Read an Excerpt
The Forces of Economic GrowthA Time Series Perspective
By Alfred Greiner Willi Semmler Gang Gong
Princeton University PressCopyright © 2005 Princeton University Press
All right reserved.
Chapter OneEconomic Growth in Historical Perspective
1.1 Historical Perspective
Over the past two hundred years, countries have varied widely in their patterns of economic growth. In the nineteenth century, the United Kingdom was the leading industrialized country, with Germany and France catching up, and then the United States leapfrogged the European countries around the turn of the century. In the period after World War II, per capita income in Japan and Germany increased dramatically. Out of these spurts of growth emerges a long-term historical trend: the United States and other countries that now belong to the Organization for Economic Cooperation and Development (OECD) have seen a persistent increase in per capita income of roughly 2 percent per annum over the last century. Yet during the same period, other countries have continued to languish in poverty. This marked difference in economic performance is not accidental, for in some countries major forces of growth were set in motion that were lacking in other countries.
The problem of economic growth has been studied for as long as economics has been a recognizable discipline. In the eighteenth century, Adam Smith (1976) saw thatthe forces of growth were released by freeing market agents from external restrictions. He predicted that the increasing size of markets, as well as increasing returns and externalities due to a rising division of labor, would spur development. Early in the nineteenth century, David Ricardo (1951) emphasized investment in machinery as a cause of the increase in per capita income. Karl Marx (1967), following Ricardo, also saw investment in machinery and capital accumulation as major sources of growth. John Stuart Mill (1900), by contrast, emphasized education and the sciences as engines of growth. All of the classical authors understood that economic activity, carried out by private agents in markets, must be complemented by social and public infrastructure.
The classical economists also knew that the development of market forces and economic growth would likely be accompanied by inequality. As economies expand, traditional sectors and traditional methods of production are rendered obsolete, the workforce is deskilled, and the income of some groups is depressed-while other agents grasp opportunities, create wealth, and accumulate fortunes. Joseph Schumpeter (1935) in particular perceived economic growth as a process of "creative destruction" in which some actors gain and others lose.
In addition to recognizing the divergence in income between sectors and groups, the classical economists (as did Schumpeter) conceived growth as a process that converges in the long run toward a stationary state of per capita income. In the modern period, after John Maynard Keynes (1936), growth theory was furthered by the seminal contributions of Roy Harrod (1939, 1948) and Evsey Domar (1946, 1957) and then of Robert Solow (1956, 1957) and Trevor Swan (1956) and of Nicholas Kaldor (1956, 1961, 1966). Kaldor (1961), taking a position contrary to the classical economists, was the first to state that persistent growth of income per capita is one of the major stylized facts (that is, phenomena that can be observed in a number of countries over a long period) of advanced countries. The revival of growth theory, with important contributions by Hirofumi Uzawa (1968), Robert Lucas (1988), Paul Romer (1986, 1990), and Robert Barro (1990), has taken roughly the same view as Kaldor in identifying the causes of persistent economic growth. Classical forces of growth have been rediscovered and presented in formal models by building on intertemporal behavior and the dynamic optimization of economic agents.
As Angus Maddison (2001) shows, forces of economic growth were set in motion in western Europe a long time ago through its encounter with parts of the world where high cultures had developed. The major sources of growth since the Renaissance, as Maddison demonstrates, have been learning from others, education, collecting and diffusing technological knowledge, and improvement of scientific methods. The diffusion of new knowledge and new technology was, in western Europe, accelerated by the interaction and institutionalized cooperation of scientists in institutions of higher learning and scientific academies, which encouraged discussion, collection, and publication of theoretical and practical research. In European countries this was always a matter of public discourse and public policy.
It is well understood by now that different forces of economic growth characterize each stage of development. This book takes a time series perspective on development, employing dynamic and time series methods to study the major sources of growth. We concur with recent criticism of cross-country studies that maintain that the forces of growth are the same at all times and in all countries. In taking a time series perspective, we support the view that in earlier stages, learning from others, externalities, and increasing returns are major sources of economic growth. At a later stage, education and the build-up of human capital are important, as growth effects are visible that appear to be proportional to efforts devoted to education.
However, such scale effects of education and human capital may not hold for still later stages of development. Nonlinearities now seem to be at work, since educational efforts show less than proportional effects on growth rates in advanced countries. A growth model with human capital, such as the Uzawa-Lucas model, might be an appropriate one to describe the stage of development at which the creation of human capital is effective in increasing per capita income. At a later stage, the creation and diffusion of knowledge and new technology through research and development (R&D) spending and a high proportion of scientists and engineers in the total working population seem to become important. Only countries at the forefront of such efforts may be successful in keeping growth rates high. The Romer model, which analyzes some of these forces, seems to be suited to describe this stage of growth. Social and public infrastructure appears to be important for all stages of growth, yet each stage may need specific social and public infrastructure. Last, the connection between economic growth and inequality, to which a great many theorists, both classical and contemporary, have alluded (see Aghion 2002), appears to be an important factor at all stages of development.
Basing our conclusions on a time series perspective and allowing for nonlinearities, we will discuss the implications of our study for policy. We note, however, that economic growth may not only increase potential per capita income for future generations but may also create negative externalities by reducing renewable and nonrenewable resources, as well as by degrading the environment. Although this is an important problem in the context of a study on economic growth, it will be left aside here. Finally, in taking a time series perspective in our modeling and estimation strategy, we are very much aware of thresholds in development and growth. Only countries that have crossed those thresholds may enjoy the stages of growth sketched above.
1.2 New Growth Theory and Cross-Country Studies
As we have mentioned, important studies of the persistent growth of per capita income were provided by Harrod (1939, 1948), Domar (1946, 1957), and Kaldor (1961, 1966). Harrod and Domar were primarily concerned with the stability of the steady-state growth path. The knife-edge problem stated by Harrod and Domar was contested by Solow (1956), who, assuming smooth factor substitution, could demonstrate global stability and convergence toward the steady-state path. Kaldor (1966) obtained stability results by referring to different saving rates from class income with changing income levels. However, the growth theory of the 1950s and 1960s did not sufficiently identify the major sources of growth. In Solow, growth of per capita income occurs only because of exogenous technical change. Modern growth theory, by contrast, attempts to explain economic growth endogenously.
The new growth theory started with Romer's 1986 paper. This model explains persistent economic growth by referring to the role of externalities. This idea had been formalized earlier by Arrow (1962), who argued that externalities, arising from learning by doing and knowledge spillover, positively affect the productivity of labor on the aggregate level of an economy. Lucas (1988), whose model goes back to Uzawa (1965), stresses the creation of human capital, and Romer (1990) and Grossmann and Helpman (1991) focus on the creation of new knowledge as important sources of economic growth. The latter authors have developed an R&D model of economic growth. In the Romer model, the creation of knowledge capital (stock of ideas) is the most important source of growth. In Grossman and Helpman, a variety of consumer goods enters the utility function of the household, and spillover effects in the research sector bring about sustained per capita growth. A similar model, which can be termed Schumpeterian, was presented by Aghion and Howitt (1992, 1998). In it the process of creative destruction is integrated in a formal model; the quality grades for a product are modeled as substitutes; in the extreme case the different qualities are perfect substitutes, implying that the discovery of a new intermediate good replaces the old one. Consequently, innovations are the source of sustained economic growth. Perpetual growth can also arise due to productive public capital or investment in public infrastructure. This line of research was initiated by Arrow and Kurz (1970), who, however, only considered exogenous growth models. Barro (1990) demonstrated that this approach may also generate sustained per capita growth in the long run. Numerous empirical studies have been generated by the new growth theory. The first round of empirical tests by and large focused on cross-country studies. There are a great many cross-country empirical estimations of recent growth theory, using either an extended Solow-based approach or endogenous growth theory. Here we do not exhaustively survey the cross-country studies on endogenous growth theory. Their success or failure is reviewed by Sala-i-Martin (1997) and Durlauf and Quah (1999). However, we have to point out that criticism has been raised against cross-country econometric studies. It has been demonstrated that these studies, by lumping together countries at different stages of development, may miss the thresholds of development (Bernard and Durlauf 1995). Moreover, cross-country studies rely on imprecise measures of the economic variables involved, and the results are amazingly nonrobust (Sala-i-Martin 1997).
In addition, cross-country studies imply that the forces of growth, as well as technology and preference parameters, are the same for all countries in the sample. When estimating the Solow growth model using a sample consisting of, say, one hundred countries, the obtained parameter values are identical for each country. However, if the countries in this sample are highly heterogeneous in their states of development, different parameter values will characterize their technology or preferences.
It is also to be expected that different institutional conditions and social infrastructure in the countries under consideration will affect estimations and will make the countries heterogeneous, leading to differences in the estimated parameters. Brock and Durlauf (2001) therefore argue that cross-country studies tend to fail because they do not admit uncertainty and heterogeneity of parameters into the model.
An influential cross-country study that assumes an exogenous growth model is the paper by Mankiw, Romer, and Weil (1992), who augment the Solow-Swan exogenous growth model by integrating human capital. The production function then is given by
Y(t) = K[(t).sup.[psi]] H[(t).sup.[omega]] [(A(t)L[(t)).sup.1-[psi]-[omega]],
with Y(t) aggregate output, K(t) physical capital, H(t) human capital, L(t) labour, and A(t) the level of technology, which grows at an exogenously determined rate. Physical capital and human capital are formed by saving a certain fraction of output, which is then devoted to the formation of these capital stocks. Denoting with [s.sub.k] and [s.sub.h], [s.sub.k] + [s.sub.h] < 1, the constant fraction of aggregate output in the formation of physical capital and human capital, the evolution of the capital stocks is given by
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],
where y(t) = Y(t)/A(t)L(t), k(t) = K(t)/A(t)L(t) and h(t) = H(t)/ A(t)L(t) give quantities per effective unit of labor. n is the exogenous growth rate of the labor force; [delta] is the depreciation rate of physical and human capital, which is the same for the two stocks; and [g.sub.A] = [??](t)/A(t) is the exogenous growth rate of technology.
Assuming diminishing returns to scale in physical and human capital, that is, [psi] + [omega] < 1, this economy converges to a steady state, which is defined as a rest point of the two equations [??](t) and [??](t). Setting [??](t) = [??](t) = 0 and solving these equations simultaneously gives the steady-state values for k and h as
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Inserting [k.sup.*] and [h.sup.*] in Y(t) and taking logarithms yields an equation that gives aggregate per capita output at the steady state. This equation is given by
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Mankiw, Romer, and Weil (1992) estimate this equation using a cross-country sample of ninety-eight countries. They assume that all economies are at their steady-state position. The result is that the augmented Solow model explains almost 80 percent of the variation in income in the countries of their sample. The implied physical capital, human capital, and labor shares are about one-third each. Mankiw, Romer, and Weil conclude that these findings cast doubt on endogenous growth models and claim that the augmented Solow exogenous growth model is able to explain much of the cross-country differences in income.
Yet in this analysis, structural parameters are posited to be the same, independent of whether a highly industrialized country or developing country is considered. This aspect is taken into account by Durlauf and Johnson (1995), who use the same data set as Mankiw, Romer, and Weil (1992) but allow for different aggregate production functions depending on 1960 per capita incomes and on literacy rates. Durlauf and Johnson use a regression-tree procedure6 in order to identify threshold levels endogenously. They find that the Mankiw, Romer, and Weil (1992) data set can be divided into four distinct regimes: low-income countries, middle-income countries, and high-income countries, with the middle regime divided into two subgroups, one with high, and one with low, literacy rates. The result of this study is that different groups of countries are characterized by different production possibilities, implying different coefficients on inputs in the aggregate production functions. Further, in contrast to Mankiw, Romer, and Weil (1992), initial conditions matter for long-run incomes, a result that is in line with endogenous growth models but in contrast to exogenous growth models. This outcome questions the empirical validity of the augmented Solow growth model, implying that cross-country differences in income cannot be explained entirely by differences in the rates of growth of physical capital, human capital, and population.
Excerpted from The Forces of Economic Growth by Alfred Greiner Willi Semmler Gang Gong Copyright © 2005 by Princeton University Press. 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.