Development and Underdevelopment under

Macroeconomics of Poverty:

Development and Underdevelopment under Globalization

David Mayer-Foulkes

Copyright © 2017 David Mayer-Foulkes All rights reserved. ISBN: 1975751744 ISBN-13: 978-1975751746

DEDICATION For a democratic Economics

CONTENTS Acknowledgments

i

1

Development and Underdevelopment under Globalization: A Historical Overview

1

2

Technological Change, Convergence and Divergence

38

3

Human Development

76

4

Trade, Economic Growth, and Divergence

99

5

FDI, Current Globalization, and the Single Global Market Economy

129

6

The Mass Market Economy

165

7

The Present Economic Scenario

199

References

204

ACKNOWLEDGMENTS To CIDE, for the unencumbered freedom of research. To Oded Galor, Peter Howitt and Phillipe Aghion, for their working welcome when I was on sabbatical at Brown University. To William Nyogen Yeo, for endless conversations on human realities, and his emphasis on the extents of greed and suffering. To Cesar Enrique Uribe Ortiz, for his invaluable help in editing and designing the book for print.

i

1 DEVELOPMENT AND UNDERDEVELOPMENT UNDER GLOBALIZATION: A HISTORICAL OVERVIEW

Our macroeconomic perspective on poverty begins by considering economic development over the last three centuries. This has occurred as a global process. While in many ways countries form the unit of macroeconomic analysis, in fact the histories and economies of countries are sufficiently interrelated for the economic process to take place as a global phenomenon. This chapter provides a descriptive overview of the current process of economic growth and its history. This process of economic development also generated persistent poverty which shall be modeled in terms of multiple equilibria models in the following chapters.

I.

Cross-country economic growth, 1960-2010

Cross country data for a large number of countries is not available much deeper in the past than 1960, also the approximate date of independence of many African and Asian colonies. A first qualitative approximation to the characteristics of economic growth and development is found in Figures 1.1a and 1.1b, which show the average trajectory of the logarithm of income per capita for 100 countries, and the average trajectory of life expectancy for 184 countries, both from 1960 to 2010.

1

DAVID MAYER-FOULKES

Figure 1.1ª Average trajectory of the logarithm of Income per capita for 100

Figure 1.1a. Average trajectory of the logarithm of income per capita for 100 countries, 1960‐2010.

12

11

Log income per capita

10

9

8

7

6

5 1960

2010

Source: Author’s elaboration with data from World Development Indicators

Figure 1.1b. Average trajectory of life expectancy for 184 countries, 1960-2010

Figure 1.1b. Average trajectory of life expectancy for 184 countries, 1960‐2010 85

75

Life Expectancy

65

55

45

35

25 1960

2010

Source: Author’s elaboration with data from World Development Indicators 2

DEVELOPMENT AND UNDERDEVELOPMENT UNDER GLOBALIZATION

Figure 1.1a is like a window into the process of economic growth that provides a view into a Income per capita and centuries-long process displaying wide differences life expectancy are in income and growth rates. Many countries are both important following an approximately stratified, parallel indicators of pattern of growth. There are also quite a few economic countries that have grown much faster, in what is development. The often referred to as miracle growth. Other, logarithm of especially low income countries, have experienced income per capita deep plunges in their economic performance, provides an usually because of the impact of epidemics such as indicator that takes HIV, or because of war. Other ways of visualizing into account that the information show that the process of increments in economic growth is also very volatile, with income as well as countries’ economic performance changing technology tend to continually across time both in the long-term and be significant in the short term. An important degree of according to their divergence is present across these fifty years. The proportion. By coefficient of variation (the standard deviation contrast, life divided by the mean) of log income per capita expectancy cannot increased from 0.15 in 1960 to 0.21 in 2000 but be increased then decreased somewhat to 0.18 in 2010. At the proportionally. average US growth rate over the period, the bottom 50 countries lagged between 106 and 264 years behind the US in 2010. An underdeveloped country is

underdeveloped in levels if its

long-term growth rate is similar to developed countries (at least 2% a year) and

underdeveloped in growth rates

if its long-term growth rate is significantly lower than that.

The window that Figure 1.1b provides into the process of economic development by looking at life expectancy is different. By comparison, here we see a smoother process of generalized catching up. Although health disasters have definitely occurred, a systematic improvement in life expectancy is observed. An important degree of convergence is also present. The coefficient of variation of life expectancy decreased by about 39.7% between 1960 and 2010. The improvement in life expectancy has also been more uniform than the growth of income. The coefficient of variation of the income growth rate during the period was 0.76, while the coefficient of variation of life expectancy change was 0.47.

3

DAVID MAYER-FOULKES

Of course an important difference between income and life expectancy is that income has evolved into a process of permanent growth which life expectancy does not admit.

[12.1,12.55]

[11.65,12.1]

[11.2,11.65]

[10.75,11.2]

[10.3,10.75]

[9.4,9.85]

[9.85,10.3]

[8.95,9.4]

[8.5,8.95]

[8.05,8.5]

[7.6,8.05]

[7.15,7.6]

[6.7,7.15]

[6.25,6.7]

[5.8,6.25]

[5.35,5.8]

0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 [4.9,5.35]

Frequency

Log Income per Capita Intervals 1960

1985

2010

Figure 1.2a. Histograms for the logarithm of income per capita for the years 1960, 1985 and 2010

0.3

Frequency

0.25 0.2 0.15 0.1 0.05 [85,90]

[80,85]

[75,80]

[70,75]

[65,70]

[60,65]

[55,60]

[50,55]

[45,50]

[40,45]

[35,40]

[30,35]

[25,30]

[20,25]

0

Life Expectancy Intervals 1962

1987

2012

Figure 1.2b. Histograms for life expectancy for the years 1962, 1987 and 2012

4


Figures 1.2a and 1.2b portray the distribution of log income per capita and life expectancy across countries for the years 1960, 1985 and 2010. In the case of life expectancy the years are 1962, 1987 and 2012, because for 1962 and 1987 these correspond to the original data rather than estimates. The distribution of both log income per capita and life expectancy are multiple peaked. This implies that the process that has formed the distributions has grouped countries at different levels of income and life expectancy. In the case of income the peak to the right corresponds to development. The peaks to the left indicate that there are two kinds of underdevelopment, one more backward than the other. As growth has proceeded, the lower of the three peaks has grown the slowest, while the higher of the three peaks has grown the fastest, also increasing in the number of countries. These three types of income performance are obtained theoretically in Chapter 2 and defined as development, underdevelopment in levels and underdevelopment in growth rates. In the case of life expectancy two peaks are apparent. The lower peak has tended to disappear, while the higher has increased in number, indicating a third group of countries transiting towards higher life expectancies. Vollmer et al (2013) give recent evidence for these three human development clubs. Figure 1.3a portrays the distribution of the average growth rate of income per capita from 1960 to 2010, for countries initially below or above the log income mean. The growth rate of countries initially above the mean has a more concentrated distribution. The growth rate of the countries initially below the mean is mostly concentrated at almost stagnant or low rates of growth, consistently with divergence, and includes a higher frequency of both negative growth rates and sustained, high, miracle growth rates. Figure 1.3a portrays the distribution of the average growth rate of income per capita from 1960 to 2010, for countries initially below or above the log income mean. The growth rate of countries initially above the mean has a more concentrated distribution. The growth rate of the countries initially below the mean is mostly concentrated at almost stagnant or low rates of growth, consistently with divergence, and includes a higher frequency of both negative growth rates and sustained, high, miracle growth rates. Figure 1.3b shows the distribution of the change in life expectancy from 1962 to 2012 for countries with life expectancy below and above 50 years in 1962, approximately the median. Here what is remarkable is that the two distributions, each with about ninety countries, are almost identical.

5

DAVID MAYER-FOULKES

Figure 1.3a. Histograms for the average growth rate of income per capita from 1960 to 2010 for countries initially below or above the mean

Figure 1.3b. Histograms of the change in life expectancy from 1962 to 2012 for countries with life expectancy below and above 50 years in 1962. 6


It is as if all of the countries underwent a common process after 1962, perhaps the implementation of modern health methods ranging from sanitation and vaccinations to maternal care and modern hospital medicine, which in low health countries had not been not widely implemented before 1962. By contrast the obstacles for achieving high income levels were more substantial.

1750

L o g G D P p e r c a p ita

10

United Kingdom

1820

Western Europe

9

Western Offshoots 8

East Europe Latin America

7

Asia 6

1914

1945

Africa 2000

1950

1900

1850

1800

1750

1700

1650

1600

1550

1500

5

Figure 1.4. Main economic characteristics of globalization in the period 1500-2015, and the Great Divergence. Source: Author's elaboration using Maddison’s (2001) historical data. The figure depicts the Great Divergence (Pritchett, 1997), and the annotations represent the findings of diverse authors (see text). The United

II. Globalization, 1500-2015 The fifty year process of divergent income growth examined in the previous section is part of a longer historical process, as can be seen in Figure 1.4, which graphs log income per capita by selected regions over the period 1500-2000, according to Maddison’s (2001) historical data. While the

7

DAVID MAYER-FOULKES

history of trade goes back much further in many regions, systematic global contact began with the Great Discoveries. Figure 1.4 depicts the Great Divergence, as it was called by Prittchet (1997). Starting in Britain in the second half of the 18th Century, the Industrial Revolution led to modern economic growth and raised the incomes of currently developed countries from almost ancestral subsistence levels to unprecedented levels of prosperity, albeit in a history also marked by poverty and war. Meanwhile, underdeveloped countries, most of which had been colonized by European powers, lagged behind, in a process of economic divergence which as we have seen continues to this day. Industrialization was intimately linked with trade and globalization – as was underdevelopment. To see this we continue with an overview of the history of globalization over the period 1500-2015, which puts the current period of globalization into context, and is indispensable for a full flavor of the process of economic growth. The main economic features of this history are annotated on Figure 1.4. 1500-1750: Foundations of Global Trade The history from 1500 to 2015 is analyzed in two stages, shown in Figure 1.4 and Table 1.1. In the first, approximately from 1500 to 1750, transport technology, institutions, and other capabilities slowly developed, partly due to the incentives of trade, while the structure of production remained essentially autarkic. In the second, approximately from 1750 to the present, global trade and specialization emerged, were interrupted over the period 1914-1982, and continued. The initial transition to the second stage occurred from about 1750 to 1820, when systematic innovation based on importing raw materials and exporting industrialized products began in Britain, the leading commercial power. World trade and the Industrial Revolution took off simultaneously. Then world incomes bifurcated. The processes of development, underdevelopment and divergence appeared and have persisted to the present day. Maddison’s (2001) data provides population and income estimates for the year 1500, which is the approximate date of the Great Discoveries, including Columbus’ crossing of the Atlantic to America in 1492, Vasco da Gama's voyage to India in 1498, and Magellan’s circumnavigation of the globe completed in 1522. These voyages opened trade between Europe, Asia and America. They were motivated by the spice trade, and the process of global economic integration can be thought to begin at this time. The economic incentives generated by trade in the Age of Discovery modified 8


Table 1.1. Periods of Trade, Hegemonies, and Economic Growth, 1500-2015  Large scale transportation slowly developed  Europe consolidates initial advantage Industrial Revolution  Large-scale trade and specialization British commercial and  Systematic innovation industrial leadership  Free trade embraced I  The Great Divergence begins 2nd Industrial Revolution  US and Germany overtake Britain and New Imperialism  Catch-ups in Europe and Japan Climax of colonial competition Specialized 1914World Wars Emergence of socialism Global 1945 Inter-war protectionism The Great Depression Trade Keynesian Economics  Fall of colonialism US commercial and  Rise of socialism industrial leadership Cold War  Persistence of underdevelopment and 1945Single developed capitalist divergence 1982 trading block  Reconstruction of multilateral free trade Stagflation leads to new  Miracle growth in East Asia, several classical economics European countries, Israel  Accelerated globalization: free trade and foreign direct investment  Prolonged growth in China Fall of socialism  Concentration of production 1982Global market economy 2015 First Global Crisis (2008):  Persistent, deepening inequality  Employment challenges in Europe the Great Recession  Global economy beyond the control of individual economies  Global sustainability challenges Although Spain obtained a large empire, trade was not amongst its objectives, concentrating instead on extraction, mainly of silver. This did not incentivize the type of economic, institutional and legal development that occurred in Holland and Britain (Coatsworth and Tortella, 2002). Guilds resisted industrialization; there were agrarian problems related to the expulsion of the Moors; and the overextended Spanish colonies were costly. Thus Spain declined over the period 1598-1700 (Vives, 1970). Incipient Global Trade Transition

15001750

1st Colonial Wave

the course of European history. Commercial dominance shifted from Venice and other Italian cities to Atlantic countries (Findlay and O'Rourke, 2002). Portugal, Spain, Holland and Britain became dominant colonial and commercial powers in turn. In the process, they engaged in a navigation technology race and, to make trade feasible, developed military power, financial institutions, and integrated their economies. Nevertheless, it took time for trade to grow. Findlay and O'Rourke (2002) show that the composition of trade from 1500 to 1780 consisted of goods not competing between continents, such as spices, silk, silver, slaves; goods with extremely 9

DAVID MAYER-FOULKES

Table 1.2. Institutional innovation, Hegemonies, and Trade, 1500-1750 1415-1612 Portugal

 Dominates Spice trade 1584-1702 Dutch Golden Age

 Dutch republic 1581  East India Company and Amsterdam Stock Exchange 1602 (stocks and bonds)  Bank of Amsterdam pays interest 1609 1700-1750 1500-1700 Emerging British Commercial, political and military consolidation of Power England  Commonwealth of England 1529-1660 – towards nation  Kingdom of Great Britain 1707 state  Patents, monopolies from 15th C. Statute of Monopolies  England dominates England trade with India 1623 1717  East India Company 1600  Navigation acts restricting foreign shipping (1651, 1660,  Merchant capitalism …) lead to Anglo-Dutch wars  Lloyd’s insurance for shipping 1688  Bank of England founded: bank of banks 1694 After its independence from Spain in 1581, Holland gained commercial ascendancy. The Dutch East India Company, with a base in Batavia (now Jakarta, Indonesia), was founded together with the Amsterdam Stock Exchange in 1602. It was the first company to emit stocks. In 1609 the Bank of Amsterdam introduced interest bearing debt. The success of the Dutch East India Company rested on its institutional and financial innovations (Harley 2002; p. 3). In 1641, Holland acquired Malacca from the Portuguese and took control of the spice trade. The Dutch Golden Age came to end in the period of the Anglo-Dutch wars (1652-1654, 1664-1667, when the British gained what became New York, 1672-1674, 1780-1784)., when Dutch industrial production declined. By 1799 the Dutch East India Company, formerly the world's largest went bankrupt. Holland

high value to bulk ratios. These authors also find no evidence of secular, intercontinental commodity price convergence before the 1820's, as should occur for sufficiently voluminous trade. O'Rourke and Williamson (2002) also consider 1500-1750 as an essentially autarkic period. Thus 1500-1750 is a period of incipient global trade, as compared to aggregate production. the course of European history. Commercial dominance shifted from Venice and other Italian cities to Atlantic countries (Findlay and O'Rourke, 2002). Portugal, Spain, Holland and Britain became dominant colonial and 10


commercial powers in turn. In the process, they engaged in a navigation technology race and, to make trade feasible, developed military power, financial institutions, and integrated their economies. Nevertheless, it took time for trade to grow. Findlay and O'Rourke (2002) show that the composition of trade from 1500 to 1780 consisted of goods not competing between continents, such as spices, silk, silver, slaves; goods with extremely high value to bulk ratios. These authors also find no evidence of secular, intercontinental commodity price convergence before the 1820's, as should occur for sufficiently voluminous trade. O'Rourke and Williamson (2002) also consider 1500-1750 as an essentially autarkic period. Thus 1500-1750 is a period of incipient global trade, as compared to aggregate production. Trade and institutional development were closely intertwined in England as well. Her colonial history begins with policies by King Henry VII (reigned 1485-1509) aimed at consolidating its merchant marine system in relation to wool trade. The Commonwealth of England, unifying England, emerged in 1529-1660, together with a strengthened navy. The British East India Company was founded in 1600. The statute of monopolies (on patents) was enacted in England in 1623. As British naval power rose, Navigation acts were imposed restricting foreign shipping (1651, 1660, ), leading to the Anglo-Dutch wars and eventually wresting commercial leadership from Holland. Insurance for shipping became available from Lloyd's in 1688. The Bank of England was founded in 1694. By revolutionizing the credit markets, it lowered interest rates, in turn strengthening Britain's military power, which depended on the borrowing capacity of the crown. In 1707, the Kingdom of Great Britain was consolidated. British presence in India was favored by the Moghul tax exemption in Bengal in 1717. In 1757 the British East India Company took control of Bengal, gaining supremacy in India. The Dutch East India Company, in its time the largest company in the world, began its decline. By 1799 it was bankrupt. Summarizing, during the period 1500-1750, the leading European countries exploited their superior technology in navigation, shipbuilding and armaments to develop international trade through monopolistic trading companies. In the earlier case of Spain, mercantile aims and colonial policy were not merchant capitalist but more akin to those of ancient imperialism. Having squeezed out the plunder, Spain declined. But in the Netherlands, France, and the UK, the overseas empire of the merchant capitalist period had a more beneficial effect on the productive capacity of the domestic economy, because it not only augmented capital resources but helped considerably to enlarge the size of markets. Trade led to institutional development including economic integration and financial development (national union, banks, the stock exchange, insurance, patents). In the 11

DAVID MAYER-FOULKES

leading countries, income per capita rose through the period. By contrast, colonialism retarded institutional development in the colonies. In the case of colonies such as Brazil, it was only until independence that the country could create its own banking system (Maddison, 1982, 2001). Thus, during the period 1500-1750 the foundations for global trade were established. To a great extent due to their commercial engagement, the leading countries reached higher institutional and technological initial conditions for the next period. Table 1.2 summarizes gives an outline of institutional innovation, hegemonic succession and trade over this period. The history of development and underdevelopment is intimately linked with the history of colonialism, and colonialism itself is closely linked with the formation of institutions, through the types of government and law that the dominant powers introduced. For example, colonies dedicated to settlement were more likely to reproduce home institutions, while colonies dedicated to extraction were more likely to have authoritarian, oppressive institutions imposed. Since institutions are quite persistent, colonial history continues through them to have a strong impact today. Institutional change itself responds to the balance of political power resulting from political institutions and the distribution of economic resources (Acemoglu, Johnson and Robinson, 2001, 2005). 1750-1914: Industrialization and underdevelopment Although the Voyages of Discovery led to the emergence of global trade flows, the type of goods being traded evolved only gradually in the 16th and 17th Centuries. At first, traded goods were ‘non-competing’, in the sense that they were available in some continents but not in others (Findlay & O'Rourke, 2002). We refer to this period as one of incipient global trade. It was only until the late 18th Century that true specialization began to emerge. British firms in a few key industries developed technological superiority over producers elsewhere and captured world markets (Harley, 2002). Eventually, the transport revolutions of the 19th Century led to inter-continental trade in bulk commodities which could be produced anywhere, such as wheat, iron and steel (Findlay & O'Rourke, 2002). Thus, the transition from incipient global trade to specialized global trade occurred in the second half of the 18th Century, beginning around 1750 (O'Rourke and Williamson, 2002). This transition coincides with protoindustrialization and then with the First and Second Industrial Revolutions. Trade, the division of labor, and innovation facilitated each 12


other. The higher level of production for larger markets through trade facilitated the division of labor, before the invention of engines powered by water, wind or steam, as reported by Adam Smith (1776). Incentives provided through world trade made innovation possible in a preexisting context of specialization. Finally, innovation increased exports and made cheaper imports available.

Institutions and development The dictionary definitions of market economy and capitalism include: “an economic system in which prices are based on competition among private businesses and not controlled by a government” and “an economic system in which investment in and ownership of the means of production, distribution, and exchange of wealth is made and maintained chiefly by private individuals or corporations owners for profit, rather than by the state.” Putting together such a system requires an institutional setting that maintains private property rights, contracts and so on. This institutional setting must be embedded in the governance both of the economy and of other affairs. At a minimum government must at the same time be strong enough to enforce the institutional setting, for example property rights, and democratic enough that its agents do not abuse private property rights. Of course, not all goods are private; indeed, the existence of private property can be considered a public good. Thus the capacity to organize the provision of public goods is part of what is required to make an institutional setting optimal. Institutional optimality is discussed further in Chapter 6, section IV.

The acceleration of trade during the mid 1700’s is shown in Figure 1.5. Table 1.3 gives a synopsis of hegemonies, trade, the demographic transition, and the diverse trajectories of economic growth that occurred now restricted to this period, which we subdivide into three subperiods, the Industrial Revolution from 1750 to 1820, Pax Britannica and the appearance of underdevelopment from 1820 to 1870, and the US and German overtake of Britain with the Second Industrial Revolution, 1870 to 1914.

13

DAVID MAYER-FOULKES

The intensification of trade with the Industrial Revolution

Figure 1.5. Descriptive OLS regression split at the year 1750 for the tonnage transported by the British East India Company over the period 1610-1813. The coefficient for years changes significantly showing a significant change in the growth rate of trade in the mid 1700’s. Intercontinental trade to India and China accelerated in the mid 1700’s, as can be observed from the tonnage transported by the British East India Company. Figure 1.5 shows the results of the descriptive OLS regression ln(tons) = – 4.45 + 0.00826 year + 27.6 D1750 + 0.0159 D1750 x year (–4.26) (13.3) (–7.3) (7.45) Here “tons” is the total tonnage transported by the company, ”year” runs from 1610 to 1813, when the company started declining, and “D1750“ is 1 for years  1750 and 0 for earlier years. t statistics shown in parenthesis. The results show that the rate of growth increased significantly in the mid 1700’s.

1750-1820: The Industrial Revolution The period 1750-1820 was one of pivotal historical change. Britain rose to colonial and naval dominance through the Seven Years' War, the Industrial 14


Revolution, and the Napoleonic Wars. Britain emerged as a world commercial and military power, dominating the seas, and exchanging manufactured goods for raw materials. The wars also led to independence in the America's.

Table 1.3. Hegemonies, Trade, Demographic Transition, and Diverse Economic Growth Trajectories, 1750-1914 1750 – 1820 1820 – 1870 1870 – 1914  Seven Years’  Transport revolution World War  Pax Britannica  New imperialism Context  Napoleonic  Revolutions of 1848  Scramble for Africa Wars United  Industrial  Capitalism fully emerges Kingdom Revolution  Britain the “Workshop  Leadership lost  British of the world” Industry: commercial  Free trade policy 1846 dominance  Fertility declines: popDemographic  Population ulation growth peaks,  Population growth growth in UK Transition in in UK reaches 1.5%  Low education slows rises above Europe in 1820 2nd Industrial 0.9% in 1780 Revolution in Britain  2nd Industrial Rev.  “American System”  Gold standard  US independMain 1816 –1970, espeence, more  US and Germany Converging cially 1861-1932 independent overtake Britain Countries  German and Italian trade policy  Russia, Japan begin unification industrialization  India loses domestic textiles  China opened: opium wars 1834 –1843. Open  India loses  Deep levels of underdevelopment Periphery textile exports  US “Gunboat diplomacy” on Japan 1853 followed by Japanese modernization Closed  Latin American  High tariffs in Latin America Periphery independence  Intermediate level of underdevelopment

We use the dates 1750-1820 for the period in which the initially leading industrial country, Great Britain, experienced the transition to systematic innovation, and became a technological leader. The new economic dynamics became dominant by 1820, and most other countries became involved in the dynamics of development or underdevelopment. An examination of the timing of wars, specialization, trade, industrialization, and population growth in this period shows the working of a clear logic of feasibility.

15

DAVID MAYER-FOULKES

The Seven Years' War (1754 and 1756-1763) involved all major powers of Europe: Prussia, Great Britain and Hanover against Austria, France, Russia, Sweden, and Saxony. France's power in the Americas ended and Great Britain emerged as the world's leading colonial and naval power, including the strongest position within India, the jewel of the British Empire. Intercontinental trade in cotton with India, the pivotal sector of the industrial revolution, now became feasible. However, In Britain's American colonies, the Seven Years' War led to the formation of an army that later fought and won the American War of Independence (1775-1783) when Britain sought to impose new taxes. The impact of trade on market size and on the subdivision of labor and therefore productivity, was evident by 1776. This is the publication date of The Wealth of Nations, in which Adam Smith explains how markets coordinate the division of labor, determine wages and interest, and argues for free trade. Note that when he describes production Smith (1776) mentions machines frequently, but refers neither to engines nor to the use of water, wind or steam power. Britain's first true factory, a water-powered mill, was first built in 1771 by Richard Arkwright at Cromford, Derbishire. Two important patents, 1769 and 1775, were involved in achieving industrial-scale cotton production. Arkwright’s mill eventually employed more than 800 workers (Fitton, 1989). The first steam locomotive railway was built in 1804 (Rattenbury and Lewis, 2004). By increasing market size, trade promoted Smithian growth through protoindustrialization by the subdivision of labor. This context made the introduction of engines in production feasible and raised the incentives for innovation. The leading sector of the Industrial Revolution was cotton textiles. This industrial sector was closely linked with trade. The demand for raw cotton imports into Britain and for slaves to the Americas expanded in the 1780's as a consequence of the innovations in Manchester (Findlay and O'Rourke, 2002). Cotton exports rose from 6% of total British exports in 1784-6 to a peak of 48.5% in 1834-6 (Chapman, 1999). The growth of this sector and the incentives for its increased productivity were directly linked with imports of cheap raw materials from India, while manufactured textiles were exported to Europe (Broadberry and Gupta, 2005). This sector flourished through the success of labor-saving technological progress that made unit labor costs lower in Britain than in India despite a high wage differential (ibid). A measure of the transformation of this period is given by the growth of British navigable waterways, which quadrupled between 1750 and 1820 (Girard 1966, p. 223 in O'Rourke and Williamson, 1999; canals offered a transport option 50-75 percent cheaper than roads).

16


The impact of the Industrial Revolution abroad was immediate. India, the most prominent open colony, lost its textile export market during this period (Clingingsmith and Williamson, 2005). Also, population growth began to accelerate in Britain, climbing to 0.9% in 1780 and to 1.5% in 1820 (Wrigley, 1985). The French Revolution and the Napoleonic Wars (1792-1815) weakened France at a time when Britain, which was less severely damaged, emerged again as the leading commercial and military power and industrialized. British cotton pulled decisively ahead of its Indian competition after the Napoleonic Wars (Broadberry and Gupta, 2005). Napoleon's invasion of Spain (1808) was the turning point for independence in Latin America, all of which was free by 1825 except for Puerto Rico and Cuba. Brazil achieved independence in 1822. 1820-1870: Pax Britannica: Income Bifurcation With the Industrial Revolution and the growth of trade, Britain became the Workshop of the World, and capitalism fully emerged in Britain. Industrialization implied specialization wherever it spread – and equally, where it didn't – as countries traded among themselves, exchanging manufactures for primary products and vice versa. A fundamental international division of labor emerged that had not been seen before on such a scale, and also heralded the Great Divergence of incomes and productivity in the last two centuries (Taylor, 2002), in what was the First Great Age of Globalization, that ended in 1914. While world trade grew at a little over 1% per annum between 1500 and 1800 (O'Rourke and Williamson 2002a), it grew at around 3.5% per annum since 1820 (Maddison 1995) and through the 20th Century. Some historians (Hyam, 2002, Smith, 1998) refer to the period 1815 to 1914 as Britain's imperial century. During this period 10,000,000 square miles of territory with a population of roughly 400 million people were added to the empire. By 1922, the British Empire held sway over about 458 million people, one-quarter of the world's population (Maddison, 2001) and covered more than 13,000,000 square miles, approximately a quarter of Earth's total land area (Ferguson, 2004). The steamship and the telegraph, invented in the second half of the 19th Century, underpinned British imperial strength; and the sun never set on the British Empire. Even so, one of the lessons Britain had drawn from the independence of the US had been that free trade and informal empire could be more 17

DAVID MAYER-FOULKES

profitable than colonial rule and its costs of defense and administration. The slave trade was abolished in 1807 and slavery in the colonies in 1833. Britain granted self-government to white settler colonies such as Canada and Australia. Free trade, following the thought of Adam Smith and David Ricardo, was established in 1846 with the derogation of the Corn Laws, import tariffs first enacted in 1815 to protect British farmers and landowners against cheap grain imports. Average per capita income growth rose from 0.25% over the period 1700-1820 to 1.25% per year. Britain dominated overseas markets and favored a strategy of informal colonialism, epitomized by the opium wars, 1839-1842 and 1856-1860. China was forced to sign trade treaties with England, Germany, Russia, Japan, and the United States, and suffered two extreme famines twenty years after each opium war in the 1860s and 1880s. Gunboat diplomacy was also used by the United States on Japan. Commodore Perry arrived at Uraga Harbor in Kanagawa Prefecture in 1853; superior military force enabled him to negotiate a treaty allowing American trade with Japan, ending a 200-year period in which trading with Japan was only allowed to the Dutch. By contrast with India and China, this was a period of high tariffs in Latin America, where the newly independent states used taxes on trade to finance government. Britain's trade with its formal and informal empire focused innovation in Britain. An idea of the relative size of this effect is given by the following data, reflected in Figures 1.6a and 1.6b. From 1820 to 1940 Britain's formal empire had a similar population to China, the largest country in the world with 381 million people in 1820, with each of the US, Russia, France, Germany and even the whole of Western Europe without the UK (which can be thought of potentially as an area of German influence) and the whole of the non-colonized world, much smaller. In 1820 China’s production was higher than the British Empire’s. By 1870, the opposite held. An important point in comparing the forces of economic development in Britain and China is that there was no region in China where economic forces for innovation were as focused as they were in Britain, which functioned as the economic center of an economy as large as China’s. This gives a strong reason why the Industrial Revolution occurred in Britain rather than China. Data on manufacturing provide another look at these events. This was a period of pronounced deindustrialization in the periphery (Williamson, 2004), as can be appreciated in Figure 1.7a, which shows the percentage of 18


Percentage of World Population

world manufacturing output produced in the developed core, India, China, and the rest of the periphery.

40

British Empire

35

China

30 United States

25 20

Russia

15 Not colonized

10 5

France and its colonies

0 1700 1820 1870 1913 1940 1992 Year

Germany and its colonies

Percentage of World Production

Figure 1.6a. Population of various Powers and Regions as a proportion of World population. 1700-1992 (Maddison, 2001) British Empire

40 35

China

30 United States

25 20

Russia

15 Not colonized

10 5

France and its colonies

0 1700 1820 1870 1913 1940 1992 Year

Germany and its colonies

Figure 1.6b. Production of various Powers and Regions as a proportion of World population, 1700-1992 (Maddison, 2001)

19

Manufacturing Output (per cent of total)

DAVID MAYER-FOULKES

100 80

Developed Core

60

China

40

India

20

Rest of the Periphery

0 1750 1800 1830 1880 1913 1938 Year

Figure 1.7a. Proportion of manufacturing output taking place in the developed Core, China, India, and the rest of the periphery from 1750 to 1938. Source: Simmons (1985), quoted by Williamson (2004).

60,000

Length of Railway Lines (km)

50,000

Austria‐Hungary Belgium

40,000

France

30,000

Germany Great Britain

20,000

Italy

10,000

Netherlands Russia

0 1840

1860 1880 Year

1900

Spain Sweden

Figure 1.7b. Spread of railways in ten selected European countries during the period 1840-1900. Source: Halsall Paul 20


Figures 1.7a, 1.4 and 1.1a show that the concentration of manufacturing in the developed core continued well into the 20th Century, with divergence continuing to this day. 1870-1914: Second Industrial Revolution The Second Industrial Revolution (1867-1914), based on scientific innovation, generated mass-scale production in the basic manufacturing sectors such as steel, oil, mining, telephone, automobile (Smil, 2005). These manufacturing sectors, as well as the banking sector, consolidated into huge enterprises in the late 19th and early 20th Centuries, in waves of mergers also featuring vertical integration (Lipton, 2006; Lamoreaux, 1991). The Sherman Antitrust Act of 1890 gives a flavor of this era that culminated in mass production with Henry Ford's 1913 assembly line producing a Model T every 93 minutes (Domm, 2009). Electricity became the basis for manufacturing technologies and was also used for consumption goods, as were chemicals for such goods as paper, soaps, textiles, and fertilizer; synthetic dyes, and petroleum refining. The internal combustion engine was also used for factories. The Second Industrial Revolution thus introduced the large scale sector, bringing with this, together with scientific innovation, a permanent change in the nature of the industrial market economy. Mass production remains the basis of modern productivity to this day, with 51.6% of the US workforce is employed in a tiny proportion (0.3%) of firms with 500 workers or more in 2012. Transportation was a significant precursor of the Second Industrial Revolution. The first commercial steamboat services were introduced in 1807 in the US along canals, while the first commercial steam railway service between Manchester and Liverpool was introduced in 1830 (Rodrigue, 2013). Figure 1.7b shows the expansion of railways in Europe. Railroad transportation completely changed the geography and composition of trade. For example, a world market for grains emerged. In the 1830’s 8.2% of the British were dependent on imports of foreign wheat, while by the 1890’s this figure had risen to 71.2% (Ejrnæs, Persson, and Rich, 2008). By the 1860's, railways had emerged as a primary mode of transportation, Germany's system being particularly large in Europe. In the US, there was an even higher railway building boom from 1830 to 1870. The first transcontinental railway link was achieved in 1869. In both Europe and the US, railways replaced canals as the primary mode of industrial 21

DAVID MAYER-FOULKES

transportation. Railways were also important in Japan. The TokyoYokohama railway opened provisionally in 1872, and by 1907, 17 private railways were nationalized. Empire, a source of raw and agricultural materials as well as markets, was viewed at the time as an important determinant of economic growth, prosperity and power on the global scene. The scramble for Africa beginning in the 1880s and lasting until 1914 illustrates this point. However, “Invisible empire” through trade and political influence was particularly cost-effective. A significant proportion of British trade and investment in the 19th Century took place with Latin America, which was independent. Railways completely reshaped the geography of trade. Landmasses now served as trade catchment areas, without recourse to maritime or canal transportation, on which colonialism, in particular the British empire, was based. During this period Germany, the United States and Japan industrialized. Following are some comments on how industrialization proceeded in these countries. Other countries industrializing in this period were Denmark, Sweden and Italy, which experienced specific periods of rapid, catch up growth (see Pipitone, 1994 for case studies). The United States The first US factory was built in 1790, when Samuel Slater brought the secrets of British textile machinery implemented by Richard Arkwright in the 1770’s to the US (Everett, 2006). The American System was put in place from 1816 to 1970, but especially in the period 1861 to 1932, providing the elements and scale for industrialization. Bringing together a continent about as large and perhaps richer in resources than Europe, the Northern States served as a focus of trade, conducted by railway, and innovation, selling industrial products to the South and to the West Coast, in exchange for agricultural products such as cotton, and for raw materials. At the same time, Northern infant industry was protected from British competition. Germany The German Customs Union was formed in 1834, in the first steps to German unification. This partly followed Economist Friedrich List's views on protecting infant industry to achieve the Industrial Revolution. The customs union began in Prussia in 1818 and evolved through 1888, including agreements with Luxembourg, Sweden and many other areas. The 22


unification of Germany into the German Empire came in 1871. Also central to German economic growth was the University system. Germany invested more heavily than the British in science and pure research, especially in the chemical industry, and came to control 90% of the world's chemical market. Russia Russia experienced a decade of fast growth in the 1890’s, when its coal, iron, steel, and oil production tripled. It also greatly increased its railroad system, which became larger than any other nation’s except for the US. The population of St Petersburg and Moscow more than doubled. Nearly half the workers worked in factories with more than 1000 employees. The Trans-Siberian Railway was completed in 1905. Japan The Japanese Meiji restoration, from 1868 to 1912, coincided with this period. During this time, Japan started its modernization and rose to world power status. Japan employed of over 3,000 foreign experts teaching English, science, engineering, the army and navy etc.; and dispatched many Japanese students to Europe and the US. It also adopted technologies from the West, and gradually took control of much of Asia's market for manufactured goods, beginning with textiles. Japan imported raw materials and exported finished products. The Meiji rulers embraced the concept of a market economy and free enterprise. Economic reforms included a unified currency, banking, commercial and tax laws, stock exchanges, and a communications network. Establishment of a modern institutional framework conducive to an advanced capitalist economy took time but was completed by the 1890s. By this time, the government had largely relinquished direct control of the modernization process, primarily for budgetary reasons. While the government was initially involved in economic modernization, after the first twenty years of the Meiji period, the industrial economy expanded rapidly. Japan achieved power status at the turn of the Century. Rival ambitions over Manchuria and Korea led Japan to attack Russia in 1904 before the completion of the trans-Siberian Railway in 1905, inflicting the first defeat on a Western power by an Eastern power. Japan emerged from World War I as a major industrial nation. France The Seven Years' War, the French Revolution and the Napoleonic Wars (1792-1815) weakened France and her naval power at the time when the 23

DAVID MAYER-FOULKES

Industrial Revolution took off. France took a long time to industrialize and her population growth in the 19th Century was much slower than in Britain or Germany. When railroad transportation became important in the second half of the 19th Century, Germany benefited from trade more than France. Napoleon III (Emperor from 1852 to 1870) reconstructed Paris, modernized the French banking system, expanded the French railroad system and made the merchant marine the second largest in the world. He negotiated free trade agreement with European trading partners. Mostly

The demographic transition While the Industrial Revolution sparked population growth in Great Britain and other industrializing countries, eventually there was a demographic transition towards lower fertility. This has been a characteristic component of economies achieving modern growth and high per capita income levels. Western European countries transitioned from high to low fertility in the period 1870-1914, at the time when the Second Industrial Revolution, characterized by systematic, scientific innovation, generated a demand for skilled workers, and also a demand for minimum levels of education beginning with literacy and then advancing to higher curricular levels. In the US the demographic transition took a long time, from 1810 to 1930 approximately, perhaps due to a lower land constraint. Endogenous fertility theory (see for example Galor, 2005, 2012) explains how the demand for human capital in the workplace, which requires a higher parental investment in their children that results in higher income levels, motivates a shift in fertility choices towards less but better prepared children, particularly when land constraints reduce the per capita returns to technology. The demographic transition occurred, indeed is occurring, much later for underdeveloped countries.1 Endogenous fertility theory explains this delay in terms of a lower demand for human capital. This can occur together with lower incentives for technological change in backward countries. In particular, when developed and underdeveloped countries trade, for example exchanging industrial goods for raw materials, this specialization implies a specialization in skilled and unskilled labor, in turn implying a lower population growth in developed than in underdeveloped countries. Galor and Mountford (2008) argue that this is one of the causal factors of the Great Divergence, since it accentuates bifurcation rather than convergence.

24


after 1850, France rebuilt a colonial empire in Africa, Indochina, and the South Pacific. By 1900 it was the second-largest colonial empire, after the British. Convergence to industrialization in the late 19th Century Convergence is the idea that economies converge in terms of income, for example, that they converge to development. In Chapter 2 we shall see how the concept of convergence derives from economic models with stable equilibria or steady states, towards which economic trajectories tend. Here we observed that once Britain industrialized, other countries were also able to do so in the 19th Century. These countries became powers. In 1914 the great powers were: Britain, Germany, Russia, Austria-Hungary, France, Italy, Japan and the United States. It is interesting to note here that in the long 19th Century (until 1914) countries that converged to industrialization also tended to converge to world power status and the acquisition of a colonial empire. In the period 1870-1914 industrializatin and nationalism led to the age of New Imperialism. Spheres of influence had an impact on trade and economic development and were part of the competition between industrial leaders. 1914-1945 World Wars and interwar period The World Wars were in some sense the climax of imperial competition. Empires viewed themselves as competing economic and military entities and antagonism and conflict was the result. After World War I, there was no clear hegemony. Europe was devastated by war. In the US, first there was economic growth (the Roaring Twenties; boom in the automobile industry), but then came the Great Depression in 1929. In Germany, the Weimar Republic gave way political and economic turmoil, the German hyperinflation of 1923 and later the rise of Nazism. In Asia, Japan became an ever more assertive power, especially with regards to China. Other powers were the Soviet Union and Mussolini's Italy. Meanwhile in the United States there was a resurgence of isolationism. Globalization, which had tended to be fractured along hegemonic lines, retreated in the interwar period. There were no major market emergences into development during this period, and country's developed and underdeveloped status persisted through it. However, the Soviet Union emerged as an industrial and military power, together with Eastern Europe, organizing its development through the state. 25

DAVID MAYER-FOULKES

Interrupted by the First World War, the Second Industrial Revolution can be thought to continue during the roaring twenties, characterized by innovation and mergers. Large scale production became generalized in the economy. This laissez faire period ended with the Great Depression, 19291933. The introduction of Keynesian policies led to a new era in US economic history that introduced social security, was concerned with workers’ living standards, supported education and science, and lasted until the 1980’s.

World Context

Core

New Core Periphery

Table 1.4. Hegemonies, Trade, Demographic Transition, and Diverse Economic Growth Trajectories, 1945-2000 1989 – 2015 1945 – 1989  Cold War (1945-1989)  Liberalization  Single capitalist trading block  Fall of Socialism, 1989  Reconstruction of multilateral free trade  Washington Consensus, 1990  OEED, OECD, IMF, World Bank,  Accelerated globalization, WTO GATT  FDI average annual growth 1982  Fall of colonialism – 2007: 14.6%, trade 6.2%  Oil shocks 1973, 1979  US leadership; growth: 3.13%  Marshall Plan, European  G8 (1973) convergence  European Union (1992)  European Free Trade Association,  US growth 1989-2014: 2.38% 1957  European growth 3.9%  Miracle growth in Japan, Hong Kong, Taiwan, Korea, Israel  Fast growth in Cyprus, Spain, Portugal, Malta, Ireland and Iceland  Africa stagnates  Persistence of underdevelopment and  Latin America opens: mixed divergence for 168 countries results (Maddison, 2001)  Fast growth in China and India

1945-2015: US Hegemony After World War II the US emerged as the leading capitalist country. With the weakening of Great Britain, colonialism entered its decline. Most colonies gained independence by the 1960's. The era of competing empires gave way to the Cold War between the United States and the Soviet Union, and their respective allies. (See Table 1.4 for a historical outline).

1945-1989: The Cold War 26


After antagonism and conflict had led to the retreat from free trade, after World War II the leading powers sought to reconstruct multilateral free trade. At the July 1944 Bretton Woods Conference, 44 Allied nations signed agreements to set up the International Bank for Reconstruction and Development (IBRD, the original institution of the World Bank Group); the General Agreement on Tariffs and Trade (GATT, precursor of the World Trade Organization); and the International Monetary Fund (IMF). In Europe, economic union was used to promote economic growth as well as preserve and foster peace. The Marshall Plan was put into place to reconstruct Western Europe. The European Coal and Steel Community (1951) was formed, followed by the European Economic Community (1957). West Germany experienced miracle growth and Europe as a whole achieved an average growth rate of 3.9% per year over the period. The United States also enjoyed an unprecedented period of internal growth and prosperity with the baby boom generation. Miracle growth was experienced in Japan, Hong Kong, Taiwan, Korea, and Singapore. Wan (2004) provides case studies for these growth processes, emphasizing the attainment of technological know-how in conjunction with exports to a large market (the US), which formed part of the Cold War geopolitical strategy. However, underdevelopment and divergence persisted for 168 countries (Maddison, 2001). For many populations emerging from colonialism, market institutions had not offered a successful development path. This was particularly true in contexts of poor political institutions often supported by Western powers. Socialism became an attractive alternative. The USSR and China, as well as other socialist countries, made important advances in industrialization, education and health based on centrally planned economies. However, beyond these achievements socialist development reached a limit. China began to introduce market mechanisms in its economy in December 1978, and the Berlin Wall fell in 1989, ending the Cold War. 1989-present: Global Market Economy These events, together with others affecting the Western economies, contributed to produce a single global market. The Western economies were faced with the stagflation crisis of the 1970's and the first oil crisis. This led to the freeing of trade and investment by Ronald Reagan and Margaret Thatcher, and the emergence of liberalization and new classical economics, leaving Keynesian economics behind. The Washington 27

DAVID MAYER-FOULKES

Consensus (a term coined in 1989) implemented a standard liberalization reform package in any developing country that faced a crisis. The World Trade Organization was established in 1995.

The democratic transition Democracy has been associated with successful market economies, although in fact the development of democracy is a continuing process many of whose basic features occurred well into the 20th Century even for the most developed countries. Britain became a constitutional monarchy after 1688, when the supremacy of Parliament was settled. The US began developing its democracy since independence. The Revolutions of 1848 in Europe, also called the Springtime of Nations were a series of democratic and nationalist revolts against the monarchies of Europe. It occurred in fifty countries including Italy, France, Germany and the Austro-Hungarian Empire (but not Great Britain, Russia, Spain and some Scandinavian countries), had many causes and was not coordinated. Although it was unsuccessful and suppressed, it is part of the history of the formation of democracy in Europe. Current studies support the idea that the transition from authoritarianism to democracy is a complex process that has advanced across the world but perhaps has to pass through an unstable phase. To understand cross-country democratization, Epstein et al (2006) introduce an intermediate category between autocracy and democracy, “partial democracy.” They show partial democracy is more volatile than either autocracy or democracy, but that increased income per capita increases the likelihood of democratization.

The combination of events sparked a new era of globalization and restarted economic growth. Between 1983 and 1998, foreign direct investment grew at an average rate of 29% per year. The European Union was formed in 1992. for twenty five years (1982-2007) FDI grew at an average real rate of 14.6% a year, while worldwide exports grew at a rate of 6.2%, approximately doubling as a proportion of world GDP from, 14.5% in 1982 to 30.6% in 2006. Transport and communications technologies continued to play important roles, for example containerized shipping and the internet. 28


While some countries experienced fast growth, such as China, India, Cyprus, Spain, Portugal, Malta, Ireland and Iceland, results were mixed in underdeveloped countries. Latin America had mixed results; much of Africa stagnated. Results were also mixed within developed countries.

5 4 3 2 1 4/1990 12/1991 8/1993 4/1995 12/1996 8/1998 4/2000 12/2001 8/2003 4/2005 12/2006 8/2008 4/2010 12/2011 8/2013 4/2015

2 1 0 ‐1 ‐2 ‐3 ‐4 ‐5

0

St. Dev (percent annual rate of change)

Mean (percent annual rate of change)

Mean

St. Dev.

Linear (Mean)

Month

Figure 1.8a. Average and standard deviation of monthly change during 1990-2015 of the selected set of stock exchange indices (see text). The optimism of incipient new classical economics and new growth theories predicting convergence reigned until the Global Financial Crisis of 2008, the worst financial crisis since the Great Depression in the 1930s. Such large business cycle fluctuations were considered a thing of the past in view of the Great Moderation, a reduction in business cycle fluctuations experienced in developed countries from the mid-1980s, that was believed to be inevitable with free market policies. As globalization proceeded, the many market economies subscribing to the World Trade Organization began to act more and more as a single economy. Particularly a single, global market for capital emerged. Figure 1.8a shows, as an example, how the behavior of a set of well-known stock exchange indices became more correlated (Dow Jones Industrial Average, S&P 500, Nikkei225, Hang Seng, DAX, FTSE100, CAC40, BSE Sensex). The top graph shows the average monthly percentage fluctuation in the value of the set of indices, broadly indicating the business cycle, while the bottom graph shows the monthly standard deviation of these fluctuation in value of, which significantly decreases across the period, showing that world financial markets increasingly behaved as a single market. 29

DAVID MAYER-FOULKES

Anual Rate (Percent)

20 1954‐ 2008

15 10

2009‐ 2015

5

7/1954 8/1958 9/1962 10/1966 11/1970 12/1974 1/1979 2/1983 3/1987 4/1991 5/1995 6/1999 7/2003 8/2007 9/2011

0

Month

Figure 1.8b. Monthly average Effective U.S. Federal Funds Rate, 1954-2015. Source Federal Reserve Bank of St. Louis Economic Research Division, https://research.stlouisfed.org/fred2/series/FEDFUNDS/

Unemployment Rate (%)

30

Greece

25

Spain

20

Cyprus

15 Portugal

10

Italy

5

Ireland

0 2008 2009 2010 2011 2012 2013 2014 Year

France

Figure 1.9a. Unemployment in Selected European Union Countries and the U.S.. Source: Eurostat The crisis of 2008 and the Great Recession it gave rise to marks a qualitative change in the process of globalization. The new single global market economy began to feature the impacts of a global business cycle that highlighted the institutional challenge posed by a global economy lying 30


beyond the control of individual national governments. Figure 1.8b shows the Effective Federal Funds rate, monthly from 1954 to 2015. Let us first comment on the period 1954 to 1981. The interest rate shows a tendency to increase that corresponds with the inflationary aspect of stagflation that led to the demise of Keynesianism. As non-inflationary macroeconomic policies were put into place, together with liberalization and globalization, interest rates decreased. Eventually a “savings glut” occurred (Bernanke, 2005) that continues to this day (Krugman, 2015), leaving US policy makers with very little room to maneuver, and keeping the key US Federal rate very close to 0% since 2008. This is a symptom of a global investment climate that lies beyond the control of the US government. During the Great Recession, unemployment in developed countries has focused on some of the weaker economies. Figure 1.9a shows unemployment in a selected set of European countries. In Greece and Spain unemployment increased from 2008 to 2013 reaching Great Depression levels of 25% that focused on young adults. Cyprus, Portugal, Italy, Ireland and France had unemployment rates ranging from 16. 1% to 10.3% in 2014. During the Great Depression in the US, Roosevelt used Federal policies to address unemployment. During the Great Recession, there was no European-level consensus to help Greece as its banks defaulted and its population faced deep unemployment. The concurrent fiscal crisis was consistent with the generalized, cross country decrease in median statutory corporate tax rates, shown in Figure 1.9b, that may be due to tax competition in the face of international foreign direct investment (Overesch and Rincke, 2011).

31

DAVID MAYER-FOULKES

Figure 1.9b. Median statutory corporate tax rates by World Bank country income group, 1980-2010. Source: Keen and Konrad (2012).

Figure 1.10a. US trade deficit and corporate profits as percentage of GDP, 1960-2014.

32


Figure 1.10b. Scatter plot of US trade deficit and corporate profits as percentage of GDP, 1960-2014. The straight lines represent OLS regressions for the periods 1960-1989 and 1990-2015. The US trade deficit and corporate profits Following is a piecewise linear OLS regression estimate for the trade deficit based on corporate profits, with slope and intercept shifting in 1990, yields highly significant coefficients for corporate profits, with an opposite sign for each segment and an R2 of 0.70. Trade = 0.15 x Profits x [year1990] – 0.17 x Profits x [year 0. Assume for simplicity that the frontier growth rate is also constant, with 

gt = g  > 0. Then according to (2.2.20) technological dynamics would be given by a t 1 =   

1  a t  H 1 a t . 1 g

52

(2.21)


Figure 2.1a. Convergence dynamics for technological change with constant

Figure 2.1a shows a diagram for these dynamics on the (at , at 1) plane. The bold line graphs equation (2.2.21). The dashed line represents the

at 1 = at . The arrows show how to use the diagram to obtain a sequence a0 , a1 , a2 ,... which converges in this case at a unique steady state

diagonal

a =  

1 g   0,1, implying that in the long run the country’s   g

productivity (the numerator of at = At / At ) would grow at the same rate as the global frontier (the denominator of

53

at ).

DAVID MAYER-FOULKES

Figure 2.1b. Conditional convergence dynamics for two parameter vector , The absorption effect We return to the absorption effect, according to which

t

is proportional

to at rather than constant. In this case there is a countervailing disadvantage of backwardness, which implies that convergence will depend on the country’s competitiveness parameter  instead of being guaranteed. To analyze the conditions under which convergence will or will not occur, we focus on the following two equations. First, replacing the innovation rate in (2.2.20) by its equilibrium value (2.8) yields the law of motion for normalized productivity:

at 1 = Next,

using

(2.2.22)

at 1  gt and

 at    1   . 1  gt   the

fact

that,

(2.22) by

construction,

1 Gt = At 1 / At = 1 gt at 1 / at , we see that the country’s growth rate Gt depends negatively on its normalized productivity, according to an 54


equation again reminiscent of those frequently estimated in cross-country growth regressions:

 a  Gt =   1  t  .  1  gt 

(2.23)

Consider first a follower country that has no influence over the frontier 

growth rate gt . Suppose that g t converges to a limiting value g > 0 in the long run (this will be indeed implied by the model.) Then the long-run behavior of the follower country’s normalized productivity will be governed by the asymptotic limit of the law of motion (2.2.22):

at 1 =

at  at  1    at  .    1 g  1  g  

(2.24)

It is easy to verify that the right-hand side of (2.2.24) is an increasing and

at on the unit interval, equal to zero when at = 0 and less than 1 when at = 1. As shown in the two panels of Figure 2.2, this

concave function of

implies that there exists a unique long-run equilibrium value to which normalized productivity will converge from any initial position.

A follower country The country’s long-run normalized productivity depends on its competitiveness  and on the asymptotic frontier growth rate g  . An increase in the frontier growth rate will make it harder for the country to keep up with the frontier, both because sectors that don’t innovate will fall even faster relative to the growing frontier and also because, as pointed out above in the discussion surrounding equation (2.8), an increase in global

growth will reduce the fraction t of sectors that do innovate by reducing the effectiveness of entrepreneurial skills at any given current distance from 

the frontier. For both of these reasons the increase in g will shift the  curve down in Figure 2.1, thereby reducing the country’s long-run normalized productivity. On the other hand, an increase in the country’s competitiveness will raise the innovation rate in any given situation, thus resulting in an upward shift of the  curve and an increase in the 55

DAVID MAYER-FOULKES

country’s long-run normalized productivity (for the proof of  / g  0 and  /   0 see HM).

Figure 2.2. The dynamics of normalized productivity in a follower country for higher and lower competitivities  .

More specifically, if motion (2.2.24) is:

 > g then the unique stable steady state of the law of







a = 1  g  1 g / > 0,

56

(2.25)


Convergence and conditional convergence Convergence is the idea that economies converge in terms of income, for example, that they converge to development. The concept of convergence derives from economic models in the form of dynamical systems with stable steady states. An example is given in Figure 2.1a, which shows a country’s technological level converging to a fixed proportion a  of the leading edge technology. If the steady state is unique, countries with similar parameters will converge to similar steady states. This is the concept of absolute convergence. In addition, poorer countries with similar parameters will grow faster than richer countries, because change slows (is decreasing in at ) as the equilibrium is approximated:

at 1  at =   

  g 1 g 

at

(2.26)

On the other hand suppose we have a cross-country data set for countries i  1,..., N with different relative technological levels a ti  1 , following the dynamics in equation (2.2.21), but with different competitivities  i (as in Figure 2.1b). Then given variables X 1i ,..., X ni which can explain the competitivities, a regression

a ti 1  a ti =  a ti   1 X 1i  ...   n X ni  c

(2.27)

will yield a negative beta (the definition of conditional beta convergence). What this means is that the trajectories of the relative technological levels converge to individual equilibrium trajectories, which are what a steady state represents. However, these might vary widely in levels according to the variables X 1i ,..., X ni . The negative beta is in fact a very robust result for “convergence regressions” on income per capita across economies. It does not imply convergence if this means that economies truly converge in terms of income, for example, that they converge to development. It only implies that on average each trajectory converges to its own equilibrium trajectory, quite a remarkable result in itself.

57

DAVID MAYER-FOULKES

which depends negatively on

g  and positively on . This is the case 

shown in panel (a) of Figure 2.2. Substituting at = a into the growth equation (2.2.23) above verifies that the country’s productivity growth rate

Gt will equal g  in this steady state. We call this divergence in levels. It describes parallel growth. However, if  < g then there is no positive steady-state normalized productivity. In this case, as shown in panel (b) of Figure 2.2, the slope of the  curve will be less than unity everywhere, and normalized productivity will fall asymptotically to zero. According to the growth equation (2.2.23) above the country’s long-run growth rate will equal  , the value of its competitiveness parameter. We call this divergence in growth rates. It describes semi-stagnation. 

In the borderline case where  = g the only steady stage is a = 0, and the growth equation (2.2.23) implies that the country’s long-run growth rate 





will equal the frontier growth rate g . In this borderline case the follower country tends to an infinite lag but its growth rate is still asymptotically

g . These results are formalized in the following proposition:

Proposition 2.1: Long-run growth and normalized productivity in a follower country. (a)

If

  g,

then







  limt  at = 1  g 1  g /  0

and



limt  Gt = g . (b) If

  g  , then limt  at = 0 and limt  Gt = .

The frontier growth rate Suppose for simplicity that there is just one leading country, labeled country 1, whose innovations affect the evolution of the global stock of knowledge

At . Everything that was assumed above of a follower country is true also 58


of country 1. Thus its rate of innovation

 t1 depends on its

competitiveness parameter  and its normalized productivity at according to equation (2.8) as in a follower country: 1

1

t1 =

1  at1

(2.28)

1  gt

The only difference between country 1 and the others is that the global frontier depends on the rate of innovation

 t1 in country 1, according to:

g t = t1

(2.29)

where  > 0 is a spillover coefficient. These last two equations solve for the frontier growth rate as a function of the leader’s normalized productivity:

g t = g~ ( a t1 ) =

 1  4 a  1/2. 1

1 t

(2.30)

Normalized productivity in country 1, the average technology level across sectors as a proportion of the leading edge, obeys the same law of motion (2.2.22) as in a follower country. Substituting from (2.2.30) into this law of motion, yields the reduced-form law of motion: 1 t 1

a

at1  1 at1  1 =  1   (at1 ). 1  1  ~ ~ 1  g (at )  1  g (at ) 

(2.31)

~ ( a1 ) and (a1 ) are increasing and concave in a 1 , with The functions g t t t

' 0 > 1, 0 = 0 and 1 < 1 (for the proof see HM). This allows us

to show the following.

Proposition 2.2: Global growth and the leader’s normalized productivity. The leader’s normalized productivity converges monotonically to:

59

DAVID MAYER-FOULKES

  1   a 1 =  1  > 0  1   1    1   

(2.32)

and the frontier growth rate converges monotonically to:

g = where

1

 1 , 1 

(2.33)

is the leader’s competitiveness parameter.

Modern R&D We model the introduction of modern R&D by supposing that before some date t 0 all productivity advances were based on a pragmatic creativity occurring close to the production process, which we call implementation, with innovation-productivity  . Thereafter, an alternative method of technology investment, called R&D, became available to all countries. R&D produces innovations according to a more productive process than implementation, using the technology described by equation (2.2.11) above, but with an innovation-productivity   >  . To be viable, however, the new technology requires entrepreneurs to possess a skill level at least equal to some threshold value  At , which depends upon the global technology frontier. If entrepreneurs do not have this threshold level of skills then R&D is impossible, although the original process of implementation remains. Because the entrepreneurial skill in any country is At = at At , therefore R&D will be possible in the country if its normalized productivity at = At / At is at least equal to the critical value:

a c  /

(2.34)

Because the productivity of R&D innovation is a multiple of implementation’s, and because an entrepreneur’s expected payoff in any given situation depends positively on her innovation-productivity, all technology investment will take the form of R&D in countries where normalized productivity is at least equal to a c . A country with normalized 60


productivity less than a c will lack the absorptive capacity needed to perform R&D, and all technology investment in the country will take the form of implementation. Hence, both in the case of a leading or a follower whenever at  a . we simply replace  with   . c

at 1

45°

1

R&D

 3lmp 2 lmp

0

1´lmp * almp

0

aR*&D

ac

1

at

Figure 2.3. The full dynamics in the case of a follower country. In the case of a leading country, once R&D exists the growth rate will be given by (2.33) with  given by (2.2.16), using the parameters for the leading country and   instead of  . 1

In the case of a follower country, Figure 2.3 illustrates the full dynamics. Three examples ´lmp , lmp , lmp are given for the functions describing 1

2

3

the dynamics resulting from implementation. The first is a case resulting in divergence in growth rates or semi-stagnation. The second is a case resulting in divergence in levels or parallel growth. The third is a case in which after some time a lagging country graduates from implementation and achieves R&D. For simplicity of the diagram a single case of the R&D function is given in the figure, which is viable and lies above the three implementation examples in innovation productivity. Hence it is assumed for graphical simplicity that any of the three follower countries whose implementation dynamics are described would have the same R&D dynamics if their relative technological level comes to be higher than the critical level a c . 61

DAVID MAYER-FOULKES

III. Convergence clubs and transitions or miracle growth The first two cases in Figure 2.3 illustrate the concept of multiple steady states. In these cases a follower country whose initial relative technological level lies below the R&D threshold, so

a0  ac ,

* lmp ,

implementation steady state at either 0 or a

will tend to an

while if the initial level lies

above the threshold, a0  a , it will tend to the R&D steady state aR&D . Countries nearby will behave similarly, and so we have the concept of convergence clubs, with different sets of countries each converging to approximately similar growth trajectories, which however differ significantly between the sets. *

c

Development, Underdevelopment in Levels, and Underdevelopment in Growth Rates The current model explains both divergence and the existence of several types of steady states. In this and the following chapters we will study other sources of divergence in growth rates and levels; institutional differences in this chapter and trade and FDI in chapter 4. In every case we obtain three types of steady states: development, underdevelopment in levels, and underdevelopment in growth rates, the last two named according to their type of divergence. Each of these three sets of countries defines a convergence club.

In the case of our model, we have three kinds of steady states, with qualitative differences between them. The R&D steady state represents developed countries, which follow trajectories parallel to the leading country (or countries), and perform a similar kind of scientific innovation. Then, there are two kinds of implementation steady state, both representing underdeveloped countries. Countries in the first set follow approximately parallel growth paths but perform implementation rather than R&D, while countries in the second set implement but diverge in growth rates.

62


When for some reason a country is able to change steady state, for example by changing its mix of institutions and policies so that ´lmp rises to lmp , 1

2

or lmp to lmp , its trajectory will change substantially, becoming a 2

3

transition between steady states. During the main part of the transition, it will display much faster growth than before then slowing down to normal growth rates as it approaches its new steady state. These faster growth rates explain to episodes of miracle growth.

Log GDP per Capita (chained US dollars)

Convergence Clubs and Transitions Between Them: an Illustration 11

United States

10

Great Britain

9

Korea

8

Chile

7

Mexico

6

China

Year

2014

2011

2008

2005

2002

1999

1996

1993

1990

1987

1984

1981

1978

1975

1972

1969

1966

1963

1960

5

India Selected Least DCs

Figure 2.4. Real GDP for the US, Britain, South Korea, Chile, Mexico, China, India; and average GDP for 28 selected very low income countries, see text. Source: Worlb Bank per capita income data at constant dollars and Federal Reserve chained dollars.

We illustrate the concept of convergence clubs and transitions over the period 1960-2014 in Figure 2.4.2 First, two developed countries are included, the US and Great Britain. This is an example where, in spite of its somewhat volatile trajectory, Britain, growing at an average rate of 3.14% during this period, converges to the US, which grew at an annual average of 2.01%. Clearly both countries are developed, and the graph illustrates convergence between them. Next, Figure 2.4 includes two Latin American countries, Chile and Mexico. These respectively grew in approximately parallel paths at average rates of 2.71% and 3.00% over the period, faster than the US, but slower than Great Britain. From 1960 to 2014 Chile’s income per capita grew from 18.3% to 26.6% of US income per capita, while Mexico’s grew from 11.2% 63

DAVID MAYER-FOULKES

to 19.0%. It would take Mexico another 279 years to catch up with the US if this 54 year period is representative. This period included part of the Mexican Miracle, with an average growth rate of 4.2% from 1960 to 1970, and the Miracle of Chile,3 an average growth rate of 9.2% from 1985 to 1997. The period also included poor growth. Chile’s income was lower in 1985 than in 1960 (providing a potential for its later catch up), and Mexico only grew an average of 1.2% from 1980 to 2005. Thus, in spite of the different timing of the substantial policy changes that occurred through the period in both countries, the income trajectories are consistent with both Chile and Mexico belonging to broadly similar capabilities of technological change that qualifies as implementation and not R&D, which may be what situates them in a broadly similar income trajectory. This broadly similar status includes a broadly similar institutional and policy mix, and broadly similar initial state of development. Chile and Mexico are underdeveloped in levels. Next, Figure 2.4 includes South Korea, whose income per capita grew at an annual average of 12.3% from 1965 to 1989, since then growing at an average of 4.25%. This is convergence to development. It occurred endogenously as a result of a specific policy mix, not just free markets, in which government supported domestic technological development based on export markets to the US made available for geopolitical reasons. The result is a win-win situation for both countries since Korea’s development does not diminish the US but instead complements it. Finally we consider the lowest incomes. For purposes of definiteness and comparison we select from amongst the countries having data for the period, those having the lowest average GDP and lying outside Latin America, excluding China and India. These are 28 countries.4 Then we consider China and India separately. The three graphs follow approximately similar trajectories until 1993. From 1960 to 1993 China’s per capita income grew an average of 0.07% a year, India’s an average of –0.35%, and the 28 selected countries an average of –0.70%. This is consistent with stagnation. From 1993 to 2014 the average growth rates were 13.2%, 6.2% and 3.1% respectively. At least until 1993 all of these countries were underdeveloped in growth rates. In the case of India and the 28 selected countries, it is not clear yet whether these growth represents a change of steady state to underdevelopment in levels, or a period of catching up within the underdevelopment in growth rates steady state (as we saw in the case of Mexico and Chile, which remain in their steady state). China’s growth certainly qualifies as miracle growth. Its trajectory is consistent with an as yet incomplete steady state transition from underdevelopment in growth rates to at least underdevelopment in levels 64


(that is, from semi-stagnation to parallel growth). It is not clear yet whether China will become fully developed, an R&D leader, or remain mainly an implementer of technologies developed elsewhere – underdeveloped in levels. Figure 2.4 is consistent with the existence of multiple steady states. What this means is that underdevelopment in levels or in growth rates can occur in market economies in the presence of technological transference, and that policies facilitating technological change are necessary for development. Here each country’s market is independent. In Chapter 4 we consider trade and FDI.

Log Productivity

A B

0

C

0

t0

Figure 2.5. Introduction of R&D at t0 and the Great Divergence according to the model.

The Great Divergence The model presented above can be used to model the Great Divergence (see Figure 2.4 in Chapter 1) with the emergence of R&D. Suppose that before t0 all countries were on parallel growth paths, with some leveldifferences attributable to differences in country-specific parameters. The method of technological change would have been the pragmatic creativity called implementation. Figure 2.5 depicts the evolution of productivity in 65

DAVID MAYER-FOULKES

three countries, one in each convergence group. Country A has the "best" parameter values (i.e. those most favorable to innovation) and country C the worst. After t only those countries whose productivity level exceeds some critical value will have a skilled enough labor force to begin using modern R&D immediately. (In Figure 2.5 this is only country A.) Those that do will start growing faster as a result of using the more productive (R&D) method of technology investment. All others will continue using implementation, and will start to fall further behind country A. As the other countries fall further behind, technology transfer will start to pull their growth rates up, but until their growth rates have caught up with the growth rate of the frontier, the erosion of absorptive capacity engendered by their increasing technological backwardness will tend to weaken the force of technology transfer. In countries that are not too far behind to start with, absorptive capacity will remain strong enough to eventually put them on growth paths parallel to that of country A, but with a permanently bigger gap in productivity levels, as in the case of country B; the initial gap that was due to different parameter values is now amplified by the fact that country A has adopted modern R&D while country B has not. But if the country starts too far behind, as with country C, then the erosion of absorptive capacity will weaken the force of technology transfer to such an extent that, although the country will continue to grow forever, its asymptotic growth rate will be strictly less than the common long-run growth rate of countries A and B. As in the historical record, the dominant long-run pattern in Figure 2.1 is a great divergence – a widening in the distribution of the cross-country distribution of productivity – but the dominant pattern in the most recent era is club-convergence, with countries like A and B all growing at some common rate and the poorest countries like C growing more slowly. HM presents this argument formally. The article also includes more information and discussion, for example of Lucas’ (2000) position conceptualizing the Great Divergence as a transitory phenomenon, in terms of the adoption of institutions.

IV. Financial development and convergence We discussed above how the current skill level can define an absorptive capacity and therefore limit the impact of technology transfer that can generate an advantage of backwardness. We now explore the hypothesis that financial constraints can define an absorptive capacity and prevent 66


poor countries from taking full advantage of technology transfer, causing some of them to diverge from the growth rate of the world frontier and others to diverge in levels. We introduce credit constraints in the innovation function. These credit constraints derive from an agency problem that limits an innovator’s access to external finance. Specifically we assume that an innovator can defraud her creditors by hiding the results of a successful innovation, at a cost that depends positively on the level of financial development. Because of this, in equilibrium the innovator’s access to external finance will be limited to some multiple of her own wage income, which is proportional to the country’s technological level. Therefore a technological laggard can face a disadvantage of backwardness that counteracts Gerschenkron’s advantage; that is, the further behind the frontier it falls the less its innovators will be able to invest relative to what is required in order to keep innovating at a given rate. The lower the level of financial development in the country the greater will be this disadvantage.

Variable

Nt 1 n~  t

~(n)



 c



a() H 1 ( at )

H 2 (at )

Meaning quantity of general good invested in innovation inverse innovation function resulting innovation function linear parameter of inverse innovation function quadratic parameter of inverse innovation function constant in defrauding cost function, measures financial development resulting credit constraint parameter, measures financial development resulting relative technological level defining credit constraint level next periods technological level with unconstrained innovation next periods technological level with constrained innovation

Table 2.3. Additional variables used in the Aghion, Howitt and Mayer-Foulkes (2005) model of the impact of financial development on development and underdevelopment. 67

DAVID MAYER-FOULKES

We define the instantaneous economy as in the previous model, except that we set   1. Then we change the modelling of innovation so as to include the credit constraint. Innovation In each sector, let the R&D investment needed to innovate at any given rate

t

is governed by the cost function:





Nt 1 = n~t At = t  t2 /2 At  ,  > 0, where

(2.35)

Nt 1 is the quantity of general good that must be invested. As above,

~ by At to recognize the “ fishing-out” effect; the further we multiply n ahead the frontier moves the more difficult it is to innovate. We also use ~ Namely, an intermediate producer the inverse of the R&D cost function n who invests the amount n At in R&D will innovate next period with probability:





~ n  = n~ 1 (n) =  2  2n   / .

(2.36)

~0 = 0, ~n  > 0 and ~n  < 0. Finally, we assume: Note that 

 <  <    . This condition guarantees that the equilibrium probability

(2.37)

t

will lie strictly

between 0 and 1. In equilibrium payoff:

t

will be chosen so as to maximize the expected net

t At  n~t At in each sector, subject to credit constraints.

68

(2.38)


Equilibrium innovation under perfect credit markets Suppose that each innovator can borrow (from other young people) unlimited quantities at the going rate commitment to repay. Then

t

r =  1 1 subject to a binding

will be chosen so as to maximize (2.2.38)

with no constraint. This implies that

t =   , where:

  =   / ,

(2.39)

with corresponding equilibrium R&D expenditure:

 

Nt1 = n~  At = n At .

(2.40)

We are now in the case of constant  that we discussed above, with all countries converging to the same growth rate. See equation (2.2.21) and Figure 2.1a. As before, we suppose that the growth rate g of the global technology frontier is determined by the pace of innovations in the leading countries, none of which is credit constrained. For simplicity, assume there is just one leader, labeled country 1. Then:

g  =   = 

 1 1   1 1

(2.41)

where  > 0 is a spillover coefficient and the subscript 1 indicates a parameter value in country 1.

Credit constraints Now suppose that credit markets are imperfect. Each entrepreneur at the end of period t is a young person with access to the wage income

wt . Thus

Nt in an R&D project she must borrow Nt  wt . We assume that by paying a cost cN t she can defraud her creditors by hiding the to invest

proceeds in the event that the project is successful. Since she will not be 69

DAVID MAYER-FOULKES

lent so much that she will be willing to pay the cost to defraud her creditors, this implies that in equilibrium the entrepreneur cannot borrow more than a finite multiple of her wage more than hiding cost

wt

wt (see AHM), and therefore she cannot invest

in innovation, where   1 depends positively on the

c.

This credit constraint will be binding if the unconstrained optimal 

investment n At 1 is strictly greater than the innovator’s investment capacity wt , or equivalently, after dividing through by At 1, if:

n > at,

(2.42)

where    (1   ) /(1  g ). We represent financial development by the cost parameter c, or equivalently by the credit multiplier  (or by  ) , on the grounds that a highly developed financial system protects creditors by making it hard to defraud them. 

We see from (2.2.42) that: (i) for given level of technological development

at of the country, domestic firms are more likely to be credit constrained if

financial development  is lower; (ii) for given level of financial development  firms are more likely to be credit constrained the further

the country is behind the technological frontier (i.e., the smaller is at ). This is the “ disadvantage of backwardness” induced by the existence of credit constraints. Our model implies that, holding the credit multiplier  (or  ) constant, among those countries that are financially constrained external

financing (equal to (  1)wt ) is higher in those that are closer to the technological frontier. However, the opposite is true among those countries that are not constrained, as the amount of external financing is then entirely 

determined by the gap between the R&D cost n At which is proportional to the frontier productivity At and the amount of internal finance which is proportional to current domestic productivity. Thus, firms in more advanced countries with

at > n /  a 

70

(2.43)

DEVELOPMENT AND UNDERDEVELOPMENT UNDER GLOBALIZATION 

will invest the unconstrained amount n At 1 in innovation and therefore will innovate with probability with

  , whereas firms in less advanced countries

at < n /  a 

(2.44)

cannot invest more than wt = at At 1 and therefore will innovate with probability

~(at),

(2.45)

~ is given by (2.2.36). This raises the where the innovation technology  question of why a constrained entrepreneur at t  1 would not instead target a lower technology level Bt < At , which would be less expensive given the assumption that the cost of innovating at a given rate is proportional to the targeted technology level. In Aghion, Howitt and Mayer-Foulkes (2004) we answer the question by showing that this alternative would be dominated, from the entrepreneur’s point of view, by the strategy of always targeting the frontier. This relies on the fact that the ~(n) has an elasticity less than one, which in turn innovation function  ~( ) is strictly convex with follows from the fact that the innovation cost n

n~0 = 0. In the restricted case at 1 will therefore be determined

according to:

1  ~ at  a  H a . at 1 = ~  at   t 2 t 1 g

(2.46)

Three dynamic patterns

at will evolve according to the unconstrained dynamical system (2.33) when at  a() and according to the constrained system (2.2.46) when at < a(). Thus:

In general, the country’s technology gap

71

DAVID MAYER-FOULKES

at t = H (at )  minH1 (at ), H2 (at ).

(2.47)

Note that H1 is a linear function with positive vertical intercept and slope between 0

and 1.

H 2 is an increasing concave function when at  mina(),1, with H 2 (0) = 0 and H 2 (0) = 1  (1 g)1.

Countries will fall into three groups, defined by the level of financial development . The evolution of the technology gap is illustrated for each case in Figures 2.6a, 2.6b and 2.6c. Figure 2.6a shows a country for which the financial system is well developed, so that at sufficiently high levels of average technology relative to the frontier

at , the country is unconstrained. 

Therefore it will be unconstrained at the steady state level a , which will be independent of small changes in the level of financial development.

Figure 2.6a. A country with the highest level of financial development. Source: AHM. Figure 2.6b shows a country with a less developed financial system, and hence with a lower curve

H1at  . In this case the dynamics reach their 

steady state while the credit constraint holds. Hence the level a of the steady state depends on the level of financial development. However, the

72


growth rate is development.

g  and will not change with marginal changes in financial

Figure 2.6c shows a country with an even less developed financial system. In this case the relative technological level at tends to zero and we have divergence in growth rates. The asymptotic growth rate will be increasing in the level of financial development.

Figure 2.6b. A country with a medium level of financial development. Source: AHM.

73

DAVID MAYER-FOULKES

Figure 2.6c. A country with the lowest level of financial development. Source: AHM.

The proofs of these statements are in AHM. The article also contains an empirical study supporting these predictions, and showing that at low levels of financial development, countries diverge. As we mentioned before, this model also explains the existence of three kinds of steady state trajectories, representing development, underdevelopment in levels, and underdevelopment in growth rates. The steady state trajectories form a continuum that depends on the level of financial development.

This chapter is based on Howitt and Mayer-Foulkes (2005) and Aghion, Howitt and Mayer-Foulkes (2005), refered to as AHM. 2 Mayer-Foulkes (2007) has a cross-country analysis of convergence clubs and transitions for the period 1960-1995. Much of the club corresponding to underdevelopment in levels is in Latin America, while underdevelopment in growth rates is situated in Sub-Saharan Africa. 3 See Krugman (2015b). 1

74


4 Burundi, Malawi, Nepal, Rwanda, Burkina Faso, Uganda, Central African Republic, Niger, Bangladesh, Lesotho, Chad, Sierra Leone, Togo, Madagascar, Benin, Liberia, Kenya, Pakistan, Sudan, Ghana, Mauritania, Nigeria, Senegal, Sri Lanka, Cameroon, Zambia, Papua New Guinea, Zimbabwe.

75

Main ideas Since the introduction of the modern theory of economic growth in the late 1980’s, empirical studies have established that differences in income levels are mainly due to differences in technological levels, with differences in capital and human capital levels intervening as well. Endogenous technological change is introduced here, with innovators investing resources for finding new ways of increasing productivity. The return to their investment accrues as profits due to market power in the sale of new intermediate goods that incorporate this increased productivity. Innovation consists of improving the quality of the intermediate goods. These intermediate goods are bought by final goods producers. We consider a continuum of intermediate goods to smooth the uncertainty generated by the innovation process. The model considers two countries whose only contact is the exchange of ideas. The first country is a technological leader. Its innovation generates a stock of leading edge technologies. In turn, innovators in both countries use this stock of ideas to improve their intermediate goods. The second country is a follower country because it uses the stock of knowledge generated in the first country. In a first instance we consider that all innovators can perform innovation independently of their technological level, and obtain a new, leading edge technological level for their intermediate good. This introduces a convergence effect, an “advantage of backwardness.” In a second instance we consider that an absorptive capacity is also a factor of innovation. This introduces a “disadvantage of backwardness” which can generate divergence in levels or in growth rates, depending on parameters defining each country’s competitiveness or innovativity. To explain the Great Divergence we introduce two innovation functions. The first is a pragmatic creativity that we call implementation. The second is R&D, which uses scientific methods to generate new technologies. This is a more productive method of innovation, but requires a threshold knowledge level to be viable, that

76


3 HUMAN DEVELOPMENT I.

Human capital

As we mentioned in the introduction to Chapter 2, studies have shown that both skills and knowledge are fundamental factors of production and therefore of economic development. In Chapter 2 we used Aghion and Howitt’s Schumpeterian approach to endogenous technological change to show how firms produce specific knowledge for production (that is, technology), usually using knowledge produced elsewhere, by other firms or by universities. Skills, on the other hand, are obtained through a process of training or learning. This takes time for the person learning, and may use resources such as a teacher’s skills and time, and material inputs of various kinds. Overall, the process of acquiring skills is a process of investment, and skills therefore constitute a form of capital – human capital – whose yield takes the form of higher wages for skilled labor.5 As industrialization proceeded and the role of science increased, a variety of skills became important across the industrial workplace, from literacy through a whole range of technical, professional and scientific skills. Different levels of schooling spread across the population, and public policies were put into place to support it. Education became an important determinant of individual income. This process advanced much further in developed countries. The divergence in schooling was analogous to the divergence in incomes (Figure 3.1a). In the case of literacy (Figure 3.1b), because there is an upper bound (100% literacy), what we see are delayed, incomplete transitions. As we have mentioned, education is also related to

77

DAVID MAYER-FOULKES

the demographic transition. Parents intending to educate their children more also have fewer children.

12 Europe & offshoots

10

Schooling (years)

Japan, Korea, Taiwan 8

Eastern Europe

6

Latin America China

4

Other Asia

2

Africa South Asia

0 1870

1910

1950

1980

2000

Figure 3.1a. The schooling transition, 1870-2000. Based on Morrison and Murtin (2007, Table 2) as reported by Patrinos and Psacharopoulos (2011)

100

Japan, Korea, Taiwan

Literacy (%)

80

Europe & offshoots Eastern Europe

60

Latin America 40

Other Asia China

20

Africa South Asia

0 1870

1910

1950

1980

2000

Figure 3.1b. The literacy transition, 1870-2000. Based on Morrison and Murtin (2007, Table 2) as reported by Patrinos and Psacharopoulos (2011) Conversely, the access to education has become one of the determinants of lifelong income. Low education access, poverty, inequality, and underdevelopment, are different facets of delayed and divergent development. In today’s globalized world, most people’s income depends directly on their skills. 78


The impact of individual education on income has been confirmed by estimating, for example, the private returns to education investment. Citing from Patrinos and Psacharopoulos (2011), the rate of return to an investment in a given level of education can be defined and estimated as the rate of discount ( r ) that equalizes the stream of discounted costs and benefits at some point in time. In the case of university education lasting five years, for example, the formula is:

(Wu  Ws )t 5  (Ws  Cu )t (1  r )t  t t 1 ta 1 (1  r )



42

(3.48)

where Wu Ws is the earnings differential between a university graduate (subscript u) and a secondary school graduate (subscript s, the control

Cu represents the direct costs of university education (tuition, fees, books), and Ws denotes the student's foregone earnings or indirect costs.

group).

Psacharopoulos and Patrinos’ (2004) summary of the results of many studies is shown in Table 3.1. The Table shows that primary, secondary, and higher education have high returns, particularly in low and middle income countries. Per capita income group

Educational level Primary Secondary Higher Low income 21.3 15.7 11.2 Middle income 18.8 12.9 11.3 High income 13.4 10.3 9.5 Table 3.1. Social returns to investment in education by level and per capita income group (%). Source: Table 2, Psacharopoulos and Patrinos (2004). Education is not the only investment in human beings with impacts on both productivity and welfare. Nutrition and health have been found to have important impacts on long-term income and economic growth. The World Health Organization (WHO) Commission on Macroeconomics and Health (Sachs, 2001) recommended taking these impacts into account in cost-benefit analysis of health provision. Currently non-communicable diseases (NCDs) are also found to have important economic impacts. Thus a full conception of human capital includes health, education, early child development, cognitive development, and so on.

79

DAVID MAYER-FOULKES

While historical and macroeconomic studies find that health and nutrition have significant impacts on long-term income and economic growth, disentangling the underlying causal channels presented a major challenge to microeconomic research. Studies found that only a fraction of the overall impact of health on income is accounted for by the productivity effects of adult health on income. Instead, what is found is an intergenerational impact that operates through early child development. Child nutrition and health have a significant effect on education and therefore on adult income.6 This is related to research results on the ‘gradient’ of adult health along income using databases for developed countries which, first, place childhood health at the origin of the gradient, and second, show childhood health is a function of parental socioeconomic status.7 Human capital accumulation has an intergenerational nature. Parents invest in their children’s human capital, with early child development playing a key role. From the point of view of the child’s life, this investment is often suboptimal. In economic terms this means that the child would have been willing to borrow for this investment, and to repay with the increased income later on in life. In fact this is corroborated by the returns to education shown in Table 3.1, which are higher than the normal interest for capital. If government supports education from taxes on labor, in effect it is providing this credit. Thus, market failures characterize human capital investments and lie at the core of the process of human capital accumulation and its relation with economic growth. In what follows we extend the concept of human capital accumulation to a wider concept, human development, and explain with a model the persistent inequality that can characterize it. Here we use a dynamic trap, allowing for continually trapped human capital levels that can nevertheless grow over time, analogous to the concept of divergence in levels. This also allows us to examine how human capital accumulation relates to technological change and underdevelopment.

II. Human development As mentioned above, health joined education in a unified conception of human capital. Nobel Prize winning historical studies by Fogel (1991, 1994a, 1994b), Fogel and Wimmer (1992) find that a third or even one half of the economic growth in England from the late 1700’s to the late 1900’s is due to improvements in nutrition and health. Arora (2001) found 80


comparable results for seven advanced countries using one hundred to one hundred and twenty five year time series of diverse health indicators. This line of research concluded that the synergism between technological and physiological improvements produced a rapid, culturally transmitted form of human evolution that is biological but not genetic. This process, which continues in both rich and developing countries, is called techno-physio evolution by Fogel (2002). The present day role of health in raising income and education is confirmed in a series of macroeconomic studies. However, their microeconomic counterparts measuring the impact of adult health on productivity found a smaller impact than could be expected from the macroeconomic and historical studies. In fact, the indirect impact of nutrition and health on the productivity of education and therefore on income was much stronger. The momentous secular rises in stature, weight, life expectancy, and education that occurred through the Industrial Revolution give a whole new meaning to the concept of long-term human development. For example, average stature rose from 164 to 181 cm in Holland between 1860 and 2002 and from 161 to 173 cm in France and Norway between 1705 and 1975. Average weight rose from 46 to 73 kg in France and Norway in the same period. Life expectancy rose from 41 to 78 years in England between 1841 and 1998; and from 29 to 60 years in India between 1930 and 1990. Schooling rose from 2.3 to more than 11 years in England between 1800 and the 1980’s. Fogel (2002) refers to this process as techno-physio evolution, because it depends on the interaction of human beings with technology. Human development includes other dimensions such as cognitive development, fertility preferences, and ethical development, whose changes can only be surmised. There is strong evidence that a transition in population height occurs with the process of development. Population height can be measured in units of standard deviations from the mean, using normal development tables for children and adults, drawn for specific ages. These are called Height for Age Z-scores (HAZ). Figure 3.2a shows the population height transition across 43 regions of Bolivia, Brazil, Guatemala, and Peru. Comparison with the 45 degree line shows that this has been a divergent process. If we consider this diagram as the phase space of a dynamic process, the lag between the bottom and top of the graph is approximately 5 generations. The information is obtained from comparable Demographic and Health Surveys (DHS) for Bolivia (1994, 1997), Brazil (1996), Guatemala (1995, 1998), and Peru (1992, 1996, 2000). The 43 regions have representative samples in the DHS surveys. Figure 3.2b shows the correlation between daughters HAZ and the socioeconomic status of their households of origin.8 81

DAVID MAYER-FOULKES

0.5

Daughter's HAZ

0

-0.5

-1

-1.5

45 degree line

-2

-2.5

-3 -3.5

-3

-2.5

-2

-1.5

-1

-0.5

Mother's HAZ

Figure 3.2a. Intergenerational transmission of Height for Age ZScore from mothers to daughters in 43 regions of Bolivia, Brazil, Guatemala and Peru (regions with representative samples in the DHS surveys). 0.5

Daughter's HAZ

0 -0.5 -1 -1.5 -2 -2.5 -3 0

0.2

0.4

0.6

0.8

1

Socio Economic Status

Figure 3.2b. Intergenerational transmission of Height for Age ZScore according to socioeconomic status in 43 regions of Bolivia, Brazil, Guatemala and Peru (same regions).

82


The long-term rise in stature is related to long-term improvements in nutrition and health. Related research has found that most stature loss is determined irreversibly in the first two years of life, and is a predictor of life-long health and longevity. A wave of research has focused on the biological mechanisms through which these interconnections occur. These concerns have led to a focus on early child development (ECD), the combination of physical, mental and social development in the early years of life. Numerous links are now known between malnutrition, beginning in uterus, early infection, the crucial period of brain development in uterus and shortly after birth, and a series of adult health issues. Amongst the policy alternatives that have been explored to improve attained human development levels are Early Child Development (ECD) programs, commonly addressing nutrition, health, cognitive development, and social interaction in early childhood. Attention has also been directed to micronutrient deficiency. Human development is typically subject to what may be thought of as a sequence of market failures that occur as economic development proceeds. These require public action to improve the efficiency and pace of the process. Several kinds of such barriers to human capital accumulation have been modeled theoretically. A first example, occurring at very low levels of income, is the possibility of a low productivity trap due to low nutrition, addressed by the efficiency theory of wages, whose study has documented substantial effects of nutrition on labor productivity. Nutrition has substantial effects on economic growth both directly and through life expectancy, and possibly schooling. Child labor traps have also been studied. A further kind of poverty trap, occurring at later stages, when education becomes important, is found in Galor and Zeira (1993), who show that increasing returns in skill acquisition leads to multiple equilibria in the presence of credit constraints to human capital accumulation. This kind of trap can occur successively for obtaining literacy, primary education, secondary education and higher education. For example, De Ferranti et al (2003, page 86) argue that in most Latin American countries there is a ‘bottleneck’ at secondary education. In the US, neighborhood effects on diverse aspects of human development have been documented. The process of human development (including techno-physio evolution) can be understood as an intergenerational cycle of investment in nutrition, health and education that is beset by market failures. This process provides the main economic inputs for production and technological change; labor, skills and knowledge. In turn, the intensity of intergenerational human development depends on the resulting income and technology levels. 83

DAVID MAYER-FOULKES

In our model, in any given country there will be a potential level of human development to which the wealthy have access. The corresponding level of health, education and income may differ across developed and underdeveloped countries. However, the poor may not have access to the potential level of human development in their countries, trapped in a lower level of human development that lags behind. We thus model human development with a dynamic poverty trap. Similarly to underdevelopment in levels in Chapter 2, here we find high and low steady states for human development in each country. As technology improves, both the high and the low steady states are subject to growth. Nevertheless, the lag between the steady states will persist. The data shows that these lags can be very significant.

III. The human development trap model Our discussion in the previous section leads to the following stylized facts about long-term human development. First, technological improvements lead to techno-physio evolution; improvements in stature, health, longevity and education. Second, higher human development provides inputs for raising technological levels. Third, the presence of market failures inhibits the human capital investment that each new generation must make to achieve current levels of human development; only the wealthier part of the population may reach the current human development potential, while the poorer part may remain trapped away from achieving its full potential. Here we formalize these ideas in a concise economic model of a human development trap and its interaction with economic growth. A classic example of a static human capital trap is provided by Galor and Zeira (1993). This trap is based on the need for an indivisible investment for acquiring skills, and on the presence of a credit constraint not allowing the poor to make the investment. I take this model, as presented in Basu (2003, section 3.4), as point of departure to construct a dynamic human development trap model. Consider an economy with St skilled workers, each with human capital At, so that the aggregate human capital stock is Ht = AtSt, and with Lt unskilled workers. A single good is produced according to the production function

Yt  ( At Lt )  H t1   At Lt S t1  ,

84

(3.49)


with technology level also At. Wage rates for labor and human capital are therefore given by

wtL   At ( S t / Lt )1  , wtH  (1   ) At ( Lt / S t )  ,

(3.50)

and are proportional to At. Note that in contrast to Galor and Zeira’s (1993) model, we do not assume increasing returns to human capital. Instead, what is assumed is that scarcity of human capital, which will result from barriers to its acquisition, results in a higher, endogenous wage gap between skilled and unskilled wages. In period t, each young person receives a bequest xt, and decides on investing hAt units to acquire an indivisible package of At+1 units of human capital consisting of ECD, health and education (“becoming skilled”).9 The L

alternative is to work earning η wt , where 0 < η ≤ 1 represents the ratio of child to adult wages. Any remaining bequest is saved and earns an interest rate rt. For simplicity, a full credit constraint is assumed: it is impossible to borrow to acquire nutrition, health and education. In the second period of H

life, each adult earns yt+1, the sum of her skilled or unskilled wage ( wt 1 or

wtL1 ) and accrued savings. Each adult consumes ct+1 and decides on a bequest bt+1 to her one child, who receives a bequest xt+1 = bt+1. Young people maximize: 

1

max Ut = ct 1bt 1   t u At , 0 < α < 1, H

(3.51)

subject to the budget constraint ct+1 + bt+1 = yt+1. The utility function contains two terms. The first represents preferences over consumption and bequests. The second term represents the non-labor benefits of acquiring health and education. t takes the value 1 if skills and health capital were acquired, and 0 otherwise, while uHAt (with uH constant) measures the nonlabor benefits of human capital. For simplicity, so as not to include consumption in both periods, the interest rate is assumed to be exogenous. Suppose that, independently of the ratio of skilled to unskilled labor, parameters are such that anybody receiving a bequest xt ≥ Ath will decide to acquire human capital. Suppose further that given a technological level At, descendants of skilled workers can acquire skills, and that their attainable human development is bounded. For now, suppose At+1/At = 1 + gt is constant and exogenous and define the deflated variable ~ x t = xt/At. 85

DAVID MAYER-FOULKES

x High in Proposition 1. Under these assumptions, a higher steady state ~ x which skills are acquired exists. A lower steady state ~ in which skills are not acquired also exists (see Figure 3.3) if the skilled to unskilled workers ratio lies below the critical level Low

1

 (1  g t  (1   )(1  r ))h  1  St  .   Crit   Lt  ( 1  )(  ( 1 r ) 1 g )     t  

(3.52)

In this case St and Lt remain constant. For higher growth rates gt, higher skilled to unskilled ratios are necessary to escape the trap, and returns to education are higher. Proof. Maximization of the utility function assigns the following consumption and bequests: ct+1 = αyt+1, bt+1 = (1 – )yt+1. The utility levels are given by Ut = ε()yt+1 + δtuHAt, , with ε() ≡ (1 – )1–. Unskilled and skilled workers’ income is given by:

ytL1 = (xt + η wtL )(1 + r) + wtL1 , ytH1 = (xt – Ath) (1 + r) H

+ wt 1 .

(3.53)

Note that if at any time the ratio of skilled to unskilled labor rises to St/Lt > (1 – β)/β, then human capital becomes oversupplied. Some skilled workers L

will work at unskilled jobs, and wages will equalize at wtH = wt = ε(β)At. L

In general, wt ≤ ε(β)At ≤ wtH . Human capital will be acquired just for its non-labor benefits if uH is large enough:

utH1 – utL1 ≥ 0  ε(){–(Ath + η wtL )(1 + r) + wtH1 – wtL1 } + uHAt ≥ 0  uH ≥ ε()(h + ηε(β))(1 + r). H

(3.54)

L

This is assumed. Note that since wt 1 – wt 1 is proportional to At+1, returns to skills rise with economic growth, which is given by function G. For the skilled workers’ young to afford skills, it is necessary that b(Ath) ≥ At+1h  (1 – )(1 – β)At+1(Lt+1/St+1)β ≥ At+1h

86

(3.55)


 (1 – )ε(β) ≥ h. This is also assumed. Hence anybody receiving a bequest xt ≥ Ath will acquire human capital, and the full bequest dynamics are given by: xt+1 =

(1   )(( xt  wtL )(1  r )  wtL1 ) bt  b( xt )   H  (1   )(( xt  At h)(1  r )  wt 1 )

Deflated bequest to next generation, x ~ xt 1  t 1 At 1

~ ~ xt 1  b ( ~ xt , S t / Lt )

xt  At h, xt  At h.

(3.56)

~ xt 1  ~ xt St/Lt < Crit St/Lt = Crit St/Lt =/(1–)

St/Lt =

/(1–) St/Lt = Crit

High Low High x NoTrap x WithTrap x WithTrap ~ h ~ ~

Deflated inheritance ~ xt  xt / At

Figure 3.3. Diagram for the human development trap. Children who inherit more than hAt (deflated to h) chose to be skilled. They earn higher wages and bequest more to their children. Wages for skilled and unskilled workers are endogenous and depend on the aggregate ratio of skilled to unskilled workers St/Lt. If this ratio lies below the critical level Crit, there is a trap, which otherwise disappears. Near the critical level Crit, workers receiving bequests below hAt will go through a prolonged transition towards the higher steady state.

87

DAVID MAYER-FOULKES

x Low exists when the unskilled Note b(0) > 0. Hence the lower steady state ~ cannot acquire human capital, that is, lim x t  At h  b ( x t )  hA t  1 .

(3.57)

Under (3.55) and (3.57), the skilled to unskilled ratio remains constant, so St+1/Lt+1 = St/Lt. Hence we can write the deflated bequest dynamics

    1 r S S xt ( t )1  xt  h,  ( t )1 , ~ (1)(~ 1 L g L  S  ~ t t t   t   ~ xt1  b(~ xt , )   (3.58) Lt  ~ St 1  1 r ~ xt  h, (1)(xt h) (1 )( ) ,  1 gt Lt    ~

This is depicted in Figure 3.3. The trap holds if b ( h , S t / Lt )  h , yielding condition (3.52). The right hand side of (3.52) is an increasing function of gt, so the trap’s hold becomes tighter if the rate of growth rises. For the higher

x High to exist, it is still necessary that the gradient of steady state ~ ~ xt  h be less than 1, that is (1 – )(1 + r)/(1 + gt) < 1. b (~ x , St / Lt ) along ~

Together with (3.55), this means human development is bounded and viable, as assumed.■ At both steady states, wages rise with the technological level At, implying growing levels of nutrition and health even for the unskilled. However, the unskilled population is trapped away from reaching the full current human development potential, which is only accessible at the higher steady state. In this sense, the model defines a dynamic poverty trap. The rising indivisible human capital investment may be thought to correspond to successive investments at different stages of development. For example, successive stages of education, such as literacy, and the equivalents of primary, secondary, tertiary and higher education, have at different periods been the privilege of the rich. In Mexico, current unskilled wages may be thought to include, for many, free public schooling for children up to 6 or 9 years of education, as well as some level of sanitation. But they do not necessarily afford a full measure of ECD for children, nor access to higher education. As long as the trap exists, children would like to borrow so as to obtain education. Thus, investing in education offers higher returns than the

88


interest rate. Under these conditions, we say that education offers extraordinary returns to investment. If the dynamic trap condition (3.52) does not hold, a prolonged transition may nevertheless take place, in which several generations must pass before the offspring of a poor dynasty achieve a full human development status. Under a prolonged transition, market failures slow the transition to a steady state and result in systematic, prolonged, underinvestment in human capital. The transition corresponding to the perfect market counterpart would be faster. B

A

C D

D D

A

A

1984

1989

1994

C

C B

C

B

B

B

D

C

D

D B

A

A

2000

A

2006

2012

Figure 3.4. Distribution of households according to female and male spouses' schooling (data from ENIGH 1984, 1989, 1994, 2000, 2006, 2012). Households restricted to two spouses of opposite sex ages 25 to 30. The axis pointing left and away according to the perspective represents education of the female spouse and the axis pointing right of the male spouse. The axis ticks represent the following schooling intervals: [0,2], [3,5], [6,8], [9.11], [12, 14], and 15 or more years. At points marked A, B, C and D, both parents have 3-5, 6-8, 9-11, and 15 or more years of schooling respectively.

IV. Evidence for a Human Development Trap in Mexico In an empirical section, Mayer-Foulkes (2008) gives evidence for the following assertions for the case of Mexico, using data ranging over the 89

DAVID MAYER-FOULKES

period 1985-2000. 1) Education has increasing returns for adult income. 2) Early childhood health has a strong, probably causal relation on permanence in school throughout the educational career. 3) The population is classified itself into two social classes, those with lower secondary schooling or less, and those above. Investment in schooling in the lower group mainly responds to public investment in education. Each of these findings is supported by other studies. Figure 3.4 shows updated histograms for the distribution of households according to female and male spouses' schooling data from ENIGH 1984, 1989, 1994, 2000, 2006, 2012.

Figure 3.5. Distribution of households according to female and male spouses' schooling (data from ENIGH 2006, 2012) for rural and urban areas, see caption for Figure 3.4. Rural is defined as a locality with less than 2,500 inhabitants.

90


Consistently with a dynamic poverty trap, we see here a two-peaked distribution of households, with the bottom peak advancing from the interval [3,5] to the interval of [9,11] years of schooling, and membership of the higher peak at 15 or more years of schooling increasing. Through the years portrayed, the lower peak moves forward from lower and completed primary school to completed secondary school (which corresponds to 9 years of education). This level of schooling became obligatory in Mexico in 1993, so the transition was achieved through the application of public policies. The higher education peak also rises through these years. Public policies also supported higher education. Figure 3.5 shows similar histograms for with ENIGH data for 2006, 2012, selected for rural and urban areas, to see how closely the poverty trap mechanism interacts with the rural-urban divide. The poverty trap is broadly similar but somewhat deeper in rural areas, with a smaller higher peak and with the lower peak subdivided between completed primary and completed secondary schooling. Urban areas have a higher proportion of the population in the higher peak, and the lower peak is centered at completed secondary school. There were only slight changes between 2006 and 2012. 100

80

60

40

20

0 Europe & Central Asia

East Asia & Latin America Middle East & South Asia Pacific & Caribbean North Africa

Sub‐Saharan Africa

Communicable, maternal, perinatal, and nutritional conditions Chronic or noncommunicable diseases

Figure 3.6 Worldwide percentage of deaths by cause and World Bank region (excluding high-income countries, 2002) Source: Mathers et al. (2003) 91

DAVID MAYER-FOULKES

V. Non Communicable Chronic Diseases and Human Development Non-communicable chronic diseases (NCDs) have become important disease burdens in both developed and underdeveloped countries, and in both rich and poor populations (see Figure 3.6). They have important microeconomic and macroeconomic impacts. The World Economic Forum and the Harvard School of Public Health estimate that NCDs will cost the world economy $47 trillion (equivalent to 75% of global GDP) over the next 20 years. Recent macroeconomic analysis demonstrated that each 10% rise in NCDs is associated with 0.5% lower rate of annual economic growth. The fiscal costs are expected to be higher than retirement costs.10 NCDs represent the single largest global cause of mortality in people of working age. Over half of the deaths in the world are due just to these four chronic conditions – cardiovascular disease (30% of total global mortality), cancers (13%), chronic respiratory disease (7%) and diabetes (2%). These diseases —primarily heart and stroke disease, cancer, chronic obstructive pulmonary disease and diabetes—caused an estimated 35 million deaths in 2005, 80% of which occurred in low and middle income countries (or LAMICs, World Bank classification), which had 84% of the world population. These conditions were responsible for 50% of the disease burden in 2005. Death rates from NCDs are considerably higher in LAMICs than in high-income countries: estimated to be 56% higher in men and 86% higher in women in 2005. NCDs are responsible for significant premature mortality and morbidity throughout the world. In LAMICs the disease develops at an earlier age than in high-income countries. The epidemiological transitions from communicable to non-communicable diseases, itself the result of improved health, is differentiated across socioeconomic levels. This clearly places NCDs in the framework of the Human Development Trap model, as a new stage of techno-physio evolution driven by higher income or technology levels and affected by market failures. Amazingly, premature mortality and morbidity due to the four most prominent NCDs are largely preventable and can be greatly reduced. Cancer, cardiovascular disease, chronic obstructive pulmonary disease and diabetes share modifiable and preventable risk factors related to lifestyles, predominantly tobacco use, unhealthy diet, physical inactivity and the harmful use of alcohol, to which are added pollution and stress. What is striking about these harmful factors is that many of them are actively promoted through the production and advertising of harmful consumption by large, global corporations. This includes not only intensive 92


advertising of tobacco and alcohol, but also the nutrition transition, the replacement of a traditional diet rich in fruit and vegetables with a diet rich in calories derived from animal fats, and lower in complex carbohydrates, which is transforming global agriculture and is promoted especially amongst the young. These activities are known to impose large future costs on the general public and on the governments of a multitude of countries. Constructing a responsible environment in this respect poses a huge challenge in global governance.

Benefits

These risk factors reflect the major forces driving social, economic and cultural change, including globalization, urbanization and the general policy environment. Further, global forces such as those in trade and marketing are increasing the causal entrenchment of chronic diseases in all of the regions. For example, one of the health-related effects of globalization is the ‘nutrition transition’ tending to replace a traditional diet rich in fruit and vegetables with a diet rich in calories derived from animal fats, and lower in complex carbohydrates. Except for countries with a less developed infrastructure, this transition is well underway in almost every corner of the globe. Such a diet, when combined with a low level of physical activity, regular tobacco use and alcohol consumption, sets the scene for a population-wide atherosclerosis and global distribution of chronic diseases.

Unwholesome Good Immediate Positive Benefits

Normal Good

Negative future benefits subject to uncertainty and/or imperfect information

0

Time

Figure 3.7. Unwholesome goods and externalities often have an immediate benefit, as a normal good, but also negative future benefits subject to uncertainty and/or imperfect information.

93

DAVID MAYER-FOULKES

The impact of the multinational company advertising on cultural change is an important part of the dietary transition. Nestle spent $3.1 and Coca Cola $2.9 billion in advertisement in 2013, and Mars, Inc., PepsiCo, and McDonalds Corp. another $8.3 billion. For comparison, WHO’s organization-wide budget for the two years 2014-2015, was US$1.99 billion per year. Large diversified food companies, the second-largest ad spending category after cars, where expected to spend $30.7 billion on ads in 2016, up slightly from $30.4 billion in 2015, fifteen times the full budget used by WHO to deal with world health issues. Advertising by the pharmaceutical industries was expected to be $21 billion in 2016, also way beyond WHO’s capacity to regulate. Just food advertising in the US is approximately US$10 billion a year, while the cost of NCD was approximately 6.8% of GDP, or US$939 billion. To get a clearer picture of the political aspects, note that in the campaign for 2009 health reform in the US, pharmaceutical companies seeking to hold their market power and avoid government competition spent 41% more than in the previous year, when they spent about US$230 million. In 2007 their lobbying was worth US$225 million, compared to advertising worth maybe US$6.5 billion. This means that the drug industry spent on obtaining the favor of its regulator, the US congress, approximately 3.5% of what it spent on advertising: not much! The food and beverage industry lobbied even less, $26 million in 2009 (the highest ever for them).

Potential Unwholesome

Human Development

Trapped Trapped and unwholesome t Wholesome Human Development Unwholesome Human Development Figure 3.8. Wholesome and unwholesome human development. Unwholesome consumption and externalities reduce and distort human development, with impacts differentiated by socio economic levels. 94


Three of the main risk factors of NCDs, tobacco use, unhealthy diet, and the harmful use of alcohol, fall into the category of consumption. The fourth main risk factor, physical inactivity, can be considered an externality of urban living. We can define the unwholesome component of NCD risk factors as those that are consequences of human actions. Unwholesome consumption and externalities often have the characteristics shown in Figure 3.7. Unwholesome consumption and externalities have impacts that diminish the quality of human development, and that are related with irrationality in consumption, irresponsible advertising, information problems, and so on. Thus, instead of seeing an epidemiological transition that reflects health improvements in communicable diseases, what is occurring is an unwholesome epidemiological transition: health and demographic transitions impacted by unwholesome consumption and externalities. This has consequences for health, health costs, education and income across the lifecycle. Figure 3.8 explains diagrammatically what this implies for human development. Universal health coverage is an integral part of the new stage of health development. In a previous stage, a series of low cost measures such as vaccinations, sanitation and micronutrients could achieve important improvements in health. Today, some cost-effective measures for reducing NCDs are also available, such as reducing the consumption of salt, sugar, alcohol and tobacco. However, medical attention covering a series of conditions and choices has now become an important component of life expectancy and health. Effective health policies also call for sectoral collaboration to achieve a healthy food sector, healthy working environments, an informed public making healthy lifestyle choice, affordable medicine, and so on.

VI. The human development trap with endogenous technological change Let us now follow Chapter 2 to complete the Human Development Trap model with an endogenous rate of technological change. Let Nt = St + Lt be the total number of workers. Here the economy’s skill level is

(St / Nt ) At , the proportion of skilled workers in the economy multiplied 95

DAVID MAYER-FOULKES

by the technological level. Hence we can substitute   S t / N t in the model. Let us suppose for simplicity that the frontier growth rate is constant, with gt = g   ( At 1  At ) / At > 0. Equation (2.21) now reads

a t 1 =

1 1 a t    (1  at ) ,  1 g 1 g

(3.59)

where now according to equation (2.16) 1

1

  1   =       1

1

St . Nt

(3.60)

We have assumed that the leading country does not have a poverty trap, so all people acquire human capital for its non-labor benefits, even those working in unskilled jobs. The poverty trap exists in the follower country

for values of St / Nt below Crit /(1 Crit ) . Once the proportion of skilled workers reaches this critical value, the proportion of skilled workers immediately jumps to 1, with the proportion employed in skilled jobs equal to α. For definiteness we suppose that in this case in equation (3.2.11).

St / Nt is replaced by α

The steady state value a   (1  g  )  /( g    ) for dynamics (3.2.18) is increasing in  . Hence the less deep the poverty trap, so that the average level of skills is higher, the higher the technological level relative to the leading technological edge. This implies:

Other things being equal, a country with a deeper human development trap will lag further behind technologically and will therefore be less developed.

Conflict of interest between skilled and unskilled workers Consider a country with a human development trap and suppose that the government can promote both technological change and human capital investment for unskilled workers. 96


By investing in human capital for unskilled workers, some proportion of the population can become skilled, in a policy applied successively. Their wages will rise permanently for them and for their descendants. Also, the rate of technological change will rise because of the increased level of skills in the economy

Figure 3.9. Human development trap model combined with endogenous technological change. Horizontal axis is relative technological level at. Vertical technological level is proportion of skilled workers St/Nt. If this proportion is initially less that ηCrit/(1 + ηCrit), there is a human development trap and St/Nt remains constant. Otherwise the human development trap disappears and St/Nt immediately jumps to 1 in the next period, with a proportion α employed as skilled workers. Technological dynamics occur along *

horizontal line and there are three types of steady states, a0 ,

a1* , a2* ,

which lie on the heavier line, together with the point ( a2 ,  ). If the *

*

government supports technological change, the growth rate at a0 *

rises, and ai becomes ai*,TechPolicy , for i = 1, 2. If the government supports human development, the trajectory follows the dotted line. Once St/Nt reaches ηCrit/(1 + ηCrit), the human development trap 97

DAVID MAYER-FOULKES

disappears and the government switches to a technological policy, so the economy reaches a2*,TechPolicy . However, it the number of skilled workers increases (and therefore their supply), the skilled wage decreases. Even though technological change increases because of the increased level of skills in the economy, it is likely that by subsidizing technological change directly, skilled workers’ wages would increase faster. If this is the case, there can be a conflict of interest between skilled and unskilled workers, the former preferring policies for technological change, and the latter policies for human capital investment. Different optimal trajectories will result from weighting these preferences differently and with different discount rates. Figure 3.9 shows how if human capital investment is pursued, following the unskilled workers’ preferred policy, eventually the poverty trap disappears and the government’s policy switches to technological change. At some point the trap disappears and there is a relatively large jump in steady state which brings on a period of miracle growth. The result is a two-stage trajectory that first eliminates the low human capital trap (and inequality) and then embarks on technological change, reaching a higher final technological level than a policy for technological change not addressing the human development trap. Institutional difficulties may arise in carrying out this later, substantial policy shift. This two stage trajectory is reminiscent of the history of ex-socialist countries. These countries first achieved enormous advances in literacy, health and education (also accompanied with a series of technological achievements). Nevertheless, once these human capital improvements were completed, economic progress slowed. Serious technological lags began to appear, perhaps due to poor organization for technological change. For these and other reasons, important policy changes were adopted to accelerate technological change. These included the use of market mechanisms.

98


Main ideas Human capital is an important factor of production. Together with the divergence in economic development that occurred after the Industrial Revolution, there was a divergence in schooling and life expectancy levels. Rising productivity and per capita incomes also gave rise to health improvements reflected in increasing nutrition, stature, weight and life expectancy. This process can be understood as physio-techno evolution. However, it does not occur equally across society. Instead the poor lag often four or five generations behind in the achievement of the potential levels of health and education of their societies. These socioeconomic differences are transmitted intergenerationally and reflect the existence of market imperfections in the acquisition of different kinds of health and education. Society can divide into skilled and unskilled populations. We call the whole process human development, and model it as a dynamic poverty trap analogous to divergence in levels. Low levels of human development slow down technological change and deepen underdevelopment. A conflict of interests may arise between skilled and unskilled workers for government policies supporting human capital accumulation or technological change. Eventually, technological change is a necessary policy. However human capital policies can at first decrease poverty faster and at raise technological levels in the short and in the long-run.

5 This chapter is partly based on Mayer-Foulkes (2008). See the paper for further references throughout. 6 Mayer-Foulkes (2008) discusses this aspect in further detail. 7 Case, Fertig, & Paxson, 2003; Case, Lubotsky & Paxson, 2002. 8 See Mayer-Foulkes and Larrea (2007) for the study on Racial and Ethnic Health Inequities: Bolivia, Brazil, Guatemala, and Peru that originated this data. 9 I assume technological change transforms both production and human capital. Thus, A t units of human capital acquired in youth become A t+1 units of adult human capital. Alternatively, let the package of At+1 units of human capital available to the young cost hAt+1, and substitute h with hgt+1 in the results. 10 This section is based on Mayer-Foulkes (2011), Mayer-Foulkes and PescettoVillouta (2012), and updates in DCP3 (see http://dcp-3.org/resources/cost-

99

DAVID MAYER-FOULKES

effectiveness-ncds-and-their-risk-factors-paho-region-systematic-review-literature). For further references and details see these articles.

100

4 TRADE, ECONOMIC GROWTH, AND DIVERGENCE11 Globalization consists of two principal elements, international trade, and international investment. While trade is widely held to be a force for convergence between countries, we explain how in fact trade can simultaneously be a force for development and for underdevelopment, and that this has been the case since the Industrial Revolution and the Great Divergence. To do so we provide a historical discussion of the First Great Age of Globalization during the 19th Century and until 1914, and a theoretical model of trade and economic growth that displays multiple steady states, representing development and underdevelopment, that can be applied both in the longer term or contemporaneously. Two sources of divergence are considered, that are inherent characteristics of the process of technological change. The first is the need for absorptive capacity in taking advantage of leading edge technologies. The second is the existence of innovation externalities between goods, which implies that the more goods are engaged in R&D in any country, the more productive R&D is. We explain below how these are essential characteristics that are implicit in the process of technological change as described by Aghion and Howitt through their work (e.g. Aghion and Howitt, 1992, Aghion et al, 1998, Howitt, 2000, and so on). Trade has played a major role in modern economic growth since its origins. It forms a major strand in Maddison’s (2001) description of the economic ascension of Western Europe through Venice, Portugal, Spain, the Netherlands and Britain, from the year 1000 to the present. The Great 101

DAVID MAYER-FOULKES

Discoveries were motivated by the spice trade. Cotton exports in late 18th and early 19th Century England, widely recognized as the Industrial Revolution’s leading sector, rose from 6% of total British exports in 1784-6 to a peak of 48.5% in 1834-6 (Chapman, 1999). The growth of this sector and the incentives for its increased productivity were directly linked with imports of cheap raw materials from India at this initial juncture of the Great Divergence (Broadberry and Gupta, 2005). More recently, the rapid growth of Germany, Israel, Cyprus, Spain, Portugal, Malta, Ireland and Iceland were intimately linked with trade. More spectacularly, the development of Japan and the East Asian countries was also inextricably linked with trade. Foreign direct investment (FDI) became a major actor towards the end of the 19th Century, with the Second Industrial Revolution. Wilkins (1970) identifies 1875-1914 as the period of “…the rise of truly large-scale foreign investments in the private sector” including “…more foreign direct investments than most subsequent commentaries have recognized” (cited in Lipsey, 2001). Investments in the colonial and dependent countries became a source of extraordinary profits, due to very cheap labor and raw materials. By 1899 giant corporations such as the United Fruit Company controlled 90 per cent of US banana imports. By 1914 Royal Dutch/Shell accounted for 20 per cent of Russia’s total oil production and Standard Oil of New Jersey, Singer, International Harvester, Western Electric, and, Ford Motor Company had major production facilities outside the United States (Beaudreau, 2004). British assets abroad amounted to between 124 and 180% of its GDP in 1914. Taking British investment as a whole, between 1865 and 1914 approximately as much went to underdeveloped Africa, Asia, and Latin America (29.6%) as to the UK itself (31.8%) (Ferguson, 2003). Svedberg (1978) estimates that some 44 to 60% of the $19 billion of accumulated investment in developing countries in 1913-14 consisted of foreign direct investment. In this Chapter we discuss and model the role of trade, and in the next the role of FDI, in economic growth and divergence. In both cases we use models that have simultaneous application to long-term and to current globalization. In the case of trade we focus our examples on the 19th Century. In the case of FDI we focus on the 20th Century and the present.

I.

Free trade in the history of development

Notwithstanding Adam Smith’s and David Ricardo’s arguments for free trade, both Britain and the US adopted Free Trade only after they gained 102


industrial supremacy. The repeal of the Corn Laws in Britain in 1846, allowing the import of grains, was based on the power of steam engine run industrial production to pay. Similarly, while US opposition to free trade had been a staple of 19th and early 20th century foreign policy, the metamorphosis to unwavering support for Free Trade occurred until 1934, when Roosevelt signed the Reciprocal Trade Agreements Act, based on electric power-based mass production supporting US industrial supremacy (Beaudreau, 2004). To understand how industrial producers look at trade it is necessary to consider the dynamic impacts of trade on productivity. These dynamic impacts of trade on productivity are also reflected in the histories of weaker trading partners. Table 4.1 compares the initial conditions of lagging countries India, China, Mexico, Brazil, and the US in 1820, when Great Britain had become the industrial leader of its time. Both India and China were opened to trade by force, India in 1757 and China in 1842. From the point of view of sovereignty, their institutions were weak, and they were forced to follow policies serving British and other foreign interests. We already mentioned the deindustrialization of China and India (Figure 1.7a). Mexico and Brazil, with much smaller populations, completed their independence in 1821 and 1823, with institutions inherited from Spain and Portugal. As in Latin America generally, the main cities were inland and governments used tariffs to finance themselves. Thus Mexico and Brazil were relatively closed to trade. By 1820 the US was very well placed to imitate the British advances. Its first factory in 1790 was based on British textile machinery secrets brought by Samuel Slater to the US (Everett, 2006). The American System was advanced by Henry Clay and others after 1812. It implemented high tariffs to protect American infant industries from British industrial supremacy, and promoted trade between North, South and West through transportation improvements. The South, having access to markets for its cotton, had no incentives to join the System, one of the causes of the Civil War (Spannaus and White, 1996; Salisbury, 1992). Thus India and China were essentially open through the long 19th Century until 1914, while Mexico, Brazil and the US were relatively closed to trade during that time. What is interesting here is that we see two quite different rates of divergence. Between 1820 and 1980, India and China both multiplied their income per capita by 1.8. On the other hand, Mexico and Brazil multiplied 103

DAVID MAYER-FOULKES

their income per capita by 8.3 and 8.0. (Note that during this period India, China, Mexico, Brazil, and the US multiplied their populations by factors of 2.6, 3.2; 10.4, 27.3; and 22.8, the last two importantly through migration.) The US, which multiplied its income per capita by 14.8, overtaking the industrial leader, Great Britain, which grew by a factor of 7.6. These figures support are consistent with divergence in levels and divergence in growth rates, as well as the possibility of convergence and overtaking. It is also noteworthy that the lagging countries that were open to trade, India and China, diverged more than the countries closed to trade, Mexico and Brazil, and that a country capable of overtaking the leader, the US, chose to remain closed. A dynamic model of trade must therefore be able to simultaneously explain: comparative advantage and its benefits, the advantages of being open for the technological leader, why lagging countries might benefit from being closed, the possibility of divergence under trade for lagging countries, and under what circumstances lagging countries can converge to development under trade (such as the Asian tigers, discussed in the next chapter). A further test of a dynamic trade model is to be able to explain the main features of the “Colonial Diktat” imposed by colonial powers on their colonies (See Bairoch (1997), who asserts that the colonial diktat was the main cause for the non-transmission of the Industrial Revolution outside Europe). The typical “colonial diktat” consisted of (a) colonies could import only products from the metropolis and tariff rates had to be low, normally 0%; (b) colonial exports could be made only to the metropolis, from where they could be re-exported; (c) production of manufactured goods that could compete with metropolitan products was banned; and (d) transport between colony and metropolis was conducted only on metropolis ships. (See Beaudreau, 2004, who cites Bairoch, 1997). To model the impact of trade on economic growth and divergence, first I extend the model in Chapter 2 to open, trading economies. The first source of divergence I consider is the need for an absorptive capacity for taking advantage of leading edge technologies. The second source of divergence that is modelled is innovation externalities between goods. Here the idea is that the more goods are engaged in R&D in any country (the unit of analysis is really the knowledge system), the more productive R&D will be. More advanced and more populous countries will produce a wider variety of goods and will therefore have dynamic advantages in R&D. This idea

104


also applies to specialization in sets of goods where more innovation can take place, for example industrial production versus raw materials.

Table 4.1. Comparison of Initial Conditions in Several Peripheral Countries in 1820, and Economic Growth from 1820 to 1913 and 1980 India and China India: 200M China: 381M Weak, serving British interests

Mexico and Brazil Mexico: 6.5M Brazil: 4.5M

Open by force: India: 1757 China: 1842

High tariffs, many cities inland

1820 GDP per capita‡ 1913 GDP per capita Growth factor 1820-1913

India: $533 China: $600 India: $673 China: $552

Mexico: $759 Brazil: $646 Mexico: $1,732 Brazil: $811

1980 GDP per capita†

India: $938 China: $1,067

Growth factor 1820-1980

 1.8  1.8

1820 Population Institutions, governance Trade

 1.3  0.9

Weak, closed

Mexico: Brazil:

 2.3  1.3

Mexico: $6,289 Brazil: $5,198 Mexico: Brazil:

†UK

 8.3  8.0

United States† 10 M Strong, inherited from Britain Discretionary: “American System” based on industrial protection $1,257 $5,301  4.2

$18,577

 14.8

population in 1820: 21M. UK income: $1,250 in 1700, $1,706 in 1820, $12,931 in 1980. ‡1990 International Geary-Khamis dollars. Source: Maddison (2004)

105

DAVID MAYER-FOULKES

II. Free trade and innovation We introduce trade in the model in Chapter 2. Consider two countries, Country 1 and Country 2, that produce a set of tradable goods indexed by [0,1. We will assume that Country 1 is the technological leader. We use a continuum of tradable goods that is analogous to the set of intermediate goods in Chapter 2. At time t , Country 1 produces on the set

[0,1t ] and

(1t ,1], where

Country on the set

1t  2t = 1.

(4.61)

The sets [0,1t ] , [1t ,1] of goods which each country produces are assigned by trade through international competition by prices. This involves both the technological level and a comparative advantage in the production of each good. To simplify the model we assume that for each good produced by a country, there is a single infinitely lived incumbent who innovates with certainty. For each good the incumbent produces as a monopolist, subject to a competitive fringe producing with a smaller productivity by a factor of  >1. Therefore the incumbent can sell at a price which is a multiple  of her cost. For simplicity we assume this factor is the same in both countries. Since innovation proceeds with certainty, we assume that all goods in each country share the same technological level. To express some of the economic quantities symmetrically between countries, we let

2 =11

2 [0,2t ) = G2t

and consider variables

1 [0,1t ] = G1t ,

indexing the goods each country produces. When

instead we use an index

 [0,1] for the whole interval of goods, when

 [0,1t ] it represents a good produced by Country   (1t ,1] , it represents through a change of 2  1 [0,2t ) produced by Country 2. The production functions for good

i  Git

i = 1,2, are 106

1, while when variable

good

produced by Country i , with


yit (i ) = (1 i ) Aitlit (i ), i = 1,2. All goods produced by Country i share the same technological level

(4.62)

Ait at

lit (i ) units of labor at time t in good i . Finally, each good i has a fixed productivity effect 1  i , where 0    1 . This means that goods are ordered in a gradient of comparative any time t . They employ

advantage that goes in opposite directions for each country. The parameter  is introduced to be able to compare degrees of comparative advantage. The case  = 0 means there is no comparative advantage, while the case  = 1 is the maximum (in this representation) for which productivity is positive under autarchy for all goods (except for a single good ). The technological level will also have an impact on competitive advantage and the equilibrium under trade. Relative wages between countries will adjust in equilibrium so that each country employs its labor in goods more cost effectively than if the other country attempted their production. We assume there are no transportation costs. The cost of production of each good i in country i at time t is therefore:

 it ( i ) =

wit

(1   i ) Ait

, i = 1,2

(4.63)

where wit is the wage paid for labor in each country. Since the incumbent sells at a multiple  of the cost, the price is pit ( i ) =

 wit , i = 1,2 . (1   i ) Ait

(4.64)

U = U (Ct ) depend on a subutility function Ct for an agent consuming ct () units of good  goods,  [0,1] , according to the Cobb-Douglass function

Let the instantaneous consumer utility

ln( C t ) =

1

 ln( c ( ) ) d  . t

0

(4.65)

This preference function takes the place of the final good production function in Chapter 2. We use this Cobb-Douglass function for 1) 107

DAVID MAYER-FOULKES

consumption preferences and also for 2) the composite good research inputs, defined as

ln( X t ) =

X t used for

1

 ln( x ( )) d .

(4.66)

t

0

Hence world expenditure across goods will be constant for both consumption and innovation. Since at each time t expenditure constant across goods, pit ( i ) yit ( i ) =

wit (1   i ) Ait lit ( i ) = wit lit ( i ) = z t , (1   i ) Ait

zt is (4.67)

lit (i ) = lit is independent of i . In each country i at each time t an equal number lit of workers produces

It follows in this stylized model also each good. Profits in each good are:

 it = (1   1 ) zt . Let

L1 , L2

(4.68)

be the number of workers in each country, which we assume

constant for simplicity. In Country i ,

lit workers work on a measure it of

goods, so  it li = Li , and therefore t

lit = Li /it , i = 1,2.

(4.69)

Observe by equation (4.2.3) that prices pit (i ) are increasing in i . For this reason the goods that each country will cease to produce first

correspond to higher values of i . For low productivity countries, wages decrease to the point were employment takes place on some interval of goods. Under free trade the boundary of this interval is defined by some unique good

1 = 1t

for which both countries produce at the same price.

The corresponding good in symmetric notation is Country 2. Hence the trade equilibrium occurs when

108

2 = 2t = 11t

in


p1t 1t  = p2t 1  1t   Country 1 produces on the set

w1t w2t = . 1  1t A1t 1   1  1t A2t

(4.70)

[0,1t ] and Country 2 on the set (1t ,1]. (I

have simply assumed that sector

1t ,

as a point a set of measure 0, is

produced by Country 1.) Since expenditure is independent of goods, and payrolls are a fixed proportion

 1

i

across

of sales income, it follows

w1tl1t = w2tl2t . Multiplying each side of the second equation in (4.70) by li / li and using (4.69), we obtain 1  1t

1t

L1 A1t =

1   1  1t  L2 A2t . 1  1t

(4.71)

Definition 4.1. Let the relative technological and population levels between the two countries, and the ratio of their effective labor levels, be

at =

A2t L ,  = 2 , bt = at . A1t L1

We assume Country 1 is the technological leader so

(4.72)

at 1.

Proposition 4.2. Under trade, the sectors of production are assigned as a function of

bt , the relative size of the economies. There exist functions

1t = 1(bt )  (0,1), 2t = 1 1(bt ),

(4.73)

that define the intervals on which each country produces and satisfy

1 (0) = 1, 1 (1) = 12 , lim 1 (bt ) = 0, 1(bt ) < 0. bt 

Proof. (4.71) implies

109

(4.74)

DAVID MAYER-FOULKES

1  1t  2t = 1  1t 1  1t  = b . t 1t 1  2t  1t 1   1  1t 

(4.75)

This function on the left hand side (LHS) decreases from infinity to zero on

the interval 1t  (0,1] , and has non zero derivative. Hence a unique differentiable inverse function exists as in (4.73), with properties (4.74). Note that

1bt  solves a quadratic equation.

Note that the smaller the relative economic size of Country 2, as measured by A2t L2 / A1t L1 , the smaller the number of goods it produces in.

Theorem 4.3. For any given technological levels A1t , A2t , production, wages and profits are higher under trade than under autarchy. Wages and incomes

Y1t , Y2t of each country, and total income Yt are: w1t =

A1t e 1  1t 

1



w2t =

Y1t =

1  2t 

1

1

1

e 1  1t  1  2t  Y2t =

,



A2t

1  1t A1t L1 , 1 1 e1  1t  1  2t  Yt =

1

1

(4.76)

.

1  2t A1t L1 , 1 1 e1  1t  1  2t 

1  1t A1t L1  1  2t A2t L2 . 1 1 e1  1t  1  2t 

Prices can be expressed

110

(4.77)

(4.78)


1  1t   1    1 1  e1    1    1t 2t pt   =  1  2t  1   1     1 1   e1  1t  1  2t 

  [0, 1t ], (4.79)

  (1t ,1].

Proof. Equation (4.70) now implies the wage ratio is

w2 t 1   2 t  = at = 2 t . w1t 1  1t  1t

(4.80)

 1   2 t bt = 2 t . 1t 1  1t

(4.81)

since using (4.75) ,

The total production of each good is:

yit (i ) = (1 i ) Aitlit (i ), i = 1,2.

(4.82)

We now calculate the aggregate products of each country. We use the following notation for functions considered for both countries simultaneously.

 y ( )   [0, 1t ], yt ( ) =  1t  y2t (1   )   (1t ,1], and similarly for other variables such as

111

pit (i ) :

(4.83)

DAVID MAYER-FOULKES

w1t   (1   ) A 1t pt ( ) =  w2t   (1   (1   )) A2t

  [0, 1t ], (4.84)

  (1t ,1].

Given these prices, let Yt be the amount of composite good that can be produced using all of the goods produced. The part of the integral (4.66) involving goods from Country 1 is



1t

0

ln( y t ( ) ) d  =



1t

0

ln(( 1   ) A1t l1t ( )) d 



=  1t ln(1   )d  1t ln ( A1t 0

L1

1t

)



L   1  1t =  (1   ) ln(1   )     ln (( A1t 1 ) 1t ) 1t   0

1 L  =  (1  1t ) ln(1  1t )  1t  ln (( A1t 1 ) 1t )



1t

11t   A L   = ln (1  1t )  ( 1t 1 ) 1t    e1t  

Adding to this the component from Country 2, 

Yt =



( A1t L1 ) 1t ( A2t L2 ) 2t 11t

e1  1t 



12t

1  2t 



1t

 2t

.

1t 2t

(4.85)

The income of the inputs of the composite accruing to Country 1 is

w1t L1.

Hence, taking the price of the composite as numeraire,

1=

w1t L1  w2t L2 Yt

112

(4.86)


  w1t L1 1  2t  1t =

11t 12 t    e1  1t   1  2t   1t1t  2t2t  ,   ( A1t L1 ) 1t ( A2t L2 ) 2t

where we have used (4.80), so

w1t =

A1t



1t bt 2t

 e1   1 1t 1   1 2t  1t  2t 1t 2t 1t 2t

.

(4.87)

Note that it follows from (4.86) that the total expenditure in all goods is Yt . Hence this is aso the total expenditure in each good, since the total measure for all goods is 1. Because the intermediate consumption utility

Ct and the

X t used for innovation are defined using the same homogeneous Cobb Douglass function of degree 1, the proportions ct () , composite good

xt () of total production yt ( ) of each good used for consumption, or for innovation, do not differ between goods, and we will have

Ct  X t = et  Yt .

Equations (4.75) imply

1  1t  2t = b  1  1t 2t  2t2t t  1t 1  2t  1t 2t 1  2t  2t



= bt 2t

   1  1t 1t1t  2t2t  = 1t bt 2t .  1t  2t 1  1t  1  2t 

Hence expression (4.87) simplifies to (4.76), the expression for w2 t being obtained similarly. Using the first equality in (4.86), the total income

Yi = wit Li , of each country can be written as in (4.77), and the total

income (4.78). Expressions (4.76), (4.77), (4.78) show wages and production rise under trade. This can be seen as follows. For Country 1, under autarchy

1t = 1 , 2t = 0 . Hence

113

DAVID MAYER-FOULKES

Y1autarchy = t

A1t L1 e1   

1



1

.

Hence 1

1

1  1t  1  2t  < 1, Y1autarchy t = 1 Y1t 1    1 1

(4.88)

since 1    1 is a decreasing function in  . Finally, using expressions (4.76), for real wages, prices (4.84) can be expressed as in (4.79). 1

Wages are higher in countries with higher technology. Also they are higher in countries with less goods in production, because more of their production will take place in goods with a higher comparative advantage. Hence they are higher for countries with a smaller population. From the static point of view, that is, taking technological levels as given, each economy has higher income and wages under trade than under autarchy.

Our setup reproduces the standard results, from the static point of view. Taking technological levels as given, each economy has higher income and wages under trade than under autarchy. However, an additional result appears. The number of sectors under production in each country is proportional to the size of the economy, which itself is proportional to the population and technological levels. This means that larger and more advanced countries will produce in more sectors than smaller and lagging countries. This can have an impact on technological change, and therefore on the long-term benefits of trade. We turn to these issues now.

Technological change Throughout their work, Aghion and Howitt describe the process of technological change as follows: 114


1)Resources are used for innovating a new production technology for some good. If this research effort is successful, a higher level of technology will result for producing this good. Research is conducted taking into account new technologies that are becoming available. Successful research is often modelled as implementing new, leading technologies. 2)The leading edge technology is a measure of the externalities or spillovers that results from all of new technologies that research in the economy makes available. Often a separate variable represents this as an aggregate, “the leading edge technology.” Now, the resources used for raising the technological level can consist of knowledge or goods. Absent credit, the resources available on average in different countries are proportional to their level of production and knowledge, both of which are closely related to their technological level (e.g Howitt & Mayer-Foulkes, 2005). If credit is imperfect, available resources for innovation are also proportional to income and therefore to the technological level (e.g. Aghion, Howitt and Mayer-Foulkes, 2005). We can think of these combined resources available for innovation as the absorptive capacity. In single country models, for which the reference technological level of the economy is constant, the implicit absorptive capacity is often ignored. However, in multiple country models the idea that absorptive capacity intervenes in technological change and is proportional to the technological level is natural and follows from the inherent structure of the innovation process. Divergence can follow from this, and thus follows inherently from the properties of innovation, without any other assumption about what can make a countries backward. Turning to the leading edge technology, this is really an aggregate construct for research spillovers or externalities (or technological transfer) available from all of the research taking place in the economy. That is, it represents positive innovation externalities across goods. In Howitt (2000) the leading edge technology is defined as the maximum technological level in all countries and sectors. In fact, though, real-world aggregation of these spillovers is much more complex. For example, spillovers have greater and more immediate impacts within than between countries, especially between developed and underdeveloped countries, which are more distant in several ways. In effect the “leading edge technology” is in constant formation, as an aggregation of technological spillovers transmitted across different channels. (See Coe and Helpman, 1995 for a study of international R&D spillovers.) Below we consider a leading edge technology defined as the positive innovation externalities of goods whose innovation takes place in the same 115

DAVID MAYER-FOULKES

country. This is a reference case representing spillovers that are more easily transmitted between researchers in the same country and at the same technological level. This component of the leading edge technology externality can be a force leading to divergence. We turn to the innovation model. As mentioned above, for each good there is a single, infinitely lived innovator who can produce an innovation for the next period.12 We consider a continuous model in which innovations are incremental, or smooth. Observe that good  will only be in production in one country, either 1 or 2, because under the equilibrium wages it will have a slight comparative advantage in this country. Hence it also has a slight advantage for production after innovation, and therefore innovation in good  will only occur in the country that produces it. We also assume that when the boundary 1t shifts so that some goods start being produced in the other country, the incumbent from that country immediately takes over production and innovation at the country’s own technological level. I use a continuous model because it is amenable to two variables. Here only one variable is used. This leaves scope for other applications.

Definition 4.4. An agent has perfect myopic expectations if she can predict economic variables over a horizon  t when this horizon tends to zero.

In this case we will consider an innovator with perfect myopic expectations who maximizes profits in the short term  t by choosing innovation inputs, and then let t  0 to establish her decision at any time t . This eliminates the need for considering the second set of variables that is needed for infinite foresight. In fact a discrete model with two periods also has a similar type of shortsighted foresight. First I consider the role of technological absorptive capacity, which as we saw in Chapter 2 can explain divergence. Then I consider a second source of divergence: positive innovation externalities across goods when innovation takes place in the same country (or knowledge system). The effectiveness of innovation investment of the good  entrepreneur will have three components. The first is an absorptive capacity derived from knowledge and is proportional to the skill level S jt = A jt that she has been able to accumulate in production, which we assume is the 116


technological level of her firm. This generates a disadvantage of backwardness. The second component considers the impact of nascent,

positive externalities from other firms’ technological edge, (1   ) A1t . This term represents externalities from research for other goods, presenting itself in diverse forms as nascent possibilities, either in the form of ideas or embodied in the use of other firm’s goods at time t  t . It is analogous to the formation of a leading technological edge through innovation externalities in Chapter 2, which we eliminate, simplifying the model by one variable. The difference ((1   ) A1t  Ajt )t measures how far back our innovating firm is from these nascent possibilities, contributing an advantage of backwardness that generates convergence. The effectiveness of these combined inputs is inversely proportional to the level of the nascent possibilities, the fishing out effect. The third component is a material input vt . Innovation occurs with certainty combining these components to obtain a technological level

Att according to:

 1    A1t  A jt S jt t   vt 1  , = A jt   j  1   A1t   

At  t

j  1,2, FDI .

(4.89)

This means that, as in Chapter 2, the impact of innovator’s skill on the technological change that a firm can obtain is proportional to the skill level, proportional to its distance to the nascent technological frontier, and inversely proportional to the nascent technological frontier. In additional, the knowledge impacts combine with material inputs according to a CobbDouglass function. The parameter

 j represents the innovation productivity of the combined

inputs. Innovation externalities will be introduced by modifying this parameter. Using myopic perfect foresight, so as t  0 any firm correctly expects the new technological levels Ajtt , the profits level of an individual firm innovating to a technological level

 t  t

Att is:

 A jt  t = (1    At  t

117

1

  

)Yt  t ,

DAVID MAYER-FOULKES

since as we saw, the total expenditure in each good is Yt , where

Ajtt / At t measures the comparative reduction in costs. Hence the profit maximizing rate of innovation investment is obtained by maximizing: 1

max e

t

v

where

 A jt  t   (1    At  t v  

)Yt  t  1   j vt ,

et is the discount factor, and  j  0,1 represents an innovation

subsidy, a (positive or negative) proxy for all distortions and policies affecting the incentives to innovate.

Theorem 4.5. The rates of technological change of the leading and lagging countries are given by: 1 1 A1t  ~ A a   = 1at   , 2t = ~2 1  t at   . A1t 1   A2t  1  

(90)

The rates of change of the relative technological level is:

a t   a =  1  t at   1  

1   ~ ~ 1  at   .  2  1   

(4.91)

Here 1

 1   1      ~j =  1      1  j   at  =



1

j , 

1  1t L1  1   2t at L2 . Yt = 1 1 A1t e1  1t  1   2t 

(4.92)

(4.93)

Proof. Writing At t = Ajt t since firms in sector j are symmetric expost, the first order condition is:

118




e t 1   b jt  1 A jt  t A2jtt

 1    1

j



 1   A1t  A jt S jt t   vt   Yt  t t = 1   j t.  1   A1t   

Letting  j =

1   1      1 1   j 

vt 



j

, material inputs

v are given by:

 1   A1t  A jt S jt t  Yt  t  , b jt  j  1   A1t  A1t  t  

=e

t

Domestic innovators decide

At  t

 1   A1t  Ajt S jt t  t Yt  t   e  j  = A jt   j  1   A1t A1t  t   

1 



,

j  1,2.

(4.94)

Note also that since yt t depends on at a relative scale effects is introduced that complicates the dynamics. This aspect is simplified by using ~ =   (1 ) /  , which is continuous myopic foresight. Now set:  j j j equivalent to (4.92), noting that this final innovativity parameter for each economy is decreasing in market power  , because, as can be seen by following the derivative above, the higher the market power, the relatively lower the input costs and therefore the lower the impact of technological improvement on profit. Taking the limit as t  0 , and writing At t = Ajtt , 1  1   A1t  Ajt Ajt L1at   , A jt = ~ j 1   A1t

j  1,2.

(4.95)

j =1 yields the rate of growth of technology in Economy 1. This gives a scale effect in L1 for the

where, using (4.78), (4.93) is obtained. The case

global growth rate that depends on the size of the global economy relative to

A1t . The case j = 2 yields the rate of growth of technology in 119

DAVID MAYER-FOULKES

Economy 2. The difference yields the rate of growth of the relative technological level

at .

A technical note on myopic innovators We have considered innovators with perfect myopic foresight. They maximize profits in the short term  t by choosing an innovation input flow, and then letting t  0 . This is equivalent to defining perfect myopic foresight as having perfect knowledge of time derivative of the current economic variables. The myopic agent uses this knowledge to maximize the current time derivative of her profits. To see this, fix a time t and suppose that a firm can spend v units of inputs per unit time for a period t to obtain at the end of this short period a profit (t  t, v) . Suppose this function is twice continuously differentiable so that we can approximate it by a first order Taylor series in t of the form

(t  t, v)  (t, v)  ( X (t, v)  v)t, where   t

 (t , v )  t  0  X (t , v )  v

(4.96)

is written as benefits minus costs (recall the costas are v units per unit time). Then by the definition of perfect myopic foresight, the firm’s first-order condition at time t (after dividing by t ) is:  v

X (t, v)  1.

(4.97)

On the other hand, if the firm maximizes the expected time derivative of its profits (4.96), it obtains the same first order condition (4.97). To know the expected time derivative of its profits the firm needs knowledge of the expected time derivative of all relevant economic variables We can now describe the technological dynamics in the case when an absorption capacity is needed for technological change.

120


Theorem 4.6. Country 1’s technological level

A1t

grows at a rate

1  g at  = 1 ~1at   that is increasing in the size of the global economy, and

higher than under autarchy. If Country 2’s innovativity

~ < a) satisfies  2

~1 , Country 2 diverges in growth rates with Country 1; ~  [  ~ , ~ ] , Country 2 diverges in levels with Country 1 and b) satisfies  2 1 1 1 ~ converges to a steady state a  = 1     ~1 ; 2 ~ > ~ ., Country 2 overtakes Country 1 c) satisfies  2 1 

1

Y autarchy

Proof. For the first statement, note (at ) = A1tt  A11tt  1t A1t . For the performance of Country 2, note that (1  1at ) ~2  1 ~1 is decreasing in Y

at and equal to



1

Y

( ~2  ~1 ) at at = 1 . Thus Country 2 overtakes

~ > ~ . On the other hand the same expression is negative Country 1 if  2 1

~ <  ~ . Hence under this condition Country 2 diverges at at = 0 if  2 1 1 in growth rates with Country 1, with 1 1 A 2t ~ A  ~ = 2 0  < lim 1t = 10  . t  A t  A 1  2t 1t

lim

In the intermediate cases Country 2 diverges in levels with Country 1, to the given steady state.

Innovation externalities Recall that more advanced and more populous countries will innovate in more goods. It is likely that innovation for some goods makes innovation for other goods easier. For example, as experience was obtained in the industrialization of some goods, it became easier to industrialize others. We now turn to modelling innovation externalities. Consider instead of (4.89) the innovation function  1   A1t  A jt t   vt 1  , = A jt   j h jt  1    

At  t

121

j  1,2,

(4.98)

DAVID MAYER-FOULKES

in which we have eliminated the expression S it / A1t using skills as absorptive capacity (compared to the leading technological edge due to the fishing out effect) and instead suppose that there are positive externalities h ( jt ) in innovation related to the number of goods  jt being innovated

h(0) = 0 (so almost no research is possible if research is done for almost no goods), h(1) = 1 (this sets the scale), and that h is

for. We assume that

increasing in  jt . We show that divergence in levels always happens between identical countries, so that a lagging country with the same parameters can never catch up without policies to compensate for the lack of externalities, and that divergence in growth rates is possible under certain conditions on h at  jt  0 .

Proposition 4.7. Technological change in Countries 1 and 2 follows

(1   ) A1t  A jt A jt = ~ j h ( jt )  ( at ) (1  ) /  , 1

j  1,2.

(4.99)

Hence

A1t  ~ = 1h( jt )(at )(1 ) /  , A1t 1   A 2t ~ 1  (1 ) /   = 2 h( 2t ) at1  . (at ) A2t 1   

(4.100)

Hence the rate of growth of the relative technological level at is: 1   at  ~ 1   ~  =  2 h2t  at1  1h1t at   .  at  1   1   

(4.101)

Proof. Following the same profit maximization process we obtain (4.99) from (4.98) instead of (4.2.19) from (4.89). The remaining statements follow directly.

122


We study two cases. The first is when at is near zero, to see if there can be divergence in growth rates. The second is for two identical economies, for ~ = ~ . Here the question is whether the which at = 1 ,  = 1 , and  1 2 steady state at = 1 is unstable, so that if Country 2 is lagging only slightly, it will permanently diverge, at least in levels. By considering identical economies, we simplify the mathematical considerations.

Lemma 4.8. The derivatives of  1 t ,  2 t with bt , at bt = 0 are:

d1t dbt

= bt = 0

d 2 t dbt

1 , 1 

= bt = 0

1 . 1 

In the case of identical economies, so bt = 1 , where 1t =  2 t =

d1t dbt

bt =1

d 2 t dbt

1  =  , 4 8

= bt =1

1 2

,

1   . 4 8

d2t d =  1t , at bt = 0 , where dbt dbt

Proof. Differentiate (4.75), recalling

1t = 1 and  2t = 0 . The first result follows almost directly, obtaining  1   

d1t dbt

= 1 because there are so many zeros. For the second, bt = 0

recall at bt = 1 , 1t =  2t =

1 d   1t 2 dbt

1 . The derivative yields the quotient 2

 d  (1  ) 1t 2 dbt b =1 t

 bt =1

d1t dbt

 1 (1  ) 2 2

from which the result follows.

123

 1 d (1  )   1t 2 2 dbt b =1 t

bt =1

= 1,

DAVID MAYER-FOULKES

We can now describe the technological dynamics in the case when in each country there are innovation externalities between goods.

Theorem 4.9. 1) Country 1 grows faster under trade than under autarchy if h 1Yt  h  1t Yt autarchy , that is, if innovation improves more from the gains of economic scale than it looses in innovation externalities as some of the innovation goes abroad. 2) Country 2, grow faster under autarchy than under free trade if h 1Y2autarchy  h  2 t Y2 t , that is, if it gains more from innovation externalities when t it closes than from the economies of scale under trade. 3) Countries identical in their innovation parameters and populations will diverge if they start from slightly different technological levels. 4) Divergence in growth rates is possible if innovation externalities grow sufficiently slowly for small economies, that is:

h0 < 1   

 1 

~1 . ~2

Proof. 1) By equation (4.100), A1t / A1t will be higher under trade than

under autarchy if h(1) ( at )  h(1t ) 0 , which is equivalent to the stated condition. 2) Equation (4.99) for the rate of technological change of Country 2 shows that by resorting to autarchy h(2t ) will increase to h(1) . The convergence

term ( a t1  11 ) will remain the same while the scale effect  ( at ) will be reduced. Thus if h (1)Y2autarchy  h ( 2 t )Y2 t growth will be faster under t autarchy.

124


Figure 4.1 Two examples of phase space diagrams for the dynamics of relative technological change given by (4.101). In the first there is only divergence in growth rates. In the second there is also a steady state with divergence in growth rates. Additional configurations are possible.

125

DAVID MAYER-FOULKES

3) For identical economies

H 1 = ~1[h 12 (1  11 )  h 12  11  1 h 12   1 h 12 ] =  ~1h 12  < 0, so if a0 < 1 initially, at cannot reach 1. Hence Country 2 must diverge in levels or in growth rates. 4) Let H ( a t ) = ~2 h ( 2 t ) (1  1at )  1 ~1h ( 1t ) a t , the function in the big bracket of (4.101) multiplied by at . Since

h(0) = 0 , H satisfies

H (0) = 0 , H (0) = ~2 h(0)  ~1 1 , which is less than zero under the given condition. Hence at = 0 is a stable steady state. Note therefore by L’Hopital’s rule

 ~ at ~ h(0) (0) (1 ) /   = 2 1h(1t ) < 0, at 0 a 1  1 t lim

so Country 2 diverges in growth rates. Of course, these results are stronger if countries initially have further disadvantages.

Discussion Theorem 4.6 extends the results in Chapter 2 and shows that divergence in levels and in growth rates are possible under trade, as well as catching up and overtaking. The growth rate is found to be higher under trade than under autarchy because of a market scale effect that increases due to the assignment of production according to comparative advantages. Whether these scale effects exist or not has been subject to discussion, mainly because growth rates do not seem to be proportional to country size. However, the scale effect is really the scale of the sphere of influence of the technological leader. Clark, O'Rourke and Taylor (2014) find that while trade had only a small impact on British welfare in the 1760s, it had a very large impact in the 1850s. The cotton textiles sector became dependent on foreign markets for about 60% of its total sales. Thus trade not only allowed Britain to specialize in manufacture, the innovating sector, but also to produce for a 126


much larger market, becoming the Workshop of the World. In turn, it depended on foreign markets, including its colonies, for both food and raw materials. In the post 1960’s context, most developed countries traded together and therefore shared the same scale effect – namely the sum of their economies. It has been shown that the scale effect might be mitigated for example when R&D is expended in increasing the number of varieties. Thus there need not be a full linear scale effect as in these models. Nevertheless, as discussed in Chapter 1, a scale effect seem to have been present for the Industrial Revolution through world trade, by focussing innovation in Great Britain; for the US catch-up to industrialization through the success of the American System by focussing innovation in the North; and for German catch-up to industrialization with the rise of railways in the mainland in Europe after 1840. Britain's trade with its formal and informal empire, together with its advantage in manufacturing, focused innovation in Britain. An idea of the relative size of this effect is given by the following data, reflected in Figures 3 and 4. From 1820 to 1940 Britain's formal empire had a similar population to China, the largest country in the world with 381 million people in 1820. Each of the US, Russia, France, Germany and even the whole of Western Europe without the UK (which can be thought of potentially as an area of German influence) and the whole of the noncolonized world, was much smaller. In 1820 China’s aggregate product was higher than the British Empire’s. By 1870, the opposite held. An important point in comparing the forces of economic development in Britain and China is that there was no region in China where economic forces for innovation were as focused as they were in Britain, which functioned as the economic center of an economy as large as China’s. This gives a strong reason why the Industrial Revolution occurred in Britain rather than China. The high degree of economic competition associated with colonialism, high enough to have featured as a cause of the World Wars, is also evidence of an economic scale effect. Note how by 1913 Western European production (without the UK) had overtaken the production of the British Empire. There may be an underlying relationship between the income dominance shifts shown in Figure 4 and the World Wars. Theorem 4.9 also shows that innovation externalities can also lead to divergence in levels and in growth rates. Here innovation externalities between goods is the simplest general concept that relates innovation in some goods to innovation in other goods. In the case of 19th Century Britain, we could also consider that its knowledge system had more 127

DAVID MAYER-FOULKES

knowledge on industrialization than other knowledge systems, and that therefore it was able to innovate with advantage in those goods most amenable to industrialization. We could let the combination of functions on comparative advantage and h represent these features to obtain the same results as in this Theorem. Let us apply this theorem to the colonial diktat. Essentially the metropolis is Country 1 in the model, and its colonies together form Country 2. Even taking into account that the metropolis could trade with additional countries, at least from the point of view of each colony its trade situation looked like Country 2 in the model, since transport between colony and metropolis was conducted only on metropolis ships, and colonial exports could be made only to the metropolis. Production of manufactured goods that could compete with metropolitan products was banned, so that the set [0, 1t ] of goods with high productivity and innovation externalities which Country 1 produced was protected. Moreover, since goods coming from other colonies had to pass through the metropolis (or at least be traded by it), for each colony these goods enlarged  1 t or at least diminished market size in some goods from the set ( 2t ,1]. The consequence was that Country 2 diverged, either in levels or in growth rates. When the US obtained its independence from Britain and set up the American System, it opted to be closed and ceased to be Country 2 of this model. More generally, Theorem 9 explains the features shown in Table I. Brazil and Mexico, which were more closed to trade than China or India, diverged less. On the other hand, for various reasons including size and institutions, which feature in the model through the scale effect and the parameter  j , Brazil and Mexico did not have the necessary conditions for catching up, as the US did. Absorptive capacity and innovation externalities are specific, but quite general, causes that can simultaneously give rise to economic growth and divergence in the context of trade. There are others, for example agglomeration externalities as in Krugman (1991). These various different dynamics can of course combine in reality. The model applies so long as Countries 1 and 2 each represent a knowledge system within which there is some common level of knowledge. If policy steps are taken to reduce knowledge boundaries between countries, then the forces for divergence are also reduced, since skill levels and externalities are shared.

128


It is worth commenting that so long as captive trade through a colonial system yielded economic advantages, convergence to development set up for advanced countries an economic pressure to compete in the colonial system. Thus convergence to development tended to also be convergence to colonial power, with an implicit military rivalry between advanced countries. Economic competition set up a dynamic of military rivalry that was an antecedent of the two World Wars. In the current scenario (2017) it may likewise be important to maintain a single rather than a fractured global economic system. Trade has played a major role in modern economic growth since its origins. It forms a major strand in Maddison’s (2001) description of the economic ascension of Western Europe through Venice, Portugal, Spain, the Netherlands and Britain, from the year 1000 to the present. The Great Discoveries were motivated by the spice trade. Cotton exports in late 18th and early 19th Century England, widely recognized as the Industrial Revolution’s leading sector, rose from 6% of total British exports in 1784-6 to a peak of 48.5% in 1834-6 (Chapman, 1999). The growth of this sector and the incentives for its increased productivity were directly linked with imports of cheap raw materials from India at this initial juncture of the Great Divergence (Broadberry and Gupta, 2005). More recently, the rapid growth of Germany, Israel, Cyprus, Spain, Portugal, Malta, Ireland and Iceland were intimately linked with trade. More spectacularly, the development of Japan and the East Asian countries was also inextricably linked with trade.

This chapter is based on Mayer-Foulkes, D. (2017). We are therefore abstracting from creative destruction. In effect this implies that we are considering creative destruction that is neutral to trade, and that the innovation effectiveness and cost parameters are net of creative destruction. 11 12

129

Main ideas Suppose that productivity depends on each country’s technological level and also on comparative advantages due to geography and other country-specific characteristics. Then when countries trade, at equilibrium each country specializes on those goods it produces best, and wages adjust to offset productivity differences due also to technological levels. Aggregate productivity is higher and both countries have a higher income. A country with higher technology or larger population will produce a wider variety of goods. Nevertheless, when technological change requires absorptive capacity, there need not be full convergence to development. Divergence in levels and in growth rates can take place instead. Under trade each country specializes in producing certain goods. If each firm carries out its own innovation, the assignment of production also implies an assignment of innovation. (The alternative case corresponds to foreign direct investment, which we examine in the next chapter.) Technological or population leaders will produce and therefore innovate in a wider variety of goods. If within each country there are innovation externalities between goods, this will produce a disadvantage to backwardness and increase the tendency to divergence. For example, identical countries with slight initial differences will diverge. Free trade was only embraced by Britain and the US when they attained technological supremacy. The policies that the US adopted to catch up with British industrial development included protection through high tariffs. Mexico and Brazil, which diverged only in levels with the US and Britain were relatively closed to trade, while China and India, which were open by force, diverged in growth rates. The models explain how the Colonial Diktat imposed by the great powers on their colonies produced economic benefits for them but generated divergence for their colonies. In the case of innovation externalities the model implies that being open is preferable to being closed so long as for innovation the dynamic disadvantage of a reduced knowledge system, implied by producing for a smaller set of goods, is less than the favorable impacts of comparative advantage and market scale.

130


5 FDI, CURRENT GLOBALIZATION, AND THE SINGLE GLOBAL MARKET ECONOMY As we mentioned before, globalization has two principal features, international trade, and international investment. We mentioned in the introduction to Chapter 4 how by the end of the 19th Century foreign direct investment (FDI) seeking cheap labor and raw materials had become a major economic actor. British assets abroad amounted to between 124 and 180% of its GDP in 1914. Between 1865 and 1914 approximately as much British investment went to underdeveloped Africa, Asia, and Latin America as to the UK itself. From 44 to 60% of accumulated investment in developing countries in 1913-14 consisted of foreign direct investment. In this Chapter we discuss and model the role of FDI in economic growth and divergence.13 The model, which is related to the one in Chapter 4, can be applied in a long-term context or more locally in time. We focus on examples from the Postwar era to the present.

I.

Globalization from the 80’s to the present

The acceleration of globalization in the 1980's began with Ronald Reagan and Margaret Thatcher's liberalization policies. Faced with the stagflation crisis of the 1970's and the first oil crisis, they restarted economic growth by freeing international trade and investment. Keynesian policies were not working and this led to a wave of New Classical economics. In November 1982, a ministerial meeting of the General Agreement on Tariffs and Trade proposed what became the Uruguay Round of negotiations, "the largest negotiation of any kind in history," launched in 1986, concluded in 1994, and signed by 123 countries, that led to the creation of the World Trade Organization in 1995. 131

DAVID MAYER-FOULKES

The negotiations addressed trade in goods and services, investment, and intellectual property. Meanwhile, the Washington Consensus (a term coined in 1989) implemented a standard New Classical reform in any developing country that faced a crisis, recommending and imposing liberalization of trade and foreign direct investment, privatization, and deregulation. In addition to Western liberalization, China's introduction of market mechanisms in December 1978, and the fall of the Berlin Wall in 1989, came together to create a global market economy.

Main features of globalization Liberalization tapped a huge potential for economic growth. For twenty five years worldwide exports grew at an anual rate of 6.2%, approximately doubling as a proportion of world GDP from 14.5% in 1982 to 30.6% in 2006. World GDP itself grew at 4.0% a year during the period 1982-2007. While the growth of trade classifies as miracle growth, FDI grew at an average real rate of 14.6% a year. Of this investment, a great part consisted of mergers and acquisitions, for example 89.3% in 2007. The 100 largest non-financial TNCs produced 14.1% of world GDP in 2007. While aggregate world exports of goods and non-factor services reached US $17 trillion dollars in 2007, aggregate sales of foreign affiliates of Transnational corporations (TNCs) reached US $31 trillion, surpassing 50% of world GDP in the years 2008-2010. In 2010, TNCs accounted for one-quarter of world GDP. TNCs have come to play a central role in the global economy. At $7.9 trillion, the gross product of foreign affiliates of TNCs reached 45.4% of the US $17.4 trillion GDP in 2014. Two thirds of world trade is TNC-related. Intra-firm trade alone amounts to one third. In fact market concentration has been the norm rather than the exception for US production during the 20th Century, one of the results of the Second Industrial Revolution. From 1935 to 1992, the average production of the four largest firms in 459 industries was 38.4% of all shipments. Similarly, from 1992 to 2002, the 200 largest manufacturing companies accounted for 40% of value added in manufacturing. Thus in the US the equilibrium level of concentration was higher than at the global level, where it therefore stands to increase further. In 2010, more than half of FDI went to developing economies. As we shall show in our model, what results from FDI between developed and underdeveloped economies is a polarized form of globalization that admits 132


the coexistence of development and underdevelopment, and is characterized by huge steady-state profit flows with very significant impacts. In 2007, FDI profits amounted to $1.1 trillion, a profit rate of about 7% of gross income. About 30% of this was reinvested. Much of the remaining 70% must have remained offshore, as indicated by the following three pieces of information. First, the US $700 billion trade deficit was of the same size as these profits, consistent with a mass of capital invested in US brokerage instruments under foreign corporate names. This approximate relationship between the profits of US foreign affiliates and the US trade deficit has held for many years, see Figure 5.1. (Figures 1.10a and 1.10b show how the relation between corporate profits and the trade deficit has changed in the US, clearly aligning corporate profits with trade deficits after 1990.) Second, the frequent lobbying for tax holidays by US corporations generating profits abroad. Third, as reported by Tax Justice International, between $21 and $31 trillion dollars in assets were held offshore at the end of 2010, that is, between 1.4 and 2.1 times the US GDP!

700

Billion US dollars

600

US Trade Deficit

500 400

US Foreign Affiliate Profits

300 200 100 0 1980

1985

1990

1995

2000

2005

Figure 5.1 U.S. Trade Deficit and Profits of Foreign Affiliates of U.S. TNCs, billions of dollars, 1980-2005.

Inequality in the US rose as globalization proceeded. Using IRS tax data, Piketty and Saez have shown that, while the bottom 90% saw their income share drop from 66.8% in 1982 to 52.8% in 2014 (a reduction of 23%), the top 0.01%, 0.01-0.1% , 0.1-0.5%, 0.5-1% and 1-5% income brackets saw 133

DAVID MAYER-FOULKES

their participation in income multiplied by 5.9, 3.1, 1.9; 1.6, 1.3 and 1.1 respectively (excluding capital gains, which are more unstable), see Figure 5.2. The really high gains were in the really high brackets. Of course the higher bracket income includes income from abroad. It is for this reason that the fraction of total US income growth captured by the top 1% over the 2009-2011 recovery was 121%.

Figure 5.2 Factor by which income share of higher income brackets was multiplied from 1970 to 2014. Source: The World Top Incomes Database, read 10/16/2015. The huge profits flows registered under globalization contributed to the global saving glut observed by Bernanke in 2005. This contributed to a long-term decrease in interest rates that continues to this day, and to the 2008 crisis. The share of national income going to wages and salaries fell to 51.6% in 2006, its lowest recorded level (since 1929). The share of national income captured by corporate profits, in contrast, rose to its highest recorded level, 13.8%. Let us note that skill biased technological change, the most accepted cause for increased inequality in the US, is directly linked to globalization in two ways. First, the outsourcing of labor intensive work raises the proportion of 134


skill intensive work, in itself a skill-biased change in the composition of production technologies. Second, R&D carried out to increase the productivity of the remaining workforce itself also constitutes skill-biased technological change. Oxfam, giving an update on the political and social consequences of the worsening degree of global economic inequality, warns that the “massive concentration of economic resources in the hands of fewer people presents a significant threat to inclusive political and economic systems." Besides huge profit flows and decreasing interest rates, another notable trend of the last thirty years has been a global decrease of statutory corporate tax rates, see Figure 1.9b. Over the past 25 years, corporate tax rates in Europe show a remarkable downward trend. In 1983, the mean statutory corporate tax rate of 13 Western European countries was 49.2%. As of 2008, the average tax rate of these countries had eroded to 27.2%. One reason for this decrease may be tax competition between countries to attract FDI. Another may simply be the increase in political influence of large corporations due to higher profits under globalization. Not only have corporate taxes diminished, there is also an important degree of tax evasion. As mentioned above, $21 to $31 trillion dollars in assets were held offshore at the end of 2010. Note tax havens favor international over national production simply because international production has an easier access to tax avoidance. As international production and value chains develop, tax avoidance provides the incentives to shift domestic production to an international context in which, at least on paper, it can appear to take place beyond the reach of tax officials.

II. Globalization and economic growth Let us brief review how globalization works and why it is a potent economic force. Economic growth is the increase in the amount of the goods and services produced by an economy over time, public and private. Globalization consists of the integration of national economies through trade and FDI. So, what forces drive the current period of globalization and how do they impact economic growth?14 First, as we saw in Chapter 4, trade is driven by comparative advantage. Countries specialize in the goods they are most productive at, given their technological levels, and thus increase their income by trading. This is a static effect that would soon reach equilibrium in the absence of 135

DAVID MAYER-FOULKES

technological change. Technological change itself is driven by the incentives for profits and market power that come with developing and owning new production techniques, product lines and so on. This market power gives rise to large corporations that subsist through sequential innovation. In addition to trade, globalization is driven by foreign direct investment. Firms invest abroad seeking cheap labor, cheap raw materials, new markets for their goods, and more efficient arrangements of production. When approximately equal partners such as European countries and the US engage in trade and FDI, there is a symmetry to how countries specialize in the goods for which they have a comparative or a technological advantage and in which they perform FDI. Production takes place more efficiently and for larger markets. The incentives for innovation increase in all countries involved, raising not only the levels of production but also longterm economic growth. Economic well-being tends to grow in parallel across these countries. This type of FDI tends to function as an extension of domestic production abroad, in other words as good-specific specialization. Its impact on production and innovation does not differ too much from the impact of trade. The case of cheap-factor-seeking FDI is different. When quite unequal partners engage in trade and FDI, specialization also occurs according to the costs of factors of production such as labor and capital. In these circumstances a polarized, asymmetric form of globalization emerges. Developed countries with high wages invest in underdeveloped countries, seeking low wages. Large transnational corporations emerge that obtain extraordinary profits as they purchase cheap labor and cheap resources in backward countries and sell their products in advanced countries. Because more workers are now using better technologies, overall productivity rises. However, extraordinary profits need not raise the incentives for innovation, because under lower costs innovation yields lower cost savings, as shown below. In the theoretical model we concentrate on cheap-factor-seeking FDI. Polarized globalization leads to imbalances both in developed and underdeveloped countries. In developed countries, workers have now to compete with lower-salaried workers elsewhere, endangering the middle class, while capital and technology tend to receive higher returns, as noted above. On the other hand, underdeveloped countries competing with each other for the investments of TNCs that export their profits may find it hard to raise taxes for funding public goods. They may also find it hard to compete in innovation. Because advanced economies are large and innovate in many goods, backward economies may remain technologically dependent, for example due to a low technological absorption capacity, or a low level of innovation externalities, reaching an equilibrium technological lag that in itself constitutes underdevelopment. Thus, it is possible for 136


globalization to generate a global economy that can remain persistently polarized into countries with high and low equilibrium trajectories and at the same time accumulate capital faster than under autarky. Alternatively, when there is enough technological transfer, “miracle” growth may occur. Indeed, the majority of countries that attained industrialization and development did so through a prolonged period of high, sustained economic growth. Such are the cases of Denmark, Sweden, Italy, Japan, South Korea, Taiwan, Hong Kong, Singapore, Ireland, Germany in the 19th century, Western Germany after the War, Cyprus, Iceland, Spain, Malta, Portugal, Israel. In Wan’s (2004) comparative case studies of the Asian Tigers’ growth experiences (mentioned in Chapter 1), the reference convergence trajectories include at least two decades of growth higher than 5%, viewed explicitly as a transition to a higher stationary state. In the case of the Asian Tigers, Wan (ibid) emphasizes the role of technology transfer. The possibility of miracle growth represents a hope for hundreds of millions of people for emerging from poverty. For this reason Wan considers the East Asian rise to development one of the mega-events of the 20th Century. However, some countries have experienced periods of miracle growth without fully reaching development, such as Argentina, India, Nigeria, Brazil, and Mexico in the sixties and seventies, and Chile from 1985 to 1997 (see the discussion relating to Figure 2.4). Perhaps what these countries were unable to achieve was enough technological transfer. After presenting the model we discuss the differences between trade and FDI regarding technological transfer.

Globalization profits as antecedent of the 2008 financial crisis When international trade was opened in the 1980s through the negotiation of a series of new trade and investment treaties, a transition began towards a globalized economy (see the model below). This transition led to large profit flows from the underdeveloped to the developed world. Eventually corporations decided to save an important proportion of this profit flow, and therefore placed it in the financial system of the developed world, mainly in the US. This could be because the transition slowed, or became more difficult, or simply be part of a steady state with permanent extraordinary profit flows. This “savings glut” led to capital account inflows in the United States, Spain, the United Kingdom, Australia, Italy and France. As financial institutions received more funds, these became available for the housing sector and led to simultaneous, substantial housing appreciation in all of these countries, setting the stage for the stock market crisis. 137

DAVID MAYER-FOULKES

Corporate profits reached what at that time was an all time high in the US and the UK, as well as for many corporations across the world. The 2008 World Investment Report highlights increased profits of foreign affiliates for 2007, notably in developing countries. It mentions that even after the crisis TNCs in most industries had ample liquidity to finance their investments, based on high corporate profits. The aggregate profits of TNC’s foreign affiliates reached $1,100 billion in 2007, with profit rates of about 7%, calculated as ratio of net income to total sales. These profits were increasingly generated in developing countries rather than in developed countries. About 30% were reinvested and the rest repatriated, in an amount that is remarkably close to the US capital account surplus. The economic magnitude of FDI profits was the same as the US trade deficit. The overall picture of high global corporate profits, high earnings for the top income levels, a falling participation of income for salaried workers in advanced countries, and transfers of 70% of affiliate profits to their home countries, is consistent with the role of FDI under polarized globalization. Repatriated income was not invested. It remained in the financial system of the developed world, overvalue it, and fed housing appreciation in many countries. Krugman (2007) wrote on this lack of investment in an editorial column, making reference to a “double disconnect:” “high profits haven’t led to high investment, and rising productivity hasn’t led to rising wages.” This disconnect is the difference between polarized globalization and a selfcontained, leading economy. The prevalence of low interest rates and a lot of money chasing a limited investment portfolio was evidenced by the derivatives bubble, that was based largely on mortgages. According to the Bank of International Settlements (BIS), world’s clearinghouse for central banks in Basel, Switzerland, the valuation of the world’s derivatives grew exponentially from $72 trillion in June 1998 to $683 trillion in June 2008, in real terms multiplying by 5.8 and growing at an annual rate of 19.5% a year. This truly astronomical figure, a sum of wagers of so called investors, compares to: US annual GDP of about $15 trillion, US government’s maximum legal debt of $9 trillion, world GDP of approximately $50 trillion, total value of the world’s real estate estimated at about $75 trillion and total value of world’s stock and bond markets at somewhat above $100 trillion. The derivatives market crashed, leading to the failure of Lehman brothers and the 2008 stock market crash. After the 2008 financial crisis, the Federal Reserve Board applied an expansionary monetary policy in coordination with many other countries. While it is very likely that these policies cushioned the impact of the stock exchange crash on the economy, signs of a global investment slowdown 138


have continued. Investment activity in the US has not risen enough for the Fed to be able to pull back its expansionary, low interest, monetary policies. The interest rate has remained below 0.22% since 2009. US corporations are making more profit than they wish to spend or invest. They therefore seek financial instruments for saving this money. One way of neutralizing the risk of financial investment is to buy back their own shares. Renick and Regan (July, 2015) mention that this boosts companies’ reported earnings per share and write as follows. Corporate buy-back of their own stocks reached a high in 2007, but continued to the present. … Since 2009 companies have spent $2.4 trillion on buybacks. … One reason buybacks are common is that U.S. companies have earned so much money in the past few years. Over the previous 12 months they’ve generated $1.1 trillion in profits—a sum that “cannot possibly be reinvested back” as capital spending or research and development, says Dubravko Lakos-Bujas, an equity strategist at JPMorgan Chase. “Cash flow generation for U.S. companies has been very robust, balance sheets have remained pretty healthy, and interest rates are still low,” he says. “With growth fairly anemic, it’s extra reason for buybacks.” Or as BTIG’s Greenhaus puts it, “Companies have to do something with their cash.” It is global private investment that has slowed down, in a cycle that follows a much slower rhythm than country business cycles. Global FDI fell by 16% to $1.23 trillion in 2014, with developing economies leading in global inflows (UNCTAD, 2015). And this year, in 2015, corporate buybacks are set to reach an annual record of $993 billion (Renick and Regan, July 2015). Meanwhile, in underdeveloped countries, “there are huge unmet investment needs for infrastructure” which “far exceed the amounts being invested by governments, the private sector and other stakeholders, resulting in a significant financing gap” (UNCTAD, 2008). There are also huge shortfalls in human capital investment; witness the Millennium Development Goals (MDGs), whose basic education and health goals may remain unmet. The 2015 Sustainable Development Goals now call for even further investments. In short, there are significant investment shortfalls in what have been traditionally domains of publicly provided complements to private investment, such as infrastructure, education, health and science, due to more than two decades of low public spending associated with liberalization policies. In the next chapter we argue that what are needed are distributive-innovative public policies which can raise both efficiency and equity. 139

DAVID MAYER-FOULKES

Case Schiller Real Home Price Index

200 175

Real Home Price Index 1890‐2014

150

Linear Predictor based on 1890‐1982

125 100 75

1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

50

Figure 5.3a. Case-Shiller real home price index since 1890 and predicted time trend based on 1890-1982 data, with a dummy for the period 1914-1945.

Case Schiller Real Home Price Index

200 180

1950‐1982

160

1982‐2014

140

Linear (1950‐1982) Linear (1982‐2014)

120 100 6

7

8

9

Log After Tax Corporate Profits (2009 chained dollars), Adjusted for Inventory Valuation and Capital Consumption

Figure 5.3b. Scatter plot of Case-Schiller real home price index versus log real corporate profits after tax adjusted for inventory valuation and capital consumption. Trend lines added for the periods 1950-1982 and 1982-2014. 140


Clearly it is not private corporations that are going to carry out these public investments, which would serve to surmount the global investment slowdown. Even worse, if the great mass of accumulated profits channeled into savings does not find through the market an investment counterpart (to clear the market for investable funds, so S  I ) and remains on paper, it can only disappear through some kind of financial loss. This is why corporations prefer to buy back their own stock, reducing their exposure to financial market risks. The excessive supply of savings can at best raise asset prices, such as housing in the US, which has become less affordable to the median household.15 Figure 5.3a shows the Case-Schiller real home price index since 1890, with a predicted time trend based on the 1890-1982 period (including a dummy for the period 1914-1945), and Figure 5.3b for a scatter plot of real home prices versus real corporate profits in the US. Real profit flows are at record levels and house prices are higher than the 18901982 historic trend. The 2012-2014 upward trend in home prices shows that the housing bubble that peaked in 2006 was not an isolated event, but part of a thirty year trend that coincides with globalization. Extraordinary profits – extraordinary waste!

III. The globalization model The objective here is to model the process of globalization, understood as trade and foreign direct investment (FDI) between economies with different steady state levels of development. These different levels of development can result from several causes. We model the case of skills as in Chapter 2, and discuss fixed productivity effects for production or innovation, such as institutions, and innovation externalities as in Chapter 4. The model explains the following main facts of globalization highlighted in the introduction, based on the prominent role of FDI. Globalization: (a) Increases aggregate economic growth. (b) Increases the economic participation of TNC profits, and lowers wage participation. (c) Admits development, underdevelopment and miracle growth. (d) Increases inequality in leading countries. (e) Leads to lower corporate taxes and more conservative policies. For simplicity I analyze the interaction of two economies, a developed and a less developed economy. These could represent the whole of the developed and underdeveloped world, the US and China, or any underdeveloped country vis a vis the developed world. It can also represent, 141

DAVID MAYER-FOULKES

in a second stage, the set of transnational corporations vis a vis the set of domestic firms. The model is similar to the one in Chapter 4. Economies 1 and 2 produce a continuum of tradable goods indexed by  [0,1], where each  refers to a good. Domestic firms in Economies 1 and 2 have different technological levels A1t > A2t , representing different levels of development, that change endogenously. Under autarchy each economy produces the full set of goods. The economies exchange ideas as in Chapter 2. Under trade and FDI (free trade agreements are also free investment agreements) production goods  [0,1] fall into three disjoint sectors. The first sector is domestic production and innovation in Economy 1, carried out by Economy 1 innovators on the set 1t = [0, 1t ] . The second sector is likewise domestic production and innovation in Economy 2 on the set  2t = (1t , 1t   2t ] . Finally, the third sector is FDI on the set  FDIt = (1   FDIt ,1], produced in Economy 2 by Economy 1 innovators. The measures of these sectors add up to one:  1t   2 t   FDIt   1t   2 t   FDIt = 1.

The three sectors are indexed with j  1,2, FDI , and mostly refer to  jt as their number. FDI can be modelled in two modes. In the first, corresponding to an initial period of globalization, the FDI sector  FDIt = [1   FDIt ,1] on which it is feasible for innovators from Economy 1 to produce in Economy 2 is exogenous. The reverse cannot occur since innovators from Economy 2 are not competitive in Economy 1. Because it combines advanced technologies with cheap labor, profits will be higher in the FDI sector. The expansion of  FDIt along time is considered exogenous, and represents a progression of trade and investment agreements and other social arrangements making FDI possible. The second way to model FDI becomes relevant when new investment possibilities appear for the TNC sector in Economy 1. This second mode corresponds to a second stage of globalization in which every economy becomes an Economy 2, having a domestic sector and hosting cheap factor 142


seeking FDI on the part of TNCs. Economy 1 now consists of all transnational corporations, without a population, but with technological level A1t . The model for this second mode of globalization is a direct application of the model for the first mode. Thus a simple, two-economy innovation-based growth model with trade and FDI is constructed. Innovation occurs by continuous increments as follows. In each economy there is for each good  [0,1], a single, infinitely lived innovator who invests in innovation and becomes a national monopolist under autarky or world monopolist under trade, producing in the presence of a competitive fringe. Innovation is cheaper for the producing incumbent than for the competitive fringe, and she therefore has an innovation advantage. Her monopoly therefore persists indefinitely, both in autarky and under trade, so long as her good is assigned by trade for production in her economy. The international assignment of production therefore also implies an international assignment of innovation, not only between domestic but also between international producers. This is equivalent to identifying the producer and innovator with the holder of market power over good  [0,1], even if she subcontracts some of the innovation and production tasks elsewhere. In each economy, domestic or FDI knowledge resides in a set of firms, each monopolizing production in some good  [0,1]. The firms are symmetric and have the same technological level A1t or A2t in each type of sector. In the case of domestic firms, entrepreneurs may draw skill inputs from their own economy, at the same technological level A1t or A2t . The firm’s installations, entrepreneurial skills and the brunt of its skilled workforce all correspond to its technological level A1t or A2t and defines its absorptive capacity. This does not exclude the use of special knowledge inputs at the leading technological level A1t , incorporated as part of the leading technological edge knowledge externality. Innovation requires private inputs. First, the firm’s own knowledge inputs we have just described, complemented in the domestic case with local skills, and second, in the form of material inputs. This applies the ideas of Chapter 2 as in Chapter 4. For simplicity, innovation occurs with certainty. Next, instead of considering a leading technological edge, itself formed as an innovation externality, I consider that other firms’ innovation has 143

DAVID MAYER-FOULKES

positive externalities on the incumbent’s innovation investment, producing nascent possibilities in proportion to their technological level. This has two advantages. First, it is unnecessary to use an additional variable to represent a global stock of leading edge technology. Second, we could consider that this models a purely private global knowledge system, which concords with the much more sophisticated knowledge currently used throughout production, which need not be available to the public, and with the diminished current public knowledge systems with much less public support for science. Private knowledge cannot be held fully watertight and diffuses through employees, technical advisers, products, and so on, provoking positive externalities from one firm to another. This diffusion accounts for Gerschenkron’s “advantage of backwardness” and generates a force for convergence. It can for example include embodied technological knowledge promoted by suppliers for use in the near future, contracting leading edge technicians to help implement a new level of firm know-how, and so on. By contrast, the dependence of innovation on the firm’s own knowledge and starting point accounts for a “disadvantage of backwardness,” and generates a force for divergence. A scale effect occurs in innovators’ incentives through the impact of the lagging economy’s future relative size on global profits. Since the time scale in which individual firms operate is short compared to the evolution of the global economy, and to avoid the additional variables involved in infinite perfect foresight, I consider a myopic decision maker who lets her time horizon tend to zero and only has perfect foresight as t goes to zero. In Chapter 4 we showed this is equivalent to assuming that decision makers have perfect foresight on the current time derivatives of the relevant economic variables. This simplifies the scale effect by bringing it to the current time. It turns out that even though FDI obtains extraordinary profits, since this occurs through lower costs, it reduces its innovation incentives below the incentives for Economy 1 domestic firms. This seems unrealistic in that a transnational corporation with lower technology would soon face international competition from other Economy 1 domestic innovators. Therefore I assume in our first mode of globalization that the FDI innovator decides to innovate at the same rate as Economy 1’s domestic innovators for strategic reasons, so as not to fall behind them. Hence both the domestic sector in Economy 1 and the FDI sector maintain the same technological level A1t . However, strategic considerations and more resources can also eventually allow FDI innovators to outcompete domestic Economy 1 innovators and to find FDI possibilities in Economy 1, thus leading to the second mode of globalization.

144


To construct the model, we describe production, trade, FDI, and innovation.

Production Let the population of Economies 1 and 2 be Lit , i = 1,2. Under autarchy at each time t two state variables will fully define the state of both economies: the technology levels A1t , A2t of each economy. Under free trade and FDI, the global economy is additionally defined by the endogenous variable  FDIt . Let us now turn to the production functions. We consider a single input, labor.

Definition 5.1. The production function for goods    jt

j  1,2, FDI is:

y jt ( ) = A jt l jt ( ),

in sectors

j  1,2, FDI .

Here y jt ( ) is the quantity produced of good    jt . Compared to the model for trade in Chapter 4, besides including an FDI sector, for simplicity we have eliminated the comparative advantage component, setting   0. We nevertheless assume that Economy 1 produces on goods to the left of Economy 2, as would occur for an arbitrarily small  . MF takes into account a fixed productivity factor q j representing the effects of such nontechnological factors as geography, institutions and policies that influence a country’s total factor productivity. A jt is the technological level in each sector type. l jt ( ) is labor employment. The FDI technological level is

AFDIt = A1t . We assume there are no transportation costs.

Definition 5.2. Define the relative variables at = A2t / A1t , t = L 2t / L1t .

145

DAVID MAYER-FOULKES

This definition coincides with Definition 4.1. For FDI to be viable we assume A2t  A1t , so 0 < at  1. The cost of production of each good  in each type of sector j at time t is therefore:

 jt   =

w jt A jt

,

j  1,2, FDI .

The reader can find a guide to the model in Figure 5.4.

Figure 5.4. Diagram for the static economy in the globalization model. The horizontal axis shows production sectors, and vertical axis technological levels. From left to right we have Economy 1’s domestic sector, Economy 2’s domestic sector, and FDI by Economy 1 in Economy 2. In its domestic sector, Economy 2 compensates lower technological levels employing more labor for the production of each good. The domestic sector in Economy 1 and the FDI sector employ the same amount of labor per good. The proportion of income allocated to profits and wages is shown for each good, in gray and white. Each agent then consumes all goods symmetrically according to her income, implicitly defining trade volumes.

146


We assume again that the competitive fringe faced by each domestic incumbent has lower productivity by a factor   1. Domestic prices are therefore

p jt   =

 wit Ait

,

j = 1,2 .

(5.102)

Assume now that FDI technologies are beyond the reach of Economy 2 competitive fringe producers, and therefore that the competitive fringe for FDI producers is the one in Economy 1. Assume that these are small producers who can trade costlessly but cannot afford to produce abroad. It follows that FDI products are sold at the same prices as domestic products in Economy 1. Their price will therefore be given by:

pFDIt ( ) = w1t / A1t .

(5.103)

Trade and FDI Under trade and cheap-factor-seeking FDI, production responds to global demand, and global prices are formed, which in turn determine local wages. We now use the same instantaneous consumer utility U = U (Ct ) , function of a Cobb-Douglass subutility function Ct for an agent consuming ct ( ) units of good

 [0,1]

as in Chapter 4 (equation 4.5). We also use the

same composite good X t for research inputs (equation 4.6). Hence at any time t world expenditure across goods will be some constant zt including both consumption and innovation. Hence

p jt   y jt ( ) =

w jt A jt

A jt l jt   = w jt l jt   = zt ,

It follows each l jt   = l jt is independent of time t . Profits in each sector are: 

jt



j = 1,2.

in each sector j at each

= (1    1 ) z t , j = 1,2 ,  FDIt = (1  a t   1 ) z t .

Domestic firms have normal innovation profits, as in the case of trade or autarchy. However, FDI firms produce with an advanced technology,

147

DAVID MAYER-FOULKES

charge a price

p FDIt  p1t , employ labor lFDIt  l1t but pay wages

w2t corresponding to a lagging technology and therefore obtain what we term extraordinary profits. Below we show that w2 t / w1t  at . We now work out how production is allocated across the two economies. The boundary 1t between domestic sectors in Economies 1 and 2 is determined endogenously and would shift to the right or to the left if p1t were different to p2t (by attracting more domestic goods into production in the economy offering the cheaper price) except possibly in the boundary cases 1t  {0,1   FDIt } . Now 1t > 0 , because otherwise labor in Economy 1 would not be employed, making w1t very low and additional production possible, so the only boundary case is 1t = 1   FDIt , when all labor in Economy 2 is employed in the FDI sector. The “Banana Republic” refers to this case, when a great part of labor in an economy is employed by FDI. In the extreme case employment in domestic production in Economy 2 is not competitive with employment in FDI goods, so there is no domestic supply and there is no price p2t . Without loss of generality we can set p2t = p1t .16 Now similarly pFDIt cannot be more than p1t , otherwise FDI goods would loose their markets to domestic goods in Economy 1. On the other hand, the competitive fringe for FDI goods is in Economy 1, so pFDIt will be at least p1t . Hence pFDIt = p1t . It follows that all prices p1t , p2t , pFDIt are equal. Let pt  p1t = p2t = pFDIt . Since each good  has the same price, the cheapest way to produce one unit of composite good X t (equation 4.6) is by using one unit of each good This costs

1

 p d = 0

t

.

p t . Let the composite good be the numeraire. Then

pt = 1 . Since expenditure is constant across goods, it also follows that production quantities are equal. Hence: Proposition 5.3. Prices and quantities of production are constant across goods  ,

p1t = p2t = pFDIt = 1,

y1t = y2t = y FDIt = yt .

Using (5.102) and (5. 103) it now follows that wages are

148

(5.104)


w jt =

Ait



,

j = 1,2 .

(5.105)

It is now clear that w2 t / w1t  at . This is the cost ratio of FDI to domestic producers in Economy 2. Proposition 5.3 implies that the price and cost of domestic goods is the same in both economies, and that these goods are produced in the same quantities. Therefore w1t l1t = w2t l2t , and lFDIt = l1t , so

l l1t w2t A2t = = = at , FDIt = at . l2t w1t A1t l2 t

(5.106)

To complete the instantaneous description of the economy, observe:

Remark 5.4. The labor market clearing conditions, setting demand equal to supply, are:

1t l1t = L1t = L1t ,  2t l2t   FDItlFDIt = L2t  LFDIt = L 2t . Using (5.106) it follows from these conditions that

1t =

L1t L ,  2t   FDIt at = at 2t . l1t l1t

Since the measure of goods adds up to 1,

L1t L  qat 2t   FDIt 1  at  = 1. l1t l1t Define the “FDI multiplier”

 FDIt =

1

1   FDIt 1  at 

.

Proposition 5.5. Employment levels in each type of sector are given by: 149

(5.107)

DAVID MAYER-FOULKES

l1t = L1t  atL 2t  FDIt , l2t =

l1t , lFDIt = l1t . at

(5.108)

Hence the measures of the domestic sectors are:

1t =

t at 1 ,  2t = a . 1  t at  FDIt 1  t at  FDIt t FDIt

(5.109)

Income Define the gross world product as Yt =

1

 y d = y 0

t

t

. Since quantities of

production are constant across goods, it is enough to calculate the quantity of production for domestic producers in Economy 1. Using the expressions for wages (5.105) and employment (5.108),

Yt = y1t =  1 w1t l1t = A1t L1t  atL 2 t  FDIt = LGt  FDIt . Define the aggregate global labor in efficiency units,

LGt = A1t L1t  A2 t L 2 t . Theorem 5.6. The aggregate global product is:

Yt  LGt  FDIt 

A1t L1t  atL 2t  . 1   FDIt 1  at 

(5.110)

The aggregate global payroll is:

w1tL1t  w2tL 2t =

1



LGt .

(5.111)

Aggregate profits are:

 1 t = Yt  Wt =   FDIt  LG .  

150

(5.112)


In the absence of FDI, when  FDIt = 0 , the FDI multiplier is  FDIt = 1 , and the aggregate global product equals aggregate labor in efficiency units LGt . If we now admit “normal” innovation market power  > 1, wage participation diminshes.17 Now also admitting FDI, so  FDIt > 0 , the use of more advanced technologies in Economy 2 increases gross output by a factor FDIt, the FDI multiplier. This increase accrues only to aggregate profits, since wages rates are local. Thus aggregate profits are increasing in both market power and FDI expansion (an increase in  FDIt ).

Technological change The income distribution between wages and profits described for globalization is a consequence of the technological lag at (and therefore also of any institutional lags reflected in at ). If underdevelopment is persistent, so these lags are steady state features, and the distribution between wages and profits also becomes a steady state feature. Here we show that globalization is consistent with persistent underdevelopment modelled as divergence in levels (tending to an equilibrium proportional lag) and divergence in growth rates (tending to a lower growth rate). Economic growth can occur in the steady state through technological change. But growth can also occur through FDI expansion, or through technology transfer to the lagging countries. FDI expansion increases the number of sectors  FDIt that can be subject to FDI. We just saw in Theorem 5.6 that the benefits of this kind of growth are skewed towards profits.  FDIt can increase for example by introducing new trade and investment agreements between countries. On the other hand technology transfer to the lagging country increases its productivity and wages, and decreases the participation of profits. It also increases the world growth rate through its scale effect. Figure 5.5 presents the three kinds of growth. The contrast between FDI and technological transfer is discussed below in the section Trade, FDI and convergence in practice. We now introduce technological change in the model. As mentioned above, we suppose that for each good in each economy there is a single, infinitely lived innovator who can produce an innovation for the next period.18 As in the trade model in Chapter 4, each good  will only be in production in one economy, either 1 or 2, because under the equilibrium wages it will 151

DAVID MAYER-FOULKES

have a slight comparative advantage in this economy. Since innovation is continuous, it also has a slight advantage for production after innovation (except that the boundary 1t can shift), and therefore innovation in good  will only occur in the economy that produces it. I consider a myopic innovator in a continuos time model, who maximizes profits in the short term t by choosing innovation inputs, and lets t  0 . We simplify the calculation using the result shown in Chapter 4 that this is equivalent to having perfect foresight on the current time derivatives of the relevant economic variables.

Figure 5.5. Steady state growth, FDI expansion and technology transfer in the globalization model. Steady state growth (arrows with vertical lines) exhibits trajectories with extraordinary profits based on cheap wages resulting from steady state underdevelopment in levels or in growth rates. FDI expansion (arrow with horizontal lines) increases the number of goods  FDIt that can be produced through FDI, and expands both production and the participation of profits in income. Increased technology transfer to lagging countries (white arrow) expands production and wages, and reduces the participation of profits.

152


We use the same innovation function (4.31) as in Chapter 4, this time written as:

 1   A1t  A jt S jt d At =  j  1   A1t dt 



 1   v , 

j  1,2, FDI .

(5. 113)

The innovator chooses at what rate to change her technological level At (a choice which will turn out to be identical to other innovators in her sector by symmetry) by selecting her investment rate v. As in Chapters 2 and 4, the result depends on her (and her Economy’s) level of skills S jt = A jt , an advantage of backwardness convergence effect 1   A1t  A jt , and her

productivity of innovation  j . The term 1   A1t represents the fishing out effect, and  technological externalities from the leading edge generating growth. Assume that the competitive fringe advances technologically at the same rate as the economy does ex-post, based on technological externalities. Write  j (t   t , v ) for the profit level of an individual firm in sector j investing at rate v on the time interval [t, t   ] , with We can write the first order approximation

 t t = (1 

b jt Ajt  t

 ( Ajt  t dtd At )

t (0, ),   0 .

) yt  t .

(5.114)

Here b1t  b2t  1 , while bFDIt  at . The incumbent maximizes

max ( v

 j  t

Her first order condition is therefore

(t , v )  v )

 2 j vt

(t , v )  1 . Since we are using

twice continuously differentiable functions, we can first differentiate with v and then with t , evaluating at t  0 :

153

DAVID MAYER-FOULKES

  1   A1t  A jt S jt  b jt A jt  t t j (1   ) 2  j 1   A1t    (t , v) =  vt t  ( Ajt  t dtd At ) 2  

     v    yt  t    t  0

 (1   )b jt  1   A1t  A jt A jt      v yt .  j 1   A1t A jt   

Since this quantity must equal 1, it follows that

  (1   )b jt yt v j  A jt  Substituting in (5.113), and writing are perfect,

1/ 

   

1   A

1t

 A jt A jt

1   A1t

d dt

.

At = A jt , since myopic expectations

 ((1   ) A1t  Ajt ) Ajt  (1   )b jt yt    A jt =  j    (1   ) A1t Ajt    1

1 



,

j  1,2, FDI .

(5.115 )

Note that FDI firms have lower incentives to innovate even though their profits are higher, because they face lower cost ratios. This holds so long as strategic competition with Economy 1 firms is not considered. As mentioned before, we therefore assume for the first mode of globalization that FDI firms decide to innovate at the same rate as Economy 1 domestic firms, maintaining AFDIt = A1t . Since yt  t depends on at , a relative scale effect in the future level of Economy 2’s size is introduced, that is simplified by using continuous myopic foresight, bringing it to the present. We write

at  =

L1t  atL 2t yt = . A1t 1   FDIt 1  at 

Define also the effective innovativity parameter

154




~ j = 1    1

1 





1

j , 

j  1,2.

These final innovativity parameters are decreasing in market power, because the higher the market power, the relatively lower the input costs and therefore the lower the proportional impact of technological improvement on profit. The cases

j = 1,2 yield the technology growth rates in Economies 1 and 2,

1  A1t  ~ 1at   , = A1t 1  

1  A2t ~  a  = 2 1  t at   . A2t  1  

Hence the growth rate of relative technological level at is: 1  at   a   ~ =  1  t ~2  1 at   . at   1    1  

We can now describe the technological dynamics.

Theorem 5.7. Economy 1’s technological level 1 

g at  = 1 ~1at  

A1t

grows at a rate

that depends on the size of the global economy relative to

A1t , and is increasing in  FDIt . If ~2 < 1 ~1 , Economy 2 diverges in growth rates

with Economy 1. If ~2  [ 1 ~1 , ~1 ) , Economy 2 diverges in levels converging to a steady state a = 1    ~1 / ~2 . Equality between the two economies is achieved

~  ~ . Finally, if ~ > ~ Economy 2 overtakes Economy 1. If ~ is the when  2 1 2 1 1 optimal innovation productivity, policies leading to the development of Economy 2, for ~  ~ , lead to the highest world growth rate g 1 . example by rendering  2 1 a Proof. Note that 1  1t  ~  ~  at a = 1 .   1

2

1

t

 ~   ~ is decreasing in a and equal to t  2 1 1 Thus Economy 2 overtakes Economy 1 if

~2 > ~1 . On the other hand the same expression is negative at at = 0 if 155

DAVID MAYER-FOULKES

~2 < 1 ~1 , implying Economy 2 diverges in growth rates with Economy 1, with 1  1  A2t ~ A1t  ~   .     =   0 < =   0 lim lim 2 1 t  A t  A 1   2t 1t

In the intermediate cases Economy 2 diverges in levels with Economy 1, to the given steady state, exactly catching up at equality. Finally, if we suppose ~ , the relative that innovation productivity cannot be higher than  1 technological level at is bounded by 1, and the highest possible world growth rate is given by g (1) , since (at ) is increasing in at .

Transnational corporations as leading sector Since TNCs have more resources than domestic Economy 1 firms, while the incentives for innovation derived from their cost structure in Economy 2 are lower, they have strategic incentives to innovate more than domestic firms in Economy 1, to be able to take over more sectors of production. These firms have integrated production structures (e.g. Baldwin, 2012), allowing their operations in multiple advanced economies to take the form of FDI, perhaps employing cheap human capital rather than cheap unskilled labor. In addition, they will dedicate economic and political resources to increasing  FDIt . For example non-equity modes of investment are a new way to expand FDI (UNCTAD, 2013). Thus we now consider the case when TNCs as a set constitute the leading knowledge system and can be considered as Economy 1. We analyze the relation between the TNC sector and a typical country, which constitutes Economy 2, whose own innovation is located in domestic corporations not performing FDI abroad. Economy 1 does not have a population, so L1t = 0 and  1t = 0 . Instead it employs Economy 2 labor, in the sector hosting FDI. From equations (5.108), l FDIt = L1t  atL 2t  FDIt and

l2 t = and

L1t  atL 2t  FDIt . at l2t = L 2t  FDIt .

Setting L1t  0, it follows lFDIt = atL 2t  FDIt

Income

can

now

be

calculated,

Yt  A2tL 2t  FDIt , and Theorem 5.6 holds with L 1t = 0 . 156

yielding


Discussion The model demonstrates the main facts highlighted above about globalization, understood as the result of factor exchange between economies at different levels of development. This simple hypothesis is enough to understand that globalization: (a) Increases aggregate economic growth (see Theorem 5.7); (b) Increases the economic participation of TNC profits, lowering worker participation (Theorem 5.6). (c) Is logically consistent with development, underdevelopment and miracle growth (see Theorem 5.7; miracle growth occurring when a structural reform significantly increases ~2 ). (d) Increases inequality in leading countries (Theorem 5.6, since FDI investors belong to Economy 1). (e) Leads to lower corporate taxes and more conservative policies, given that TNCs have additional resources to influence policies across economies (Proposition 8 in MF). MF also shows that if the innovativity of the competitive fringe rise, so that market power  is reduced, the participation of wages and effective innovativity rise. Also, production becomes more efficient in the case when there are goods inputs. Thus the model upholds Adam Smith’s result that market power diminishes welfare, but does not assume profits are zero. Both profits and persistent underdevelopment exist. They do not disappear under laissez faire, much to the contrary. Moreover, the model shows that these two are linked: it is precisely underdevelopment, as the persistent determinant of low FDI host-country technology and wages, that gives rise to increased TNC market power. In addition, profits are ubiquitous, not only in underdeveloped but also in developed economies. Perfect competition does not describe economies with vibrant innovation. It follows that theories that do not explain the existence of both profits and underdevelopment cannot explain the main facts and challenges of globalization. The model shows that under globalization increased TNC market power is as persistent as underdevelopment, and that it increases economic and 157

DAVID MAYER-FOULKES

political inequality everywhere, weakening national as well as international governance and democracy. One policy for meeting the twin global challenge of market power and underdevelopment is to harmonize global corporate taxes and use the proceeds for sustainable economic development everywhere (in particular raising the supply of publically provided goods), applying what in Chapter 6 are called distributiveinnovative policies. Another policy is also suggested in Chapter 6, a profit rate tax that raises no revenue in equilibrium but provides incentives for producers with market power to raise their production levels (thereby diminishing profit rates). For simplicity the model does not include comparative advantage. Were it to be included, it is clear that the trade element of the model would work similarly, with an increase in productivity from trade and therefore also of the global market size, increasing the resources and incentives for innovation. As before, the scale effect in innovation is a common effect between countries and might be mitigated by effects such as increasing variety of goods and so on. Fixed productivity effects, such as institutional differences between countries, can be incorporated by replacing the technological level A jt with

q j A jt , where q j is the institutional productivity effect. This impacts the steady state economic levels, as is shown in MF. Institutional differences affecting innovation productivity can be considered incorporated in 1 , 2. There are other impacts which can be considered. FDI is often claimed to raise technological change in host economies. However, this type of effect has been often found to depend on a local absorption capacity. We could define this as a weak FDI externality, proportional to A2 t , a strong one being proportional to A1t . Mayer-Foulkes and Nunnenkamp (2009) examine this theoretically, showing that divergence is possible when the externalities are weak, and impossible when they are strong. They also show that US FDI has impacts on growth hat depend on the technological level of the host countries, and are therefore consistent with both convergence and divergence. It is also possible for FDI to raise wages in the host country, though that tends to defeat its purpose and can cause it to migrate. Wage competition 158


between Economies 1 and 2 is also possible, and can put a downward pressure on wages in Economy 1.

Trade and FDI as development policies There is an important contrast between the trade and FDI models. When a country can produce a good for export, it can then engage in innovation to maintain its competitivity on the world market. However, if it allows a good to be produced by foreign investors, in general it will not be able to become a future producer of the good. An example is furnished by automobile production. Let us consider the situation in 1998. Automobiles represented a fairly mature product with a not particularly impressive rate of innovation (yet, 78% of world sales are due to ten corporations, three from the US; Najera, 1998). Many middle income countries were quite capable of engineering and producing a line of automobiles. However, doing so while facing the competition of established and advanced producers, as Argentina attempted, may be impossible. Table 5.1 shows countries and regions of production and consumption of automobiles. Countries in Europe and North America that developed the automobile continued to produce and trade it, both with each other and with other countries. In contrast, the new producers, mainly Japan and Korea, did not import automobiles. These countries developed their capabilities in automobile production by promoting their exports and performing a full import substitution, eliminating competition from FDI in automobiles. On the other hand, Latin American countries that supposedly “substituted for imports” by allowing FDI in automobiles (since 1916 in Argentina; Ford home page; and since 1920 in Mexico; Soto-Rodríguez, 2002), did not develop their own industries.

159

DAVID MAYER-FOULKES Consumers

South America

Other Europe and Turkey

Other Asia, Pacific

S Korea

Japan

European Union

LDC's

9,508 3,846 350 52 611 520 1,077 1,240 European 3,636 11,881 811 10,049 1,251 88 39 Japanese 2,851 1,954 28 150 S Korea 756 Other Asia, Pacific 792 Other Europe South America Total 15,995 16,538 10,049 1,954 2,905 2,159 1,890 Table 5.1. Consolidated World Motor Vehicle Production by Nationality of Origin Consumption, 1998 (thousand units) Data Source: ILO (2000).

LDC's Developed

Producers

NAFTA

Developed

American

Total

14,367 18,354 15,089 2,132 756 792 and

Automobile production thus provides an example of innovation crowding out by FDI. On the set of FDI goods  FDIt , Economy 1 producers outcompete Economy 2 producers, and tend to remain the incumbents permanently. Even if Economy 2 had very little expertise in producing the good originally, for example producing horse carriages, if it had made the necessary investments to upgrade to automobiles it could have become a permanent producer, while if it allows FDI, this possibility might be permanently lost. With respect to the opening to trade it is often mentioned that there must be some winners and some losers in production. The same holds for innovation. Economy 2 must have enough winners in innovation if it is to develop. If winners and losers are sufficiently balanced, everyone gains since the world economy grows faster, as shown in Theorem 5.7. Note that under FDI Economy 1 innovates in 1t   FDIt goods while Economy 2 innovates in  2t goods. We can write down the following result about innovation crowding out.

Theorem 5.8. The larger the size  FDIt of the FDI sector, the more sectors Economy 1 innovates for and the more innovation crowding out there is in Economy 2.

160


Proof. From (5.107),

1  at  FDIt =  0. Hence  FDIt 1   FDIt 1  at 2

 2t t at  FDIt =  a  0, 2 1  t at  FDIt  FDIt t  FDIt while

 1t   FDIt   1   2t  =  0.  FDIt  FDIt Note that the case  FDIt  0 reduces to the trade model with   0.

Theorem 5.8 will extend to the case   0 of FDI plus comparative advantage, since Economy 2 will produce in less domestic sectors when it hosts FDI. Theorem 5.7 about innovation crowding out helps explain why underdevelopment can be persistent in the presence of FDI. It also implies that the analysis in Chapter 4 on innovation externalities is stronger when there is FDI. The analysis goes through simply by replacing h(1t ) with

h(1t   FDIt ) throughout, for example in equation (4.36). Theorem 4.4 then applies, writing “trade and FDI” instead of “trade”.

Trade, FDI and convergence in practice We turn now to the Asian tigers for further comparison of trade and FDI as development policies. Wan’s (2004) remarks: “The acceleration and then deceleration of East Asian growth is one of the mega- events of the 20th Century”, suggesting that “for all developing economies, ... ‘fully developed’ status is within reach.” This remark contextualizes Wan’s (2004) detailed comparative analysis of East Asian growth experiences. For him, the experiences of Japan, Korea, Taiwan, Hong Kong, Singapore, and lately China, constitute strong empirical evidence that development and underdevelopment are distinct steady states, and are explicitly modelled as such. The policies that these countries applied throw light on the barriers they surmounted and the economic forces they harnessed to reach development.

161

DAVID MAYER-FOULKES

Japan and Korea, the largest countries (in 1999, the East Asian populations were, in millions: Japan, 126.6; Korean Republic, 46.8; Taiwan, 21.9; Hong Kong, 6.9; Singapore, 3.2), concentrated on the creation of large industrial firms with scale economies, and combined export promotion with import protection. They used trade to integrate with the US production chain and were careful to obtain technological transfer. Both avoided depending on FDI, and they promoted innovation rather than imitation, attaining dominance in the supply of new product lines. Japan depended on domestic saving. Some separation in government economic powers helped it to maintain e ciency in the support of carefully selected and changing infant industries. Korea was more authoritarian and more dependent on foreign saving, leading to some ine ciencies and financial instabilities. Taiwan concentrated on small and medium enterprises, avoiding reliance on FDI, and only subcontracting from FDI under specific regulation for an emphasis in technology transfer, also integrating with the Japan-US production chain. Backward integration with technological absorption was closely promoted. In e ect, the local portion of what would be FDI, as well as the backward linkages, were locally owned. This form of ‘subcontracted’ FDI transfers innovation incentives to the local firms. Summarizing, Japan, Korea and Taiwan, avoided reliance on FDI, and emphasized technological transfer. Cheap labor favored, and even subsidized, local firms rather than FDI. Hong Kong’s manufacturing growth was also triggered by trade, in this case triggered by natural advantages including Hong Kong’s position as an outlet for China and its well located harbor. Cheap food imports from China also played a role. Singapore, the smallest of these countries, is an exception in that its economic policy was to use low wages and taxes, and an excellent location, to attract FDI. It relies on an authoritarian regime including ‘bonded scholars’ to attract cheap-human-capital-seeking FDI, with income rising on the basis of human capital accumulation. China began a process of fundamental reform by introducing market mechanisms into its economy in December 1978. After an initial and successful period of transition, it also established trade and FDI policies to gradually open its economy. In parallel with other Asian countries, China combined export promotion with import protection. Since import protection can raise costs for exporters and make the home market more attractive for producers (Flatters and Harris, 1994), China also neutralized 162


this anti-export bias by allowing exporting sectors to import goods duty free. China’s policy towards FDI (authorized since 1979) as a source of foreign capital, technology and management skills, has also been selective. The Twenty Two Regulations on FDI (1989) promoted export oriented and technologically advanced projects (Bransteeter and Feenstra, 2002). FDI was encouraged in sectors targeted for import substitution, while it was severely constrained in other sectors, especially its access to the domestic market. The Chinese government has limited the ability of foreign firms to compete in the Chinese market by imposing export, localization, and technology transfer requirements as well as di culties in recruiting and retaining key personnel. China’s policies were extremely successful. Its average annual GDP growth increased from 5.3% between 1961 and 1979 to 9.5% between 1980 and 2003 (Fan and Felipe 2005). Exports rose form 6.4% of GDP in 1979 to 21.8% in 1999. China has become the second largest host of FDI, after the US. Distortions that may be inevitable under any industrial policy have also appeared in China. Since the early 90’s, it has gradually lowered both tari protection and FDI restrictions. (This paragraph is based on Lemoine, 2000). In Japan, Korea, Taiwan, Hong Kong, Singapore and China, specific policies or natural advantages were indispensable to harness trade and FDI to drive capital accumulation and technology transfer. It is instructive to compare these experiences with those of countries that did not apply such policies but, instead, opted for straightforward liberalization, such as India, Brazil and Mexico. India underwent liberalization in 1991 after following a strategy of importsubstitution policies since independence in 1947. Following a macroeconomic crisis, a stabilization and structural adjustment program supported by the IMF and the World Bank reformed trade and FDI (Nagesh, 1995). The New Industrial Policy abolished most of the industrial approval system that had existed until then. These policies met with success, and India’s growth rate rose between 1981 and 1991 from the “infamous ‘Hindu’ rate of growth of 3.5%” (Srinivasan, 2002, page 3), to 5.4 %, and then to 6.0% between 1992 and 2003 (Fan & Felipe, 2005). FDI in power generation, telecommunications, petroleum exploration and transportation sectors enjoyed special incentives, given the importance of these sectors for trade and industrial development (Sharma, 2000), although these did not specifically promote technology transfer. The Indian FDI stock increased to almost US$ 39 billion by 2004 (UNCTAD online database). Nevertheless, India lags behind China in many economic indicators (Chakraborty and Nunnenkamp, 2006). Our model o ers a clear explanation for this comparatively mixed success. India’s GDP (a measure 163

DAVID MAYER-FOULKES

of its productive capacity) in 1991 was the sixth largest in the world, after the US, Japan, China, the USSR (less involved in trade) and Germany (Maddison, 2003). Therefore, its transition could be expected to be one of convergence, tending towards the capture of a sizeable set of export sectors. However, India did not significantly apply policies supporting technological transfer. Hence, it did not achieve the Chinese rates of growth. The comparison of Brazil and Mexico with India is interesting. These countries have applied similar policies. They have liberalized trade and received relatively large flows of FDI, and have not significantly applied technological transfer policies in relation to trade and FDI (for Brazil, see Rothmuller, 2003; for Mexico see López-Córdova, 2002). Yet, they have been less successful than India, achieving less economic growth. The model suggests, through the role played by country size in capturing production goods, that one reason may be that these countries are smaller. In 1995, Brazil’s GDP lagged behind the US, China, Japan, India, Germany, Russia, France, the UK, and Italy, while Mexico also lagged behind Indonesia and Canada. Although Brazil was not too far below these European countries, these also enjoy the benefits of outward FDI flows. Thus, Brazil and Mexico are at a disadvantage in the allocation of innovation sectors through free trade, and, in addition, su er innovation crowding out through FDI. Nevertheless, since they are not too far in size from these technologically autonomous European countries (for which convergence under trade occurs, or at least seemed to before the crisis) perhaps Brazil and Mexico are not situated too far below the convergence threshold under free trade. An additional push through technology transfer policies could lead to their transition to development. This general comparison of trade, FDI as development policies in practice shows that NIC-style policies (NIC for Newly Industrialized Country) are quite different from free trade and investment agreements, which may not elicit enough technological transfer to create sovereign know-how in lagging countries (see Figure 5.5 for a comparison of FDI and technology transfer). While such know-how is experienced by leading country incumbents as a threat, in fact when additional countries rise to development, they become equal partners, welfare rises, countries specialize in different sectors of production, workers from advanced countries cease to face competition from low salaries, and more resources become available for innovation, raising the world economic growth rate. This was the case in the relation between the Asian Tigers and the US, which brought mutual prosperity.

164


Main ideas The present wave of globalization began with Ronald Reagan and Margaret Thatcher's liberalization policies since 1982, which led to the creation of the World Trade Organization in 1995. In addition to Western liberalization, China's introduction of market mechanisms in December 1978, and the fall of the Berlin Wall in 1989, came together to create a global market economy. Globalization led to economic growth through the increase of world trade and Foreign direct investment (FDI). Much of the increased international production involved the combination of factors of production, capital and technology from developed countries and cheap labor from underdeveloped countries. This combination produced large investment and profit flows, with much of trade occuring as intrafirm trade. Profits from globalization contributed to a global rise of inequality. Also, there came a point when these profits could not all be reinvested. In 2005 a savings glut appeared, reported by Bernanke, Chairman of the US Federal Reserve in 2005. This glut fed investment bubbles that led to the 2008 financial crisis. High profit flows and low interest rates still continued in 2015. A model of economic growth combined with trade and FDI between developed and underdeveloped countries explains the main features of this asymetric, polarized form of globalization. Introducing any model for underdevelopment, such as the need for an absorptive capacity for technological change, innovation externalities, or simply institutional differences, completes an explanation for how in the steady state globalization: (a) Increases aggregate economic growth; (b) increases the economic participation of TNC profits, and lowers wage participation; (c) admits development, underdevelopment and miracle growth; (d) increases inequality in leading countries; (e) leads to lower corporate taxes and more conservative policies. Moreover, a second stage of globalization is possible in which transnational corporations as a set conform the technological leader, with all countries becoming follower economies. The policies which led the Asian tigers or NICs to economic development were different. They emphasized not only larger markets through trade, but also technology transfer, which is weaker through FDI.

165

DAVID MAYER-FOULKES

13 This chapter is based on Mayer-Foulkes (2015a), referred to as MF. References and an expanded exposition can be consulted there. The model presented here is a simpler version. 14 It is tacitly understood that economic growth with increasing welfare for all includes sustainability as a basic criterion. 15 The Case-Schiller index for house prices was 154.9 in mid-2015, 27.5% above its three year high of 121.4 in 1979. http://www.mybudget360.com/the-magical-2housing-ratio-between-median-nationwide-home-prices-and-household-income/, read 10/24/2015, reported that while the ratio of median home price to median annual household income was near 2.2 from 1950 to 2000, it rose to 3.3 in 2010. 16 Note if all sectors are involved in FDI then wages are not fully defined domestically, and neither is the home technological level. We will assume below that home knowledge does continue to exist and follows the same dynamics, although other analyzes are possible. 17 MF shows that when goods are used as production inputs, the market power distortion introduced through prices in the input ratio between labor and goods also reduces aggregate output. 18 As in Chapter 4, we are therefore abstracting from creative destruction. This is equivalent to considering that creative destruction is neutral to trade and FDI, and that the innovation effectiveness and cost parameters are after creative destruction.

166

6 THE MASS MARKET ECONOMY19 Mass production truly came into its own with the Second Industrial Revolution (1867-1914). Truly large enterprises producing for the masses, way beyond anything Adam Smith could have imagined, only finally appeared when electricity began to power the conveyor belt, as in Henry Ford's 1913 assembly line producing a Model T every 93 minutes. As mentioned before, mass production remains the basis of modern productivity. From 1935 to 1992, in 459 industries the four largest firms produced on average 38.4% of all shipments. Similarly, from 1992 to 2002, the 200 largest manufacturing companies accounted for 40% of manufacturing value added. Today, more than 51.6% of the US workforce is employed in firms with 500 or more employees. The purpose of this chapter is to show that mass production brought with it a qualitative change in the behavior of the market economy that makes it necessary to take technological change into account in order to understand the short term, or static, equilibrium. We show that the idea that perfect equilibrium models with their optimality properties are an approximation to the short term, while technological change is an explanation of the longterm, does not adequately explain the income distribution and efficiency properties of a mass production economy. It therefore does not appropriately guide government policy, nor does it adequately frame a discussion of good institutions. The idea “market economy” is too closely linked with the idea “perfect competition”. Historically it was a significant challenge to develop the theory of general equilibrium, treated by Walras (1874) in his Elements of Pure Economics. The existence of general equilibrium was rigorously 167

DAVID MAYER-FOULKES

established in 1954 by Nobel Prizes Arrow and Debreu, using complex mathematical techniques from differential geometry. To establish their intertemporal results they assumed that profits are always zero. The theory of endogenous technological change, including the optimization of resources dedicated to R&D, appeared in 1989 (Romer, 1989; Grossman & Helpman, 1989; Aghion & Howitt, 1989). Long-term innovation is motivated by profits, and the innovation rate establishes an equilibrium level of profits. In the present chapter we combine the two kinds of model to understand the market economy with mass production. While it is possible in any context, static or dynamic, to make the assumption that competition reduces profits to zero (and this makes the mathematics more tractable) in real life significant profits exist. The layman would be truly surprised to hear economists assume that profits by and large are zero. It is more realistic to think that profits break the equilibrium and lead to new developments. Chapters 2, 4 and 5 showed that models of technological change including the need for skills or credit for technological absorption, or involving research externalities between innovation sectors, can explain both convergence and divergence across countries, both with and without trade and FDI. Note how these explanations for current income differences are based on long-term technological dynamics. Note too that these long-term processes produce through FDI a sizeable stream of profits every day. We show in this chapter that in developed economies too technological change is an essential element for defining efficiency and distribution in the current static equilibrium. Since the Second Industrial Revolution technological change has become so systematic that sectoral differences in the feasibility of technological change have come to shape productivity differences within countries. These directly intervene in the distribution of income through the distribution of profits and the determination of equilibrium wage levels. To highlight these insights we concentrate on a leading economy. Some of the questions we are addressing are the following. How can wages remain low while technological levels rise? Wages are predicted to be proportional to the average technological level, both in general equilibrium models, and in models of endogenous technological change. The market economy often does not absorb enough of the population into good jobs, both in developed and in underdeveloped countries. In the context of full employment models, what this means is that wages are low compared to the potential technological level. This is related to the prevalence of poverty and poverty belts. Suffice it to mention that in 2010 there were 14 states 168


(plus the District of Columbia) in the US where 30 percent or more of the population lived in poverty areas (Bishaw, 2014). In 2013, 14.5% of the population of the US was poor, 2.0% more than in 2007 (DeNavas-Walt and Proctor, 2014). How can income concentration reach such high levels in the context of market competition? That 85 people can own approximately as much wealth as 2.37 billion people, the poorer half of the world population (as mentioned in Chapter 1), reflects on the dynamics of the market economy more than on specific persons. High technologies and high profits are linked with large enterprises, and poorly employed labor with small firms or self-employment (often in the streets). We find that the answer to our questions is related with the interaction of small and large firms. The coexistence of large and small enterprises is a feature of both developed and underdeveloped countries. In the US, in 2012 51.6% of the workforce was employed in the 0.3% of firms with 500 or more employees, where they eraned 58% of the wagebill. Meanwhile 89.6% of the 5,726,160 firms had less than 20 employees, employed 17.6% of the workforce, and earned 14.2% of the wagebill. In Mexico, in 2008 out of a total of 3,626,954 firms, the 1000 largest firms (that is, 0.027% of firms) employed 19.5% of workers, paid 40.5% of the wage bill, and produced 65.3% of value added. We model a mass market economy20 as consisting, for simplicity, of two sectors, one innovative, with monopolistic competition, and the other absorptive (we refer interchangeably to technological diffusion, absorption or adoption), with perfect competition. The model is related to the ones presented in Chapters 4 and 5. Note that the institutions that support markets allow for the coexistence of competition both with and without market power. Moreover, two core productive processes generate market power, innovation and large scale production. Yet at the same time millions of small firms produce in a context approximating perfect competition. Thus to understand the full features of the industrial or mass market economy, we consider how production with and without market power combine. We explain how the interaction of a) mass production with technological innovation under monopolistic competition, and b) small scale production with technological absorption under perfect competition, can explain the prevalence of both high income concentration and relatively low wages.

169

DAVID MAYER-FOULKES

Let us turn to the small scale sector. By now, in a country like the US, in general terms anything that can be mass-produced will be mass-produced. This implies that small firms are producing goods that for diverse reasons cannot be mass-produced. Examples are medical services, individual house construction, some forms of commercial distribution and services, etc. So long as individuality is valued and people live in their own houses, have their own relationships, care for their own children, and perhaps pursue an extraeconomic meaning to their life, a fully mass-produced society can hardly be conceived. Nevertheless, the productivity of mass technologies clearly shapes society, for example our rural-urban social habitat, and deeply interacts with identity (e.g. Lunt and Livingstone, 1992, Akerlof and Kranton, 2000). We characterize the small scale sector as covering goods for which innovation cannot be financed by obtaining sufficient profit margins over a significant proportion of their market. This could be either for technological reasons, or because improvements cannot be appropriated. This setting of technological change in the small scale sector implies several sources of inefficiency. First, the technological absorption that is conducted is repeated by all producers, and restricted in each case to a small scale effort. These efforts are not pooled to produce better results. Second, inappropriable innovation is not pursued, including the use of mass production techniques when feasible. We refer to these sources of inefficiency as the public good nature of technological absorption. They imply that public policies can be applied to raise the productivity of technological absorption in the small scale sector. While large scale production and innovation generate productivity and growth, through their market power they also generate income concentration and inefficiency. The detrimental impacts of market power on efficiency and distribution are clear when analyzed from a static point of view. Here we also study the dynamic context, where market power also presents detrimental impacts on innovation. First, price distorsions due to market power make the input mix less rich in innovative goods, both in consumption and investment. Second, market power, expressed as a mark up, implies that costs appear relatively lower and so there are lower incentives to innovate. We discuss a public policy for reducing market power when this can increase innovation rates, namely a profit rate tax presented below. This also applies to the case when some market power is optimal, as in Aghion et al’s (2001, 2005) inverted U-curve result for innovation as a function of market power. Summarizing, the aggregate product of the mass market economy is a function of the technological levels of both the large scale and small scale sectors. While the innovative sector leads economic growth, it also 170


generates inequality and inefficiency in both production and innovation. At the same time, the small scale sector absorbs technologies inefficiently. These inefficiencies in production, innovation and absorption explain how wages can lag behind their potential level, since the overall level of wages is a positive function of the technological levels of both sectors and a negative function of market power.21 In turn the profit concentration in large scale firms explains income concentration at high income levels. It follows that free market policies are not optimal, as we discuss after the model. We can now assert that the dynamic properties of innovation and absorption determine some of the main static properties of the economy, such as relatively low wages and income concentration. Technological change dynamics are an essential determinant of the current static equilibrium.

III. The mass market economy model Let us define a mass market economy consisting of two sectors. The heart of the mass market economy is an industrial, technological, mass production sector characterized by ongoing innovation. Innovation is motivated by the acquisition of market power and generates firms spanning important portions of their markets. (We consider even small, specialized, innovating firms part of the innovation, mass production sector if they produce for an important portion of their market.) These large scale firms generate income concentration. Now, for various reasons innovation cannot be financed for every type of good by obtaining sufficient profit margins over a significant proportion of its market. An important proportion of the working population is employed in a second sector, consisting of many small firms that do not innovate significantly and operate competitively. These small, non-innovating firms (including self-employment and informal economic activity), improve their productivity by expending effort on absorbing technologies developed by the industrial sector, which functions as its technological leader. Relative to each other, the large and small scale sectors display opposite characteristics. While the first is innovative and has market power, the second absorbs technologies and is competitive. From the static point of view, the innovative sector is at the same time physically more productive, employing higher technologies, and economically less efficient, diverting resources from production of innovative goods through high prices. By contrast, the small scale sector is physically less productive, employing lower technologies, and economically more efficient, since it is more competitive. 171

DAVID MAYER-FOULKES

For simplicity, we keep the distinction between goods belonging to the large and small scale sectors exogenous, a function of the kind of good in question. Similarly we abstract from horizontal innovation including the appearance of new small or large scale goods, or of goods that have their origins in small enterprises that become large. Consider an economy with two sectors L and S that produce a continuum of tradable goods indexed by   [0,1] , where each  refers to a good. Large scale sector goods   L = [0, ) use a mass production technology and are therefore modelled with all production concentrated on a single large producer that is able to make a profit, while small scale sector goods    S = [ ,1] are produced on the small scale, with constant returns to scale, and modelled with infinitely many small, identical, competitive producers. We assume  > 0 for some sectors to innovate, and  < 1 since not all sectors are amenable to mass production.22 In each sector technological change is endogenous, with differences due to the different competition structures. For simplicity we abstract from innovation uncertainty and assume that innovation is symmetric within each sector L and S . Thus we are assuming goods    j in each sector j L, S have the same technological level A jt . Innovation occurs as follows. In the large scale sector L there is for each good   L a single, infinitely lived innovator who invests in innovation and produces monopolically in the presence of a potential competitive fringe. For simplicity we assume that innovation is cheaper for the producing incumbent than for any other innovator, and she therefore has an innovation advantage.23 Her monopoly therefore persists indefinitely. By contrast, in the small sector S anybody can innovate, so as to reap the productive benefits of new technologies, increasing the returns to production factors such as labor. We assume that small producers can produce any good, while large producers can only produce goods in sector  L for which mass production technologies are available that are more productive than small scale technologies.

172


Production, consumption and prices For simplicity we exclude capital from the production function and limit ourselves to innovation as the only source of market power. Thus we only distinguish the two sectors by their competitive context.

Definition 6.9. The production function for goods    j in sector j  L, S is:

y jt ( ) = Ajt l jt ( ),

j  L, S .

Here y jt ( ) represents the quantity produced of good

   j . A jt is the

technological level in each sector. l jt ( ) is the quantity of labor input. We assume that the small scale sector can produce any kind of good, but the large scale sector is ahead in productivity, ALt > ASt . We use the same preferences as in Chapter 4 (equation 4.5). Suppose a L

S

consumer has a budget zt for purchasing quantities ct , ct of goods produced in the large and small scale sectors. We assume large and small scale sector goods   L ,  S are symmetric within sectors so have common prices pSt , pSt . Since the subutility function Ct is CobbDouglass, consumers dedicate the same budget zt to each good. Thus, the quantity purchased of each type of good is:

z ctL= t , ctS= p Lt

zt . pSt

Hence the subutility is given by ln Ct =  ln

Ct =

zt  1   ln p Lt

zt , that is, pSt

zt . Let us set the price level so the cost of a subutility amount pLt p1St 

Ct = 1 is 1. Then

173

DAVID MAYER-FOULKES

pLt p1St = 1.

(6.116)

Let wt be the domestic wage level. Note that therefore aggregate net 1



income is Z t = zt d = zt . 0 In the case of small producers one unit of good    S is produced competitively by infinitely many firms. Wages equal the income from selling the product of one unit of labor, so the price can be written

pSt =

wt . ASt

(6.117)

For each small scale good    S let lSt   be the aggregate employment of all of the firms producing this good. Since the number of units produced is c Sjt =

zt = ASt lSt   , the labor quantity is constant in  , so we can pSt

drop  from the notation, and

lSt =

zt z = t. pSt ASt wt

(6.118)

In the case of the large scale sector, each producer has two types of potential competitors. The first type of competitors are small-scale producers, who can produce good  using a technological level ASt . Hence it will always be necessary that p Lt  p St , mass production just being feasible at equality. The second type of competitor, in the competitive fringe, has a lower technological level

 1 ALt , implying large scale firms

can have a markup  > 1 , since potential producers in the competitive fringe are just unwilling to enter at zero profit. The level of production considered by the incumbent and her competitor are given by the aggregate expenditure level on this good, z t =

pLt ( ) y Lt ( ) . The markup of the will be  . Unless we are considering a transition for which mass-production comes into existence with low levels of technological advantage, the usual case will be when under the full 174


markup  nevertheless p Lt  p St . The incumbent will drive her industrial competitor to the zero profit limit, and will therefore act as if her productivity were ALt / . Hence instead of (6.117) we have

pLt =

wt ALt

.

(6.119)

The incumbent produces the same quantity but employing less labor,

l Lt = therefore at a cost

zt  1 zt = p Lt ALt wt

(6.120)

 1 zt , hence making a profit  Lt = (1   1 ) zt .

(6.121)

The wage level can now be obtained by substituting of (6.117), (6.119) in 



(6.4.86), so that 1 = p Lt p

1 St

 w   w  = t  t   ALt   ASt 

1

. Hence

wt =   ALt ASt1 .

(6.122)

Substituting back in (6.117), (6.119) and simplifying, we can solve for the prices in terms of the current technological levels, 

  1 ALt  pSt =   , A  St 

 A  pLt =  1St    ALt 

1

.

(6.123)

Hence for large-scale production to outcompete small-scale production at a mark up level  , the technological levels must satisfy

pSt  1 ALt = > 1. p Lt ASt

(6.124)

We will keep to the case where this condition holds, not just ALt > ASt .

175

DAVID MAYER-FOULKES

Labor and income Let the population of the economy be L . Suppose L L and L S are the aggregate employment levels in sectors L and S , with L L  L S = L . Then if employment levels for each good are l Lt , lSt , when the labor market clears,

l Lt = L L , (1   )lSt = L S , l Lt  (1   )lSt = L. Now wt l St = zt , since the participation of labor equals income in sector S , while wt lLt =  zt in sector L . It follows that 1

lSt = , l Lt as also follows from (6.118) and (6.120). Hence, we can solve

l St =

 1L L , l = . Lt  1  1     1  1   

(6.125)

From wages and employment income now follows. Using equations (6.122) and (6.125),

  ALt ASt1 L zt = wt l St = 1 .    1   

(6.126)

The average wage participation is

wtL =  1  1   . zt

(6.127)

Wage participation in the large scale sector is lower than in the small scale sector, so as  rises, wage participation drops.

176


Efficiency and equity under market power While in some situations there may be a trade-off between efficiency and equity, market power simultaneously results in less efficiency and less equity. This also holds at the macroeconomic level in a mass market economy. The static distortions due to the presence of market power are the following.

Theorem 6.10. Market power distorts the mass market economy as follows: 1) Aggregate income is decreasing in market power. 2) The profit to income ratio is increasing in market power. 3) Wages and aggregate wage participation are decreasing in market power. 4) Employment intensity l Lt in the large scale sector is decreasing in market power, while employment intensity lSt in the small scale sector is increasing in market power.









d  1     1    =   1 1   1 1    > 0 , so from d dZ t (6.126), < 0 . 2), 3), 4) follow from (6.121), (6.122) and (6.125). d

Proof. 1)

When the proportion  of mass producing sectors increases, the presence of market power implies that wages do not rise in proportion to the increased productivity. Let us examine how the relative size of the large scale sector  affect wages in the presence of market power  .

Theorem 6.11. When the size of the large scale sector increases, wages respond as follows:

  ln wt   ln    ln ALt  1   ASt  = ln ALt < ln ALt . =  ASt ASt  Proof. Differentiate (6.122) and note (6.124). 177

DAVID MAYER-FOULKES

Note that the impact of mass production on wages can be low if market power is near its maximum feasible level  = ALt / ASt , when the large scale sector faces low large scale competition. Furthermore, if new large scale sectors do not face competition from small scale sectors, so that  can be larger than ALt / ASt , the impact on wages could be negative.

Efficiency and inequality in the presence of capital The simplest stylized model for a mass market economy, as presented here, does not require the inclusion of capital. However, there are some interesting points that can be made if capital is included. First we show theoretically that the economy wide capital to labor ratio and the corresponding wage level can be considerably distorted by market power in the innovating, mass production sector. Let us suppose that we replace Definition 6.9 with:

Definition 6.12. In the presence of capital let the production function for goods    j , j L, S be:

y jt ( ) =

k  1

jt

( )

 A l 

jt jt





( ) ,

j  L, S ,

where    = 1 ,  =     . Here k jt ( ) represents units of composite good as defined in Chapter 4 in equation (4.6). I assume capital markets are perfect so that the interest rate equals the marginal return of capital. Writing K t for aggregate capital, it can be shown that the interest rate and wages are given by: 



 K t    1  L   , wt =  rt =    ALt ASt1   ALt ASt . K t   L  





These can be verified simply by observing that the large scale sector acts as if its technological level were ALt / , so that the effective average technological level across goods is

  ALt ASt1 . 178

DEVELOPMENT AND UNDERDEVELOPMENT UNDER GLOBALIZATION 

Suppose for this discussion that the equilibrium interest rate r is determined by intertemporal preferences setting r  =  . Then the optimal capital to labor ratio is given by:

K    ALt ASt1 = . 1 L 



At this level of capital per worker the corresponding wage level is: 



1 St



 

w =  ALt A  . 

Hence we have shown:

Theorem 6.13. A market power level  reduces both

K and w  by a factor L

  .

The interaction of innovation profits in the large scale sector with the interest rate on capital in the small scale sector provides a context for understanding the role of the stock market in bringing forward innovation profits, capitalizing innovation income streams according to the prevailing interest rate, and concentrating them on innovators. In the presence of capital, innovation investment yields a profit rate  Lt net of payments of interest rate rt to capital. A capital market provides innovators with an instrument to bring their profit flow to the present. They can sell through the stock market a project producing an income flow through their innovation. Small investors will purchase this income flow capitalized at a value determined (net of risk) by the interest rate. This brings the innovator’s profit flows to the present, included in the project’s value. The price of the innovative goods will still reflect the original markup. However, the project’s book values will not register innovation profits, only a cost for the purchase of technology that already includes the profit accrued to the innovator. Examples when profits are brought forward are: when a company goes public, a start up is sold, or mergers or 179

DAVID MAYER-FOULKES

other reorganizations occur. Hence the study of operating profits through accounting books may not address the full impact of innovation profits. There is a considerable, controversial, literature on the efficiency costs of monopoly power. In a well known paper, Harberger (1954) concentrates on the misallocation costs of monopoly, and arrives at a very low estimate of 0.1% of GDP. The data is obtained from accounting books for seventy three manufacturing industries for the period 1924-1928. In that paper a benchmark operating profit rate of 10% is considered normal and its efficiency costs are not estimated. The paper concentrates on the allocation impact of profit rates varying above and below the 10% level. Because it studies only the innovating sector, only estimates the partial impact of these allocation adjustments, and takes its information from accounting books, this paper does not address the issues we raise here. Cowling and Mueller (1978) weaken Harberger’s (1954) assumptions and arrive at social cost estimates of 7 to 13%. Our model goes quite a long way in explaining the inequality pointed out by Piketty (2014) for mass market economies. The reasons are the following. First, in our model Piketty’s interest rate r in fact refers to the profit rate, which is even more easily greater than the growth rate g . This means that the wealth of people receiving profits can grow much faster than average wealth. Second, the concentration process we describe works in terms of the returns to real investments. It is not only that large financial accounts can get a preferential rate of return. It is also that large real investments can access the profit rate through innovation rather than just the competitive interest rate through capital investment. While discussing the historical aspects of Piketty’s (2014) work is beyond the scope of this paper, we would hypothesize that convergence to equilibrium inequality levels or capital to income ratios is faster than posited by Piketty, and responds significantly in a couple rather than in quite a few decades to substantial changes in profit level determinants (see Figure 5.2). Thus, while the two World Wars may have had the most salient negative impacts on capital accumulation, other changes such as the rise and fall of the economic framework of the Great Prosperity (including taxes on profits, human capital investment, financial regulation and welfare), or epochal changes in globalization, have also had highly significant impacts on income distribution.

180


A profit rate tax It is customary for a model proving inefficiencies to propose a tax than can restore efficiency. Our results show that the presence of market power implies an inefficiency in production levels and wages. If incentives can be found for producers not to diminish their production level so as to raise prices and profits, aggregate economic efficiency will rise. Let us suppose that a series of conditions not modelled here imply the social, economic or political convenience of designating some positive profit rate for large scale production, which is lower than in an unregulated market. For example, such a profit rate might be the optimal one for innovation, whose efficiency may follow an inverted-U relationship with the profit rate (Aghion et al, 2001, 2005). We define a profit rate tax whose incentives are for the producer to decrease her exercise of market power down to the socially designated profit rate. No taxes are levied at equilibrium. Instead the effect is to require the producer to increase production, while still making a profit, therefore reducing prices and improving both efficiency and equity. Suppose some markup  is prevalent for large scale producers. For any lower markup   (1,  ] profits will be  Lt = (1   1 ) zt , and the profit to input rate   1 . Let    be the income tax schedule

 0    0   L    =  L 

  0,  <  0.

(6.128)

Besides the constant tax rate  L , above the profit rate  0  1 , where

 0  (1,  ) , taxes rise with markup rates. The point is that from  0 up, net profits decrease if the producer decreases production.

Theorem 6.14. 1) Under a tax schedule  ( ) , if  0 > 1 /( 0 ( 0  1)) , the economy behaves as if market power has lowered to  0 =  0 . In this example the marginal tax on profits as the markup increases at  0 is less than 1 so long as the profit rate is less than 61.8% .

181

DAVID MAYER-FOULKES

2) The economy can approximate the first best for which  = 1 . Define instead tax schedule (6.128) using  0 = 1 . To avoid the tax, large scale production adjusts to a markup    0  = 1 

1

0

, which also tends to 1 as  0   .

Proof. 1) Below  0 , since  0 <  , the incentives are to raise prices to increase profits. Hence firms will select the mark up  0 . Above  0 ,

(1   ( )) Lt ( ) = (1   0 (   0 ))(1   1) zt has negative derivative with respect to  if 0 >  0 (1   1)  (1   0 (   0 )) 2

 0 >  0 ( 2   )  1   0 (   0 ). For  0 > 1 to satisfy the inequality we need  0 ( 02   0 ) > 1 , that is,  0 > 1 /( 0 ( 0  1)) . The inequality remains valid for  >  0 since the next derivative with  ,   0 (2  1)   0 < 0 is negative for these values. Observe that the marginal tax on profits at  0 is d [(1   ( )) Lt ( )] d  Lt ( )

=  =

 0 2

1   01

0 , that is, so long as  0 < 1

5

2) Let  0 = 1 . Then the derivative of (1   ( )) Lt ( ) is negative if

0 >  0 ( 2   )  1   0 (  1) = 1  (1   ) 0 (  1), that is, for  >   ( 0 ) = 1 

1

0

. 182


Technological change We define the process of endogenous change for the technological levels ALt , ASt in this two sector leading economy. We begin with the large scale sector, the technological leader. As mentioned above, there is for each mass produced good a single, infinitely lived, innovator who can innovate smoothly and with certainty. We consider an innovator with perfect myopic foresight, as in Chapters 4 and 5. In the case of the small scale sector, we assume producers absorb technologies just to keep abreast of competition. However, these small firms with limited resources can only apply a limited set of techniques to produce their technological change. The entrepreneur might for example dedicate some of her time to search for new techniques and solutions to adapt to his productive context. Although we exclude human capital from our model, we can think of an entrepreneur who has abilities or who hires labor for this purpose. Recall that each small scale sector is characterized by the property that innovation cannot be financed by obtaining sufficient profit margins over a significant proportion of its market. Thus the nature of this sector makes it inviable to establish large research crews using more sophisticated techniques, and excludes from consideration the techniques of large scale or mass production.24 Steady state productivity therefore lags behind in the small scale sector. We use the same innovation function (4.31) as in Chapter 4, written as (5.12) in Chapter 5,

 ((1   ) ALt  A jt )S jt d At =  j  (1   ) ALt dt 



 1   v ,  

j  L, S .

(6.129)

(we have dropped the  used above briefly for capital). Again the skill levels are S jt  A jt , j  L, S . Here  S is analogous to L , except that it reflects a limited kind of innovation that can be carried out on a small rather than a large scale, so  S <  L . This is analogous to the distinction between implementation and R&D of Chapter 2, in that in the small scale innovation is unlikely to use an R&D lab, employ scientists, and so on, and is more likely simply to 183

DAVID MAYER-FOULKES

implement technologies created in the large scale sector. (These new technologies may be partly embodied in capital or material inputs, although we abstract from these in this simplified model.) S St is the skill level of the firm (entrepreneur, workers and installed productivity), which we consider equal to ASt . Here, however, the small scale sector, which in this setting always lag behind the large scale sector, experiences a technological spillover from the large scale technology ALt , represented by the factor

1  ALt  Ajt 1  ALt , which in the case of the large scale sector represents within

sector spillovers generating growth. Recall that the defining characteristic of the small scale sector is that firms cannot obtain sufficient profit margins over a significant proportion of their market. Thus a significant level of market power cannot be achieved, and we assume producers are price takers. However, they cannot be infinitesimally small and still invest in technological absorption. Thus we assume there is some large number of firms N , which represents an approximation to perfect competition. For simplicity all small scale firms are the same size. Therefore their sales are zt = zt / N . Let us consider the optimization of the large scale producer. Let  L ,  L  (0,1) represent a profit tax and an innovation subsidy, positive or negative proxies for all distortions and policies affecting profits and the incentives to innovate, let  = (1   ) /  , and define the effective innovativity: 

 (1   )(1  L )  1   . ~L =    L  1   L   Proposition 6.15. Large scale incumbents set their technological rate of change at:

d  ~  zt ln ALt = L  dt 1    ALt



  . 

Proof. The incumbent’s mark up, at time t  t will be

(6.130)

AL (t  t , v)

. ALt  t Thus, using myopic perfect foresight, at any given time t she maximizes her expected rate of change of profit 184


 1    d  1    (1   AL t  t , v   ) z   1 j max  dt L t  t   ALt  t   v      t =0







 v . 



Hence the first order conditions in Chapter 5 work if we replace zt with

(1  L ) zt /(1  L ). Material inputs v are given by 1/ 

 1  L zt   vL   ˆ L (1   ) (1  L )  ALt  

 1 

ALt ,

(6.131)

where we add a subscript L for reference, and the resulting technological rate of change is given by (5.14), which is written as (6.130). Let us now consider the optimization of small scale producers. Let  S  0,1  represent an innovation subsidy there are no profits), and let

1  1  ~S =  N  1  S



 1   S . 

Proposition 6.16. Small producers set their technological rate of change at:

d (1   ) ALt  ASt ln ASt = ~S dt (1   ) ALt



 ALt zt    .  ASt ALt 

(6.132)

Proof. Small scale innovation is analogous to large scale innovation except that

L becomes  S , zt becomes z t = z t ,   1 and we consider an N

innovation subsidy S  0,1 , but not a profit tax  S . Hence the same derivation yields, after simplification, material inputs given by 



1/ 

 z  vS   ˆ S t  ASt  

((1   ) ALt  ASt ) ASt , (1   ) ALt

185

(6.133)

DAVID MAYER-FOULKES

and rate of technological change (6.132). Since zt depends on both ALt and ASt a relative scale effects is introduced that is simplified under continuous myopic foresight.

The steady state We now find the steady state growth rate and the steady state relative lag of the small scale sector. Definition 6.17. Define the relative state variable at = ASt / ALt . We can write income (6.126) in the form

zt  1 at1 L = . ALt   (1   ) 

(6.134)

Substitute in (6.130), (6.132) to express the rates of technological change in terms of the relative technological level at ,

 ~    at1 L  d  ln ALt = L  1      (1   )   dt 1    at d ln ASt = ~S 1  dt





  1 at L       (1   )   .  

The dynamics of the relative technological level at between small and large

  1 L d ln at = H at       (1   )  dt 

  ~S (1    at ) at  ~ L  at 1   .  1 

scale production can now be written,

186


Observe that H (at ) is a strictly decreasing function, with lim at  0 H ( at )  . For simplicity we now assume that condition (6.124) is maintained if it holds initially: the small scale sector cannot overtake the large scale competitive fringe. This means that large scale innovators can always use a markup  . The condition is given by H at  < 0 for at >  , that is,

~S (1     1 )  ~L 2 .

Theorem 6.18. Suppose that the small scale sector cannot overtake the large scale competitive fringe. The relative technological level at of the small to the large scale sector has a unique positive steady state a    1 with growth rate growth rate

 =

d dt

ln ALt a = a given by t

~La  =  ,  ~  S (1    a )

(6.135)

1    a    1 a (  )L     = ~S 1      (1   )  









=

   L  ~  1    La . 1      (1   )  

(6.136)

The steady state a  and growth rate   satisfy:

a  > 0,    < 0, 

a  = 0,  d  > 0, d

a  > 0, ~S   > 0, ~S

a  < 0, ~L   > 0. ~L

(6.137)

~ can be obtained by addressing the public good nature of small scale Increases in  S innovation, mentioned above.

187

DAVID MAYER-FOULKES

Proof. Since limat 0 H ( at ) =  , H  < 0 and H (  1 ) < 0 there is a unique steady state a  (0,  1 ) given by H (a ) = 0. Hence the steady state is given by ~ (1    a  )  ~    a  , which is equivalent to S

L

(6.135). Since the LHS of this equation is increasing in a  and the RHS in

,

a  > 0 . The economy’s growth rate is given by 

 =

d ln ALt dt

at = a

=

d ln ASt dt

at = a

,

which yields the first expression for   in (6.136). The second is obtained by applying (6.135) to it. Now, the first expression is decreasing in a  , and also decreasing in because ,

d  1  1     1 a  =  < 0 . Since > 0 , it follows d   (1   )         2

that

  a  < 0 . Regarding changes in  , note from (6.135) that =0.  

Then from (6.136)

d  (   1)   da  1  ln   =  ln( a  )   ln     0  d   (1   )   1    a  a   d , The RHS is zero since a < 1 so ln( a  )  0 , the quotient is positive, and

~La  a   ~ Next, by (6.135), so, differentiating by   = 0. S = 1    a 

~S ,   = ~L

d a a  ( ) . The middle derivative expression is da 1    a ~S

positive since the quotient is increasing in a  . Hence the second expression in (6.136) yields

188

a  > 0 . From here ~S

  > 0 . Similarly from (6.135) ~S


a   a ~ = so, differentiating by L , < 0 . Hence, ~L ~S (1    a ) ~L differentiating

the

first

expression

in

(6.136)

with

~L ,

    a  = >0. ~L a  ~L

Inefficiency of innovation Is the private assignment of innovation resources optimal in the mass market economy? We answer this question by examine the innovation incentives for a benevolent government. We show that it is possible to improve income growth by subsidizing innovation, and explain under what conditions this subsidy can be paid for by taxing profits. In accordance with perfect myopic foresight, let the government maximize Zt t , deducting expenses in innovation incurred for raising Zt t . Note that this optimization assumes market exchange takes place in the presence of market power, so the question posed is only seeking a second best. More precisely, at any time t the government maximizes

max

v Lt ,vSt

 t

~ Z t  t , vLt , vSt 

t = 0

 vLt  1   NvSt .

~

Here Z t  t , vLt , vSt  , where t > 0 , is an income trajectory envisaged by the government over a small time interval into the future, given innovation investments vL for each large scale good, and vS in each of the

N firms or each small scale good. The maximization is subject to the physical equations for technological change (6.129). Note that the N small firms still repeat innovation in this government maximization. d ln Z t d d =  ln ALt  1    ln ASt . dt dt dt The government maximizes vLt , vSt , independently. When it maximizes Now, using expression (6.126),

income due to each large scale good, it maximizes income minus cost. Since incumbent’s profit is a multiple 1   1 of income, the government performs the same maximization as the incumbent except that instead of 189

DAVID MAYER-FOULKES

maximizing a multiple of (1  ( approximation AL t  t ,v  ALt  t

is

 d ALt dt

AL t  t ,v  1 ALt  t

) ) zt  t , whose first order

AL t , v t , it maximizes a multiple of

zt  t , which has the same first order approximation but with

  1 and with no subsidies or taxes. In the case of the small scale sector, the government simply omits the subsidy from the calculation. Therefore the government allocates the innovation investments given by (6.131) and (6.133), but with

1  L  1, (1  L ) 

1 = 1. 1  S

(6.138)

It follows that if the government assigns subsidies and taxes so that these equalities are satisfied, then the private assignment of innovation will be efficient, without taking into account distorsions in prices as given in (6.123) and in the static equilibrium, as shown in Theorem 6.10. Large scale easier it is innovation.  setting 11L

innovation is not efficient. The reason was stated above. The to make profits, the relatively less it is worth to invest on Hence it is optimal for the government to raise innovation by =  in the large scale sector. This implies that the subsidy rate

L

for innovation is higher than the tax rate for profits,  L >  L . These results would change if creative destruction were included in the model. They would provide additional incentives for private innovation, and could give rise to an inverted U curve for innovation rates as a function of market power (as in Aghion et al, 2001, 2005). It could also be that for strategic reasons firms perform more than the optimal level of R&D, to make sure they have a stock of product development lines, or creating more than the socially optimal level of variety. In any case there may be an optimal market power that takes these factors as well as static efficiency and the distribution of income into account. The profit rate tax mentioned above provides a tool for lowering market power to any given level. In the case of the small scale sector, private innovation is efficient except for the N -fold repetition of absorption because these efforts are not pooled. Also, non approrpiable innovation is not pursued. The following results on efficiency and appropriate government innovation policy in the large and small scale sectors can now be stated.

190


Theorem 6.19. 1) As market power becomes small and   1 , private innovation tends to efficiency. 2) When the profit rate tax is applied, as  0  1 , case 1) is approached in the limit. 3) Suppose that in the large scale sector profits are quantitatively higher than optimal innovation investment. Then taxes and subsidies  L ,  L  (0,1) exist for which the government’s budget is balanced and innovation is optimal. If profits are not that high, a lump sum tax on wages is needed to obtain optimal innovation with a balanced budget. 4) The steady state trajectories of both ALt and ASt lag behind what is economically feasible. Proof. 1) When L = L = 0 and   1 ,

 1 L

(1L ) 

 1 so innovation tends

to efficiency. 2) Under a profit rate tax,  is replaced by  0 . Thus in the limit the previous case applies. 3) Observe that the function L = f (L ) = 1   L  L (for which 1  L

1  L

=  ) satisfies f (

 L 1 L

) = 0 , f (1) = 1 and f (L ) =  > 1 . The

government surplus or deficit in establishing taxes and subsidies  L , L is given by

G(L ) =  [ f (L )(1   1 ) zt  L vLt ]. Let us evaluate this government surplus or deficit at L = 1   1 and

L = 1 . In the first case L = 0, while L > 0 , so G(

L 1 ) < 0 . In the L

second case G(1) =  [(1   ) zt  vLt ]. Since this quantity, aggregate profits minus optimal innovation costs, is positive by assumption, 1

 

G L =  [  (1   1) zt  vLt ] >  [(1   1) zt  vLt ]  0 by the same assumption. Hence by the Intermediate Value Theorem there exists L  (1   1 ,1) for which the government budget is balanced. At 191

DAVID MAYER-FOULKES

this value  L ,  L  (0,1) . If instead G (1) < 0 , a lump sum tax on wages is needed to obtain optimal innovation with a balanced budget. 4) The previous statements show this for ALt . As for ASt , in Theorem 6.18 we showed that by addressing the public good nature of technological ~ , the small scale sector technological absorption and therefore raising  S steady state could be raised.

IV. The mass market economy – discussion Policiers for equity and efficiency in mass market economies Let us call the set of policies that can improve efficiency and equity in the mass market economy distributive-innovative. In particular, we have discussed two examples of public economic policies that can improve on free market policies for the mass market economy, by reducing market power and improving innovation and absorption in the large and small scale sectors. The first promotes both industrial and small scale technologies by supporting innovation using taxes on profits, and diffusion using these or other taxes. In the US, the high taxes on profits and the support for science and human capital formation applied during the Great Prosperity, which served to promote both innovation and diffusion, corresponds to this combination. These policies are of course independent of Keynesian macroeconomic policies. The second addresses the essential source of inefficiency and inequity in the mass market economy: the incentives that producers have to underproduce so as to make higher profits. We define a profit rate tax. By taxing profit rates above some determined level, the tax provides disincentives for any further underproduction. The profit rate tax generates incentives that reward production rather than profit rates. Moreover, its equilibrium taxation revenue is zero. The result is higher efficiency in production and innovation, and higher equality. The profit rate tax makes it possible to reduce market power to a targeted optimal level if necessary, as mentioned above. It is worth mentioning here that, by facilitating mergers, the stock market provides financial mechanism for raising market power. These mechanisms can create additional market power, therefore generating profits, and could raise market power above the optimal level for innovation, or provide protection from creative destruction. The profit rate tax could control this kind of market power creation without reducing efficiency. 192


Neither a general competitive equilibrium model nor a standard model of endogenous technological change include the full set of features we have described. Yet these features are necessary ingredients for understanding the macroeconomic functioning of the industrial, or mass production, market economy. Understanding these can help in addressing a series of urgent issues such as pro-poor growth, global income concentration, the increased political influence of large corporations under deregulation, sustainability in the face of both poverty and corporate power, and the global economic business cycle.

Brief history of market power and governance in the US We mentioned that the introduction of mass production in the Second Industrial Revolution (1867-1914) brought with it a qualitative change in the market economy. The qualitative change consists of the coexistence of, on the one hand, a small percentage of large, mass producing firms that innovate, employ a significant portion of the workforce, and have significant market power with, on the other hand, a large number of small firms that are essentially competitive, and improve their technologies through technological diffusion, absorption or adoption. Several factors combine to make the economy unequal and inefficient. The presence of market power distorts the allocation of resources towards the small scale sector, overprices the goods produced with high technologies, reduces wages, and reduces the incentives to innovation. The small scale sector also has inefficient innovation incentives, since innovation in this sector has the nature of a public good. Finally, profits are concentrated in a few owners of large scale, innovation firms. It follows that there are public policies that can improve the performance of the mass market economy as compared to free market policies. Since the mass market economy emerged quite some time ago, the question naturally arises as to how these results tally with the historical record, particularly of the US, which eventually emerged as the leading economy. We give a summary of the economic history of the US as it relates to the history of the mass market economy. The rapid transformation brought about by industrialization and the application of scientific methods expressed itself in many walks of life. The period from the 1890s to the 1920s is called The Progressive Era. Progressives favored gradual social, political, and economic reform in response to the social changes brought by industrialization and as a way of opposing, controlling and compromising with the extreme impacts of 193

DAVID MAYER-FOULKES

business in politics (Kolko, 1963, questions their success). Progressives tried to eliminate corruption from government, supported women's suffrage, and brought to bear scientific, medical and engineering solutions to as many issues as possible. This was the time when the social sciences, Sociology, Economics and Political Science, first appeared as such. Policies adopted by progressives included the Federal Reserve System in 1913 (Kazin, Edwards and Rothman, 2011, p.181). Regarding the evolution of market power in the US, Lipton (2006) provides a history of merger waves for the US, “Merger Waves in the 19th, 20th and 21st Centuries,” which will serve to contextualize our history. Lipton's first wave of mergers and acquisitions took place from 1893 to 1904. We have mentioned this before as the period that saw the birth of the main steel, telephone, oil, mining, railroad and other giants of the basic manufacturing and transportation industries. The late 19th Century was also the time when a wave of mergers radically transformed the banking sectors of Boston and Providence (Lamoreaux, 1991). The Sherman Antitrust Act, meant to prevent the destruction of competition through the formation of cartels and monopolies, dates to 1890, indicating the same time period in general terms and also a previous tendency for competition based on consolidation. The second merger wave, from 1919 to 1929, saw the emergence of the major automobile manufacturers, featuring vertical integration that in the case of Ford reached all the way to the iron and coal mines. These two merger waves occurred in a context of light regulation and strong innovation, as the Second Industrial Revolution matured, implementing scientifically based technologies and creating economy-wide market power. This period of technological change and economic concentration climaxed with the Roaring Twenties and drew to a close with the Great Depression. Perhaps near the end of the boom there was a desire for higher saving of profits than met the necessary investment, as we mentioned for the present day in Chapter 5, leading to a fall in the prices of the assets resorted to for saving. A new era of economic policies emerged in the US as it responded to the Great Depression. With Franklin D. Roosevelt's New Deal came the New Deal Coalition, an alignment of interest groups and voting blocks that supported the New Deal and kept Democratic Presidents in office from 1932 to 1969 (except for Dwight D. Eisenhower in 1952 and 1956), while the opposing Conservative Coalition largely controlled Congress from 1937 to 1963 (Keko, 2010). The Democratic Party promoted since then a socially liberal and progressive platform. While the “First New Deal” (1933-34) provided diverse sectors of the economy with emergency help, the “Second New Deal” (1935-38) included the Wagner Act to promote labor unions, 194


the Social Security Act, and the Fair Labor Standards Act of 1938, which set maximum hours and minimum wages for most categories of workers. These labor market institutions, some based on law and some on social transfers and expenditures, set the stage for what was for a long time perceived as a level playing field between labor and industry in the US. Large trade unions negotiated with large corporations, distributing wealth derived from innovation and ushering in an era of prosperity that lasted some fifty years. This prosperity included not only growing incomes for a large middle class, but also social security and fair labor standards, benefits the markets had not provided under laissez faire. Prosperity also included a government role in infrastructure, education and health with such programs as Medicare and Medicaid, the No Child Left Behind Act, Federal Student Aid, and the social programs already discussed above.

It is necessary to pause here and recall that ever since the advent of factories in the First Industrial Revolution, long social struggles took place to establish essential features of the developed world, such as the eight hour day, child labor laws and social security. These were etched into social consciousness in the Chicago strikes of May 1st, 1886. In the US these struggles only culminated until after the Great Depression with President Roosevelt's Social Security Act of 1935 and Fair Labor Standards Act of 1938, both introduced as measures to alleviate unemployment.

Also important was the Banking Act of 1933, meant to improve financial stability by establishing the Federal Deposit Insurance Corporation and limiting commercial bank securities activities. Turning back to the history of merger waves (Lipton, 2006), only one merger wave took place in this era, the third wave, between 1955 and 1969-73. At this time the conglomerate concept took hold of American management. Note that this mechanism was organizational rather than financial. It is beyond the scope of our study to discuss how industrial, labor and financial regulation shaped or perhaps somewhat inhibited the formation of mega market power during this time. Perhaps it is simply that market power was already developed enough at the national level but was not operating on the global scale. Contrast with the merger waves that occurred once liberalization and globalization began in 1982. Following Lipton’s (2006) narrative agan, the 195

DAVID MAYER-FOULKES

fourth merger wave, or takeover wave of the 1980s, inaugurated the era of hostile takeover bids by major investment banks. It featured a new set of financial mechanisms such as junk bond financing and leveraged buyouts. The fifth wave (1993 to 2000) was the era of the mega-deal. From a modest $342 billion in 1992, worldwide merger volume reached $3.3 trillion in 2000. This wave ended with the NASDAQ collapse that year, but by 2006 a sixth merger wave was in full swing. By mid-2008, just before the financial crisis, five banks were emitting 97 percent of all derivative assets, globally. Under globalization, global mergers and global, mega market power became the standard. The history of the mass market economy sudivides into three parts (Figure 6.1). The first, a period of loose regulation, begins with the Second Industrial Revolution and ends with the Great Depression. The second begins with the New Deal and ends with the Reagan and Thatcher liberalization policies. The third is the current globalization era that began with this in the 1980s. Thus the first and third parts by and large correspond to free market policies. The second part was characterized by progressive income tax and regulation, and with active government support for social security, labor, education, and science. This policy mix has the elements that the model shows are required to improve equity and efficiency, and are an example of a set of distributive-innovative policies. The support for science and education in fact supports innovation both in the large scale and small scale sectors. The application of these policies required a political base that was provided, as mentioned above, by the New Deal Coalition, an alignment of interest groups and voting blocks that supported the Democratic party and kept the economic policy context in a different place in the political spectrum than it is at today. The New Deal Coalition supporting the distributive-innovative economic policy mix was motivated by macroeconomic stability, social security, and economic progress, while the New Classical Economics that spearheaded the change in the economic policy context in fact addressed the Keynesian economic cycle policies which had ceased to function, also claiming that free market policies were optimal for distribution, innovation and even business cycles. Here we propose that, in fact, long-term postwar prosperity and the characteristics of the income distribution in the US were a result of how the Democratic distributive-innovative policy mix acted on the mass market economy. Complementing the inequality data shown in Figure 6.1, note that the growth rate of the national average wage index of the Social Security Administraton of the US was 5.3% from 1951 to 1982 and 3.6% from 1982 to 2014.

196

DEVELOPMENT AND UNDERDEVELOPMENT UNDER GLOBALIZATION 25 Active government support for social security, labor, education, science.

15 10 5

1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 The Great Depression 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

20

Second Industrial Revolution (1867‐1914) Roaring 20s

Loose regulation

0

Retreating gov‐ ernment support for previous programs

Progressive income tax, regulation

Liberalization, globalization, low corporate taxes

Figure6.1. The three periods of mass market economy in the U.S., their general economic policy context, and the history of income inequality (top 1% income share from The World Top Incomes Database, Facundo Alvaredo, Tony Atkinson, Thomas Piketty and Emmanuel Saez, http://topeconomics.gmond.parisschoolofeconomics.eu/).

In both periods of laissez faire, under the impacts of innovation and finance, market power rose to levels implying high concentrations of wealth as well as financial instability, which however is beyond the scope of our model to analyse. What can be affirmed is that economic competition on its own was not enough to control the market power resulting from innovation and consolidation, which in addition tends to combine with the political power resulting from concentrated wealth. What was necessary to control this market power was an economic policy context supported by an adequate coalition of interest groups and voting blocks that came about through the crisis resulting from the Great Depression. The current levels of income concentration, wages and poverty are not the only possible ones. The model shows that efficiency and equity in production and innovation can be promoted together by reducing excessive market power—the essence of Adam Smith’s democratic insights on competition—and by recognizing the public good nature of innovation in the large scale sector and technological absorption in the small scale sector.

197

DAVID MAYER-FOULKES

A comment on institutional optimality Institutions are the social arrangements that allow a society to function, in particular its economy. For example, to function well, participants in a market economy need to be sure of their property rights and of the contracts they engage in. This requires the necessary enforcement of these rights. But this is only the beginning. When are institutions optimal? To study this question it is necessary to take into account how the economy works. For example, if we believe that general equilibrium describes the economy, then the enforcement of property rights and contracts should be enough to obtain the optimality properties described in the Welfare Theorems. However, we have noted how market institutions allow for the coexistence of firms and competition both with and without market power. If instead the economy functions as a mass market economy, then to approximate optimality we need institutions with enough of a democratic mandate to implement the distributive-innovative policies described above. Just the enforcement of property rights and contracts will not be optimal, and will support an unequal and inefficient economy. There is of course a whole series of additional market and political failures that good institutions need to be able to address. A wide and informed democracy that is resilient to moneyed interests is a necessary part of well-functioning institutions.

198


Main ideas In the industrial economy, especially since the Second Industrial Revolution, an important portion of production and innovation takes place in large scale firms. On the other hand , there are also many small scale firms in an approximately competitive context, that employ a considerable portion of the population. The small scale sector is characterized by producing goods for which innovation cannot be financed by obtaining sufficient profit margins over a significant proportion of their market. We develop the mass market economy model to describe an economy with these properties. The large scale sector is the technological leader, and the small scale sector lags behind. This economy has several sources of inefficiency. Market power in the large scale sector introduces significant static inefficiencies reducing wages and capital accumulation, and can also be suboptimal for innovation. Technological absorption in the small scale sector has public good properties and is therefore innefficient. Wages are also lower because they depend on an inefficiently low small scale sector technological level. Profits are concentrated on the relatively few owners of innovation, who earn a profit rate higher than the interest rate for capital. Thus in the mass market economy, free market policies are suboptimal, there is income concentration at top levels, and wages are lower than technological levels warrant. Public policies are available to increase both efficiency and equity in this economy. Examples are public support for innovation in the large scale sector and for adoption in the small scale sector. Another policy option is the profit rate tax, with incentives for the producer to decrease her exercise of market power down to a socially designated profit rate.

199

DAVID MAYER-FOULKES

19 This chapter is based on the model in Mayer-Foulkes (2015b), which includes capital. The article has been extended to Mayer-Foulkes and Hafner (2017), which includes an empirical estimation. 20 To be precise we would have to say “market economy with mass production.” I hope that the name mass market economy communicates the idea nevertheless. 21 For the purposes of our discussion, we assume that there is a single, perfect labor market including skilled and unskilled labor. This means that there is a single expected wage level that takes human capital investment costs into account. We abstract from adjustments to the demand and supply of specific skills. These can generate both wage differentials and the creative destruction of employment across skills. 22 Thus the mass market model is in the interior of the continuum lying between competitive general equilibrium and endogenous technological change models. 23 This means that creative destruction does not produce incentives for innovation in the model. While this impacts the efficiency of innovation incentives under market power, we include in our discussion consideration of optimal levels of competition such as Aghion et al (2001, 2005). This also means that the incumbent can benefit from any idea that anybody has that increases productivity in her sector. 24 Franchises may arise when an innovator has devised a way to transform a small scale sector into a large scale sector.

200

7 THE PRESENT ECONOMIC SCENARIO Economic growth is like an orchestra. Which part is most important? Shall we say that it is the director, or the violins, or the drums? If any part of it is deficient, the whole will suffer, in levels and in growth rates. Institutions, demographic transition, human capital, trade, technological change, capital, finance. Would we leave any of these out? Serious deficiencies are possible in all of them, and will often lie beyond the power of markets to correct. History has a tendency to move faster than our knowledge of economics. We must learn quickly. Today we face a single global market economy with very significant challenges in development and sutainability, and with increasing inequality. The origin of globalization in the 1980s was a global invitation by the United States to join in the development club through trade and investment. It does not quite work out like that. The historical discussion and model in Chapter 5 show that market forces on their own do not ensure the technology transfer that is necessary for development. We discussed in that Chapter the contrasts in technological change between the policies used by the NICs to industrialize and the FDI that has been promoted in the current wave of globalization. FDI has generated extraordinary profit flows for transnational corporations that in addition skew the policy climate in favor of their interests, concentrate enormous wealth in very few hands, and generate savings that for over a decade have not been properly channelled to investment, since the global savings glut began25.

201

DAVID MAYER-FOULKES

The model in Chapter 6 shows that market economy with mass production has intrinsic problems in efficiency an equity, that include unnecessarily low wages and concentrated wealth, that as we have seen are accentuated under globalization. Free market policies are suboptimal for the mass market economy. Distributive-innovative policies are necessary to improve equity and efficiency, and were in fact tested during the Great Prosperity. Applying such policies represents a significant challenge under globalization, which presents an institutional deficit simply because governments are national. Since clearly a global government is out of the question, what is necessary is to lay the groundwork for international cooperation between responsible entities so as to make sustainable, distributive-innovative policies feasible. A single global market economy has emerged which urgently requires governance and is beyond the control of single national governments. Even the traditional methods to control the economy lie beyond the reach of single national governments. The 2008 crisis was only mitigated because all of the major economies acted together to control it. And yet the crisis is not over, significant anomalies persisting such as 0% interest rates in the US, and high unemployment and fiscal instability in Europe. Demand and supply in currency exchange markets are distorted by the unequal flows of profits, and attempts for exchange rate stability impact government debt. Under globalization, large, transnational corporations acquire global market power that poses significant governance challenges. Nations are faced with strong pressures to acquiece to the interests of transnational corporations in many areas of policy. Even in the leading developed national economies, market competition has historically been insufficient to adequately control these coalesced forces of self-interest. Quite the opposite. Competition and profit-seeking can lead to extremes in extraction. This is particularly significant in the context of sustainability. Moreover, the traditional reasons for a role for government in the economy, such as the provision of public goods, infrastructure, education, health, and social security, continue to hold. These require a sufficient, harmonized tax base. In Chapter 5 we showed that a second, more imbalanced mode of globalization is possible, in which transnational corporations form Economy 1 and all countries have Economy 2 status. This second mode of globalization has already been apparent in attempts to develop its legal framework. Important trade treaty proposals have been negotiated in secret, the TPP (Trans Pacific Partnership agreement), which came to be known through wikileaks, the TTIP (Transatlantic Trade and Investment Partnership) and the TISA (Trade in Services Agreement, most secret of the 202


three), see Figure 7.1. Only large transnational corporations have been well informed of the contents of these negotiations. The treaties establish an investment rights regime through Investor State Dispute Settlement (ISDS) procedures, a parallel dispute system available only to foreign investors to sue national governments, that makes it onerous for sovereign governments to change basic policies for example in minimum wages, health, environmental regulation and so on. This constitutes a parallel judicial system that tends to override domestic law and is very difficult to change.26 Very notable in this project has been the exclusion of the BRICS countries, China, India, Russia, Brazil and South Africa from these treaties, with 42% of the world population and 20% of the world’s GDP. The subdivision of the globalization process poses the risk that significant economic rivalries can develop between blocks of countries and transform competition into conflict. As Dwight D. Eisenhower warned in 1961, these risks have the tendency to be increased by the influence of the military industrial complex. We discussed in Chapter 4 how colonial competition was an antecedent factor for the two World Wars. The European Union was created in part to try to put behind Centuries of war. So long as there is a single integrated economy, spheres of influence or perhaps even changes in economic hegemony, or the demise of economic hegemony itself, as mentioned in Chapter 1 (see the discussion of Figure 1.11b), need not lead to armed conflict between economic powers, although other problems arise that need to be dealt with, such as sustainability, inequality and macroeconomic stability. The only country featured in the intersection of TPP, TISA and TTIP, given the information so far, is the US. In so far as this global treaty system is designed to further the specific interests of the United States, the US is retreating from its global invitation to all nations to join the development club through trade and investment. What has to be recognized is that the market economy – national or global – works best when it is complemented with a government – or governance – with substantial expenditures in public goods and equality. The reason is that government promotes essential values that markets cannot provide – not least infrastructure, education, health and science. It is perfectly possible to design a global free trade and investment regime that also gives a place to national governments, so as to promote expenditures in public goods and equity. All that is necessary is to agree, as part of the WTO, on a common basic level of taxation that will serve to sustain government activities in all countries and put a floor to the race to the bottom in taxes caused by competition for transnational investment. If 203

DAVID MAYER-FOULKES

a country does not collect these taxes, compensatory tariffs can be imposed on its goods. The Brexit vote and Trump’s nationalistic appeal are symptoms of the underlying imbalance between markets and governance caused by a globalization based too exclusively on free markets. A basic agreement on a minimum level of taxation can provide not only the funding to approach national problems effectively, but also a platform to develop global cooperation for the solution of global problems.

Figure 7.1. Countries in the proposed Trans Pacific Partnership agreement (TPP), Transatlantic Trade and Investment Partnership (TTIP) and Trade in Services Agreement (TISA). Own elaboration based on: https://www.youtube.com/watch?v=Rw7P0RGZQxQ&feature=y outu.be [watched 11/20/2015].

204


We discussed in Chapter 5 how the successful, win-win phenomenon of mutual development that took place between the Asian tigers and the US, could well have been perceived by multinational corporations as a threat, and in Chapter 6 how the implementation of distributive-innovative policies for equity and efficiency need a sufficiently democratic economic policy context that is resilient to moneyed interests. Private corporations seeking profits under competition can hardly be expected to be responsible for public welfare. The need for an adequate balance between the public and private domains is as basic as the separation of powers in democratic government. Strengthening US corporations vis a vis US workers and foreign corporations will only deepen the problems of globalization. Human evolution occurred in the context of a relatively stable planetary environment. Lipton and Bhaerman (2009) explain how fitting the environment is an essential part of evolving, particularly of conscious evolution. Today, there are very significant threats to sustainability, such as climate change and ocean pollution, that to be met require coordinated global action. To be met locally, as in sustainable agriculture and mining, wholesome food production and built environments, they also require adequate balances to global economic power. Perhaps a threshold level for a politically coordinated global governance can be reached by establishing international agreements for the peaceful functioning of a regulated supply in the fossil fuel global market that is consistent with sustainable, reduced emissions and ceases to make the Middle East a battleground between the superpowers. Adequate global governance includes meeting the hopes for sustainable development for all of the world population. For many nations seeking the improvement of their institutions for the equitable provision of wellbeing, the unbalanced power of concentrated economic interests presents a formidable challenge. Sometimes the combination results in exacerbated local problems, tending to cause a local public power vacuum for their solution that can make emigration the only alternative. Sustainable development is a win-win scenario. Worsening conditions leading to mass migrations are showing that polarized globalization can be, on the contrary, a lose-lose scenario. An understanding of the problems of the market economy, exemplified by human capital and technology traps in the context of globalization and the mass market economy, points directly to the problems of how to advance global cooperation, to generate the necessary global and local public goods, global wellbeing, and global equity. 205

DAVID MAYER-FOULKES

REFERENCES

Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The Colonial Origins of Comparative Development: An Empirical Investigation. American Economic Review, 91:1369-1401. Acemoglu, D., Johnson, S., and Robinson, J.A. (2005). Institutions as a Fundamental Cause of Long-Run Growth. in: Philippe Aghion & Steven Durlauf (ed.), Handbook of Economic Growth, volume 1A, chapter 6, pages 385-471 Elsevier. Acemoglu, Daron, Philippe Aghion, and Fabrizio Zilibotti. (2002) “ Distance to Frontier, Selection and Economic Growth.” unpublished, MIT. Aghion, Philippe, and Peter Howitt. (1998) Endogenous Growth Theory. Cambridge, MA: MIT Press. Aghion, P. and P. Howitt (1992). “A Model of Growth through Creative Destruction.” Econometrica, (March), 60 (2), pp. 323-51. Howitt, P., and Aghion, P. (1998) “Capital Accumulation and Innovation as Complementary Factors in Long-Run Growth.” Journal of Economic Growth, 3(2):111-30. Aghion, P.; Howitt, P.; Mayer-Foulkes, D. (2005). “The Effect of Financial Development on Convergence: Theory and Evidence” , Quarterly Journal of Economics, 120(1) February. 206


Aghion, P; Howitt, P; and Mayer-Foulkes, D (2004). “The Effect of Financial Development on Convergence: Theory and Evidence.” NBER Working Paper 10358, March. Aghion, Philippe; Harris, Christopher; Howitt, Peter; Vickers, John (2001).” Competition, Imitation and Growth with Step-by-Step Innovation,” Review of Economic Studies, LXVIII, 467– 492. Akerlof, George A; Kranton, Rachel E (2000).” Economics and Identity,” The Quarterly Journal of Economics, Vol. 115, No. 3 (Aug), pp. 715-753. Arora, S. (2001). Health Human Productivity and Long-Term Economic Growth. Journal of Economic History, 61 (3), 699-749. Arrow, Kenneth J. (1962). “The Economic Implications of Learning by Doing.” Review of Economic Studies, 29 (June), pp. 155-73. Bairoch, P. (1997). Victoires and déboires (3 vols.). Paris: Gallimard. Baldwin, Richard (2012).” Global supply chains: Why they emerged, why they matter, and where they are going,” CEPR Discussion Papers 9103, CEPR. Discussion Papers. Barro, R. J. and Sala-i-Martin, X. (1995), Economic Growth, McGraw-Hill, Inc. Basu, K. (2003). Analytical Development Economics: The Less Developed Economy Revisited. Cambridge, MA: MIT Press. Basu, Susanto, and Weil, David N. (1998) “Appropriate Technology and Growth.” Quarterly Journal of Economics, 113, 1025-54. Beaudreau, B. C. (2004): World Trade. iUniverse, Inc., New York. Bernanke, Ben S (2005). "The Global Saving Glut and the U.S. Current Account Deficit," Sandridge Lecture, Virginia Association of Economists, Richmond, Virginia; The Federal Reserve Board, March 10. Available at http://www.federalreserve.gov/boarddocs/speeches/2005/200503102/ , read August 26, 2015.

207

DAVID MAYER-FOULKES

Bishaw, Alemayehu (2014). Changes in Areas With Concentrated Poverty: 2000 to 2010, American Community Survey Reports 27, US Census Bureau. Branstetter, L. G., & Feenstra, R. C. (2002). Trade and foreign direct investment in China: a political economy approach. Journal of International Economics,58(2), 335-358. Case, A., Fertig, A., & Paxson, C. (2003). From Cradle to Grave?: The lasting impact of childhood health and circumstance. NBER Working Paper 9788. Cambridge, MA: National Bureau of Economic Research. Case, A., Lubotsky, D., & Paxson, C. (2002). Economic Status and Health in Childhood: The Origins of the Gradient. American Economic Review, 92 (5), 1308-1334. Caselli, Francesco, Gerardo Esquivel, and Fernando Lefort. (1996) “ Reopening the Convergence Debate: A New Look at Cross-Country Growth Empirics.” Journal of Economic Growth, 1, 363-389. Chakraborty, Chandana & Nunnenkamp, Peter (2006) “Economic Reforms, Foreign Direct Investment and its Economic E ects in India,” Working Papers 1272, Kiel Institute for World Economics. Cowling, Keith and Mueller, Dennis C. (1978).” The Social Costs of Monopoly Power,” The Economic Journal, Vol. 88, No. 352 (Dec), pp 727748. Credit Suisse (2014). Global Wealth Report 2014, Research Institute, Credit Suisse. DeNavas-Walt, Carmen and Proctor, Bernadette D. (2014). Income and Poverty in the United States: 2013, Current Population Reports, United States Cencus Bureau, September. Easterly, William, and Ross Levine. (2002) “ Tropics, Germs and Crops: How Endowments Influence Economic Development.” NBER Working Paper 9106. Ejrnæs, Mette, Persson, Karl Gunnar and Rich, Soren, (2008). “Feeding the British: convergence and market efficiency in the nineteenth-century grain trade,” Economic History Review, 61, issue s1, p. 140-171, 208


http://EconPapers.repec.org/RePEc:bla:ehsrev:v:61:y:2008:i:s1:p:140171. Epstein, David L; Bates, Robert; Goldstone, Jack; Kristensen, Ida; O'Halloran, Sharyn (2006). “Democratic Transitions”, American Journal of Political Science, Volume 50, Number 3, July, pp. 551-569(19). Evans, Paul. (1996) “ Using Cross-Country Variances to Evaluate Growth Theories.” Journal of Economic Dynamics and Control, 20, 1027-49. Everett, Griff; Hitchcock, Stepanie H; Middleton, Jane; Timms, Rosemary H (2006). Samuel Slater - Hero or Traitor?: The Story of an American Millionaire’s Youth and Apprenticeship in England, Maypole Promotions (Milford). Fan, Emma Xiaoqin and Felipe, Jesus (2005). “The Diverging Patterns Of Profitability, Investment And Growth Of China And India, 1980-2003,” CAMA Working Papers 2005-23, Australian National University, Centre for Applied Macro- economic Analysis. Ferguson, Niall (2003). British Imperialism Revisited: The Costs and Benefits of “Anglobalization”, http://www.originofnations.org/British_Empire/british_ empire_ and_ globalization.htm Feyrer, James. (2001) “ Convergence by Parts.” unpublished, Brown University. Fitton, RS (1989). The Arkwrights: spinners of fortune, Manchester, UK; New York: Manchester University Press; New York, NY, USA. Flatters, F. and Harris, R.G. (1994), “Trade and Investment: Patterns and Policy Issues in the Asia Pacific Rim”, in Pacific Trade and Investment: Options for the 90s, Dobson, W. and Flatters, F., editors. Fogel, R. W. & Wimmer, L. T. (1992). Early Indicators of Later Work Levels, Disease, and Death, NBER Historical Working Papers 38, National Bureau of Economic Research. Fogel, R. W. (1994). Economic Growth, Population Theory, and Physiology: The Bearing of Long-Term Processes on the Making of Economic Policy. American Economic Review, 84 (3), 369-395. 209

DAVID MAYER-FOULKES

Fogel, R. W. (2002). Nutrition, Physiological Capital, and Economic Growth. Pan American Health Organization. Frankel, M. (1962). “The Production Function in Allocation and Growth: A Synthesis.” American Economic Review, 52, pp. 995-1022. Galor, O., & Zeira, J. (1993). Income Distribution and Macroeconomics. Review of Economic Studies, 60, 35-52. Galor, Oded (2005). "From Stagnation to Growth: Unified Growth Theory," Handbook of Economic Growth, in: Philippe Aghion & Steven Durlauf (ed.), Handbook of Economic Growth, volume 1A, chapter 4, pages 171-293 Elsevier. Galor, Oded (2012). “The Demographic Transition: Causes and Consequences,” ZA DP No. 6334. Galor, Oded; Mountford, Andrew (2008). "Trading Population for Productivity: Theory and Evidence," Review of Economic Studies, Oxford University Press, vol. 75(4), pages 1143-1179. Gilens, Martin & Page, Benjamin I. (2014). “Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens.” Perspectives on Politics, 12, pp 564-581. doi:10.1017/S1537592714001595. Grossman, Gene M. & Helpman, Elhanan (1989). "Quality Ladders in the Theory of Growth," NBER Working Papers 3099, National Bureau of Economic Research, Inc. Hall, Robert E. and Charles I. Jones. (1999) “ Why Do Some Countries Produce so Much More Output per Worker than Others?” Quarterly Journal of Economics, 114, 83-116. Halsall, Paul (1997). Modern History Sourcebook: Spread of Railways in 19th C entury, available at http://myphumanities.zischina.com/files/2013/09/Spread-of-Railwaysin-19th-Century.pdf, read 8/11/2015. Harberger, Arnold C. (1954).” Monopoly and Resource Allocation, The American Economic Review, Vol. 44, No. 2, Papers and Proceedings of the Sixty-sixth Annual Meeting of the American Economic Association (May), pp 77-87. 210


Harris, DJ (1978). Capital Accumulation and Income Distribution, Stanford University Press,1978. Henry, James S. (2012). The Price of Offshore Revisited, New Estimates for "Missing" Global Private Wealth, Income, Inequality, and Lost Taxes. Tax Justice Network, July. Howitt, P. and Mayer-Foulkes, D. (2005). “R&D, Implementation and Stagnation: A Schumpeterian Theory of Convergence Clubs”, Journal of Money, Credit and Banking, 37(1) February. Howitt, Peter and Mayer-Foulkes, David (2002). “R&D, Implementation and Stagnation: A Schumpeterian Theory of Convergence Clubs,” NBER Working Paper No. 9104, August. Islam, Nazrul. (1995) “ Growth Empirics: A Panel Data Approach.” Quarterly Journal of Economics, 110, 1127-70. Kazin, Michael; Edwards, Rebecca; Rothman, Adam (Editors) (2011). The Concise Princeton Encyclopedia of American Political History. Princeton University Press. Keko, Don (2010) "FDR's New Deal Coalition: 1932-1968," Society & Culture, January 12, http://www.examiner.com/article/fdr-s-new-dealcoalition-1932-1968. Klenow, Peter J. and Andrés Rodrguez-Clare. (1997) “ The Neoclassical Revival in Growth Economics: Has it Gone too Far?” In NBER Macroeconomics Annual1997, edited by Ben Bernanke and Julio Rotemberg, pp. 73-103. Cambridge, MA: MIT Press. Klinger, Scott; Anderson, Sarah (2014). "Fleecing Uncle Sam: A growing number of corporations spend more on executive compensation than federal income taxes," Center for Effective Government and Institute for Policy Studies, available at http://www.ips-dc.org/wpcontent/uploads/2014/11/IPS_Fleecing_Uncle_Sam_Report_Nov2014 .pdf, read 8/26/2015. Knight, Malcolm, Norman Loayza, and Delano Villanueva. (1993) “ Testing the Neoclassical Theory of Economic Growth: A Panel Data Approach.” IMF Staff Papers, 40, 512-41.

211

DAVID MAYER-FOULKES

Kolko, Gabriel (1963). "Triumph of Conservatism," The Free Press. Krugman, Paul (1991). "Increasing Returns and Economic Geography." The Journal of Political Economy 99.3: 483-499. Krugman, Paul (2007).” Another Economic Disconnect,” Op Ed, The New York Times, April 30. Krugman, Paul (2015a). "A Moveable Glut," The Opinion Pages, The New York Times, August 24. Krugman, Paul (2015b). "Fantasies of the Chicago Boys," The Opinion Pages, The New York Times, August 28. Lamoreaux, Naomi R. (1991). "Bank Mergers in Late Nineteenth-Century New England: The Contingent Nature of Structural Change," The Journal of Economic History, 51(3), Sept. Lemoine, Francoise (2000). “FDI and the Opening Up of China’s Economy,” Working Papers 2000-11, CEPII Research Center. Lipsey, Robert E. Foreign direct investment and the operations of multinational firms: Concepts, history, and data. No. w8665. National Bureau of Economic Research, 2001. Lipton, B., & Bhaerman, S. (2009). Spontaneous evolution: Our positive future (and a way to get there from here). Hay House, Inc. Lipton, Martin (2006). "MergerWaves in the 19th, 20th and 21st Centuries," The Davies Lecture, Osgoode Hall Law School, York University, September 14, available at http://osgoode.yorku.ca/. López-Córdova (2002). “NAFTA and Mexicos Manufacturing Productivity: An Empirical Investigation using Micro-level Data”. Mimeo. Lucas, Robert E., Jr. (1988). “On the Mechanics of Development Planning.” Journal of Monetary Economics, 22, 1 (July), pp. 3-42. Lunt, PK ; Livingstone, SM (1992). Mass consumption and personal identity, Open University Press, Buckingham, UK.

212


Maddison, Angus. "Development centre studies." The World Economy. A Millennial Perspective. OECD and www. oecd. org/dev (2001). Mankiw, N. Gregory, David Romer and David N. Weil (1992). “A Contribution to the Empirics of Economic Growth.” Quarterly Journal of Economics, 107, 2 (May), pp. 407-37. Mathers CD, C Bernard, KM Iburg, M Inoue, D Ma Fat, K Shibuya, C Stein and N Tomijima (2003). ‘The Global Burden of Disease in 2002: data sources, methods and results’. GPE Discussion Paper No. 54. Geneva:World Health Organization. Available at http://www.who.int/evidence. Mayer-Foulkes, D. (2006b). The Impact of Free Trade and FDI: Banana Republic or Miracle Growth? Unpublished manuscript. Mayer-Foulkes, D. (2007). “Global Divergence”, in Severov, Gleb, International Finance and Monetary Policy, Nova Science Publishers. Mayer-Foulkes, D. (2008). “The Human Development Trap in Mexico”. World Development, Vol 36(5), 775-796. Mayer-Foulkes, D. (2011). “A Survey of Macro Damages from NonCommunicable Chronic Diseases: Another Challenge for Global Governance.” Global Economy Journal: Vol. 11: Iss. 1, Article 6. Mayer-Foulkes, D. (2015a). “The Challenge of Market Power under Globalization,” Review of Development Economics, Vol 19 (2) 244264. Mayer-Foulkes, D. (2015b). “Mass Production and Competition” (Documento de Trabajo de Economía, Centro de Investigación y Docencia Económicas, No. 581). Mayer-Foulkes, D. (2017). “Convergence and Divergence under Global Trade,” International Journal of Economics and Finance, Vol 9, No 3. Mayer-Foulkes, D. and Larrea, Carlos (2007). “Racial and Ethnic Health Inequities: Bolivia, Brazil, Guatemala, Peru”, in Racial and Ethnic Disparities in Health in Latin America and the Caribbean, edited by Antonio 213

DAVID MAYER-FOULKES

Giuffrida, Inter American Development Bank, Social Protection And Health Division, Country Department, Andean Group. Mayer-Foulkes, D. and Nunnenkamp, P. (2009). “Do Multinational Enterprises Contribute to Convergence or Divergence? A Disaggregated Analysis of US FDI”, Review of Development Economics, 13(2), 304–318. Mayer-Foulkes, D. and Pescetto-Villouta, Claudia (2012). “Economic Development and Non-Communicable Chronic Diseases,” Global Economy Journal, Vol. 12: Iss. 4. Mayer-Foulkes, D., & Hafner, K. A. (2017). “The technology Gradient in the Market Economy” (Documento de Trabajo de Economía, Centro de Investigación y Docencia Económicas, No. 606). Nagesh, Kumar (1995) ”Industrialization, Liberalization and Two Way Flows of Foreign Direct Investment: The Case of India,” Discussion Papers 03, United Nations University, Institute for New Technologies. Najera-Flores, R. (1998). “Patrones de localización de la industria automotriz”, thesis, Facultad de Economía, UNAM, Mexico. OECD (2007). Corporate Saving and investment: recent trends and prospect, Chapter 3 in Economic Outlook 82, PDF, http://www.oecd.org/eco/outlook/39727868.pdf, read 11/22/2015. Oxfam (2014). Working for the Few, 178 Oxfam Briefing Paper, Jan 20. Parente, Stephen L., and Edward C. Prescott (1994). “Technology Adoption and Growth.” Journal of Political Economy, 102, 298-321. Parente, Stephen L., and Edward C. Prescott (1999). “Monopoly Rights: A Barrier to Riches.” American Economic Review, 89, 1216-33. Pipitone, Ugo (1994). La salida del atraso: Un estudio histórico comparativo, México, CIDE, Fondo de Cultura Económica. Prescott, Edward C. (1998) “Needed: A Theory of Total Factor Productivity.” International Economic Review, 39, 525-51.

214


Pritchett, Lant (1997). “Divergence, Big Time.” Journal of Economic Perspectives, 11(3), Summer 1997, pp. 3-17. Psacharopoulos, G and HA Patrinos (2004). “Returns to Investment in Education: A Further Update.” Education Economics 12(2): 111-134. Patrinos, H. A., & Psacharopoulos, G. (2011). Education: past, present and future global challenges. World Bank Policy Research Working Paper Series 5616. Rattenbury, Gordon; Lewis, M. J. T. (2004). Merthyr Tydfil Tramroads and their Locomotives. Oxford: Railway & Canal Historical Society. ISBN 0901461-52-0. Renick, Oliver J; Regan, Michael P (2015) “Getting High on Their Own Supply, Corporate stock buybacks put a floor under earnings,” Bloomberg Businessweek Reprints, July 16, http://www.bloomberg.com/news/articles/2015-07-16/corporatestock-buybacks-make-earnings-look-better read 10/15/15. Rodrigue, Jean-Paul (2013). The Geography of Transport Systems, Third Edition, Routledge. Romer, Paul M. (1986). “Increasing Returns and Long-Run Growth.” Journal of Political Economy, 94, 5 (October), pp. 1002-37. Romer, Paul (1989). "Endogenous Technological Change," NBER Working Papers 3210, National Bureau of Economic Research, Inc. Romer, Paul M. (1990). “Endogenous Technological Change.” Journal of Political Economy, 98, 5 (October), part II, pp. S71-S102. Rothmuller, Leandro (2003) “Does FDI Matter for Trade in Brazil? An Application of the Gravity Model,” Anais do XXXI Encontro Nacional de Economia [Proceedings of the 31th Brazilian Economics Meeting] c71, ANPEC – Associação Nacional do Centros de Pos-graduação em Economia [Brazilian Association of Graduate Programs in Economics]. Salazar-Torreon, Barbara (2015). “Instances of Use of United States Armed Forces Abroad, 1798-2015,” Congressional Research Service, available at http://fas.org/sgp/crs/natsec/R42738.pdf, read 9/19/2015. 215

DAVID MAYER-FOULKES

Salisbury, W. Allen (1992). The Civil War and the American System, America’s Battle With Britain, 1860-1876, EIR. Schumpeter, J. A. (1934). The theory of economic development: An inquiry into profits, capital, credit, interest, and the business cycle (Vol. 55). Transaction publishers.

Sharma, Kishor (2000) “Export Growth in India: Has FDI Played a Role,” Working Papers 816, Economic Growth Center, Yale University. Simmons, C. (1985), “De-Industrialization, Industrialization, and the Indian Economy, c. 1850-1947,” Modern Asian Studies 19 (3): 593-622. Soto-Rodríguez, Mauricio (2002). “Integración Automotriz en América del Norte”. Revista Mexicana de Estudios Canadienses. Invierno, nueva época, no 5. Spannaus, Nancy and White, Christopher (1996) The Political Economy of the American Revolution, Executive Intelligence Review. Srinivasan, T. N. (2002). Trade, Finance, and Investment in South Asia, Berghahn Books. Svedberg, Peter (1978). The Portfolio-Direct Investment Composition of Private Foreign Investment in 1914, Revisited, Economic Journal 88, No. 352, December: 763-777. UNCTAD (2013). World Investment Report 2012. United Nations, New York. UNCTAD (2015). World Investment Report 2015. United Nations, New York. Uzawa, Hirofumi (1965). “Optimal Technical Change in an Aggregative Model of Economic Growth.” International Economic Review, 6 (January), pp. 18-31. Vollmer, S., Holzmann, H., Ketterer, F., Klasen, S., & Canning, D. (2013). The emergence of three human development clubs. PloS one, 8(3), e57624.

216


Wan Jr., H. Y. (2004). Economic Development in a Globalized Environment: East Asian Evidences. Kluwer Academic Publishers, The Netherlands. Wilkins, Mira, The Emergence of Multinational Enterprise: American Business Abroad from the Colonial Era to 1914, Cambridge, MA, Harvard University Press, 1970. Williamson Jeffrey G (2004). “De-Industrialization and Underdevelopment: A Comparative Assessment Around the Periphery 1750-1939,” mimeo, Harvard University, read 8/7/2015 at http://dev.wcfia.harvard.edu/sites/default/files/Williamson_DeIndustr ialization.pdf. See the article “Global Savings Glut” in Wikipedia, which begins by citing a series of references: “Global saving glut (also global savings glut, GSG,[1][2] cash hoarding,[3][4][5][6][7][8] dead cash,[9] dead money,[10] glut of excess intended saving,[11] shortfall of investment intentions),[11] describes a situation in which desired saving[notes 1] exceeds desired investment.[12]” The glut is first reported for 2002 in OECD (2007) “For the aggregate OECD corporate sector, the excess of gross saving over fixed investment (i.e. net lending) has been unusually large since 2002.” 26 The information given in this paragraph was taken from: https://www.youtube.com/watch?v=Rw7P0RGZQxQ&feature=youtu.be [watched: 11/20/2015]. 25

217

ABOUT THE AUTHOR David Mayer-Foulkes, Mathematics PhD, has been Economics Research Professor at CIDE in Mexico City since 1991. He has also conducted research for PAHO, UNDP and the Mexican Commission on Macroeconomics and Health, amongst others. His research asks the question, how can poverty be compatible with the basic Welfare Theorems in Economics? He focuses this question in the context of economic development, and provides answers in a series of dynamic poverty trap models in human and economic development. These explain the main features of globalization including the coexistence of development and under development and the 2008 crisis. The same type of the model shows that free markets are inherently unequal and inefficient in industrial economies with mass production. Innovative and distributive policies can improve both national and global welfare.

218