that the management of operational risk started in earnest in the 1990’s and should, therefore, after twenty-five years be recognised as a reasonably matured risk management discipline in its own right. It seems that most organisations accepted the Basel Committee on Banking Supervision’s (BCBS) definition of operational risk as the risk of losses due to inadequate or failed internal processes, systems, or people, or because of external events. This definition also includes legal risk, but excludes reputational and strategic risks. (BCBS, 2003). Since the Basel Committee on Banking Supervision (2006) promulgated the regulatory framework for the banking industry, providing guidelines to link a minimum capital to risks, most organisations focused on the embedding of a structured approach to risk management. This is also true for operational risk management in the sense that banks, for example, must also allocate a capital charge for this risk type. In this regard, the BCBS (2006) reiterated that a bank should develop a framework for managing operational risk and evaluate the adequacy of capital. According

to Gregoriou (2009), the framework on Capital Measurement and Capital Standards for the banking sector has now gone live in most parts of the world and includes the covering of operational risk. It is therefore, imperative that banks and all other corporate organisations should have an operational risk management framework to ensure that the approach to operational risk management is sound and structured. A risk management framework is described by the Australian/New Zeeland Standard (AS/NZS) (2004) as a set of elements of an organisation’s management system concerned with managing risk. Young (2014) mentions that the aims of an operational risk management framework are to identify and establish a structured approach to the management of operational risk and to serve as a guideline on how to achieve the following goals: the establishment of an integrated risk management environment; development of cultural awareness of risk management; development of roles and responsibilities relating to risk management; and providing a common understanding of operational risk. Girling (2013) states that a strong risk framework provides transparency into risks in the firm, therefore allowing for informed business decision-making. In addition, Girling (2013) mentions that with such a strong operational risk management framework a firm can avoid bad surprises and equip itself with tools and contingency planning to be able to respond swiftly

881


when an event does occur. An important part of a risk management framework is a formalised and embedded risk management process. According to Chapman (2011), to implement a risk framework activity within the overall risk management framework includes the implementation of the risk management process. As such, it can be deduced that an organisation should ensure that it has an embedded risk management framework, which by implication also means an effective risk management process. According to the International Organisation for Standardisation (ISO 31000) (2009), it is recommended that organisations develop, implement and continuously improve a framework whose purpose is to integrate the process for managing risk into the organisation's overall governance; strategy and planning; management; reporting processes; policies; values; and culture. In view of the aftermath of the financial crisis, a concept that is currently under scrutiny is the concept of integrated reporting. According to Makiwane and Padia (2012), integrated reporting is a new concept not only in South Africa but all over the world. In the King Report on Governance for South Africa (King III) (2009), it is defined as an integrated representation of the company’s performance in terms of both its finances and its sustainability. According to Verschoor (2014), integrated reporting focused on the combining of financial reporting with responsibility reporting concerning social issues, governance and the environment. Integrated reporting can be regarded as the integration of the annual financial report with various sustainability reports. Eccles, Krzus and Tapscott (2010) define integrated reporting as the process of environment, social and governance integration into the annual report. According to James (2014), a trend towards combining sustainability and financial reporting is emerging and referred to as integrated reporting. It seems that although most organisations are conforming to the concept of integrated reporting, the concept is still new (and sometimes vague) and to establish an integrated reporting process, it should be clear what must be included in such a process and subsequent report. Risk reporting forms an integral part of sustainability reporting, but it sometimes seems that organisations perform risk reporting without a clear objective in mind. It is imperative that risk reporting (including operational risk reporting) should be managed by reporting criteria in order to ensure that reports are adequate and will add value as part of an integrated reporting process. Therefore, the research question applicable to this research is: are there clear guideline criteria for operational risk reporting as an input to an integrated reporting process? In order to address the research question, the focus of this article is on operational risk reporting which can be regarded as an essential component of a risk management process. Therefore the purpose of this article is to provide guiding criteria for effective operational risk reporting which could add value to a

proactive approach to manage operational risks and to serve as a valuable input for integrated reporting. Various views on risk reporting will be analysed in order to identify guiding criteria for organisations to ensure effective and timely operational risk reports. According to King (2014), “reporting has become far more complex since the days when financials were the only area on which organisations needed to report. This has led to increased pressure for a model that enables reporting across a broad spectrum of functions”. In this sense and in terms of the purpose of this paper, the concept of risk reporting will be emphasised as an integrated part of a risk management process. As such, to identify, the criteria for operational risk reporting, it is necessary to deal with the operational risk management process as the underlying concept for effective risk reporting. 2 Operational risk management process ISO 31000 (2009) infers that the risk management framework assists in managing risks effectively through the application of the risk management process. Therefore the framework should ensure that information about risks is derived from the risk management process and it should be adequately reported and used as a basis for decision-making and accountability at all relevant management levels. Many authors and institutions identified different, but mostly similar, components of an operational risk management process. For example, the AS/NLS standard (2004), indicates that risk management involves the establishing of and applying a logical and systematic method of establishing the context, identifying, analysing, evaluating, treating, monitoring and communicating risks. The ISO 31000 (2009) indicates that a risk management process is a systematic application of management policies, procedures and practices to the activities of communicating, consulting, establishing the context, and identifying, analysing, evaluating, treating, monitoring and reviewing risk. According to Young (2014:46), the operational risk management process can be defined as the systematic application of risk policies, procedures and practices by means of the identification, evaluation, control, financing and monitoring of operational risks. Girling (2013:219) mentions that an operational risk framework is designed to identify, assess, monitor, control and mitigate operational risk. It is clear that there is mostly a common understanding of the components of a risk management process. However, according to Chapman (2008:11), a way of exploring the mechanisms for implementing a risk management process is to break it down into its component parts and examine what each part should contribute to the whole. As such he (2008:11) proposes that the risk management process be broken down into six components, namely analysis, identification, assessment, evaluation, planning and management. It

882


is however apparent that from most of the aforementioned views, it seems that communication is a crucial part of an operational risk management process and without this component such a process will not be able to function. As such, communication

can be regarded as a common component ensuring the success of a risk management process. Based on some of the mentioned components of a typical risk management process, it can be illustrated as set out in figure 1.

Figure 1. Components of an operational risk management process C Risk identification

Risk assessment C

C

Risk mitigation & control

C

Continuous risk monitoring

C

= Communication

It is clear that communication creates a link between all the components of a typical operational risk management process. Firstly, it is important that there is a clear communication process between risk identification and risk assessment in order to ensure that the identified risk exposures can be assessed and evaluated to determine the residual risks that must be mitigated and controlled. Secondly, once the residual risks are determined, it must be mitigated in order to prevent the risk or minimise the effect should the risk event occur. This mitigation process also requires effective communication to ensure that the correct control measures are identified and implemented. An important part of this process can be regarded as the communication to the risk owners who must ensure the implementation of the risk control measures. Finally, effective communication is required during the continuous monitoring process to ensure the effectiveness of each risk component as part of the total risk management process. Therefore, it is essential that the results of each process of the components be communicated because an effective risk management process is dependent on the success of each component’s own internal process. In terms of the abovementioned discussion it can be deduced that an embedded operational risk management process is an essential category for effective risk reporting. Based on the aforementioned, it can also be emphasised that risk communication is an essential component of a risk management process. It is directly linked and can ensure the successful execution of the processes involved in each of the risk management

components, such as risk identification, risk assessment, risk mitigation and control and monitoring. Before exploring the concept of operational risk reporting, it is necessary to deal with the broad concept of communication in more detail in order to identify additional categories which can be used as a platform to identity guiding criteria for effective risk reporting. 2.1 Risk communication Risk communication can be regarded as the process to ensure that the right and timeous information is received by the appropriate individual or group to ensure effective decision-making and implementation of the decisions. In addition, Cleary and Malleret (2006:127) state that risk communication is a process of exchange of information and opinion among individuals, groups and institutions. In order to ensure an effective communication process throughout the organisation, it is imperative to ensure that the right people are involved in terms of generating and receiving information. According to Chapman (2011:245) a business should establish internal communication and reporting mechanisms in order to support and encourage accountability and ownership of risk and opportunity management. There should be an open channel to maintain a dialogue with key stakeholders and others to aid the implementation of risk management. Cleary and Malleret (2006:126) state that one reason why risk must be communicated

883


is that there is a need to ensure that the risks identified, assessed and intended to be managed within an enterprise risk management system are properly communicated to the people in the organisation who need to know about them so that they may act. In order to establish a successful risk communication process, it is essential that all the relevant stakeholders be involved and is aware of the process. According to ISO 31000 (2009:10), the introduction of risk management and ensuring its ongoing effectiveness require strong and sustained commitment by management of the organisation as well as strategic and rigorous planning to achieve commitment at all levels. In this regard, it can be derived that the roles and responsibilities in terms of communication should be clearly defined. The importance of risk communication is further emphasised by Holmquist cited by Davis (2007:280) when he infers that communication is critical to effective risk management. He (2007: 280-281) mentions several aspects of improving communication that are beneficial to risk managers, such as: managers should be able to quantify the related risks and build suitable controls to ensure that critical information is available and accurate; and risk managers should make information available in a form that is useful to the right people. As such, management should communicate the benefits of risk management to all stakeholders (ISO 31000, 2009). Effective risk communication will ensure that the right information reaches the right individuals or group to make timeous business decisions. In addition, the ISO 31000 (2009) indicates that an organisation should establish internal communication and reporting mechanisms to support and encourage accountability and ownership of risk. Based on the aforementioned, it is possible to identify the following crucial criteria regarding risk communication which could form part of guiding criteria for risk reporting:  Timeous and correct risk information is essential.  Accurate risk information must be channelled to the correct individuals, groups or institutions.  Internal and external communication and reporting mechanisms should be established.  Effective risk communication must indicate accountability and ownership.  Risk communication should enhance dialogue between all stakeholders.  Risk communication should establish a commitment of all role-players to effective risk management.  Effective risk communication is beneficial for risk management in terms of: o Risk quantification and appropriate risk control measures o Ensure the availability of critical risk information for decision-making

o Accurate and useful risk information to the right target group In order to add to the abovementioned criteria, each component of the process will be analysed in more detail in terms of risk communication. 2.1.1 Risk identification Risk identification aims to identify the operational risk exposures of an organisation which could potentially have a negative influence on the business objectives. According to Chapman (2011:159), risk identification is a transformation process where experienced personnel generate a series of risks and opportunities, which are recorded in a risk register. This process requires the analysis of business processes in terms of its objectives and potential inherent risks. As such, this process requires information from various avenues to identify the inherent operational risks. The data required for this process is usually qualitative in nature and can be sourced from, for example, loss incidents, process flow analysis and scenarios. The primary responsibility for the execution of the risk identification process lies with the business owners (who are also the risk owners). The outcome of this process is a risk register of the identified operational risks which is, according to Chapman (2011:162), a key communication tool as it is referred to and incrementally developed throughout the overall risk management process. The risk registers containing the identified risks serves as a platform and input for the next process, namely the risk assessment process. 2.1.2 Risk assessment Risk assessments can be regarded as the follow-up process from the risk identification process. Croitoru (2014) states that operational risk assessments aim to detect vulnerable operations carried out according to the probability of occurrences and the potential financial impact on the organisation. According to Chapman (2011:197), risk evaluation (assessment) is to assess both the identified risks and opportunities to the business in terms of their aggregated impact on the organisation. Thus, the assessment process involves the analysis of the identified risks (risk register) to determine the potential likelihood and impact of the risks by means of a rating matrix. It furthermore includes the evaluation of risk control measures in place to deal with the identified risks. After evaluating the control measures the rated residual risks are determined. The outcome of this assessment process is an updated risk register consisting of rated risks in terms of probability and impact. The updated risk register, indicating the high-level residual risks can then be used to define the key risk indicators, which can be escalated to responsible persons to manage. Once again, the primary role-players in this process are the business owners. It is also important that this

884


register be communicated to serve as an input to the next process of mitigation and control. 2.1.3 Risk mitigation and control Various authors view risk control as an important component of a risk management process and it is therefore important to understand this concept. According to Croitoru (2014), risk control is carried out with the aim to transform uncertainties into an advantage for the organisation, limiting the level of threats. Olsen and Wu (2008:73) state that risk

control is the activity of measuring and implementing controls to lessen or avoid the impact of risk elements. This can be reactive, after problems arise, or proactive, expending resources to deal with problems before they occur. Young (2014:47) states that risk control involves the application of techniques to reduce the probability of loss. It aims to eliminate or minimise the potential effect of the identified risk exposures. In addition, Chapman (2011:294) states that the controls need to be meaningful in terms of significant issues or events, and relate to the key business objectives. He (2011:294) also, states that timely controls are necessary so that there is sufficient time to act before negative events turn into terminal events. Based on the aforementioned, it is apparent that the control component of an operational risk management process is crucial to either prevent a loss from occurring or to minimise the effect should such an event occur. It is also clear that to be proactive, it is essential that timeous decisions are made at the right management levels. In order for management to make these decisions, they must be provided with the correct and accurate information, which they can obtain by means of an effective risk communication process. This process should involve appropriate risk reports. Olson and Wu (2008:73) state that risk reporting communicates identified risks to others for discussion and evaluation. According to Blunden and Thirlwell (2013:25), reports on risk should be linked to relevant controls and actions so that recipients can use them to remedy control failures, review risk appetite and perhaps remove controls. Cleary and Malleret (2006:127) state that risk reporting is essential in making decisions. It, furthermore, enables people to participate in deciding how risks should be managed; is a vital part of implementing decisions; and informs and advises people about risks. In addition, it is stated that operational risk reports play a key role in clearly identifying the operational risk strategy and how to achieve it (Blunden and Thirlwell, 2013:152 – 155). Risk control as a component of a risk management process can also be regarded as the final step in the finalisation of the risk register, which will then include the rated control measures and the residual risk exposures which should be managed according to its rating. However, it is important to note that risk management is a dynamic process and

the risk register should be updated according to changing circumstances. Therefore, it is essential that a continuous risk monitoring process should form part of an operational risk management process. 2.1.4 Risk monitoring According to Dowd, cited by Alexander (2003:46), the result of the identification and assessment process is likely to generate a number of indicators through which operational risk may be monitored on an ongoing basis. If operational risk is to become embedded within a risk management culture of the organisation, then monitoring should be conducted on a frequent and regular basis. According to ISO 31000 (2009), both monitoring and review should be a planned part of the risk management process and involve regular checking or surveillance. Chapman (2011:234) states that the primary goal of monitoring is to monitor the performance of risk response actions to inform the need for proactive risk management intervention. The monitoring and review process will be sufficient when it has satisfied the following sub-goals:  Early warning indicators have been developed.  Internal and external context are monitored to establish the current analysis of opportunities and risks.  Risk actioners and managers are implementing the risk and opportunity responses for which they are responsible in a timely manner.  Risk register are regularly updated in terms of actions.  Reports are issued on a regular cycle, providing visibility of the progress made in the success or otherwise of the risk management actions.  Contingencies are revised to reflect the current risks, opportunities and their assessment. In addition and according to ISO 31000 (2009), the organisation's monitoring and review processes should encompass all aspects of the risk management process for the purposes of:  ensuring that controls are effective and efficient in both design and operation;  obtaining further information to improve risk assessment;  analysing and learning lessons from events, changes, trends, successes and failures;  detecting changes in the external and internal context, including changes to risk criteria and the risk itself which can require revision of risk treatments and priorities; and  identifying emerging risks. According to Cleary and Malleret (2006:79), management must ensure that it has effective procedures in place to monitor the events giving rise to the risks it has accepted, so that it has early warning of changes that suggest that the risk is increasing, and that these observations are communicated rapidly to officials who can make proper decisions about how to

885


deal with the changes. It is clear that monitoring plays a crucial role during the operational risk management process. However, it is essential that communication by means of risk reports should be embedded in the process. According to ISO 31000 (2009), the results of monitoring and review should be recorded and externally and internally reported as appropriate, and should also be used as an input to the review of the risk management framework. In addition, Dowd (2003:46) states that it is expected that these reports should cover the results of monitoring activities, such as trend analysis and compliance reviews. In order to provide more clarity on risk reporting, the next section will analyse the concept in more detail. 2.2 Risk reporting

Risk reporting can be seen as one of the more important aspects of risk management in order to effectively communicate various risk information to stakeholders. Haubenstock, cited by Alexander (2003:253) states that reporting should satisfy the requirements of individual business managers as well as offering a consolidated view for senior management. A key objective is to communicate the overall profile of operational risk across all business areas and types of risk. Ong (2007: 627) states that the objectives of management reporting are to inform management about their operational risk experience, trigger actions and resource allocations where necessary, and assure management about the effectiveness of the risk management process. Hain (2009:285) states that sound operational risk management critically depends on the support of employees and their willingness to provide adequate and true information. As such, risk reporting plays a crucial role in risk management and internal and external risk reporting is vital to ensure the provision of adequate and accurate risk information for decision-making and risk management. Olsen and Wu (2008:73) state that risk

reporting communicates identified risks to others for discussion and evaluation. According to Chapman (2011:342), risk reporting is a sub-goal of communication and reports must be prepared on a regular basis advising of changes to the risk exposure and the degree of success being realised by risk response activities. Dowd (2003:46) states that an organisation must implement a system of internal reporting of operational risk with the reporting mechanism geared to the needs of the end user. This is essential if the organisation’s operational risk policy is to be established and evaluated. According to Dowd (2003:46), the board of directors should receive enough information to understand the organisation’s overall operational risk profile and its material risks. Once senior management

receives risk reports they will be able to become involved in operational risk management and make appropriate risk decisions. Blunden and Thirlwell (2103:33) state that good operational risk reporting will also generate management involvement and consensus, which will drive the ongoing identification, assessment and control of operational risk. It is clear that risk information is not only an upward reporting process, but also requires a top-down communication. In this regard Dowd (2003:46) state that reporting should not be viewed as a one-way street, with information only being passed upwards, equally important is downward dissemination or feedback. In addition, Croitoru (2014:29) states that the organisation must ensure that adequate information flow both vertically and horizontally. However, it is crucial that the risk information flowing from topdown and bottom-up should be adequate and sensible in order to lead to decisions or actions. In this regard Blunden and Thirlwell (2013:23) infer that risk reports and the information in them should lead to action. The key to good reporting is to tailor it to the needs of the reader at every management level. In addition and according to COSO (2004:33), reliable reporting provides management with accurate and complete information appropriate for its intended purpose and should support management’s decision-making and monitoring of the organisation’s activities and performance. It is clear that the flow of operational risk information is crucial for effective risk reporting. This information should stem from the operational risk management process and can be generated from the applicable methodologies. According to Girling (2013: 234), reporting will usually include analysis

of internal loss data, external loss data, risk and control self-assessment results, scenario analysis results and capital. In order to quantify and qualify the operational risk exposures, the following popular methods (also mentioned in the New Basel Accord (Basel II 2003)) can be used (Young 2014a):  Loss history. This methodology involves the use of loss data (external and internal) to identify the risks based on events that happened in the past which can be used to avoid or manage similar risk incidents.

Haubenstock (2003:256) states that events are the operational losses (internal and external) that provide the historical base for risk analysis and quantification. The primary report is a summary of statistics from the losses indicating trends of total losses and mean average losses. Reporting often includes any relevant external losses, industry trends or news related to regulation, competitors or other risk factors that might be of interest. Reports on loss data can be used as an input to determine the inherent risks of an organisation when compiling the risk register. Information can also be reported by means of an incident report, reflecting the detail of a loss incident such as the detail on what occurred, those involved and the actual loss.

886


 Risk and control self-assessments (RCSA). According to Young (2014), this method is a bottomup approach to evaluate operational risk. Selfassessments are performed by the business areas and results are aggregated to provide a qualitative profile of risk across the organisation and related action items. The results are communicated with a combination of risk maps, graphic results, issues and initiatives (Haubenstock, 2003:253).The selfassessment process involves the identifying and rating of the inherent risks and existing control measures in order to determine the residual risks that are critical to be managed. This method focuses on potential future risk exposures that should be managed and the results of the RCSA process can be reported on and incorporated into the risk register. It can furthermore form the basis for determining the key risk indicators.  Key Risk Indicators (KRIs). The identification of KRIs can result from the RCSA process and should be managed on a regular basis in order to focus on the current risk exposures and to serve as an early warning of a potential risk incident to management. According to Haubenstock (2003:256),

key risk indicators may also be reported, including related escalation criteria, explanations of any excesses and identified trends. Many KRIs are customised at a business unit level, but some may be common and reported in a consolidated fashion. Davis (2007: 7) cited Grandfield who stated that KRIs are not the only component of risk reporting; there are a large number of data elements that need to be combined to make some meaningful picture of the overall risk landscape for a business, such as: current issues and status of risks; audit and regulatory examinations; and key initiatives. However, seeing that KRIs focuses on the current risk exposures, it is crucial that regular risk reports be generated to the appropriate management levels to make decisions should a pre-set threshold be breached or a trend of an increase in risk is determined. Haubenstock (2003:253) states that reporting communicates the overall level of risk and highlights key trends or exceptions that may require particular attention. Typical reports in this regard could include various forms of graphs.  Scenarios. The use of scenarios involves the expert opinions, concerns and experience of key roleplayers in the organisation to identify potential threats and risk exposures for the organisation (Young 2014). Reports on the scenarios on future potential risks can serve as an input for the risk register. It is apparent that the operational risk methodologies play an important part in internal risk reporting. According to Girling (2013:219 – 220), there are many ways to ensure that the reporting

drives action and to protect against the danger of producing worthless reports and the abovementioned methodologies can be used in this

regard. However, it is crucial that the reports reach the right management level (manager) to ensure timely and adequate decisions; appropriate actions; and to ensure an updated operational risk profile. In order to ensure the operational risk profile, it is important that there is an integrated approach to the reports from each risk methodology. This is illustrated in Figure 2. Cleary and Malleret (2006:204) state that an integrated approach between risk assessment, risk management and risk communication is essential. For a variety of reasons, many large organisations and the people who succeed in them, are often much better at analysis, measurement and the formal processes of management than they are at communication. As such it is imperative to explicitly define the processes which will ensure an integrated operational risk reporting approach. The diagram (Figure 2), illustrates that during the operational risk management process, the loss history (internal and external loss data), Risk and Control Self-Assessments (RCSA), and Scenarios can be used to determine the past and future risk exposures, resulting in a risk register and incident reports. The KRIs can be determined from the risk register and can be managed to determine the status of the current risks and serve as early warning of potential loss incidents. By means of an integrated reporting process, the risk information can be used to determine the operational risk profile. However, according to ISO 31000 (2009), relevant information must be derived from the application of risk management and should be available at appropriate levels and times; and there should be processes in place for consultation with internal stakeholders. It is therefore necessary to identify responsible individuals who must either compile the risk reports or to take the necessary actions/decisions. The risk profile is an essential result of the risk management process and could serve as an input for various activities such as the business planning process, annual business reports and a general view of the organisation’s operational risks. According to ISO 310000 (2009), responsibilities for risk management should be clearly defined. Blunden and Thirlwell (2013:152 – 155) state that any risk report should enable management to take ownership of the information. They (2013:23), also add that risk ownership and control ownership can be clarified through good reporting and assist in identifying priorities for enhancing controls and the organisation’s operational risk profile. In terms of the ISO 31000 (2009), an organisation should establish internal communication and reporting mechanisms which will support and encourage accountability and ownership of risk. These mechanisms should ensure that: there is adequate internal reporting on the framework, its effectiveness and the outcomes; relevant information derived from the application of risk management is available at appropriate levels and times; and that there are processes for consultation with internal stakeholders.

887


According to Bolton and Berkey, cited by Davis (2005:238), there should be regular reporting of pertinent information to senior management and the board of directors. In terms of regular reporting, Blunden and Thirlwell (2013:152 – 155) state that operational risk reporting is a continuous evolving process due to the dynamic nature of good risk reporting. Haubenstock (2003:253) states that

reporting is necessary for all levels of the organisation, but the exact content and frequency of the information must be tailored to each business area. In this regard, Dowd (2003:46) states that in

level information. However it is important that the

risk information makes sense and is applicable for decision-making. According to ISO 31000 (2009), decision makers at all levels of the organisation, should ensure that risk management remains relevant and up-to-date. Alexander (2003:23) cited Swenson (2003:23) cited by Alexander, mentioned that there must be regular reporting of relevant operational risk data to business unit management, senior management and the board of directors and that the board and senior management must be actively involved in the oversight of the operational risk management process.

general the board of directors should receive higher-

Scenarios Future

Key Risk Indicators

Current

Current operational risk exposures and threats to the business

Risk Actions/Risk Information

Future

Potential operational risks in terms of the strategic business objectives expressed in terms of likelihood and impact

Integrated operational risk report

Risk and control self-assessments

Past

History on financial losses due to operational incidents

Operational risk profile

KRI reports

Loss history (Internal and External Loss Data)

Risk register & Loss incident reports

Figure 2. Integrated reporting from operational risk methodologies

Operational Risk Management Process On the other hand, risk reporting mechanisms should also cater for external stakeholders and should incorporate formal risk disclosure processes. It is important to establish an external reporting process or disclosure to ensure that relevant risk information regarding an organisation’s risk profile reaches all stakeholders. Hain (2009:291) states that gathering risk information and communicating it inside the institution supports effective risk management, allows for the consideration of risk in business decisions and is the basis for reporting the firm’s operational risk to stakeholders. Cleary and Malleret (2006:204) state in this regard that it must be ensured that relevant information are communicated in appropriate ways both to the people who are responsible for dealing with the threat and to those outside the firm who may

be affected by it. However, it is important that risk information which is reported to external stakeholders is considered in terms of the sensitivity of the information, in order not to compromise the competitiveness or the reputation of the organisation. According to ISO 310000 (2009), the organisation should develop and implement a plan as to how it will communicate with external stakeholders. This should involve:  engaging appropriate external stakeholders and ensuring an effective exchange of information;  external reporting to comply with legal, regulatory, and governance requirements;  providing feedback and reporting on communication and consultation;

888

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10  using communication to build confidence in the organisation; and  communicating with stakeholders in the event of a crisis or contingency. Disclosure of operational risk to external parties should be carefully monitored, because different stakeholders require different risk information. Hain (2009:288) states that the motivation of external parties regarding monitoring corporate decisions differs among stakeholders. For example regulatory authorities focus on social welfare, the capital market requires information for investment decisions, insurance companies try to calculate fair premiums and rating agencies as well as public accounting firms assess firms as part of their business. AIRMIC (2010:16) states that external risk reporting should be designed to provide external stakeholders with assurance that risks are being adequately managed. In addition Hain (2009:288) states that higher transparency leads to improved risk management of the firm. According to Phillips (2010:36), risks are best managed when information is transparent – that is, timeously and widely available to those who need it. According to ISO 31000 (2009), risk management should be: transparent; appropriate; and ensure the timely involvement of all stakeholders. Involvement also allows stakeholders to be properly represented and to have their views taken into account in determining risk criteria. Bolton and Berkey (2005:238) state that banks, for example, should make sufficient public disclosure to allow market participants to assess their approach to operational risk management. Furthermore, AIRMIC (2010:16) concludes that risk disclosure is a more forwardlooking activity that could anticipate emerging risks. From the aforementioned it is apparent that internal and external risk reporting should be an intrinsic part of an operational risk management process. In order to ensure a streamlined operational risk reporting process and based on the aforementioned literature review, it is possible to identify guiding criteria, which are stipulated in the next section. 3 Guiding criteria for operational risk reporting The guiding criteria for operational risk reporting aim to assist organisations in managing operational risk and to ensure that it adds value. A non-exhaustive list of guiding criteria for risk reporting can be sorted into the following main categories:  Risk management process to generate appropriate risk reports (Risk identification, risk assessment, risk control).  Governance.  Internal risk communication.

 External risk communication and disclosure.  General characteristics of sound operational risk reports. Derived from the literature, the criteria for operational risk reporting are included in Table 1 grouped per category. In order to substantiate the applicability of the guiding criteria identified in this article, a survey was undertaken to confirm the criteria for operational risk reporting and to determine the current status of risk reporting assessed against these criteria. 4 Research methodology In order to confirm the appropriateness of the identified guiding criteria for operational risk reporting, it was decided to identify a group of respondents from the Guideline Biztech database who are involved in risk management projects across various industries and sectors who mainly operates at middle and top management levels. The Guideline Biztech database holds information on a variety of risk-related projects as well as those involved in these projects. As such, it can be reasonable accepted that these individual role-players have a good understanding ad knowledge of risk management. The data was collated by means of a closed questionnaire which was distributed electronically as well as physically to pre-identified role-players involved in operational risk management. The target population was identified at the following management levels: member of the board of directors, executive management, business management, risk management, compliance management, internal audit and financial management. The main reason for distributing the questionnaire to the aforementioned was that these positions can be regarded as the main role-players in an organisation’s risk management processes. The aim of the questionnaire was, firstly, to determine the appropriateness of the guiding criteria for operational risk reporting and to determine the current status of each criterion to ensure a streamlined risk reporting process. The questionnaire requested respondents to indicate on a 5-point Likert scale their views and experiences regarding specific questions on the identified criteria for operational risk reporting and to indicate its current use. The response was analysed in terms of descriptive statistics according to the following scale: 1. To no degree 2. To some degree 3. To a moderate degree 4. To a degree 5. To a full degree

889


Table 1. Criteria for operational risk reporting Category Risk management process

Governance

Internal risk communication External risk communication

Characteristics of sound Risk reports

Criteria  Qualitative risk data is required for risk identification sourced from loss incidents, risk and control self-assessments, key risk indicators and scenarios.  Risk reporting should include the overall operational risk profile of the organisation, based on the results of the operational risk methodologies.  Continuous monitoring and risk reports are essential for proactive risk management.  Risk reports during the risk monitoring process should report on the effectiveness of risk controls.  Risk reports should include information on internal and external operational risk losses.  Risk reports should indicate potential risks derived from risk and control selfassessments.  Risk register forms the basis for risk assessments.  Risk reports should include risk trends to serve as early warning as part of a key risk indicator management process.  Risk reports should provide assurance to management about the effectiveness of the operational risk management process.  Risk reports should indicate potential operational risks derived from scenarios.  Risk reports should result from an efficient internal risk communication process.  Risk reporting process should be included in the organisation’s risk management policy.  Business owners should be responsible for operational risk management and reporting process.  Risk reporting is essential for decision-making.  Risk reporting mechanisms should indicate accountability and ownership of risks.  Risk reporting should include a bottom-up dissemination of operational risk information.  Risk reporting should include a top-down communication of feedback and decisions.  There should be a system of internal risk reporting.  Risk reporting should ensure high-level risk information to the board of directors.  Risk reporting should cater for external disclosures on operational risks to stakeholders.  External risk reporting should include relevant information to support stakeholders in business decisions regarding the organisation.  External risk reporting should comply with legal, regulatory and governance requirements.  External risk reporting should be customised according to the needs of different shareholders.  Risk reports to external stakeholders must not compromise the competitiveness and reputation of the organisation.  Effective proactive risk management decisions should result from reliable, accurate and appropriate risk reports.  Risk reports should be issued on a regular cycle in order to monitor risk management actions.  Risk reports should be internally and externally available.  Risk reporting should be informative on operational risks.  Risk reporting should be based on adequate and true information.  Risk reporting is a continuous process.  Risk reporting should be flexible and allow for customisation to suit the needs of the receiver of the risk information.  Risk reports should include relevant controls and actions.  Risk reports should include resource allocations.  Risk reports should include information that will ensure revision of risk treatments.

890

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 5 Research results The questionnaires were randomly distributed to various role-players listed on the database who were indicated as middle and senior management involved in risk management across a variety of industries and business sectors in South Africa. A total of 85 questionnaires were distributed and 45 were returned on the due date which represents a 52.9% response which is acceptable for analysis purposes using a descriptive statistical approach. Those members who responded reside from a variety of business sectors in South Africa and are indicated in table 2 below: Table 2. Business sectors of respondents Business sector Banking Financial Services Government Departments Insurance Other Total

Response percent 11.1% 6.7% 8.9% 6.7% 66.6% 100%

Response count 5 3 4 3 30 45

Although most of the respondents reside from other sectors than those specifically listed, it can be deduced that operational risk management are being managed in a variety of business sectors, such as municipalities, education, mining, agriculture and consulting firms. Eleven per cent of the respondents are from the banking sector which can be regarded as one of the leading business regarding the management of operational risk in South Africa, mainly due to the implementation of the Basel guidelines, which were adopted by the South African Reserve Bank. Figure 3 indicates the positions of the respondents, while Figure 4 indicates the years of experience. Sixty per cent of the respondents fall in the top management and business management categories, indicating that most respondents should be familiar with risk reporting and should know the role and responsibilities of top management. According to the years of experience, 55.5% of the respondents have more than 10 years’ experience, while 31% have between 5 to 10 years’ experience, indicating a vast level of experience in the relevant organisations and exposure to risk management and reporting.

Figure 3. Positions of respondents Member of the board 8

10

Executive management Business management

3 1

12 4 7

Risk management Compliance management Financial management Other

Figure 4. Years of experience

According to the feedback 74% of the respondents indicated that operational risk is being managed as an independent risk type in their organisation, while 26% indicated that it is still managed to a moderate degree. It can therefore be derived that operational risk is being managed by most organisations as an independent risk type which

confirms the importance of managing it according to a structured approach and process. The inclusion of an operational risk reporting process in an operational risk management policy, however, seems to still be at a developmental level. Thirty-five per cent of the respondents indicated that the risk reporting process is included into the risk policy to a moderate degree,

891


while 17% indicated that it is incorporated to a degree. Forty-eight per cent responded to the adequacy of the risk reporting process being incorporated into an operational risk management policy. It can therefore be assumed that although most organisations manage their operational risk as an independent risk type, the actual reporting process still requires attention in order to ensure an adequate reporting process. According to the respondents the basic operational risk management tools are being used to

manage operational risk. Figure 5, indicates the response in terms of the agreement that the respective tools are being used at an acceptable level. The response indicates that the use of KRIs seems to be the most popular (27.1%) followed by loss history (25.9%) and risk and control self-assessments (25.9%), with scenarios at 21.1%.

Figure 5. Use of operational risk management tools

In order of priority, the respondents indicated that the use of KRIs is the most significant, followed by risk and control self-assessments, loss history and scenarios. The most important deduction regarding the response is that all respondents indicated the use of these tools to manage operational risk. In the literature review, it was also determined that these tools are pertinent to provide information for an effective operational risk reporting process. According to the response 79% agreed that a KRI management process provides risk trends which could serve as early warning during operational risk reporting. Therefore most of the respondents indicated that KRIs are used during the risk reporting process. The literature indicated that the use of KRIs to identify risk trends and to serve as early warning during the management of operational risk is an important part of risk management. As such, it can be reasoned that although some organisations indicated that KRIs are the most popular risk management tool, it still requires some development in terms of its actual benefits such as trend analysis and early warning. Sixty-three per cent of the respondents indicated that the use of scenarios to indicate potential risks is to no degree or to a moderate degree being used for risk management and reporting. Only 5% indicated that scenarios are adequately used as an operational risk management tool. Therefore, it can be deduced that most organisations are aware of the use of scenarios as an operational risk management tool; however, it can still be exploited in terms of its benefit to proactively identify operational risk exposures for an organisation. On the other hand, 58% of the respondents indicated that risk and control-self-assessments are used to report on potential operational risks. Twentyone per cent of the respondents indicated that it is used

to a moderate degree. As such, it can be assumed that risk and control self-assessments play a crucial role in operational risk management and reporting. According to the response, 79% indicated that a risk register is compiled from an operational risk identification and assessment process. It can thus be readily accepted that risk registers are being compiled as a result of an operational risk management process which is in itself an important risk communication tool as indicated in the literature review. Seventy-nine per cent of the respondents indicated that operational risk reports are used to report on internal and external risk losses. According to the literature, risk reports should include information on losses suffered as it serves as the basis to determine the inherent risk exposures which should be managed as part of the risk management process. In this regard, it can be concluded that operational risk reports still requires attention to ensure the adequate reporting on internal and external losses. A reason for this lack of adequacy in reporting might be that organisations are not reporting all losses due to a potential negative influence on their reputation. However, this situation could hamper the effectiveness of operational risk management and negatively influence sound decision-making when top management relies on accurate risk reports to make these decisions. It is therefore imperative that all risk losses be reported accurately and timeously to serve as an input during the risk management and decisionmaking process. Respondents indicated a 42% agreement that risk reports include a report on the effectiveness of risk control measures, 26% to a full degree and 32% to no or some degree. It can therefore be deduced that risk reports do not adequately report on information

892


relating to risk control measures. In terms of the literature review, it is important that risk mitigation be communicated to risk owners who must ensure the implementation of the risk control measures. Therefore it seems that risk reports still require attention in order to include risk control measures to ensure effective risk management. Fifty-two per cent of the respondents indicated that risk reports provide assurance to management on the effectiveness of the operational risk management process. It can thus be deduced that according to 48% of the response, risk reports can be improved to provide the assurance to management that the operational risk management process actually contributes to the effectiveness of operational risk management. On the other hand, 78.9% of the respondents indicated that risk reports provide an operational risk profile of the organisation. This could indicate that risk reports mostly concentrate on the overall results of the operational risk methodologies instead of detail information. In order for management to make decisions, it is necessary to include detail information instead of only an overall risk profile in order to support management decisions in terms of risk management. Although, it is essential for risk reports to provide the overall operational risk profile of the organisation, it is required that risk reports should include various levels of detail for different management levels. According to the literature, business managers are the risk owners and should be responsible for the risk and reporting process. Eightyeight percent of the respondents indicated that business owners are responsible for risk management and the reporting process to a moderate and full degree. Therefore, it can be accepted that the risk reporting process is an important responsibility of business owners. On the other hand the 83.4% of the respondents also indicated that risk managers are to a degree responsible for the risk management and reporting process. It is clear that there is a dual responsibility regarding the risk management and reporting process between business managers and risk managers, although the emphasis should differ. Business managers should be ultimately responsible for risk management, while risk managers play a supporting role to ensure the effectivity of the risk management and reporting process. Although internal audit plays an important role in providing assurance that the risks are being managed, they play a limited role in the actual reporting of risks. This is supported by 61.1% of the response that indicated that internal audit is to a lesser degree involved in risk reporting. However, it seems that some organisations (38.9%) do involve internal audit in the risk reporting process. Although this approach is not the ideal, it seems that some smaller organisations depend on the expertise of internal auditors to assist in the risk reporting processes. According to the response, 77.8% agreed that risk reports result from an efficient risk

communication process. Therefore, it can be confirmed that an efficient risk communication process should be embedded in an organisation to ensure adequate risk reports. Eighty-three per cent of the response indicated that a system of internal risk reporting is embedded in the organisation, emphasising the importance of a risk reporting system. In addition, 88.9% of the respondents indicated that risk reporting is essential for decision-making. Similarly, 88.8% of the respondents agreed that risk reporting is essential for proactive risk management. Regarding the bottom-up dissemination of operational risk information, 66.7% of the respondents indicated that the process is inadequate or only effective at a moderate degree. Thirty-three per cent of the respondents indicated that the process is adequate. It can therefore be deduced that although the bottomup reporting process to disseminate operational risk information is in place, it still requires attention to ensure the development of an adequate reporting process. On the other hand, 50% of the respondents agreed that a top-down risk communication process includes feedback and decisions by top management, 33.3% indicated that it is at a moderate degree. As such, it seems that the top-down communication of risk management feedback and decisions is more embedded than the bottom-up risk reporting of information. However, from the response 88.8% of the respondents agreed to a degree that risk reports contain high-level risk information to the board of directors. This indicates that although the risk information from a bottom-up approach still requires attention, the reports to the board of directors are adequate. Therefore, it can be deduced that risk reporting still requires attention in terms of detailed operational risk information. Seventy-seven per cent of the respondents agreed that the operational risk reports provide information concerning regulatory and compliance information. In addition, 94.5% of the respondents agreed that operational risk reports comply with legal, regulatory and governance requirements. Similarly, 88.9% of the respondents agreed that operational risk reports cater for disclosures on risk management to stakeholders. In this light, it can be deduced that operational risk reports are mostly driven by regulatory and compliance requirements as well as general risk information for disclosure purposes and could still be expanded to include more management information to enhance internal business decision-making. Regarding the inclusion of operational risk information in risk reports to support stakeholders to make business decisions, 83.3% of the respondents agreed to its importance. This response emphasises the importance of including relevant operational risk information disclosed to stakeholders that will assist effective business decisions. However, it is imperative that external operational risk reports should not compromise the organisation’s competitiveness and reputation as indicated by 87.6% of the respondents.

893


The rating of the criteria for operational risk reports is indicated in Figure 6. According to the response all the criteria were rated as applicable for effective operational risk reporting. The criterion that was rated the highest was the inclusion of information that will ensure revision of risk treatments, followed by the criterion to ensure that operational risk reports must be based on adequate and true information. This is followed by the criterion that there must be ensured that risk reports must include relevant controls and

actions. Risk reporting as a continuous process is the next important criterion, followed by the inclusion of resource allocations. It is clear that the first five criteria for effective risk reporting, relates to adequate information, which reflects controls and actions on a continuous basis. As such it can be deduced that according to the response, there is a need for operational risk reports at a lower level which could add value to the actual management of operational risks.

Figure 6. Rating of criteria for operational risk reports

Table 3. Checklist to evaluate operational risk reports # 1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18

Guiding criteria Operational risk reporting should be incorporated into the organisation’s operational risk policy. Operational risk reporting mechanisms should indicate accountability and ownership of risks. Operational risk reporting should assure management about the effectiveness of the operational risk management process. Operational risk reports should be based on adequate and true information. Operational risk reporting should be informative on operational risks. Operational risk reports should provide adequate and accurate risk information for decision-making. Operational risk reporting should trigger actions and resource allocations. Operational risk reporting should communicate the risk profile of operational risk to all business areas. Operational risk reports should include potential risks which were derived from the risk methodologies (Risk and control self-assessments; loss history, key risk indicators and scenarios) and illustrate the risk profile of the organisation. Operational risk reporting should be a continuous process to ensure regular risk reports. Operational risk reporting should include a bottom-up dissemination of operational risk information. Operational risk reporting should include a top-down communication of feedback and decisions. Operational risk reports to external stakeholders must not compromise the competitiveness and reputation of the organisation. Operational risk reporting should be flexible and allow for customisation to suit the needs of the receiver of the risk information. Operational risk reporting should ensure high-level risk information to the board of directors. External operational risk reporting should comply with legal, regulatory and governance requirements. External operational risk reporting should include relevant information to support stakeholders in business decisions. Operational risk reporting should ensure the revision of risk treatment.

In conclusion to the empirical analysis of the response, the guiding criteria for effective operational risk reporting, identified by the literature review, became evident.

6 Conclusion This study provided some insights on risk reporting as an essential part of an operational risk management

894


process. During the literature review it became evident that operational risk reports should add value and form part of an integrated reporting approach. It also became clear that operational risk reports stem from the operational risk management methodologies implemented during a risk management process to identify, assess, mitigate and control and monitor operational risks. The use of these methodologies namely: risk and control self-assessments; key risk indicators; loss history; and scenarios proved to be vital for the effective communication of risk information. The primary conclusions drawn from the empirical analysis can be summarised into a nonexhaustive checklist that could serve as a guideline to evaluate the effectiveness of operational risk reports for corporate organisations (Refer to Table 3). The abovementioned guiding criteria could add value to address current uncertainties on operational risk reporting and therefore also addresses the research question of this article namely: are there clear guideline criteria for operational risk reporting as an input to an integrated reporting process? To address this research question, the purpose of the article was to provide guiding criteria for effective operational risk reporting, based on a literature review, to add value to a proactive approach to operational risk reporting. The criteria can also be used to ensure that operational risk reports are effective, achieve its objective and reach the right target audience. Effective operational risk reports, based on the guiding criteria, can all add value by serving as an input for integrated reporting, a concept currently being widely researched. References 1.

2.

3.

4.

5.

6.

7.

Association of Insurance and Risk Managers (AIRMIC). 2010. A Structured Approach to Enterprise Risk Management. www.airmic.com (1-18). Australian/ New Zealand Standard: Risk Management. Joint Technical Committee OB-007. 3rd Edition, 21 August 2004. Basel Committee on Banking Supervision. 2003. Sound Practices for the Management and Supervision of Operational Risk. Bank for International Settlements. Basel Committee on Banking Supervision. 2006. International Convergence of Capital Measurement and Capital Standards: A Revised Framework. Bank for International Settlements. Blunden, T & Thirlwell, J 2010. Mastering Operational Risk: A practical guide to understanding operational risk and how to manage it. 1st edition. Edinburgh: Pearson. Bolton, N & Berkey, J. 2005. Aligning Basel II Operational Risk and Sarbanes-Oxley 404 Projects. Operational Risks: Practical Approaches to Implementation. Edited by Davis E. Published by Risk Books, a Division of Incisive Financial Publishing Ltd. Haymarket House. Chapman, RJ. 2008. Simple tools and techniques for enterprise risk management. John Wiley & Sons Ltd. West Sussex, England.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

895

Chapman, RJ. 2011. Simple tools and techniques for enterprise risk management. 2nd Edition. John Wiley & Sons Ltd. West Sussex, England. Cleary, S & Malleret, T. 2006. Resilience to Risk. Business success in turbulent times. Human & Rousseau, A Division of NB Publishers (Pty) Ltd. Pretoria. Committee of Sponsoring Organizations (COSO) of the Treadway Commission. 2004. Enterprise Risk Management – Integrated Framework. Dowd, V. 2003. Measurement of Operational Risk: the Basel approach. Operational Risk. Regulation, Analysis and Management. Edited by Alexander, C. Pearson Education Ltd. Harlow. Eccles R., Krzus, MP, & Tapscott, D. 2010. One Report: Integrated Reporting for a Sustainability Strategy. Wiley Publishers. 1st Edition. Girling, P. 2013. Operational Risk Management. A complete guide to a successful operational risk framework. John Wiley & Sons, Inc. New Jersey. Grandfield, A. 2005. Operational Risk Management – The View from the Trenches. Operational Risk. Practical Approaches to Implementation. Edited by Davis, E. Published by Risk Books, a Division of Incisive Financial Publishing Ltd. Haymarket. Gregoriou, G.N. 2009. Operational Risk Toward Basel III. Best practices and issues in modelling, management and regulation. John Wiley & Sons, Inc, New Jersey. Hain, S. 2009. Managing Operational Risk: Incentives for reporting and disclosure. Journal of Risk Management in Financial Institutions Vol 2, 3 284300. Henry Stewart Publications. Haubenstock, M. 2003. The operational risk management framework. Operational Risk. Regulation, Analysis and Management. Edited by Alexander, C. Pearson Education Ltd. Harlow. Holmquist, E. 2007. Operational Risk. Driving Value Creation in a Post-Basel II Era. Operational Risk and Communication Conduits. Edited by Davis, E. Published by Risk Books, a Division of Incisive Financial Publishing Ltd. Haymarket. King III. 2009. King Report on Governance for South Africa. Institute of Directors in Southern Africa (IOD), Pretoria. ISO 31000. 2009. Risk Management – Principles and Guidelines. International Organization for Standardization, Geneva. pp. 8 – 21. James, ML. 2014. The Benefits of Sustainability and Integrated Reporting: An investigation of accounting majors’ perceptions. Journal of Legal, Ethical and Regulatory Issues. Volume 17, Number 2, 2014 (92 – 113). King, M. 2014. Transforming Corporate Reporting. Internal Auditor. April 2014 (58 – 62). Makiwana, TS & Padia, N. 2012. Evaluation of Corporate Integrated Reporting in South Africa Post King III Release South Africa – an exploratory enquiry. Journal of Economic and Financial Sciences | JEF | July 2013 6(2), pp. 421-438. Ong, M. 2007. The Basel Handbook. A guide for financial practitioners. Published in Association with KPMG by Risk Books, a division of Incisive Financial Publishing Ltd. Haymarket. Olson, DL & Wu, DD. 2008. Enterprise Risk Management. World Scientific Publishing Co. Pty. Ltd. Singapore.


26.

27.

Phillips, K. 2010. Transparency creates accountability. Corporate Governance. Enterprise Risk October 2010 (36). Swenson, K. 2003. A qualitative operational risk framework: guidance, structure and reporting. Operational Risk. Regulation, Analysis and Management. Edited by Alexander, C. Pearson Education Ltd. Harlow.

28.

29. 30.

896

Verschoor, C.C. 2014. Integrated Reporting: Lags in the U.S. Strategic Finance, December 2014. Institute of Management Accountants. Young, J. 2014. Operational Risk Management. 2nd Edition. Pretoria. Van Schaik Publishers. Young, J. 2014a. Practical Guidelines to Formulate an Operational Risk Appetite Statement for Corporate Organisations: A South African Perspective. Journal: Corporate Ownership and Control. Sumy, Ukraine. Volume 12, Issue 1, Autumn 2014 (47 - 63).


THE MARKET CONCENTRATION AND BANKING INDUSTRY PERFORMANCE Mohammed Salameh Anasweh* Abstract This study examines the structure-profit relationship in the Qatari banking industry. The study sample consists of all the local banks operating in the market (13 banks) listed in Qatar Stock Exchange (QSE) over the 2009-2014 period. The hypotheses related to the market power structure which includes the traditional Structure-Conduct-Performance Hypothesis (SCP), and the traditional Efficiency Hypothesis (EH). The empirical results generally support the (SCP) Hypothesis in Qatari banking industry. Thus, the main implication of these results for the policymakers, of Qatari banking sector, is to expand the ongoing deregulation efforts with the aim of reducing the industry concentration and enhancing the market competitiveness. Keywords: Market Structure, Performance of Banks, Qatari Banking Industry *College of Business Administration, Mutah University, Jordan

1 Introduction All countries need efficient financial institutions to promote and support economic growth. Starting with King and Levine (1993), research on the link between finance and economic growth reveals that countries with “better” financial systems tend to grow faster. However, the existence of financial institutions per se is not enough; the quality and efficiency of these institutions are crucial for the transmission of funds in the economy. Financial institutions multiply and allocate society’s savings, and the efficiency with which they intermediate capital has substantive repercussions on economic performance (Jayaratne and Strahan, 1996), (Demirgiic-Kunt and Maksimovic, 1998) , (Rajan and Zingales, 1998) and (Levine et al, 2000). Wheelock and Wilson (1995), Studies on banking efficiency are relevant in constantly changing economies such as the Mexican case. Countries that undergo significant transformations in their financial institutions face different challenges from one year to another, and only efficient institutions will be able to face them successfully. The study of banking efficiency is important since efficiency measures are indicators of success. Banks, as any other firm, face numerous sources of competition from both other banks and other firms inside and outside their industry. An open and flexible banking environment not only provides more credit, but also better allocation of credit, leading to the funding of more positive net present value projects that contribute to economic growth (Diaz, 2011). Financial intermediation is essential for economic development. The international banking

industry has undergone substantial structural reforms over the last two decades. There have been fundamental changes in the behavior of banks with more emphasis on profitability and comprehensive asset management in recent period. It is particularly important for emerging countries to ensure that banking system is stable and efficient. Such a banking development should lead to private and infrastructural projects being financed effectively and allocated efficiently. As Albertazzi and Gambacorta (2009) argue, because of phenomena such as globalization, growing international financial markets, deregulation and advances in technology, identifying the determinants of bank performance is an important predictor of unstable economic conditions. Athanasoglou et al (2008) also point out that a profitable banking system is likely to absorb negative shocks, thus maintaining the stability of the financial system. In this respect, it is important to investigate the effectiveness of emerging banks. How banks are affected by increased competitive pressures, depends partly on how efficiently they are run. Banks can increase their profitability through either improvement of their cost efficiency or exerting their market power. The latter approach to make profit can reduce total social welfare (Mirzaei et al, 2011). The market structure performances on the banks are rare and typically insufficiently robust as they are based on a limited number of countries only. Traditionally, market structure indicators, such as the number of banks and banking concentration, have been considered the major determinants of competition in the banking sector. This study aims to investigate the market structure-profit relationship on banks listed in (QSE). The specific objective of the

897


study is to analyze this relationship in terms of market share and banking concentration. Specifically the study will answer the following: Is there a statistically significant effect of the concentration or market share in the performance of banks listed in (QSE)?. Is there effect of concentration and market share in the performance? 2 Literature review and developments of hypotheses From the theoretical point of view, the effect of banking development on the volatility of output is ambiguous. Morgan et al (2004) suggest that improved access to banking finance allows firms to smooth out their idiosyncratic shocks. However, the effect of banking development on volatility of economic growth can be affected by the stage of the development of the country (Aghion et al, 2004), the type of shocks that the economy faces, such as monetary or real shocks (Bacchetta and Caminal, 2000), or whether the economy faces credit demand versus credit supply shocks (Morgan et al, 2004), (Hoxha, 2013). As a step toward understanding the relationship between bank market structure and economic activity, first investigate the relationship of concentration in state banking systems on the growth of manufacturing industries in the first three decades of the twentieth century. As noted earlier, previous studies have used national bank concentration ratios to investigate the effects of banking market structure on industrial growth across countries. In this research focuses on variation within a single country, the United States, which allows us to control better for differences in financial development. Studies that examine crosscountry variation, such as Rajan and Zingales (1988) and Cetorelli and Gambera (2001), assume that financial development is uniform within a country and attempt to identify cross-country differences. Following another studies, assumed that the level of financial development is uniform within the United States (although some of our fixed-effects specifications allow for the possibility that financial development varied across states). The importance of the financial intermediaries for growth has not been established until at least the last two decades. In fact, Schumpeter (1911) argued that financial intermediaries are essential for technological innovation and economic development; however, for most of the last century financial development has been observed as being correlated with economic growth. One of the first studies that established causality between financial development and growth is King and Levine (1993), which was followed by Levine and Zervos (1998) which argued that bank credit and growth are positively correlated. One view suggests that markets with concentrated and less competitive banks are not growing at their best

potential, since firms do not have access to credit, which leads to less growth, (Pagano, 1993 and Guzman, 2000). According to conventional wisdom, the increase of competition should warrant an expectation for lower prices on bank services, and greater availability of credit, which would make it affordable for the small firms to borrow and invest more. Many empirical studies support this view, finding that higher concentration and more restrictions on competition lead to less new firm creation, and less economic growth (Berger, Hasan and Klapper, 2004. Cetorelli and Strahan, 2006). Allen and Gale (2000) find that an increase in bank market power leads to higher loan rates charged to borrowers, while Claessens and Laeven (2005) using a cross-section estimation method for bank competition, find that banking competition is important for the growth of industries dependent on external finance. The study of Hamdan et al,(2014), aimed to investigate the relationship between banking market structure and profitability of banks of Bahrain and Kuwait , the study sample included local banks in the two countries, (23) bank during the period (20052010). Results of the analysis in general have confirmed support to the hypothesis structurebehavior- performance hypothesis explained the relationship between market structure and profitability of Bahraini banks, while the results did not provide support for the hypothesis structure - behavior Performance in Kuwaiti banking market, and then exclude the alliance between most banks hypothesis concentrated, and the results do not support the hypothesis of conventional efficiency in the Kuwaiti banking market.In other study to Hamdan(2014), aimed to understand the restructuring of the banking sector in the United Arab Emirates and factors are instrumental in revenues, in terms of competition and monopoly and levels of efficiency, the study sample large proportion of UAE banks (96%) , during the period (2007- 2012) .The study found experimental evidence to support absence concentration hypothesis banking in the UAE banking market. The study suggests work in conditions of full competition, and other evidence supports excellence UAE banks efficiently cost and efficiency standard profit, that explain the returns of sectors. This confirms the absence of the banking monopoly conditions in the United Arab Emirates, while the banking sector returns interpreted through the structure of efficiency and not through force market. The recommendations of this study was to prevent concentration and monopoly by encouraging access to the market to encourage competition and to support the legislation that limit the emergence of any monopolistic practices policy, and in addition to maintaining the status of the banking market balance. While in Brazil Resende (2005), investigated that the structure–conduct–performance (SCP) relationships in the context of the Brazilian manufacturing

898


industry in 1996. For that purpose, it considered a system with four equations pertaining concentration, advertising, R&D and profitability that was estimated with simultaneous equation models. In addition to the usual explanatory variables proxying barriers to entry and demand conditions, the article considered organizational practices and incentive schemes variables. The evidence indicated an important role for variables related to barriers to entry in affecting market structure, an important and nonlinear effect of concentration on advertising, a relevant impact of firm-size on the propensity to exert R&D effort and finally a significant positive impact of concentration on profitability, and were similar to the previous evidence for developed countries. Additionally, no important roles were detected for organizational practices and incentive schemes on the SCP relationships. Lam at el, (2007) , mentioned Market concentration on the major container shipping routes has the potential to reduce contestability, impede effective competition and, as a consequence, inhibit the positive relationship between trade and economic growth. This development could also hamper the ability of economic regions to realize their respective competitive and comparative advantages. Within this context, the structure- conduct-performance (SCP) framework is used to analyze liner shipping dynamics in the transpacific, Europe–Far East and transatlantic trade routes. The analysis finds no conclusive evidence that either the increased concentration of slot capacity or the attempts by shipping lines to boost potential slot capacity (mainly through collaborative arrangements) lead to improved financial performance. To conclude that, despite high and increasing concentration among carriers on each of the trade routes analyzed, these markets remain contestable.

contribution to the market. The second part refers to (Conduct) the behavior of banks, which depend on economic characteristics, management of bank costs, and the trade-off between risk and reward, size efficiency and efficiency of the debts and obligations. The third part refers to (Performance) the level that is affected by each of the banking market structure and efficiency of the administration; it must be compared to the costs and profits of the bank (Al-Atyat, 2015). The banks concentration and other impediments to affect competition on the performance of banks in inappropriate ways and generate social loss with poor banking services and pricing, this resulting to the practice of banks market strength arising from the increased concentration levels according SCP hypothesis (Hamdan et al, 2014). The Traditional Efficiency Hypothesis was presented by (Demsetz, 1973), which is assumed that differences in organizations and dispersion within the market result in inequality in market shares, so that higher levels of efficiency associated with the largest market shares for a limited number of banks which leads to high levels of performance and then a positive relationship between market share and profit (Hamdan et al, 2014). This hypothesis suggests that the most efficient banks increase in size and market share and then increase their ability to achieve high profits through market share concentration in a limited number of banks (Al-Atyat and Hamdan, 2015). Based on SCP and TE hypothesis; the main hypothesis of the study is: “There is no statistically significant effect of the market share and concentration of market on the performance of banks listed in Qatar Stock Exchange”.

2.1 SCP and TE hypothesis in Qatar Market

3.1 Study population, resources of data

Bain (1951), mentioned that there are many concentrated markets because of low competition for reasons of alliance-type or monopolistic led to the development of inappropriate prices for consumers (for example, in the manufacture of high interest rates banks put on loans and lower interest rates on deposits compared with other competitive environment) this contribute to achieving high profits, which is known as bank concentration. According to this hypothesis, there are few monopolistic banks leads the rest of the banks towards the development of higher prices and lower costs, and then achieve the highest profit levels at the expense of consumers (Al-Zubi, 2005). The SCP is composed of three parts; the first part is the (structure) which refers to the banking market structure characteristics in terms of the number of banks, the concentration ratio and the size of their

The study sample covered is the banks listed in QSE which are 14 banks. The data was collected from the Investors' Guide by QSE and banks annual reports based on the following conditions: 1) all data is available, and 2) The bank did not merge with another bank or have been liquidated during the current study period. As a result, the final study sample became 13 banks starting from 2009 until 2014 signify 93% of the inventive study population.

3 The methodology sample

and

3.2 Study model The study used the following model to examine and express about market structure and banks performance with the addition of a set of control variables, so as to adjust the relationship between independent and dependent variables.

899


n

 i ,t   0  1Conci ,t   2 MSi ,t    k Z itk   i ,t

(1)

k 1

Where: π i,t : Performance of bank (i) in the year of (t). β0 : Constant β1,2,k : Slope or changes of markets structure. MSi,t: Market shares of bank (i) in the year of (t). Conci.t : Concentration of bank (i) in the year of (t). zitk : Control Variables, include: Bank Size, Bank Age, Number of Branches εi,t : Random error. performance of banks listed in QSE, according to hypothesis of traditional efficiency; the following model has been put to the second hypotheses:

3.2.1 The sub-models The first sub-hypothesis is designed to test the relationship of market concentration of assets on performance of banks listed in QSE, according to SCP hypothesis; the following model has been put to the first hypotheses:

n

 i ,t   0   1MSi ,t    k Z itk   i ,t

(2)

k 1

3.3 Measurement of variables

n

 i ,t   0  1Conci ,t    k Z itk   i ,t

(2)

k 1

The following Table 1 summarizes the measurement of the dependent, independent and control variables.

The second sub-hypothesis is designed to test the relationship of market share of deposits in Table 1. Measurement of variables Variables Dependent variable: Bank Performance.

π

Label

Measuring Independents Variables: Market share

MS

Concentration

Conc

Measuring Control Variables: Company size Company age Number of Branches

Size Age Branches

Definition and measurement Measured by the return on asset (ROA) and the measurement of the effectiveness of administration on using available resources and the extent of their ability to achieve return from various sources available. It reflects market share of each bank deposits (Credit Facilities) and this indicator is used to measure the traditional efficiency hypothesis (Hamdan, 2014). Measured by the Credit Facilities of bank to the total Credit Facilities of banks through this equation: Measured by the total market share of assets of each bank, according to the following equation (Ahmadov,2012) Nature logarithm of total assets. Time span of the company. Measured by number of branches that the bank owned.

4 Testing of hypothesis

4.1 Validity of data

The current section contains three parts. The first part will hold testing the validity of data utilized in the research. While the second part will include the descriptive analysis followed by the third part which will hold the empirical analysis.

At the onset we have to examine validity of data for statistical analysis. For this purpose, we used normal distribution test, Multicollinearity test, Autocorrelation test, and Homoskedasticity test. Validity of the study models representing correlation between market Concentration and banking industry performance was secured. Thus, we can say that the study models in equations numbered (2 and 3) are

900

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 and have the same probability random error εi,2; εi,3 with a variance of σ2; σ3.

accurate. All variables on the right side express nonrandom variables excluding the last one εi,2; εi,3 which is supposed to belong to natural distribution with zero average and fixed variance is expressed in σ2; σ3. All these variables are independent ones. As for variable (π) they are dependent in the two models

4.2 Descriptive statistics Table 2 summarizes the descriptive statistics of banks performance, market share and concentration.

Table 2. Descriptive analysis Variables

Label

Bank Performance

PER

Market Share

MS

Concentration

Con

Years

Minimum

Maximum

Mean

2009 2010 2011 2012 2013 2014 2009 2010 2011 2012 2013 2014 2009 2010 2011 2012 2013 2014

-46.556 -47.912 -23.391 -49.155 -40.002 -44.002 0.079 0.147 0.124 0.124 0.158 0.176 4.644 11.470 10.769 8.916 10.961 12.430

24.240 22.680 22.920 21.480 24.000 26.400 6.456 6.312 5.436 5.088 5.040 5.594 12.216 12.180 10.692 10.692 8.040 9.117

1.300 -4.693 2.704 1.716 4.524 4.976 1.846 1.846 1.846 1.846 1.846 2.049 12.116 13.195 10.413 12.883 10.556 11.971

The bank performance is measured by Return on Assets (ROA).where the maximum (ROA) was 4.98 in the year of 2014. The mean was unstable between the years and in the year 2011, so the study assumes that unstable of banks performance due to the consequences and impact of the global financial crisis. The mean of market share was stable the same in all years, which indicates that the market share of banks is difficult to move between dominant sectors and it could be a competition between the banks. The mean was increased in the second year 2010 than it started to decrease till the year 2013, This is due to that in the beginning it was concentrated in high market because the number of a few existing banks for the next year and banks was entry into the market so doing, at least the previous concentration on the banks because of the new banks enter to the market and also the stability of data from the year 2010 to 2014 this indicate that difficult to move between dominant sectors and it could be a competition between the banks. 4.3 Models testing Based on that Pooled Regression and the results of this test can be found in table (3). The study hypothesis may be tested as follow: Where the first hypothesis tested the relationship of market share of deposits in the performance of

Standard deviation 2.091 2.100 1.235 1.906 1.801 1.981 0.186 0.185 0.179 0.178 0.173 0.192 0.206 0.000 0.021 0.037 0.031 0.035

banks listed in QSE. This hypothesis will test how the banks will differ if they are having high market share and if they have low market share, and if they have high or low market share how they will impacted the performance of banks listed in QSE. This hypothesis formed based on what found in previous studies about the market share of banks and their relationship in the bank's performance. In contrast, the Efficient Structure (ES) hypothesis argues that the efficient firms outperform the others and therefore gain higher market share which results in a higher concentration of the market structure. The ES hypothesis was proposed by Demsetz (1973) and developed by Brozen (1982). According to this hypothesis, the explanation of the relationship between market structure and performance of the individual firm are the firm-specific efficiencies. This efficient bank is assumed, therefore, to gain a large market share that may result in high levels of concentration, and the Bank’s efficiencies will be the driving force behind the process of the market concentration. The second hypotheses tested the relationship of market concentration of banks assets and return on assets in the performance of banks listed in QSE, according to SCP. This hypothesis measured by the market share of the assets of each bank by (HHI). This hypothesis was formed based on what was found in previous studies regarding the market concentration of

901


banks in the performance of banks. As many studies like Molyneux (2006) states that the most frequently used measure of market structure is concentration ratio and the second most frequently used is the HerfindahlHirschmanIndex (HHI). He indicates that the GCC

banking industries are characterized by high market concentration. Overall, the high degree of concentration in GCC banking markets suggests that the strict licensing rules and restrictions on foreign bank entry have helped create these market structures.

Table 3. Pooled regression results Variables Independent variables: Constant Concentration (con) Market share (MS)

Model 1

Model 1.1

1.791** (0.046) 1.417 (0.224) 8.173*** (0.000)

3.438** (0.018)

Model 1.2 2.751** (0.015) 1.579 (0.219)

4.692** (0.000)

Control variables : Numberof Branches

4.180*** 3.025** 2.420** (0.000) (0.013) (0.010) Bank size 3.214** 6.171*** 4.937*** (0.021) (0.001) (0.001) Bank Age 1.096 2.104 1.683 (0.406) (0.279) (0.223) R 0.321 0.297 0.192 R-squared 0.103 0.088 0.037 F-statistics 9.691*** 4.521*** 3.617*** p-value (F) 0.000 0.000 0.000 Note: OLS: t-test (top), p-value (bottom); significance at: *10%; ** 5% and ***1% levels. After testing the hypothesis, the results are summarized in table (3), that t-test of Market share was positive and p-value is less than 5% , the market concentration was positive and p-value is more than 5% and by testing both market share and the market concentration found that the market share positive and p-value is less than 5% which is significant by using the three models (MS, Con ,MS &Con) in table (3). This indicates that will accept the first hypothesis because the p-value is less than the 5% which mean that is significant, while rejected the second hypothesis because the p-value is more than the 5% which means that it's not significant. By accepting the first hypotheses about market shares relationship on bank performance therefore its mean that the banks with high market share will have better performance than other banks. The rejected hypotheses is the second that is about the market concentration and its effects in the performance of banks therefore its mean that is no concentration in the market of banks listed in QSE, this is linked to the laws and regulations listed in Central Bank of Qatar that prevent monopoly in the market. As many studies like Al-Muharrami and Matthewsm, (2009), conclude that there is little evidence that banks in the more concentrated GCC markets exhibit lower technical efficiency for the period 1993 to 2002. This is in contrast to Berger and Hannan (1997, 1998), who find evidence that CR3 proxies market power and those banks with more market power are less diligent in controlling costs. The results do not support the QL hypothesis and

conclude that the empirical evidence supports the basic SCP version of the market power hypothesis that associates market concentration with profit performance. 5 Discussion of recommendations

conclusion

and

The main objective of the study is to investigate the profit relationship of market structure on performance of the banks listed in QSE. The study also aimed to analyse this relationship in terms of market share and banking concentration. There are few studies that the relationship of market structure on bank performance in GCC, one of these study by (Hamdan et al, 2014) they study Market Structure - profit relationship in Bahrain and Kuwait. These studies were supported by different theories, had different sample size as well as different model. Conducting this study in Qatar aimed to benefit shareholders, investors, bankers and other stakeholders taking financial decision. Also, it is beneficial to know what really affects banks performance in this area and whether market share and market concentration really affect bank performance. Our Study built three different regression models to study effect of market structure on bank performance. The first model used Market Share (MS) as an indicator of performance, the second model used market concentration as an indicator of performance and the third model used both market share and market concentration as an indicator of performance.

902


The Study notices that t-statistic of Market share (MS) was positive and p-value is less than 5%, this result due to banks with high market share will have better performance than other banks. Market concentration was positive and p-value is more than 5%, this result indicates that is no concentration in the market of banks listed in QSE, this is linked to the laws and regulations listed in Central Bank of Qatar that prevent monopoly in the market. Finally; there is statistically significant effect of market share on performance of banks listed in QSE. Study recommends Qatar regulators and supervisors of banking sector limit the impact of market power of concentration, through putting more legislation, to regulations and constantly update to keep pace with developments in banking business, and limit concentration of banks, as this has a significant economic and social impact. Encourage banks to efficiently manage its financial resources, and to use of advanced technologies in banking business to support and enhance their ability to compete in local and global markets and improve their returns. Our Study recommends doing courses for bankers about the impact of market structure on bank performance to be ready for any changing factors that influence banks performance. However, with a small sample size, caution must be applied, as the findings might not be generalizable. This research has thrown up many questions in need of further investigation; it is recommended that further research be undertaken in the following areas: The market structure-profit relationship in the GCC’s Banking Industry.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

References 1.

2.

3.

4.

5.

6.

7.

8.

Al-Atyat, Y., & Hamdan, A. (2015). The structure of islamic Banks Market and Its Relation to Profitability : A Comparative Study of the GCC Countries. Public Administration Journal, 55(2), 330-333. Albertazzi, U. &Gambacorta, L. 2009, "Bank profitability and the business cycle",Journal of Financial Stability, vol. 5, no. 4, pp. 393-409. Al-Jarrah, I. (2010a). The Market Structure-Profit Relationship in the Jordan’s Banking Industry. Dirasat Journal of Administrative Sciences, 37(1), 251-261. Al-Jarrah, I. (2010b). Competition in the Jordanian's Banking Sector. Dirasat Journal of Administrative Sciences, 37(2). Allen, Franklin, and Douglas Gale (2000) Comparing Financial Systems, Cambridge, MA and London: MIT Press. Al-Muharrami, S., & Matthews, K. (2009). Market power versus efficient-structure in Arab GCC banking. Applied Financial Economics. Al-Zu’bi, K., & Balloul, M. (2005). Structure, Competitiveness and Efficiency Aspects of Jordanian Banking Industry. Dirasat, Administrative Sciences, 32(1), 230-249. Ashton, J. K. (2001). A test of perfect competition in the UK retail-banking deposit market. The Service Industries Journal, 21(4), 119-132

19.

20.

21.

22.

23.

24.

25.

903

Athanasoglou, P.P., Brissimis, S.N. & Delis, M.D. 2008, "Bank-specific industryspecificand macroeconomics determinants of bank profitability", Journal ofInternational Financial Markets, Institutions and Money, vol. 18, no. 2, pp. 121-136. Athanasoglou, P.P., Delis, M.D. &Staikouras, C. 2006, "Determinants of BankProfitability in the South Eastern European Region", Journal of Financial DecisionMaking, vol. 2, pp. 1-17. Bain, J. S. (1951) The relation of profit rate to industry concentration: American manufacturing, 1936– 1940,Quarterly Journal of Economics, 65, 293–324. Berger A (1995) The profit-structure relationship in banking—tests of market-power and efficientstructure hypotheses. J Money Credit Bank 27(2):404–431 Berger, Allen N., Iftekar Hasan, and Leora F. Klapper (2004) "Further Evidence on the Link Between Finance and Growth: An International Analysis of Community Banking and Economic Performance," Journal of Financial ServicesResearch 25, 169-202. Bikker, J., Finnie, P., & Spierdijk, L. (2007). The impact of market structure, contestability and institutional environment on banking competition. DNB Working Paper. Bourke, P. 1989, "Concentration and Other Determinants of Bank Profitability inEurope, North America, and Australia", Journal of Banking and Finance, vol. 13,no. 1, pp. 65-79. Cetorelli, Nicola, and Philip E. Strahan (2006) "Finance as a Barrier to Entry: Bank Competition and Industry Structure in Local U.S. Markets," Journal ofFinance 61, 437-461. Chowdhury, T., & Ahmed, K. (2009). Performance Evaluation of Selected Private Commercial Banks in Bangladesh. International Journal of Business and Management, 4. Claessens, Stijn, and Luc Laeven(2005) "Financial Dependence, Banking Sector Competition, and Economic Growth," Journal of the European EconomicAssociation 3, 179-207. Diaz Avilez, V. (2011). Evidence on banking efficiency: An analysis of financial intermediation in mexico(Order No. 3459845). Available from ProQuest Central. (875770973). Hamdan, A. (2014). The impact of market power and efficiency in the UAE banking industry performance: An Empirical Study assessment of the Federal Emirates experience in the banking market regulation for the period 2007-2012. Emirates Center for Strategic Studies and Research. Hamdan, A., Shaheen, A., & Anasweh, M. (2014). The Market Structure - profit relationship in Bahrain and Kuwait. The Academic Publication Council, 40(152), 344-371. Ifeacho, C., &Ngalawa, H. (2014). Performance of the south african banking sector since 1994. Journal of Applied Business Research, 30(4), 1183-n/a. Kammoun, N., &Ammar, A. (2012). The evolution of competition in banking in a transition economy: An empirical analysis of the tunisian banking sector. International Journal of Business and Social Science, 3(13). King, Robert G., and Ross Levine (1993) "Finance and Growth: Schumpeter Might Be Right," Quarterly Journal of Economics 108, 717-738 Kirkwood, J., & Nahm, D. (2003). Australian banking efficiency and its relation to stock returns.


26.

27.

28. 29.

30.

31.

Kirkwood, J., &Nahm, D. (2006). Australian banking efficiency and its relation to stock returns. Economic Record, 82(258), 253-267. LAM, J., YAP, W., & CULLINANE, K. (2007). Structure, conduct and performance on the major liner shipping routes. Maritime Policy & Management, 34(4), 359-359. Levine, Ross (1999) "Law, Finance and Economic Growth," Journal of Financial Intermediation 8, 8-35. Levine, Ross, and Sara Zervos (1998) "Stock Markets, Banks, and Economic Growth," American Economic Review 88, 537-558. Mirzaei, A., Liu, G., & Moore, T. (2011). Does Market Structure Matter on Banks’ Profitability and Stability? Emerging versus Advanced Economies. Economics and Finance Working Paper Series. Rajan, Raghuram G. and Rodney Ramcharan, 2009. “Land and Credit: A Study of the PoliticalEconomy of

32.

33.

34.

35.

904

Banking in the United States in the Early 20th Century.” NBER Workingpaper no. 15083 Raoudha, A. (2014). Market structure and bank fragility: Application to the tunisian banking system. Zagreb International Review of Economics & Business, 17(2), 1-20. Resende, M. (2005). Structure, Conduct and Performance: A Simultaneous Equations Investigation for the Brazilian Manufacturing Industry. Instituto De Economia, Universidade Federal Do Rio De Janeiro. Schumpeter, Joseph A. (1911) The Theory of Economic Development, Cambridge, MA: Harvard University Press. Worthington, A. (1998), “The Determinants of Nonbank Financial Institution Efficiency:A Stochastic Cost Frontier Approach”, Journal of Econometrics, 8, pp. 279 – 289.


THE LEVEL AND STABILITY OF INSTITUTIONAL OWNERSHIP AND ITS INFLUENCE ON COMPANY PERFORMANCE IN SOUTH AFRICA Waldette Engelbrecht* Abstract In this article, the level of international institutional shareholding, domestic institutional shareholding total institutional shareholding, and the variance of institutional shareholdings as the proxy variable for ownership stability were investigated. In addition, return on assets (ROA) and Tobin’s Q were used as performance variables for the company. This study can provide valuable information for regulators of emerging markets when determining policies regarding institutional ownership of companies, especially foreign institutional ownership, as foreign institutions usually play a controversial role in these markets. Against the background sketched above, research was undertaken to determine whether company performance is influenced by the level and stability of its institutional shareholders. Keywords: Institutional Ownership, Performance, South Africa *Stellenbosch University

1 Introduction High-profile accounting scandals and bankruptcies of the past two decades have highlighted the risks involved in the ownership of companies (Bhasin, 2013). The owners or shareholders of a company rely on the directors of that company to control and respect their interests. This is referred to as the agency theory. The ownership structure of a company is therefore an important tool in the corporate governance system (Perrini et al., 2008) which can most often resolve the agency theory (Jensen and Meckling, 1976; Shleifer and Vishny, 1986), thereby improving company performance (Alipour, 2013: 1138). The abilities of institutional ownership collectively exceed and are substantially stronger than the joint efforts of smaller and diverse investors (Chen et al., 2008: 109). In many countries institutional ownership has evolved over time and represents a significant part of the shareholdings of listed companies. In the United Kingdom (UK), the institutional ownership increased from 27.7 % in 1963 to 59.4 % in 2012, based on the latest available statistics (UK Office of National Statistics, 2013). In the US, corporate ownership became dispersed as early as 1930, resulting in the agency theory. Individual share ownership continued to be the dominant form of share-ownership up until the 1980s. These individuals were rarely actively engaged in corporate governance and corporate boards were mainly made up of insiders. During the 1980s, macroeconomic growth slowed down and the US economy was under pressure. Institutional ownership came to

the fore as shareholders of companies with hostile takeovers and pension funds investing in companies. These institutional investors began to participate in the affairs of the companies in which they had shares and became active players in the corporate governance of companies (Jackson, 2010). In the UK, in the 1940s and 1950s important changes in the capital structure of companies occurred. Following a number of financial scandals, minority interest protection was strengthened in the 1940s. There was also a sharp increase in institutional ownership. By 1960 about one third of listed companies had a majority of institutional ownership (Mork, 2007: 584). Unlike companies in the US and UK, shares in the typical Asian company are mostly held by family members. Companies are often affiliated by business groups controlled by the same family. In addition to the family ownership, government controls a significant number of listed companies in several countries including Singapore and China. Individual or institutional shareholders have a minority interest in corporate shares in developing China (Claeseens and Fan, 2002). Due to the political environment in South Africa between 1961 and 1994, the country was totally isolated from the global economy. Political reforms led to the collapse of apartheid in 1994 which in turn led to the lifting of sanctions imposed by the international community on South Africa (Malherbe and Segal, 2001). Since 1994, it appears that the institutional share ownership of South African corporations has evolved through two phases. An

905


early phase of institutional reform occurred from 1994 to 2001 because the government sought to encourage corporate participation in the process of incorporating the socially disfavoured into the mainstream. Arya and Zhang (2009) argue that developments such as the Mineral and Resources Development Act of 2002, followed by the Black Economic Empowerment Act of 2003, signalled another phase of institutional reform. 2 Problem statement Several past studies investigated how the proportion of institutional shareholders affected managers and company performance. This explanation is not complete without looking at other factors, namely stability of institutional shareholdings, which encourages long-term performance and corporate governance (Bushee, 1998; Bushee and Noe, 2000; Chen et al., 2007; Elyasiani and Jia, 2010; Elyasiani et al., 2010). The findings of these studies show that the higher and more stable the institutional holdings, the stronger the incentives to monitor directors. Institutional investors can provide expertise, experience, and resources for stock market analysis, which are more pronounced with foreign institutional investors. Barber, Lee, Liu and Odean (2009) and Huang and Shiu (2009) found evidence that stocks with higher foreign institutional holdings provide better returns than those with lower foreign institutional holdings. These authors also indicate differences between the performances of companies with foreign institutional investors and those with domestic institutional investors. In this article, the level of international institutional shareholding, domestic institutional shareholding total institutional shareholding, and the variance of institutional shareholdings as the proxy variable for ownership stability were investigated. In addition, return on assets (ROA) and Tobin’s Q were used as performance variables for the company. This study can provide valuable information for regulators of emerging markets when determining policies regarding institutional ownership of companies, especially foreign institutional ownership, as foreign institutions usually play a controversial role in these markets. Against the background sketched above, research was undertaken to determine whether company performance is influenced by the level and stability of its institutional shareholders. 3 Structure of the remainder of the article The remainder of this article is structured as follows: first, the limitations of the study are provided. These are followed by a review of prior research and the development of hypotheses. Next, a discussion of the methodology follows; it includes the sample selection and a presentation of the research model. The

empirical results are then discussed, followed by the conclusion and suggestions for future research. 4 Research limitations The study had specific limitations. The assessment is limited to the annual reports of 71 of the Top 100 companies listed on the Johannesburg Stock Exchange (JSE) for the 2009 to 2013 reporting period as sourced from the INET McGregor BFA. Specific market-based and accounting-based performance measures which were sourced from the financial ratio and financial model function of INET McGregor BFA were used. The measures were selected based on prior studies (e.g. Huang and Shiu, 2009; Bhattacharya and Graham, 2009; Alipour, 2013). The usage of other performance measures could possibly have led to different results. 5 Theories, prior research and hypothesis development A wide range of research on institutional shareholding was used during this study to identify how institutional shareholders influence company performance, as reported below. The agency theory forms the theoretical perspective behind the hypotheses for the study on which this article is based. Jensen and Meckling (1973) and Berle and Means (1932) explained agency theory as the separation of corporate ownership and control which has the potential to lead to agency problems in the form of self-interested actions by directors and other managers within a company. It is suggested that as a result of their independence, institutional shareholders can assist in monitoring and controlling management due to their large shareholding, whereby the agency problem is reduced (Choi et al., 2012: 269). Pound (1988) discusses further theories describing the relationship between institutional shareholders and firm value. The efficient monitoring theory states that the larger the institutional shareholding, the more efficient the monitoring of that shareholder, which leads to a positive correlation between institutional ownership and firm value. In contrast, the strategic alignment and conflict-ofinterest theories state that large institutional shareholders maintain strategic relationships with directors, which influences their voting behaviour. These two theories predict a negative correlation between firm value and institutional shareholding. Navissi and Naiker (2006) conducted a study which showed it is possible that institutional investors, similar to directors, could decrease company value once their shareholdings exceed a certain level. This is referred to as the convergence-of-interest theory. At higher levels of share ownership, institutional shareholders may engage in decisions that could be harmful to the company; this is the entrenchment

906


theory. The mixture of the two latter theories leads to a further non-linear relation between the institutional shareholders and the value of the company. In the twenty-first century, institutional ownership in companies has emerged as an important tool in monitoring and controlling business interests of the shareholders of a company (Chen et al., 2008: 109). Earlier research supports the view that company value is dependent on the shares allocated to directors and outsiders (Berle and Means, 1932; Jensen and Meckling, 1976). Shleifer and Vishny (1986) explain that large institutional investors will have a positive effect on the market value of the company because they act as an effective monitoring tool. Barclay and Holderness (1990) proved that there are positive excess returns in a company when it became known

that institutional investors would acquire large share positions. Yuan, Xiao and Zou (2008) used the Chinese market to investigate the relationship between mutual funds and company performance and provided evidence that long-term investors have stronger monitoring capabilities and positive effects on company value. Institutional shareholders, unlike individual shareholders, are equipped for efficient monitoring of the directors and reducing the costs; thus there is a positive relationship between institutional ownership and performance. This finding is supported by Shleifer and Vishny (1997), and Filatotchev, Lien and Piesse (2005), who found evidence of a significant positive relationship between institutional ownership and performance.

Table 1. Findings of studies relevant to international institutional shareholding and company performance Authors Douma, George and Kabir (2002)

Wei, Xie and Zhang (2005)

Filatotchev et al. (2005) Gűrbűz, Aybars and Kutlu (2010) Mizuno (2010)

Aggarwal, Erel, Ferreira and Matos (2011) Choi, Park and Hong (2012)

Mi Choi, Sul and Kee Min (2012)

Country and number of companies tested India (1 005 companies listed on the Capitaline 2000 database) China (5 284 companies listed on the Shangai or Shenzen Stock Exchange) Taiwan (228 companies listed on the Taiwanese Stock Exchange) Turkish (164 companies on the Istanbul Stock Exchange) Japan (189 companies listed on the Tokyo Stock Exchange) United States of America (Various companies – classified as US and non-US) Korea (301 companies listed on the Korean Stock Exchange)

Korea (896 companies listed on the Korea Stock Exchange) Source: compiled by authors

Findings There is a significant positive relationship between foreign institutional shareholders and company performance when measured by the Tobin’s Q ratio. Foreign ownership is significantly positively related to Tobin’s Q.

Link with the current study Supports hypothesis 1.

Supports hypothesis 1.

Foreign institutional shareholders are associated with improved company performance. Foreign institutional investors improve the performance of companies.


International institutional shareholders have a positive influence on corporate governance, which in turn influences company performance. International institutional ownership positively influences corporate governance mechanisms.


International institutional shareholders are positively associated with technological innovation performance and played a positive role in improving corporate governance in the company. International share ownership is positively correlated with company value.


Some studies investigating the influence of institutional ownership on firm performance used return on assets and return on equity as the accounting profit rates to test for firm performance (Filatotchev et al., 2005; Cornett et al., 2007; Chen et al., 2007). Other studies used Tobin’s Q to measure firm performance (Wei et al., 2005; Mi Choi et al., 2012).


Indirectly supports hypothesis 1.


Most studies employed both measures (Douma et al., 2002; Chen et al., 2008; Yuan et al., 2008; Huang and Shiu, 2009; Bhattacharya and Graham, 2009; Alipour, 2013). Chen, Blenman and Chen (2008) reports that Tobin’s Q is a common measure of efficiency and future opportunities of a company and thus forward looking, while accounting profit rates, such as return

907


on assets and return on equity are backward looking. The accounting profit rates only partially incorporates estimates of future events such as valuations of goodwill and depreciation. Tobin’s Q is however based on investor’s psychology surrounding world events, which include business strategies (Demsetz and Villalonga, 2001). In the sample tested in this research, both measures were utilised, namely return on assets as the accounting profit rate and Tobin’s Q as a forward-looking rate. A correlation is expected between Tobin’s Q and return on assets. This can be seen from Table 7 and a 65% correlation rate is reported. 5.1 International institutional ownership and company performance Ferreira and Matos (2008) reported that foreign and independent institutions, with potentially fewer business ties to companies, are involved in monitoring corporations worldwide (Bhattacharya and Graham,

2009: 370–371). It has been noted that foreign investors will invest abroad when the investment relates to their main business and if found it will provide a competitive advantage. Especially in the emerging economies, an increase in international shareholders is associated with positive and successful business growth (Choi et al., 2012: 273). International shareholders lead to efficient management of companies by improving corporate governance due to their relative independence from other domestic shareholders (Mi Choi et al., 2012: 208). Further, a company’s reputation in the market is improved by the presence of international shareholders and encourages additional investors (Mi Choi et al., 2012: 212). A summary of the main findings of studies relevant to international institutional shareholding and company performance is presented in Table 1. How these link with the current study is also shown.

Table 2. Findings of studies relevant to domestic institutional shareholding and company performance Authors Douma (2002)

et

al.

Country and number of companies tested India (1 005 companies listed on the Capitaline 2000 database)

Filatotchev et al. (2005)

Taiwan (228 companies listed on the Taiwan Stock Exchange)

Huang and Shiu (2009)

Taiwan (523 companies from the Taiwan Economic Journal database)

Gűrbűz (2010)

Turkey (164 companies on the Istanbul Stock Exchange) United States of America (Various companies – classified as US and non-US)

et

al.

Aggarwal et al. (2011)

Hsu and Wang (2014)


Tsai and Tung (2014)


Findings A positive relationship exists between domestic institutional shareholders and company performance although not of the same magnitude as foreign companies. The effects of domestic banks as institutional shareholders were positive yet not as significant when compared to foreign banks. Company shares with high foreign ownership outperform stocks with low foreign ownership. This provides evidence that foreign institutional investors have an advantage over domestic investors. Domestic institutional investors improve the performance of companies. The correlation between corporate governance and institutional shareholders is driven by the nationality of the institutional investor. The correlation between domestic institutional ownership is positive and significant. This study showed differences between the performances of companies with foreign institutional investors and those with domestic institutional investors. This study reported a positive effect on companies by forcing directors to oversee a company in a more efficient manner when there are more domestic institutional shareholders.

Link with the current study Supports hypothesis 2. Indirectly supports hypothesis 2. Indirectly supports hypothesis 2. Supports hypothesis 2. Indirectly supports hypothesis 2.

Indirectly supports hypothesis 2. Supports hypothesis 2.

Source: compiled by authors In summary and based on the studies indicated above, it can be accepted that company performance increased with an increase in international institutional shareholding.

It is therefore expected that: H1: There is a positive correlation between international institutional ownership and company performance

908

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 5.2 Domestic institutional ownership and company performance Hsu and Wang (2014) reported distinct differences between the performances of companies with foreign institutional shareholders and those with domestic institutional shareholders, both showing a positive and significant correlation. While international shareholders are an integral component in company shareholdings in emerging countries, it does not make up the largest portion thereof (Gűrbűz et al., 2010: 3), hence it is necessary to look at domestic shareholding. A summary of the main findings of studies relevant to domestic institutional shareholding and company performance is presented in Table 2. In summary and based on the studies above, it can be accepted that company performance increased with an increase in domestic institutional shareholding. It is therefore expected that: H2: There is a positive correlation between domestic institutional ownership and company performance. 5.3 Total institutional ownership and company performance Navissi and Naiker (2006) found that shareholding by active institutional investors of up to 30% positively influences corporate value, and Cornett, Marcus, Saunders and Tehranian (2007) found that the percentages of institutional investor involvement in a

company, as well as their numbers, are associated with better operating cash flow returns. However, the findings only hold when the investors have no business relation with the company (Jafarinejad et al., 2015: 208). Institutional shareholders are also becoming more involved in corporate governance principles, especially in under-performing companies (Cornet et al., 2007: 1773). Gillan and Starks (2000) found that corporate governance initiatives endorsed by institutional investors were favourably received in contrast to those of independent individuals. Hartzell and Starks (2003) found that institutional ownership has a negative correlation with the level of executive remuneration and is positively correlated with pay-forperformance sensitivity. In conclusion, it was found that large institutional ownership in a company deters directors from pursuing opportunistic earnings (Chung et al., 2002) and thus it is better to follow their lead (Jafarinejad et al., 2015:207; Jensen and Meckling, 1976; Shleifer and Vishny, 1986). A summary of main findings of studies relevant to total institutional ownership and company performance is presented in Table 3. Based on studies listed in Table 3, there is evidence of a positive correlation between total institutional shareholders and company performance. It is therefore expected that: H3: There is a positive correlation between total institutional shareholding and company performance.

Table 3. Findings of studies relevant to total institutional ownership and company performance Authors Filatotchev et al. (2005) Navissi and Naiker (2006)

Cornett (2007)

et

al.

Chen, Harford and Li (2007)

Chen, (2008)

et

al.

Country and number of companies tested Taiwan (228 companies listed on the Taiwan Stock Exchange) New Zealand (123 companies listed on the New Zealand Stock Exchange) United States of America (676 companies listed on Standard and Poor)

United States of America (2 150 merger deals on the Mergers and Acquisitions database of Securities Data Company) New Zealand (259 companies listed on the New Zealand Stock Exchange)

Findings Institutional shareholders are associated with improved company performance.


Shareholding by active institutional investors improves the value of the company.


Higher institutional shareholding is associated with better operating performance and confirmed that institutional ownership promotes monitoring of corporate directors. Independent, long-term institutions invest in order to monitor and make beneficial long-term adjustments instead of short-term profits.


The Q ratio is positively related to the total institutional shareholding ratio and ownership promotes strong corporate performance.


909



Table 3. Findings of studies relevant to total institutional ownership and company performance (continued) Authors Yuan (2008)

et

al.

Huang and Shiu (2009)

Country and number of companies tested China (1 211 companies listed on the Shangai Stock Exchange and Shenzen Stock Exchange) Taiwan (2 471 companies from databases maintained by the Taiwan Economic Journal)

Bhattacharya and Graham (2009)

Finland (98 companies that are publically traded)

Gűrbűz (2010)

Turkey (164 companies on the Istanbul Stock Exchange) Various companies – classified as US and nonUS

et

al.

Aggarwal et al. (2011)

Choi et al. (2012)

Korea (301 companies listed on the Korean Stock Exchange) Fung and Tsai United States (2 249 (2012) companies from the NYSE, AMEX and NASDAQ) Alipour (2013) Iran (60 companies listed on the Tehran Stock Exchange) Tsai and Tung Taiwan (137 companies (2014) listed on the Taiwan Stock Exchange) Jafarinejad, Jory United States of America and Ngo (2015) (11 882 companies on the Compustat database) Source: compiled by authors

Findings Mutual funds, as an institutional investor, have a positive effect on company performance.


Stocks with high foreign ownership outperform stocks with low foreign ownership. This is due to the monitoring and disciplinary role played by the foreign investors. A more equal distribution of the voting power among the largest institutional stakeholder may have positive effects on a firm’s performance. Institutional investors improve the performance of companies.


Institutional ownership positively influences corporate governance by acting as a disciplinary mechanism for poorly performing CEOs, which also leads to increases in company value. Institutional shareholders are positively associated with technological innovation performance. Institutional shareholders improve firm performance based on the capital expenditure decisions made. Significant, positive relationships exist between institutional ownership and ROA, ROE, and Tobin’s Q. There is an increase in company performance when there are higher levels of institutional ownership. Higher levels of institutional shareholders increase company value.


5.4 Stability of total institutional shareholding and company performance Due to the durability of their ownership, stable institutional investors have ample opportunities to learn about the investee company and have greater incentives to monitor them effectively on an ongoing basis. Firstly, this is likely to reduce the agency costs. Secondly, long-term institutional ownership makes it possible for the directors to engage in longer-term investment, leading to improved long-term performance (Jensen and Meckling, 1976), by decreasing the possibility that directors will make unsatisfactory decisions (Navissi and Naiker, 2006: 249). Based on their connection to financial markets, stable institutional investors can also help the directors to bring about increased demand for and improvement

Supports hypothesis 3


Supports hypothesis 3. Indirectly supports hypothesis 3. Supports hypothesis 3. Supports hypothesis 3. Supports hypothesis 3.

of the liquidity of its shares. Thirdly, stable institutional owners can improve corporate governance by pressuring the directors to change the executive compensation structure to better align the interests of the directors with those of the shareholders (Elyasiani and Jia, 2010: 607). Don and Ozkan (2008) further show that long-term dedicated institutions generally restrain the level of director pay and strengthen the pay–performance link. This view is confirmed by Shleifer and Vishny (1986, 1997) and Gillian and Starks (2003) who agree that larger institutional owners give them a significant incentive to become informed, involved owners. A summary of the main findings of studies relevant to the stability of institutional ownership and company performance is presented in Table 4.

910


Table 4. Findings of studies relevant to stability of total institutional ownership and company performance Authors Chen et (2007)

al.

Elyasiani and Jia (2010)

Fung and Tsai (2012)

Callen and Fang (2013)

Country and number of countries tested United States of America (2 150 merger deals on the Mergers and Acquisitions database of Securities Data Company) United States of America (1 532 companies from the Centre for Research in Equity Prices at the University of Chicago, Thomson Financial, Compustat and ExecuComp databases) United States of America (2 249 companies listed on the New York Stock Exchange, NASDAQ and Amex Various observations from the Thompson-Reuters Institutional Holdings Database

Hsu and Wang (2014)

Taiwan (647 companies listed on the Taiwan Stock Exchange) Jafarinejad et al. United States of America (2015) (11 882 companies on the Compustat database) Source: compiled by authors

Findings Where there are concentrated holdings, the institutional investors monitor directors to ensure good directors decisions in order to make long-term gains (Chen et al. 2007: 304) Stable institutional ownership is associated with better corporate company performance.



Institutional ownership improves company performance through capital expenditure decisions.


Institutional shareholdings act as monitors and the more stable the institutional investors, the greater the chance of preventing future stock price crash risk by preventing bad new hoarding by managers. Long-term institutional shareholders are associated with higher company performance. Stable levels of institutional shareholders increase company value


Based on studies listed in Table 4, there is evidence of a positive correlation between the stability of institutional ownership and company performance. It is therefore expected that: H4: There is a positive correlation between the stability of institutional shareholding and company performance. 6 Methodology 6.1 Sample and data The sample selected was 71 of the Top 100 companies listed on the JSE for the 2009 to 2013 reporting periods, based on market capitalisation as at 30 September 2013. The Top 100 companies listed on the JSE are the largest and have the most significant trading activity. These companies are also most likely to have the most institutional shareholders. The sample included only South African companies which had been listed for at least five years and had information available on the INET McGregor BFA database for the prescribed sample period. Table 5 summarises the sample selection process.

Supports hypothesis 4. Supports hypothesis 4.

6.2 Data collection Information regarding shareholding was collected from the shareholding function on the INET McGregor BFA database. Information regarding headline earnings per share, income, leverage, return on equity, and return on capital employed was gathered from the financial ratio function on the INET McGregor BFA database. Information regarding the Tobin’s Q formula was gathered from the financial models function on the INET McGregor BFA database. Table 5. Summary of sample selection process Top 100 companies listed on the JSE on 30 September 2013 Less: 1) companies primarily listed on other exchanges (non-South African companies) 2) companies listed for less than three years (listing date after 1 January 2009) 3) companies where information not available on McGregor BFA for sample period Final sample

911

100

(14) (9)

(6) 71

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 6.3 Research model The prediction found in the hypothesis is that there is a positive correlation between the proportion international and domestic institutional ownership and the stability of international and domestic institutional ownership and company performance. More specifically, higher company performance is expected for companies with a higher percentage international, domestic and total institutional ownership and where the total intuitional ownership remained stable. Institutional ownership, both internationally and locally as well as the stability of institutional

ownership can have a monitoring effect on companies and improve company performance. Good company performance also attracts stable and long institutional investors, both locally and internationally (Hsu and Wang, 2014). An ordinary least squares (OLS) model was used to test the relationship between Tobin’s Q and ROA (dependent variables for company performance) and the international, domestic and total institutional ownership and the stability of institutional shareholding. The model was estimated as follows:

TOBIN’S Q / ROA = β0 + β1SHAREI jt + β2SHARED jt+ β3SHARET jt + β4SSHARET jt + β5SIZE jt + β6HEPS jt + β7DEBT jt + β8PROFIT jt + β9GROWTH jt +ε

(1)

Where: TOBIN’S Q = Measured as market value of equity plus book value of interest-bearing debt divided by the replacement costs of fixed assets. ROA = Return on assets, measured by profit before interest and tax less total profit of extraordinary nature divided by total assets. SHAREI = Number of shares owned by international institutions divided by the total number of ordinary shares at financial year end. SHARED = Number of shares owned by domestic institutions divided by the total number of ordinary shares at financial year end. SHARET = Total number of shares owned by institutions divided by the total number of ordinary shares of the company at financial year end. SSHARET = Stability of total institutional shareholding from one financial year end to the next financial year end. SIZE = The size of the company, measured as the natural log of sales for the year. HEPS = The headline earnings per share measured as earnings attributable to the operational and capital investment activities of the company. DEBT = The leverage of the company, measured by the ratio of total debt to total assets at financial year end. PROFIT = The profitability of the company, measured by the return on equity for the year, which is measured as net profit or loss for the year/divided by equity. GROWTH = The growth of the company, measured by the yearly proportional change in sales. j and t = Company and time subscripts respectively ε = The regression residual 6.4 Dependent variables The dependent variables used in this study are Tobin’s Q (as used by: Douma et al., 2002; Wei et al., 2005; Bhattacharya et al., 2009; Elyasiani and Jia, 2010; Alipour; 2013; Jafarinejad et al., 2015) and Return on Assets (ROA) (as used by: Douma et al., 2002; Filatotchev et al., 2005; Chen et al., 2007; Huang and Shiu, 2009; Gűrbűz et al., 2010; Fung and Tsai; 2012; Alipour, 2013). 6.5 Independent variables The following independent variables were used:  SHAREI = Percentage shareholding of international institutions (as used by Douma et al., 2002; Wei et al., 2005; Filatotchev et al., 2005; Gűrbűz et al., 2010; Mizuno, 2010; Aggarwal et al., 2011; Choi et al., 2012; Mi Choi et al., 2012)

 SHARED = Percentage shareholding of domestic institutions (as used by: Douma et al., 2002; Filatotchev et al., 2005; Navissi and Naiker, 2006; Cornett et al., 2007; Chen et al., 2007; Chen et al., 2008; Yuan et al., 2008; Huang and Shiu, 2009; Bhattacharya and Graham, 2009; Gűrbűz et al., 2010; Aggarwal et al., 2011; Tsai and Tung, 2014)  SHARET = Percentage shareholding of total institutional shareholders (as used by: Filatotchev et al., 2005; Navissi and Naiker, 2006; Cornett et al., 2007; Chen et al., 2007; Chen et al., 2008; Yuan et al., 2008, Bhattacharya and Graham, 2009; Gűrbűz et al., 2010; Aggarwal et al., 2011; Choi et al., 2012; Fung and Tsai, 2012); Alipour, 2013; Tsai and Tung, 2014; Jafarinejad et al., 2015)  SSHARET = Stability of shareholding of total institutional shareholders (As used by: Cheng et al., 2007; Elyasiani and Jia, 2010; Fung and Tsai, 2012; Callen and Fang, 2013; Jafarinejad et al., 2015).

912

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 6.6 Control variables The following control variables which could have an effect on the level of executive remuneration were used:  Company size (SIZE) (measured by log of sales) (as used by Douma et al., 2002; Wei et al., 2005; Yuan et al., 2008; Cornett et al., 2012)  Headline earnings per share (HEPS) (as used by Filatotchev et al., 2005; Chen et al., 2007)  Leverage (DEBT) (used as a control measure due to the monitoring effect of debtholders) (as used by Chen et al., 2008; Yuan et al., 2008; Bhattacharya and Graham, 2009; Elyasiani and Jia, 2010; Choi et al., 2012; Mi Choi et al., 2012; Fung and Tsai, 2012; Alipour, 2013; Callen and Fang, 2013; Jafarinejad et al., 2015)  Profitability (PROFIT) (measured by return on equity) (as used by Chen et al., 2008; Bhattacharya

and Graham, 2009; Callen and Fang, 2013; Alipour, 2013)  Growth of sales (GROWTH) (measured by yearly proportional change in sale) (as used by Gűrbűz et al., 2010; Aggarwal et al., 2011; Choi et al., 2012). 7 Empirical results 7.1 Descriptive statistics Descriptive statistics regarding the companies’ dependent, independent and control variables were considered before calculating the regressions. The data was transformed to limit the skewness. Square root transformations were performed for the headline earnings per share, profitability and debt variables. Table 6 presents the descriptive statistics for the raw data (Panel A) and the transformed variables (Panel B) for years 2009–2013.

Table 6. Descriptive statistics

Variables TOBIN’S Q ROA SHAREI SHARED SHARET SSHARET SIZE (LOG) HEPS DEBT GROWTH Variables HEPS (sqrt) PROFIT (sqrt) DEBT (sqrt)

Panel A: Descriptive statistics for raw data years 2009-2013 No. of observations Mean Minimum Maximum Median 353 1.742 0.000 11.550 1.310 353 0.1347 -0.2728 0.929 0.105 353 0.027 0.00 0.659 0.026 353 0.374 0.000 0.900 0.299 353 0.401 0.000 0.944 0.339 353 0.069 -0.953 0.936 0.031 353 16.615 8.010 19.810 16.651 353 703.996 -1880.000 13772.000 399.750 353 3.284 0.000 288.970 1.000 353 0.149 -2.3451 6.4966 0.252 Panel B: Descriptive statistics for transformed variables No. of observations Mean Minimum Maximum Median 353 22.051 -43.360 117.354 19.937 353 3.732 -20.560 11.190 3.825 353 1.311 0.000 16.998 1.000

The descriptive statistics calculated for the research variables are presented in Table 6. As can be seen in this table, 353 firm year observations were studied. The average value of Tobin’s Q is 1.7 with a median of 1.3 over the five-year sample period. This is higher than Yuan et al. (2008) who reported an average Tobin’s Q for Chinese companies of 1.4 as well as a median for Tobin Q of 1.3, Bhattacharya and Graham (2009) who reported an average Tobin’s Q of 1.2 and a median of 0.9 for Finish companies and Mi Choi et al. (2012) who reported an average Tobin’s Q of 1.2 for Korean companies. Our findings are lower than those of Chen et al. (2008) who reported an average Tobin’s Q of 2.39 for New Zealand companies, and Fung and Chai (2012) who reported an average Tobin’s Q of 2.2 and a median of 1.6 for US companies. Return on assets has an average of 13.5%. This is consistent with other reported studies. For instance,

Std. Deviation 1.337 0.157 0.075 0.209 0.236 0.221 0.075 1259.650 16.053 0.584 Std. Deviation 15.622 2.746 1.259

Douma et al. (2002) reported an average ROA of 12.69% for Indian companies and Alipour (2013) reported an average ROA of 14.4% for Iranian companies. On average, international shareholders account for less than 3% of the total number of shares in issue over the five year sample period though it can be as high as 66% (maximum) in some firms. This is consistent with other reported studies; for instance, Douma et al. (2002) reported an average international shareholding of 3.62 % and a maximum of 49% for Indian companies, and Yuan et al. (2008) reported an average of 4% international shareholding with a maximum of 70% for Chinese companies. On average domestic institutional shareholders held 37% of the total number of shares in issue over the five-year sample period, rising to 90% in some of the sample companies tested. This is consistent with reported studies; for instance, Douma et al. (2002) reported an average of 35.6% for domestic

913


institutional shareholders and a maximum of 100% for Indian companies and Choi et al. (2012) reported an average of 31.3% and a maximum of 94.7% for Korean companies. The stability of the shareholding of total institutional shareholders was on average 7% over the five-year sample period. Control variables show that the standard deviation of the size of the company is 7.5, indicating that the sample covers companies that are within the same range, which can be expected from the 71 of the Top 100 companies tested. The average headline earnings per share are 704c per share. The average leverage is 3.3, indicating high debt levels for the companies tested. The average growth rate of sales for the sample companies tested over the five-year period is nearly 15% over the fiveyear period tested, compared to the average South African GDP over this period of 1.85% (The World Bank, 2015). 7.2 Correlations Pearson’s correlations were calculated to determine if there are any significant relationships between company performance (tested as Tobin’s Q and ROA) and domestic, international and total institutional ownership and the stability of institutional ownership and certain control variables. This is reported in Table

7. Significance levels of 1% and 5% were considered material to determine the relationships between dependent and independent variables. As expected, Tobin’s Q and ROA are highly related (at the 1% level) at 65.3%. The highest correlation of any explanatory variable with a performance measure for the 71 of the top 100 South African companies is SHARED and SHARET. The highest significant relationship between Tobin’s Q and any of the independent variables is with SHAREI of 12.4% (significant at the 5% level). The highest significant positive relationship is between ROA and PROFIT (59.6%) and HEPS (28.6%) (at the 1% level) and a negative relationship with DEBT at the 1% level. As expected, there is a significant relationship between SHAREI and SHARET (24.9%) and stability of intuitional shareholding (22.1%) at the 1% level. Also as expected, there is a significant relationship between domestic institutional shareholding and total institutional shareholding (97.7%) and SSHARET (30.9%) at the 1% level. A significant relationship exists between SIZE and HEPS (at the 1% level) and GROWTH (at the 5% level). A significant relationship also exists between HEPS and PROFIT (at the 5% level).

Table 7. Pearson correlation matrix TOBIN’S Q ROA SHAREI SHARED SHARET SSHARET SIZE HEPS DEBT PROFIT GROWTH

TOBIN’S Q ROA SHAREI SHARED SHARET SSHARET SIZE HEPS DEBT 1 .653** .124* .050 .075 .061 -.151** -.124* -.097 1 .021 .051 .054 .028 -.066 .286** -.184* 1 .035 .249** .221** -.174** .053 .029 1 .977** .309* -.034 .019 -.116* 1 .347** .005 .249** -.106* 1 .098 .045 -.021 1 .165** .009 1 .081* 1

PROFIT GROWTH -373** .045 .596** .008 .005 .009 -.049 -.059 .047 -.055 .031 -.080 -.053* .106* .360* .006 -.175* -.045 1 .089 1

Note: ** Significant at the 0.01 level, * Significant at the 0.05 level 7.3 Regression results The findings of the level and stability of institutional ownership’s influence on company performance of the 71 of the Top 100 companies listed in South Africa are discussed here. Table 8 shows the results of the OLS regression performed. The main interest in Table 8 is the sign and the significance of the independent and control variables tested. The adjusted square multiple regression was significantly different from zero for Tobin’s Q (F=20.258, p>001) and 33.2% of the variation of Tobin’s Q was explained by the independent and control variables. The adjusted square multiple regression was significantly different from zero for ROA (F=31.511, p>001 and 37.9% of the variation of ROA was explained by the independent and control variables.

The results show that SHAREI uniquely and significantly contributes to the prediction of Tobin’s Q at the 1% level. This demonstrates that for the South African companies tested, higher international institutional shareholding significantly contributes to the prediction of future opportunities in a company (measured as Tobin’s Q). Douma et al. (2002) (for Indian companies), Wei et al. (2005) (for Chinese companies), Huang and Shiu (2009) (for Taiwanese companies) and Mi Choi et al. (2012) (for Korean companies) found that foreign intuitional ownership also significantly contributes to the prediction of Tobin’s Q. These countries, including South Africa are considered to be part of the emerging markets, according to a study performed by Bloomberg (2014). Hypothesis 1 is therefore supported for Tobin’s Q

914


with regards to international institutional ownership for the 71 companies out of the top 100 companies tested in South Africa. SHARED, SHARET and SSHARET uniquely and significantly contribute to the prediction of ROA at the 5% level for 71 of the top 100 companies tested in South Africa. This indicates that domestic, total and stability of institutional shareholders has an influence on the historical value of the 71 of the top 100 South African companies tested. This is consistent with the findings of Douma et al. (2002) (for Indian companies), Gűrbűz et al. (2010) (for Turkish companies), Elyasiani and Jia (2010) (for US companies), Fung and Tsai (2012) (also for US companies) and Tsai and Tung (2014) (for Taiwanese companies). All these, (except for the US which is a developed country) are considered emerging markets (Bloomberg, 2014). Hypotheses 2, 3 and 4 are therefore supported for return on assets (ROA) for the 71 of the top 100 companies tested in South Africa. SIZE has a statistical significant relationship at the 1 % level with Tobin’s Q for the 71 of the top 100 companies tested in South Africa. This is consistent with the prediction and the findings of Yuan et al.

(2008) for Chinese companies, also an emerging market (Bloomberg, 2014). The regression results further indicate that HEPS is significantly related to ROA on the 5% level for the 71 of the top 100 South African companies tested. This is consistent with Filatotchev et al. (2005) (for Taiwanese companies), Chen et al. (2007) (for US companies) and Choi et al. (2012) (for Korean companies). Both Korea and Taiwan are like South Africa, considered to be emerging markets (Bloomberg, 2014). PROFIT significantly statistically contributes to Tobin’s Q and ROA on the 1% level for the 71 of the top 100 South African companies tested. This is consistent with what Chen et al. (2008) found for New Zealand companies. No relationship exists between GROWTH, DEBT and Tobin’s Q and ROA for the 71 of the top 100 South African companies tested. This is not what was predicted but is consistent with what Gűrbűz et al. (2010) (Turkish companies) and Aggarwal et al. (2011 (US companies) found. The data satisfied the assumptions of multicollinearity, normality of residuals and homoscedasticy. Stationarity is not a problem as ratios were used.

Table 8. Regression results Variable

Prediction

TOBIN’S Q Coefficients p-value 3.200 .009** -.147 .595 .100 .827 -.385 .664 .132 .005** -.100 .800 .001 .744 -018 .001** .110 .331 .349

SHAREI + SHARED + SHARET + SSHARET + SIZE + HEPS¹ (sqrt) + DEBT¹(sqrt) PROFIT¹(sqrt) + GROWTH + R-SQUARED ADJUSTED .332 R-SQUARED N 354 F 20.528 Note: ** Significant at the 0.01 level, * Significant at the 0.05 level

ROA Coefficients .240 1.879 1.361 -.520 271 -.001 .065 1.915 -.100

p-value .520 .058* .065* .034* .916 .023* .905 .000** .682 .391 .379 354 31.811

7.4 Additional analysis

8 Summary and conclusion

To test the robustness of results, an additional analysis was performed. The regression was re-tested by excluding the control variables of sales, headline earnings per share, leverage, return on equity and growth of sales. The results obtained were consistent with the original regression performed. Since no differences were found between the original regression model and the one excluding the control variables, only the original regression analysis is included in Table 8.

This study investigates the relationship between institutional share ownership (both international and domestic) and the stability of institutional share ownership and firm performance by using data for 71 of the top 100 South African listed companies. The contribution of this study includes the stability of institutional shareholding, while previous studies focused on the proportion of institutional shareholding. This study proves that both the proportion and the stability of institutional shareholding are important to ensure the monitoring effectiveness of institutional shareholders.

915


Furthermore, international institutional shareholding has an effect on future firm performance as measured by Tobin’s Q; thus hypothesis 1 is supported for the 71 of the top 100 companies tested in South Africa. Domestic institutional shareholding, total institutional shareholding and the stability of shareholding of total institutional shareholders have an effect on historical firm value measured as return on assets; thus hypotheses 2, 3 and 4 are supported for the 71 of the top 100 companies tested. Institutional shareholders who holds shares for a longer period of time have the advantage to get to know the company and incentives to act as an important corporate governess monitoring mechanism. The findings of the study emphasis the fact that directors of a company need to build positive relationships with institutional investors to boost firm performance. Future research may employ alternative firm performance measures such as return on equity, headline earnings per share, market value added or shareholding returns to test the effect of institutional shareholding on these factors. Institutional shareholdings could be further split into different institutions such as pension funds, mutual funds, insurance companies and banks.

9.

10.

11.

12.

13.

14.

References 1.

2.

3.

4.

5.

6.

7. 8.

Aggarwal, R., Erel, I., Ferreira, M., and Matos, P. (2011), “Does governance travel around the world? Evidence from institutional investors”, Journal of Financial Economics, Vol. 100, No. 1, pp. 154-181. http://www.sciencedirect.com/science/article/pii/S030 4405X10002540. Alipour, M. (2013), “An investigation of the association between ownership structure and corporate performance: Empirical evidence from Tehran Stock Exchange (TSE)”, Directors Research Review Vol. 36 No. 11, pp. 1137-1166. [Online] Available at: http://search.proquest.com.ez.sun.ac.za/docview/1447 819093/fulltextPDF/818CD53CDC894F55PQ/1?acco untid=14049. Accessed: 4 June 2015. Arya, B. and Gaiyan, Z. (2009), “Institutional reforms and investor reactions to CSR announcements: evidence from an emerging economy”, Journal of Management Studies, Vol. 46, No. 7, pp. 1089-1112. Barber, B.M., Lee, Y.T., Y.-J. Liu, Y.J. and Odean, T. (2009), “Just How Much Do Individual Investors Lose by Trading?”, Review of Financial Studies, Vol. 22, No. 2, pp.609–632. Barclay, M. and Holderness, C. (1990), “Negotiated block trades and corporate control”, Working Paper. University of Rochester, Rochester, NY. Berle, A. and Means, G. (1932), The modern corporation and private property. New York: Macmillan. Bhasin, M. L. (2013), Corporate accounting fraud: A case study of Satyam Computers Limited. Bhattacharya, P. S. and Graham, M. A. (2009), “On institutional ownership and firm performance: a disaggregated view”, Journal of Multinational Financial Management, Vol. 19, No. 5, pp. 370-394. [Online] Available at: http://www.sciencedirect.com/

15.

16.

17.

18.

19.

20.

21.

916

science/article/pii/S1042444X09000243. Accessed: 4 June 2015. Bloomberg, (2014), “List of World’s best emerging markets 2014” [Online] Available at:http://www.bloomberg.com/visual-data/best-andworst//best-emerging-markets-2014-countries. Accessed: 3 November 2015. Bushee, B.J. (1998), “The Influence of Institutional Investors on Myopic R&D Investment Behavior”, Accounting Review, Vol. 73, No. 3, pp. 305–333. Bushee, B.J. and Noe, C.F. (2000), “Corporate Disclosure Practices, Institutional Investors, and Stock Return Volatility”, Journal of Accounting Research, Vol. 38, pp.171–202. Callen, J. L. and Fang, X. (2013), “Institutional investor stability and crash risk: Monitoring versus short-termism?”, Journal of Banking & Finance, Vol. 37, No. 8, pp. 3047-3063. [Online] Available at: http://www.sciencedirect.com/science/article/pii/S037 8426613001064. Accessed: 4 June 2015. Chen, J., Blenman, L., and Chen, D. H. (2008), “Does institutional ownership create values? The New Zealand case”, Quarterly Journal of Finance and accounting, pp. 109-124. [Online] Available at: http://www.jstor.org/stable/40473470?seq=1#page_sc an_tab_contents. Accessed: 4 June 2015. Chen, X., Harford, J. and Li, K. (2007), “Monitoring: Which institutions matter?”, Journal of Financial Economics, Vol. 86, No. 2, 279-305. [Online] Available at: http://ac.els-cdn.com.ez.sun.ac.za/ S0304405X07001171/1-s2.0-S0304405X07001171main.pdf?_tid=349a49e4-0f4e-11e5-87db00000aacb35d&acdnat=1433926607_a19bec280cbafb f93edad0c8e591a543. Accessed: 4 June 2015. Choi, S. B., Park, B. I., and Hong, P. (2012), “Does ownership structure matter for firm technological innovation performance?”, The case of Korean firms.’ Corporate Governance: An International Review, Vol. 20, No. 3, pp. 267-288. [Online] Available at: http://onlinelibrary.wiley.com/doi/10.1111/j.14678683.2012.00911.x/full. Accessed: 6 June 2015. Claessens, S. and Fan, J. P. (2002), “Corporate governance in Asia: A survey”, International Review of finance, Vol. 3, No. 2, pp. 71-103. Cornett, M. M., Marcus, A. J., Saunders, A., and Tehranian, H. (2007), “The impact of institutional ownership on corporate operating performance”, Journal of Banking & Finance, Vol. 31, No. 6, pp. 1771-1794. [Online] Available at: http://www.sciencedirect.com/science/article/pii/S037 8426607000283. Accessed: 3 March 2015. Chung, R., Firth, M., and Kim, J. (2002), “Institutional monitoring and opportunistic earnings management”, Journal of Corporate Finance Vol. 8, pp. 29–48. [Online] Available at: http://www.sciencedirect.com/ science/article/pii/S0929119901000396. Accessed: 17 September 2015. Demsetz, H., & Villalonga, B. 2001. ‘Ownership structure and corporate performance.’ Journal of corporate finance, 7(3), 209-233. Don, M. and Ozkan, A. (2008), “Institutional investors and director’s pay: an empirical study of UK companies.”, Journal of Multinational Financial Management, Vol. 18, pp. 16–29. Douma, S., George, R. and Kabir, R. (2003), “Foreign and domestic ownership, business groups and company performance: Evidence from a large


22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

emerging market”, Business Groups and Company Performance: Evidence from a Large Emerging Market. Elyasiani, E. and Jia, J. (2010), “Distribution of Institutional Ownership and Corporate Firm Performance”, Journal of Banking and Finance, Vol. 34, No. 3, pp. 606–620. Elyasiani, E., Jia, J and Mao, C.X. (2010), “Institutional Ownership Stability and the Cost of Debt”, Journal of Financial Markets, Vol. 13, No. 4, pp. 475–500. Ferreira, M.A. and Matos, P. (2008), “The role of institutional investors around the world”, Journal of Financial Economics Vol. 88, pp. 499–533. Filatotchev, I., Lien, Y. C. and Piesse, J. (2005), “Corporate governance and performance in publicly listed, family-controlled firms: Evidence from Taiwan”, .Asia Pacific Journal of Management, Vol. 22, No. 3, pp. 257-283. [Online] Available at: http://link.springer.com/article/10.1007/s10490-0053569-2#page-1. Accessed: 8 August 2015. Fung, S. and Tsai, S.C. (2012), “Institutional ownership and corporate investment performance”, Canadian Journal of Administrative Sciences, Vol. 29, No. 4, pp. 348-365. Gillan, S. and Starks, L. (2000), “Corporate Governance Proposals and Shareholder Activism: The Role of Institutional Investors”, Journal of Financial Economics, pp. 275-305. Gürbüz, A. O., Aybars, A., and Kutlu, Ö. (2010), “Corporate governance and financial performance with a perspective on institutional ownership: empirical evidence from Turkey”, Journal of Applied Management Accounting Research, Vol. 8, No. 2, pp. 21. [Online] Available from: http://search.proquest.com/docview/821699251/fulltex tPDF?accountid=14049. Accessed: 15 September 2015. Hartzell, J.C. and Starks, L.T. (2003), “Institutional investors and executive compensation”, Journal of Finance Vol. 58, No. 6, pp. 2351–2375. Hsu, M. F. and Wang, K. (2014), “The level and stability of institutional ownership and firm performance: Evidence from Taiwan”, Emerging Markets Finance and Trade, Vol. 50, No. 2, pp. 159173. [Online] Available at: http://www.tandfonline.com/doi/abs/10.2753/REE154 0-496X5002S211#.Vfpp19WqpBc. Accessed: 17 March 2015. Huang, R.D. and Shiu, C.Y. (2009), “Local Effects of Foreign Ownership in an Emerging Financial Market: Evidence from Qualified Foreign Institutional Investors in Taiwan”, Financial Management Vol. 38, No. 3, pp. 567–602. Jackson, G. (2010), “Understanding Corporate Governance in the United States’. Corporate Governance, Vol. 90. [Online] Available at: http://boeckler.de/5137.htm?produkt=HBS004875&chunk=3&jahr. Accessed: 15 July 2015. Jafarinejad, M., Jory, S. R., and Ngo, T. N. (2015), “The Effects of Institutional Ownership on the Value and Risk of Diversified Firms”, International Review of Financial Analysis. [Online] Available at: http://www.sciencedirect.com/science/article/pii/S105 7521915001015. Accessed: 29 June 2015.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46. 47.

48.

917

Jensen, M. and Meckling, W. (1976), “Theory of the firm: Managerial behaviour, agency costs, and ownership structure”, Journal of Financial Economics Vol. 3, pp. 305–360. Malherbe, S., and Segal, N. (2001), “Corporate Governance in South Africa”, Development Centre Discussion Paper. Mi Choi, H., Sul, W. and Kee Min, S. (2011), “Foreign board membership and company value in Korea”, Management Decision, Vol. 50, No. 2, pp. 207-233. [Online] Available at: http://www.emeraldinsight.com/doi/full/10.1108/0025 1741211203533. Accessed: 14 September 2015. Mizuno, M. (2010), “Institutional investors, corporate governance and firm performance in Japan”, Pacific Economic Review, Vol. 15, No. 5, pp. 653-665. [Online] Available at: http://onlinelibrary.wiley.com/ doi/10.1111/j.1468-0106.2010.00521.x/pdf. Accessed: 17 September 2015. Morck, R. K. (Ed.). (2007), “A history of corporate governance around the world: family business groups to professional managers”, University of Chicago Press. Navissi, F., and Naiker, V. (2006), “Institutional ownership and corporate value.’ Managerial Finance”, Vol. 32, No. 3, pp. 247-256. [Online] Available at: http://www.emeraldinsight.com/doi/ full/10.1108/03074350610646753. Accessed: 29 June 2015. Perrini, F., Rossi, G. and Rovetta, B. (2008), “Does ownership structure affect performance? Evidence from the Italian market”, Corporate Governance: An International Review, Vol, 16, No. 4, pp. 312-325. Pound, J. (1988), “Proxy contests and the efficiency of shareholder oversight”, Journal of Financial Economics Vol. 20, pp. 237-65. Shleifer, A. and Vishny, R.W. (1986), “Large shareholders and corporate control” The Journal of Political Economy, Vol. 94, Vol. 3, pp. 461. Shleifer, A. and Vishny, R.W. (1997), “A survey of corporate governance”, Journal of Finance Vol. 52, pp. 737–782. The World Bank. (2015), [Online] Available at: http://data.worldbank.org/indicator/NY.GDP.MKTP.K D.ZG. Accessed: 25 September 2015. Tsai, M. T. and Tung, W. H. (2014), “Corporate governance, resources, FDI commitment and firm performance: Empirical analyses of Taiwanese hightech firms”, Chinese Management Studies, Vol. 8, No. 3, pp. 313-332. [Online] Available at: http://www.emeraldinsight.com/doi/abs/10.1108/CMS -08-2012-0118. Accessed: 1 September 2015. UK Office for National Statistics. (2013), Statistical Bulletin Ownership of UK Quoted Shares. Wei, Z., Xie, F. and Zhang, S. (2005), “Ownership structure and firm value in China's privatized firms: 1991–2001”, Journal of financial and quantitative analysis, Vol. 40, No. 1, pp. 87-108. [Online] Available at:http://journals.cambridge.org/action/ displayAbstract?fromPage=online&aid=4200740&file Id=S0022109000001757. Accessed: 31 March 2015. Yuan, R., Xiao, J. Z. and Zou, H. (2008), “Mutual funds’ ownership and company performance: Evidence from China”, Journal of Banking & Finance, Vol. 32, No. 8, pp. 1552-1565.


IMPACT OF STRUCTURE ON ORGANISATIONAL PERFORMANCE OF SELECETED TECHNICAL AND SERVICE FIRMS IN NIGERIA Ann I. Ogbo*, Nwankwere F. Chibueze*, Orga C. Christopher**, Igwe A. Anthony* Abstract The study aimed at establishing the impact of structure on organizational performance. Organizations today are becoming more automated and complex, hence, the need to maintain and improve performance by structuring and restructuring based on changing strategy. The study was conducted using the survey approach. The geographical scope of study was Innoson Nigeria Ltd, and Etisalat, Enugu Regional Office. Two sources of data were utilised in the study: they included primary and secondary sources. The primary source was the administration of copies of designed questionnaire to a total of eighty (80) respondents that made up the sample for the study. Out of the eighty (80) copies administered, seventy eight (78) were completed and returned. Simple percentage (%), chi-square (*a) and correlation were used in analysis of the data and in testing the three hypotheses. Findings revealed that decentralization enhanced better and more informed decision making in technical and service firms in Nigeria; that task routine affected staff productivity both positively and negatively; and that a significant positive relationship existed between narrow span of control and efficiency in organizations. The study concluded and recommended among others that managers of organizations should adopt more decentralized forms of structures as means of improving the decision making process; that managers should combine both task routine and variety in organizing employees for carrying out task in order to reap the advantages of both systems of task assignment; and that employees should be empowered to be more innovative in carrying out tasks, whether routine or not. Keywords: Structure, Restructuring, Decentralization, Task Routine

Strategy,

Automated,

Organisational

Performance,

*Department of Management, University of Nigeria, Enugu Campus, Nigeria **Department of Business Administration, Enugu State University of Science and Technology, Enugu, Nigeria

1 Introduction The business environment today is so dynamic that the decision to structure and re-structure has become paramount. Stephen and Timothy (2012) posit that structural decisions like the reconfiguration of any organization are arguably the most fundamental ones a leader has to make. This according to them is because organizational structure defines how job tasks are formally divided, grouped, and coordinated. Similarly, Nelson and Quick (2011) posit that the organization's structure gives it the form to fulfill its functions in the environment. Acknowledging the views of these authors on the indispensability of structural decisions and the ongoing debate on the interrelationships between strategy, structure and performance, one would want to agree with Joris, Brand, Marco and Zoetermeer (2002) that the outcome of the organizational design process is unmistakably an important determinant of the performance of firms. Historically, organizational structure developed from the ancient times of hunters to industrial

structures and today's post-industrial structures as pointed out by Lawrence (1982). Away from history to today's world of business, one reoccurring and widely asked question is; how does the structure of an organization affect its performance? The difficulty in answering this question hinges on the fact that the relationship between organizational structure and performance has received little attention over the past few years, especially in regards to firms with less than 100 employees. McShane and Glinow (2005) however, answer this question partly by positing that structure includes two fundamental elements: the division of labor into distinct tasks and its coordination so that employees are able to accomplish common goals. Child (2005) posits that the purpose of structure is to contribute to the successful implementation of objectives by allocating people and resources to necessary tasks and design responsibility and authority for their control and coordination. The foregoing assertion underscores the position that the structure of an organization affects not only productivity and economic efficiency but also the morale and job

918


satisfaction of the work force (Ezigbo, 2011). Additionally, Wolf (2002) believes that structure does not only shape the competence of the organization, but also the processes that shape performance. Clemmer (2003) also concludes that the performance of an organization is influenced by the structure adopted by that organization. This paper therefore will re-open the discussion on the relationship between structure and performance while placing particular emphasis on decentralization, task routine and span of control as they influence effective decision making, staff productivity and organizational efficiency respectively. 1.1 Statement of the problem In recent times, business organizations in an attempt to adopt the best type of structure with the aim of attaining maximum performance have faced a lot of problems. Also many organizational flaws can be related to an inappropriate structure chosen in order to reach a desired goal. An appropriate structure is contingent upon both the type of work to be performed as well as the environment in which the organization conducts business (Bolman and Deal, 1997). Different structures provide different strengths and weaknesses to the work to be performed and it is therefore important to find a structure suitable for the desired outcome on stability and predictability (Mintzberg, 1983). These problems encountered by business organizations in choosing suitable forms of structure are complexities associated with the recent shift from authoritarian to decentralized structures stressing employee empowerment, inability of managers to identify the best form of structure that should follow strategies adopted by their individual organizations, inability of employees to adapt to existing and changing structures, and the difficulty in maintaining a stable structure as a result of the ever changing business environment. 1.2 Objectives of the study The broad objective of this study is to examine the impact of structure on organizational performance. The key specific objectives include: 1) To establish whether decentralization improves effective decision making. 2) To determine the extent to which task routine affects staff productivity. 3) To ascertain the relationship between narrow span of control and organizational efficiency. 1.3 Research questions To effectively achieve the above objectives, the following questions were asked: 1) Does decentralization improve effective decision making?

2) To what extent does task routine affect staff productivity? 3) Does any significant relationship exist between narrow span of control and organizational efficiency? 1.4 Research hypotheses The followings hypotheses guided the study: H1: Decentralization does improve effective decision making. H2: Task routine does affect staff productivity. H3: There exists a significant relationship between narrow span of control and organizational efficiency. 2 Literature review framework of the study 2.1 The structure

concept

of

and

conceptual

organizational

The structure of an organization can be defined simply as the total of the ways in which its labor is divided into distinct tasks and then its coordination and integration is achieved among those tasks (Bernd Venohr 2007). It is the map of relationships that lets the firm orchestrate specialized experts (Thompson, 1967), and provides the basic foundation within which an organization functions (Mohammed and Saleh, 2013). Organizational structure institutionalizes how people interact with each other, how communication flows, and how power relationships are defined (Hall, 1987). It reflects the value-based choices made by the company (Quinn, 1988); it refers to how job tasks are formally divided, grouped, and coordinated and can provide the link between social and psychological subsystems (Rezayian, 2007). March and Simon (1958) expressed a more behavioural view by defining organizational structure as the „pattern of relationship and behaviours that change slowly and thus provide clarity and stability. Similarly, Ranson (1980) posits that structure is a complex medium of control, the framework of rules, roles, and authority relations that seeks to facilitate prescribed purposes by differentially enabling certain kinds of conduct, conferring support for forms of commitment and obligating those who reject the claims entailed by the framework. It is the formal system of task and reporting relationships that controls, coordinates, and motivates employees so that they cooperate to achieve on organization's goals (Underdown, 2003). Weber (1947) theorized that managers designed a structure to control the firm's activities by specifying the vertical hierarchy, formal procedures and division of labour. More precisely, structure delineates the verticality of the chain of command, breath of communication, extent of dyadic relationships and

919


relative prominence of functional, product and/or market responsibility (Rosalie, 1999). For the purpose of this paper, a working definition of structure is given as simply the way in which functions or tasks are grouped and assigned, and the manners in which relationships are coordinated between superiors and subordinates within any organization. 2.2 Dimensions structure

of

address six key elements when they design their organization's structure: work specialization, departmentalization, chain of command, span of control, centralization and decentralization, and formalization (Daft, 2010). For the sake of this paper, most of the dimensions mentioned will be discussed but emphasis will be placed on only the ones relating to the objectives of the study.

organizational 2.3 The concept performance

As a result of the position of Burns and Stalker (1961) who introduced a popular method of examining the potential dimensions of organizational structure: “mechanistic and organic” systems of organization, a large number of researchers have explored and produced different lists on the dimensions of structure. Notable among them are Pugh, Hickson, Hinings, and Turner (1968) who defined five dimensions of organizational structure: specialization, standardization, formalization, centralization, and configuration. Jackson and Morgan (1982) added a sixth dimension, traditionalism. Duncan (1971) proposed five primary features of organic structure: participation in work decisions, formalization, hierarchy of authority, impersonality, and division of labor. More recently, Daft (2003) provided a list that includes formalization, specialization, standardization, hierarchy of authority, complexity, centralization, professionalism, and personnel ratios. Additionally, the structural dimension is a tool for coordination and integration (Bernd, 2007), and managers need to

of

organizational

In broad terms, a firm's performance is determined by the success of selling products and services in the market, and, by the effectiveness of organizing and transforming inputs (such as labour and capital) into sellable products and services (Nickell, Nicolitsas, and Dryden 1997). In more specific terms, organizational performance is the ability of an organization to utilize its resources efficiently and to generate outputs that are consistent with its goals and objectives and relevant for its clients and stakeholders (Ezigbo, 2011). 2.4 Measures performance

of

organizational

Performance measures are also referred to as dimensions of performance. Several financial and non-financial performance indicators have being presented. In the table below, a summary of the most popular ones are presented.

Table 1. Performance dimensions and indicators selected Profitability

Employee satisfaction

Productivity

Net income, Turnover rate, Output level, Return on retirement plan, Product development, investment, employee training and Output per labour hour Return on equity development Source: Researchers 2.5 Decentralization and effective decision making Decentralization refers to the degree to which decision making is allowed for lower-level managers. In a decentralized organization, decision making is pushed down to the managers closest to the action. It is the term for pushing decision authority downward to lower level employees (Sablynskis, 2003) and is based on the principle of subsidiarity (Holtmann 2000). A decentralized organization can act more quickly to solve problems, more people provide input into decisions, and employees are less likely to feel alienated from those who make decisions that affect their work lives (Stephen and Timothy, 2012).

Efficiency Input-output ratio, Resource utilization, Waste minimization, Cost minimization

Effective decision making Policy implementation, Effective leadership

Similar to the views of Stephen and Timothy, research investigating a large number of Finnish organizations demonstrates that companies with decentralized research and development offices in multiple locations were better at producing innovation than companies that centralized all research and development in a single office (Leiponen and Helfat, 2001). This is due to the fact that employees in all organizations want to work in an environment of trust and respect where they feel they are making a real contribution to organizational goals and objectives (Anderson and Pulich, 2000).

920

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 2.6 Task routine and staff productivity Sustaining operational productivity in the completion of repetitive tasks is key to many organizations‟ success (Bradley and Francesca, 2011). Managers have to identify the best way to assign task over a long period of time. Task routine has both positive and negative impacts on staff productivity. For example, in recently reviewing the impact of specialization and variety on productivity, Bradley and colleague (2011) writes that when a worker completes many tasks during a day, specialization helps the worker quickly complete the focal task (Newell and Rosenbloom, 1981; Argote, 1999) and limits costly changeovers (Cellier and Eyrolle, 1992; Schultz, McClain and Thomas, 2003). They went futher an pointed out that; although limiting variety during a day may lead to improved performance the opposite may be true over many days. In this paper, we emphasize the need for managers to combine elements of task routine and variety in task assignment. This way, the benefits of both are obtained. 2.7 Narrow span of organizational efficiency

control

and

An organization characterized by narrow span of control has its managers at each level controlling few subordinates. Gittell (2001) posits that by keeping the span of control to five or six employees, a manager can maintain close control of employees. Also Hendericks (2001) commenting on the impact of narrowing span of control, writes that a reduction of span of control from 1:18 to 1:6 was found to increase productivity and profit in the company. However, Robbins and Timothy (2012) believe that all things being equal, the wider or larger the span, the more efficient the organization. They went further by pointing out that narrow spans have three major drawbacks. First, they are expensive because they add levels of management. Second, they make vertical communication in the organization more complex. The added levels of hierarchy slow down decision making and tend to isolate upper management. Third, narrow spans encourage overly tight supervision and discourage employee autonomy.

of control and the performance of organizations. They propose a multi-dimensional model in which initial increases in span of control produce increases in organizational performance, though at a decreasing rate of return. Increasing spans of control, according to Meier and Bohte, allow for greater specialization, enhancing efficiency and performance. To them, a higher manager/employee ratio reduces the manger's ability to control, communicate and coordinate leading to a decrease in performance at an increasing rate. They also propose that different spans of control can exist within one organization depending on the goal being pursued in each department of the organization. 2.8.2 Hatch’s Model of decentralization Prior to Hatch's model of decentralization, Andrew, Christiea, Marc, and Ross (2003) proposed that the theory underlying the decentralization decision is very simple. Value is increased by minimizing the total of knowledge transfer costs. Minimizing this total cost requires allocating some decision rights from the CEO's office to lower level managers of the firm. Hatch (2006) proposes that the decentralized structure allows for innovation and is thus more suitable and beneficial when used in a changing environment with high requirement on adapting to the environment. He points out that the decentralized structure is characterized by interactions that allow for redefinition of tasks and work methods. A decentralized structure uses formalization to a smaller extent than a mechanic structure, and uses horizontal communication and consulting between departments rather than vertical instructions. In such structure Hatch explains that employees rather seek advice from each other than give instructions. 2.8.3 Hierarchy-community phenotype model of organizational structure

In this section, it is intended to outline a few influential theories related to the subject matter and thus provide a background for a better understanding of the mechanism through which structure affects performance in technical and service firms.

This model was de developed by Lim, Griffiths, and Sambrook (2010). They proposed that organizational structure development is very much dependent on the expression of the strategies and behavior of the management and the workers as constrained by the power distribution between them, and influenced by their environment and the outcome. This goes to show that the extent of decentralization, task routine and specialization, and the size of span of control in an organization is dependent on the organizational strategy and behaviours of managers and subordinate within that organization. In more specific terms, the structure of an organization is dependent on and reflective of its most dominant internal and external characteristics.

2.8.1 Meier and Bohte’s model of span of control

2.8.4 Chester Barnard’s model on structure

Meier and Bohte (2000) offer the general theory on the functional form of relationship between the span

Chester Barnard was a practitioner who had read Weber's papers and was influenced by his writings.

2.8 Theoritical framework

921


But unlike Weber, who had a mechanistic and impersonal view of structure, Barnard saw organizations as social systems that requires human cooperation. For him, organizations should be structured in such a way that people will be allowed to be closer and freely communicating, and the organizations should be more flexible in adjusting to environmental changes to maintain a state of equilibrium (Mohammed, 2009). This model stresses decentralization, narrow span of control, and employee empowerment.

included primary and secondary sources. The primary source was the administration of copies of designed questionnaire to a total of eighty (80) staff that made up the sample for the study. Out of the eighty (80) copies administered, seventy eight (78) were completed and returned. Simple percentage (%), chisquare (x2) and correlation were used in analysis of the data and in testing the three hypothesis. 4 Presentation of data

3 Methodology The study was conducted using the survey approach. Two sources of data were utilized in the study: they Table 2. Responses on whether decentralization improves effective decision making

Strongly agree Agree Strongly disagree Disagree Undecided Total Source: Survey Data, 2015

Frequency 51 15 1 6 5 78

The table above shows that 65.38% of the respondents strongly agreed that decentralization

Percent (%) 65.38 19.23 1.28 7.69 6.41 100.0

improves effective decision making while only 1.28% of the respondents maintained otherwise.

Table 3. Responses on whether task routine affects staff productivity

Strongly agree Agree Strongly disagree Disagree Undecided Total Source: Survey Data, 2015

Frequency 24 30 2 8 14 78

The table above shows that 30.77% and 38.46% of the respondents strongly agreed and agreed respectively that task routine affects staff productivity

Percent (%) 30.77 38.46 2.56 10.26 17.95 100.0

while only 2.56% strongly disagreed that task routine affects staff productivity.

Table 4. Responses on whether a relationship exists between narrow span of control and efficiency

Strongly agree Agree Strongly disagree Disagree Undecided Total

Frequency 23 31 10 4 10 78

The table above shows that 29.49% and 39.74% strongly agreed and agreed respectively that there is a relationship between narrow span of control and efficiency while only 12.82 maintained otherwise.

922

Percent (%) 29.49 39.74 12.82 5.13 12.82 100.0

Corporate Ownership & Control / Volume 13, Issue 1, 2015, Continued - 10 4.1 Test of hypothesis

4.1.2 Hypothesis 2 Table 6. Test statistics (hypothesis 2)

4.1.1 Hypothesis 1 Table 5. Test statistics (hypothesis 1) Chi-Square 161.227a Df 4 Asymp. Sig. .000 Note: a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 46.6. The hypothesis that decentralization improves effective decision making was tested using the chisquare (x2) test statistic. At 5 percent level of significance, the null hypothesis was rejected, and it was therefore concluded that decentralization improves effective decision making. The conclusion is based on the fact that the critical chi-square value of 9.49 was lower than the calculated chi-square value of 161.227 at alpha level of 5 percent and at 4 degrees of freedom.

Chi-Square 33.026b Df 4 Asymp. Sig. .000 Note: b. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 15.6. The hypothesis that task routine affects staff productivity was also tested using the chi-square (x2) test statistic. At 5 percent level of significance, the null hypothesis was rejected, and it was therefore concluded that task routine affects staff productivity. The conclusion is based on the fact that the critical chi-square value of 9.49 was lower than the calculated chi-square value of 33.026 at alpha level of 5 percent and at 4 degrees of freedom. 4.1.3 Hypothesis 3

Table 7. Correlations Narrow span of control Pearson Correlation 1 Narrow span of control Sig. (2-tailed) N 156 Pearson Correlation .973** Efficiency Sig. (2-tailed) .000 N 156 Note: **. Correlation is significant at the 0.01 level (2-tailed). The hypothesis that there is a significant relationship between narrow span of control and efficiency was tested using correlation (r). The tested hypothesis gives us a very strong positive correlation coefficient of 0.973. The Sig. (2-tailed) gives a “P” value of 0.000. Since P