- Shopping Bag ( 0 items )
In Imperfect Institutions Thrainn Eggertsson extends his attempt to integrate and develop the new field that began with his acclaimed Economic Behavior and Institutions (1990), which has been translated into six languages. This latest work analyzes why institutions that create relative economic backwardness emerge and persist and considers the possibilities of limits of institutional reform.
Introduction: The Dependent Variable
In this study, economic growth is the dependent variable of primary interest, but that is not the whole story because my chief concern is with the social causes of economic failure-in particular, with the role of institutions in the pathology of economic stagnation and decline. In medicine, pathology is a discipline that studies the nature of diseases, their causes, processes, development, and consequences. The main goal of research in pathology, however, is not to study functional manifestations of diseases for their own sake but rather to find cures and develop preventive medicine. My motivation for studying economic decline is similar: the belief that knowledge of economic regress will improve our understanding of economic progress.
To assess the economic health of a nation I use a conventional but imperfect indicator, the country's national income (or product) per capita. Official national product and income statistics are incomplete solutions to the formidable task of summarizing in one figure the net output of all producers in a country. Yet data from national income accounts, adjusted for distortions caused by exchange rates, permit us with tolerable accuracy to rank countries according their output per capita and to track their performance through time. It is unlikely that future advances in measuring economic performance will significantly change the ranking of nations in terms of their prosperity or alter our ideas about the social causes of relative economic backwardness.
In the twentieth century, the distribution of national economic performance measured by average output per person has become more unequal than at any previous time in history (DeLong 2000, 17-20). I focus on the low end of this unequal distribution, but domestic inequality I consider only for its possible role as an independent variable in the growth equation-the argument being that economic inequality may be related to behavior that undermines economic growth.
Redistribution within a country from rich to poor households obviously improves the living standards of the poor, but sustained economic growth is of a different magnitude. The historical increase in income and output per person in the United States illustrates the overwhelming importance of economic growth relative to redistribution. Best estimates show that on the eve of the American Revolution (1775-83), gross domestic product per capita amounted to $765 in 1992 dollars but by 1997 gross domestic product per capita (also in 1992 dollars) was $26,847, about thirty-five times greater (Hulten 2000, 1). The various errors in these estimates partly cancel out, and the residual bias is unlikely substantially to distort the true picture.
In the vocabulary of economics, growth of per capita output (or per capita income) is known as intensive growth, and growth in the total output of an economy is known as extensive growth. In the beginning, extensive growth allowed the human species to multiply, spill out of Africa, and populate the various continents (Roberts 1997, chapter 1). Prior to the modern era, nearly all workers everywhere were primarily occupied in farming and hunting, but extraordinary sustained intensive growth, originating some 250 years ago in Western Europe, has revolutionized economic performance and daily life in the industrial and postindustrial economies. The economic revolution was fueled by technological change, yet at the beginning of the twenty-first century, the so-called developing countries in many parts of the world have not been able significantly to upgrade their methods of production and therefore have fallen behind (Jovannovic 2000, 6-7). In 2000, average income levels in the world's poorest and the world's richest nations differed by a factor of more than 100.
The modern growth experience presents us with these two basic puzzles. First there is the phenomenon of growth leaders. Why do some countries for a period of time lead the rest of the world in developing and applying superior methods of production? For example, during the Industrial Revolution (1750-1830), England held a position of technological leadership, which it later lost (Mokyr 1990, chapters 5, 10). What factors determine the timing and path of technological and economic revolutions? (Mowery and Nelson 1999, chapter 9). Second, there is the puzzling variation among nations in their ability to borrow, adapt, and apply production methods that innovating nations already have developed. What conditions and forces prevent some countries from employing modern methods of production? This study is not concerned with the nature of growth leadership, the sources of new technologies, or the strategies of second-tier nations (R. R. Nelson 1996). Instead, I tackle the second puzzle, the inability to either create or imitate new technologies.
This chapter briefly analyzes the evolution of modern growth theory in economics and looks for explanations of economic stagnation in low-income countries. According to the division of labor in the field of economics, growth theory has the task of studying long-term economic growth and is therefore a logical starting point for this inquiry into the problems of growth laggards. We will see that growth theory is concerned with equilibrium properties of successful growing economies and does not explicitly examine the role of social institutions in growth. Yet its latest version, the so-called new or endogenous growth theory, implicitly explains poor growth performance by appealing to unspecified social barriers that prevent countries from drawing on the stock of world knowledge to upgrade their production methods. In this study I distinguish between two categories of applied knowledge that are relevant for economic growth, production technologies and social technologies, and argue that inability or unwillingness to apply appropriate social technologies is the main cause of relative economic backwardness. The problem of successfully introducing new social technologies is the organizing theme of this volume.
Chapter 2 outlines the basic argument about why social technologies rather than production technologies constitute the crucial barrier to growth. The chapter also introduces key tools and concepts that I use to analyze these issues. My approach is a modified version of the new institutional economics. The traditional tools of economics are not appropriate for studying important aspects of social technologies. I believe that unblinking faithfulness to standard economics would bias the selection of variables and possibly make us turn a blind eye to important social phenomena (Stiglitz 1999). Ideally, the study of institutions and social technologies requires a robust theory of social systems, which economics and social science in general lack. The result is my eclectic approach.
Learning from Growth Theory
The Three Waves
Since it first appeared around the middle of the twentieth century, modern growth theory has evolved through three phases, all focusing on the relationship between physical inputs and outputs rather than on the social environment of producers. As a general rule, growth theory identifies two ways of raising a country's average output per worker. First, if a country's economy is inside the production possibilities frontier, which is determined by the best available production technologies, the theory suggests that the country is combining its inputs in inefficient proportions. In this instance, the country is able to raise average output per worker and move toward the production frontier by adjusting its factor (input) ratios, which usually means raising the ratio of physical and human capital to basic or unskilled labor services. Second, when a country is on its production possibilities frontier, only new technology that moves out the frontier can further raise output per capita. In growth theory, new technology is the ultimate engine of growth.
In its first two phases, the theory simply assumed that new technology followed a time trend and did not attempt to explain technological change. The third phase, endogenous growth theory, does not explain growth in terms of exogenous technological change and fails to analyze in any detail what social circumstances favor the production and application of knowledge. Endogenous growth theory, however, has abandoned the traditional assumption that all countries have ready access to state-of-the-art production technology, which implied that variation in national economic performance is always caused by different factor ratios.
In sum, growth theory tells us that countries are relatively poor because they have failed to accumulate inputs, particularly capital in its many forms, except that the recent endogenous growth theory recognizes that poor countries somehow are not able to use world knowledge to upgrade their production capacities. We now look at these findings in more detail.
The Harrod-Domar Model Creates Path Dependence
Modern growth theory originates in the theoretical world of J. Maynard Keynes (1936), the father of modern macroeconomics. The Great Depression of 1929 and the parallel ascendance of Keynesian macroeconomics changed the worldview of most economists and eclipsed the idea of a self-correcting market system. Equilibrium at low levels of employment now appeared to be an empirical and theoretical possibility, and thoughtful people wondered whether growing economies, even more than stationary ones, were prone to unemployment or overheating. Evsey Domar (1946) set out to answer this question, and his work (and earlier work by Roy Harrod ) put mainstream growth theory on a path that it has followed through the changing landscapes of the last half century.
The Harrod-Domar model emerged in response to concern about a possible imbalance between a country's growing production capacity and the increase in total demand required to keep up with increasing capacity. The reason for the uneasiness was that unemployment increases when the capacity to produce grows faster than the demand for output. As the original focus of the analysis was not on the sources of economic growth, Harrod-Domar models simply assume that some rate of technological change is given-and that it is of the labor-saving variety. The theory also makes the critical assumption that labor and capital must be used in fixed proportions because the inputs cannot be substituted for each other.
It follows from the stringent assumptions of the model that a Harrod-Domar economy contains no internal equilibrating mechanism that guarantees balanced growth. Balanced growth depends on the relationships among four variables: the saving rate, s; the capital-output ratio, v; labor saving technical change, a; and population growth, n. All these variables are determined outside the model; they are exogenous. Formally, balanced growth requires s/v = n + a, but the two sides are equal only by mere chance.
The problem of balancing aggregate supply and aggregate demand in a growing economy certainly is an important issue, but it is not the heart of the matter when we try to understand Third World poverty. Yet economists lost no time applying the Harrod-Domar model outside its intended (and in retrospect questionable) sphere of competence to provide theoretical foundations for the view that capital accumulation is the key to economic development. The premise that capital and labor are used in fixed proportions appeared to make sense for developing countries undergoing structural change, especially when combined with W. A. Lewis's (1954, 1955) famous dual-economy model of economic development with unlimited supply of labor. By merging the two images, economists made lack of saving the primary cause of poverty and supply of saving the key policy variable.
The story goes like this. Developing countries typically have a dual economy made of a modern manufacturing sector, which has a relatively high and (practically) fixed capital-output ratio as well as a traditional labor-intensive agricultural sector where the marginal product of labor is (close to) zero, indicating hidden unemployment. In such situations, economic development involves structural transformation, especially transfer of labor from the traditional sector into the modern (usually manufacturing) sector. Although the social rate of return in the manufacturing sector is high, in part because of the low marginal opportunity cost of labor, very little new investment and output expansion occurs in the modern sector. The reason is that developing countries lack investment funds (domestic saving or foreign funds), and labor cannot be substituted for capital in the modern sector. Yet if they somehow could manage to increase their investment quota, developing countries with surplus labor could for a while achieve very high rates of growth simply by expanding the modern sector and shrinking the traditional sector. Later we will have more to say about the image of insufficient investment funds as the main barrier to growth.
The Capital-Output Ratio Becomes an Endogenous Variable
The second stage in the evolution of growth theory, neoclassical growth theory, originated in the work of Solow (1956) and Swan (1956) and responded to the Harrod-Domar vision of a growing market economy as being inherently unstable and prone to overheating or unemployment. Technically, the obvious way to avoid the gloom of Harrod-Domar growth theory is to model as endogenous at least one of the four underlying variables and hope that the equation s/v = n + a will then have a solution most of the time (Solow 1994, 46). Neoclassical growth theory made v, the capital-output ratio, endogenous, but let the saving ratio, population growth, and technological change retain their status as parameters. An endogenous v implies that labor and capital are substitutes in production and that unemployment ceases to be a problem. When the labor force of a country grows faster than its capital stock, labor substitutes for capital and the flexible capital-output ratio simply shrinks. Having made v endogenous, economists waited some thirty years, or until the 1980s, before they again tried to fundamentally remodel growth theory, as the following section discusses.
Neoclassical growth theory formally models the economy of a country as a single production function. A simple production function makes national product, Y, depend on the level of production technology, A, and two inputs, capital, K, and labor, L, which we can write as Y = (A; K, L). Improved production technology-technological change-is introduced as an increase in A. When a country arrives at the optimal capital-labor ratio, technological change is the only source of increase in per capita output, but the theory makes no attempt to explain why A changes-A is an exogenous variable. Furthermore, the theory boldly assumes that all countries have access to modern technology; they all have identical production functions.
The lack of concern for both the origins and diffusion of production technologies may seem curious, but again, it is best explained by the initial motivation of the economists who developed neoclassical growth theory. The original purpose and use of the theory was to analyze long-term growth paths for mature industrial economies, and neoclassical growth theory offers some striking findings. The theory predicts that the long-run (steady-state) growth rate of a country is independent of its saving rate, that all countries will achieve the same steady-state growth rate, and that all economies eventually will converge on an identical output per head.
The original intentions were soon forgotten, and economists, in their eagerness to generalize, improperly applied neoclassical growth theory to the study of poor countries in the Third World, reinforcing the findings of the Harrod-Domar approach about the causes of underdevelopment. If all countries employ the same production function and have the same value for A in Y = f(A;K,L), differences in output per person, Y/L, must result from differences in the capital-labor ratio, K/L, assuming constant returns to scale in the production function. Developing countries are poor because on average their workers are supported by relatively few capital assets.
Excerpted from Imperfect Institutions by Thráinn Eggertsson
Copyright © 2005 by University of Michigan . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
|Introduction : opportunities lost||1|
|Ch. 1||Imperfect institutions and growth theory in modern economics||9|
|Ch. 2||Barriers to growth : institutions and social technologies||23|
|Ch. 3||Competing social models||34|
|Ch. 4||Stable poverty and unstable growth||47|
|Ch. 5||The political logic of bad economics||59|
|Ch. 6||Inefficient social norms||74|
|Ch. 7||Why Iceland starved||99|
|Ch. 8||Applying social technologies : lessons from the old theory of economic policy||127|
|Ch. 9||Degrees of freedom in institutional reform||138|
|Ch. 10||Eluding poverty traps, escaping history||152|
|Ch. 11||Minimal property rights and legal transplants||174|
|Conclusion : the subtle art of major institutional reform||191|