Printer Friendly

Program report: economic growth.

Program Report

Economic Growth

In the 1950s and 1960s, there was widespread interest among economists in the long-run rate of economic growth. Initially, economists followed the pattern set by Robert M. Solow's seminal papers and developed the theory of dynamic models in tandem with empirical work on growth accounting. (1) But by the 1970s, theoretical and empirical work on growth and development diverged. Professional interest in growth waned, despite the paramount importance of these issue. Two years ago, we created a growth project to encourage the kind of interaction between theory and evidence that once had been so successful.

Recent Trends in Empirical

and Theoretical Work

Two recent sources of data have been influential in stimulating renewed empirical work on growth. In Phases of Capitalist Development, Angus Maddison summarized much of his work on constructing comparable sets of data that extend over a century or more for a small set of developed countries. Separately, Alan Heston and Robert Summers have compiled national income accounts data for a large cross section of countries since World War II. Because these data correct for differences in relative prices in different countries, they permit more meaningful comparisons of the level of income per capita than were previously possible.

Frm the early analyses of these data, several questions emerged. (2) First, why is it that poor countries, on average, do not catch up with the rest of the world? Traditional theory suggests that convergence should take place almost automatically. Further, productivity and income per capita within the developed countries have converged to a substantial degree.

Second, who do long-run data show a correlation between the reate of savings and the rate of growth? According to the neoclassical model of growth, the only explanation for the correlation is that variation in the rate of technical change induces variation in both the rate of savings and the rate of growth. But the magnitude and importance of the observed correlation suggest other possibilities: that variation in the savings rate might cause variation in the rate of growth; or that variation in government policy migh cause variation in both the savings rate and rate of growht.

More extensive investigation of the Heston and Summers data, supplemented by data collected by Robert J. Barrio, identified other important correlations. (3) Everything else equal, a higher rate of investment by either the private or public sector, a lower share of government consumption spending, higher school enrollment rates, greater political stability, and lower fertility all are correlated with a faster rate of growth. This sample, with its large number of poor cuntries, also reveals evidence about convergence. Everything else held constant, a lower level of initial income per capita is associated with a faster rate of growth. This means that poor countries would grow faster if the other variables did not vary systematically with the level of income. But they do. Poor countries have lower rates of investment, lower school enrollment rates, higher fertility rates, and less political stability. Barro and Xavier Sala-i-Martin also find that income per capita in U.S. region has converged over the past 150 years. (4)

All of these empirical findings point to the importance of alternative models in which the long-run growth rate can be influenced by policy variables and individual preferences. It is difficult to believe that exogenous differences in the rate of technological change alone can lead to this pattern of correlations. Two complementary approaches are developing now to provide more theories linking growth, technological change, savings, and policy variables. The first assumes that even though technology is intangible, it is analogous to education or on-the-job experience and therefore can be modeled using the same perfectly competitive framework that is applied to human capital in the labor market.

The second approach emphasizes the observation that ideas that constitute a technology differ fundamentally from other economic goods. According to this second view, ideas and technological change are not transmitted in the perfect markets used in the neoclassical growth model and in the analysis of human capital. Both types of models are described in greater detail below.

Growth with Perfect Markets

When economists first assumed that technological change was determined outside the economic system, they intended this specification to be provisional. Recent work on growth with perfect markets extends neoclassical models so that all economic improvement can be traced to actions taken by people who respond to economic incentives. Sergio T. Rebelo, and Larry E. Jones and Rodolfo Manuelli, construct models in which persistent growth can arise from a convex, stationary technology if additional variables are added. (5) Rebelo's paper, which builds work by Robert E. Lucas, Jr., shows that a capital good sector with constant returns to scale can lead to persistent growth. (6) Jones and Manuelli show that even such fixed factors as labor can be introduced into this sector, provided that output is asymptotically a constant returns function of the capital stocks, as assumed by Rebelo. In both of these analyses, human capital keeps growth going.

In a paper that also uses linear technology based on human capital, Gary S. Becker, Kevin M. Murphy, and Robert F. Tamura examine the interaction between the decision by parents to invest in more children and the decision to invest in more human capital per child. (7) Adding a description of the preferences that affect fertility decisions, they show that, depending on the initial conditions, a country could end up either on a path with low growth in the population and rapid growth in per capita income or become tapped in an equilibrium with rapid growth in the population and stagnant income per capita. The model also reproduces one of Barro's strongest empirical findings: that high investment in education is associated with low fertility.

A well-known deficiency of the neoclassical model is that neither policy nor preferences can affect the steady-state rate of growth. The importance of this limitation sometimes is minimized by observing that even in the neoclassical model, policy and preferences do matter somewhat along the transition path to a steady-state growth path. Robert G. King and Rebelo have shown, however, that the transition dynamics in the neoclassical model explain only a small part of growth. (8)

Growth with External Effects

Human capital theory as developed in labor economics offers one way to model the accumulation of ideas, but there are aspects of innovation and invention that this approach does not address. When a carpenter learns a new skill, like how to hammer a nail, there is no threat that someone else will be able to free ride on the benefits of his investment in human capital. Nor does the new skill raise the productivity of his co-workers. In contrast, some discoveries can raise the productivity of many different firms or people at no cost to the discoverer, and these discoveries sometimes can be exploited by free riders who do not compensate the discoverer.

Many of the recent models of endogenous growth have captured these aspects of ideas by allowing for external increasing returns and knowledge spillovers. In the learning-by-doing formulation first outlined by Kenneth Arrow, a discovery is assumed to have no direct value to the person who makes it. Discoveries arise as an unintended side effect of some other activity and then are freely exploited by others. Mervyn A. King and Mark Robson show that the particular assumption used by Arrow, and subsequently by Romer, that the stock of capital, is quite restrictive. (9) A plausible, and more general form of knowledge spillovers can lead to complicated dynamics, important forms of persistence in rates of growth, and very different predictions about how such policy variables as taxes can influence growth.

Nancy L. Stokey uses a model based on knowledge spillovers from education and skill acquisition to show how the introduction of new goods can be linked to increases in human capital, and to address the effects of trade between developed and less developed countries. (10)

In constrast, Jess Benhabib and Boyan Jovanovic have shown that the cross-country correlation between the rate of investment and the rate of growht is consistent with a model with no external effects. (11) In their model, the underlying fundamentals are the same in each country, but persistent shocks to the technology can induce the correlation that is observed.

One of the interesting developments of the last ten years is the increasingly close connection between growth theory and the theory of macroeconomic fluctuations. The real business cycle models that evolved from the neoclassical model of growth have analogs that are based on growth models with external effects. For example, Murphy, Andrei Shleifer, and robert W. Vishny have shown how a specific type of external increasing returns could generate macroeconomic fluctuations. (12) Ricardo J. Caballero and Richard K. Lyons use time-series data for manufacturing industries to show that business cycle fluctuations are consistent with the presence of strong external increasing returns to scale. (13)

Growth in Market Power

Models with learning-by-doing and knowledge spillovers can explain how accidental discoveries arise and are exploited but cannot explain intentional attempts to make discoveries. In the microeconomic analysis of innovation and invention, market power is an essential part of the incentives that induce people to search for new discveries. Because of secrecy or explicit legal protection of patents and copyrights, the discoverer of a new idea expects to be able to charge a price for use of the idea that is higher than the opportunity cost (equal to zero) of letting it be used by others. Models with external effects depart from the perfect markets assumptions of welfare economics, but they retain the price-taking assumptions that greatly simplify the formal analysis. Models that recognize the importance of market power must abandon price taking as well, and for this reason are more difficult to analyze.

In an extension of earlier work that focused only on external factors, Romer argues for the fundamental importance of market power in the analysis of innovation and technological change, even at the aggregate level. (14) The key concepts in the analysis of ideas are the ones developed for the analysis of public goods: rivalry and excludability. In general, a good that is not rivalrous introduces a nonconvexity into the technology. Research in public finance has shown that if this nonconvexity is local, perfect price-taking competition can be supported between groups of people organized into coalitions called clubs. Jeremy Greenwood and Jovanovic analyze a growth model in which information about investment projects has this partially nonrival character. (15) In their analysis, the clubs are interpreted as perfectly competitive financial intermediaries. But if the nonconvexity is gloval, as it is for many ideas, then price taking cannot be supported and explicit attention to market power is required.

In a separate paper, Romer uses a model with market power and explicity R and D. (16) He finds that increased trade between similar regions, such as North America and Europe, will speed up growth because of the increase in market size. In a series of papers that use this model of the introduction of new goods, Gene M. Grossman and Elhanan Helpman show that trade can have complicated and offsetting effects on growth, especially when countries differ in some fundamental way, as in a product cycle model of trade between a developed and a less developed country. (17)

Innovation in these models is described as being associated with the introduction of new goods. Philippe Aghion and Peter Howitt develop an alternative model in which innovation consists of improvements in existing goods. (18) They are able to capture the dual character of innovation, which creates monopoly profits for some at the same time that it destroys monopoly profits for others. Grossman and Helpman refine this model of quality improvement, illustrate its close formal resemblance to the model in which new goods are introduced, and consider its trade implications. (19)


One of the conclusions common to all of the endogenous growth models, both the perfect markets and the imperfect markets models, is that policy choices and preferences influence long-run growth rates. In this sense, all of these models are quite distinct from the exogenous growth models. If consumers are more patient, growth will be faster. If government taxes or distortions discourage the activity that generates growth, growth will be slower. All of these models support general lessons from development outlined by Anne O. Krueger. (20) Bad government policies can affect both the level of income and the long-run rate of growth, and the welfare losses can be very large. In an explicity example, William Easterly shows how policies that affect the sectoral allocation of resources can have large effects on growth rates and welfare. (21)

Beyond these general lessons, the specific policy conclusions depend on the structure of the model. For example, in Barro and Sala-i-Martin, the optimal government fiscal policy is different in models with perfect markets, external effects, and market power. (22)

Project Activities

To support research on growth and to facilitate communication among economists doing research on empirical and theoretical topics, the growth project holds small research meetings on a regular basis, typically in the fall or spring. Because their intent is to foster interaction among different areas of economis, the meetings offer an electric collection of papers and participants with interests that include abstract equilibrium theory, history, macroeconomics, development, industrial organization, the microeconomics of research and development, and productivity analysis. The most recent meeting is described on pages 20-21 of the Summer 1990 issue of the NBER Reporter. Many of the papers from that meeting are scheduled for publication in a special May 1991 issue of the Quarterly Journal of Economics (QJE) devoted to growth theory.

To facilitate exchange of data and encourage replication of results, the project increasingly is taking responsibility for maintaining and distributing datasets. Heston and Summers have been invited to present descriptions and updates on their dataset in to different project meetings. A paper describing the most recent revision and extension (Penn World Table Mark 5) will published in the QJE special issue. (23)


The theoretical work of the last few years has established that policy can have a significant effect on growht. The empirical work demonstrates that the range of variation in growth rates is large, so that i only a small part of that variation can be traced to variation in policy, then the welfare gains from adopting better policies can be very important. Together, the empirical work and the theoretical work have outlined the set of policies that are most likely to foster growth: support for education; incentives for investment in physical capital; protection of intellectual property rights; support for R and D; international trade policies that encourage the production and worldwide transmission of ideas; and the avoidance of large government-induced distortions in the market. Ongoing theoretical and empirical work is directed at going beyond this menu of policy choices. Ultimately, economists should be able to quantify the effects of these different policies and to lead us sooner to reliable answers to the oldest question in economics: What is it that determines the wealth of nations?

(1) R.M. Solow, "A Contribution to the Theory of Economic Growth," Quarterly Journal of Economics 70 (1956), pp. 65-94, and "Technical Change and the Aggregate Production Function, "Review of Economics and Statistics 39 (1957), pp. 312-320.

(2) P.M. Romer, "Increasing returns and Long-Run Growth," Journal of Political Economy 94 (1986), pp. 1002-1037, and "Crazy Explanations for the Productivity Slowdown," in S. Fischer, ed., NBER Macroeconomics Annual 1987, Cambridge, MA: MIT Press, 1987; and W.J. Baumol, "Productivity Growth, Convergence, and Welfare," American Economic Review 76 (1986), pp. 1072-1085.

(3) R.J. Barro, "Economic Growth in a Cross Section of Countries," NBER Working Paper No. 3120, September 1989, and P.M. Romer, "Human Capital and Growth: Theory and Evidence," NBER Working Paper No. 3137, November 1989.

(4) R.J. Barro and X. Sala-i-Martin, "Economic Growth and Covergence Across the United States," NBER Working Paper No. 3419, August 1990.

(5) S. T. Rebelo, "Long-Run Policy Analysis and Long-Run Growth," NBER Working Paper No. 3325, April 1990; and L. E. Jones and R. Manuelli, "A Convex Model of Economic Growth," NBER Working Paper No. 3241, January 1990.

(6) R. E. Lucas, Jr., "On the Mechanics of Economic Development," NBER Rerprint No. 1176, May 1988.

(7) G. S. Becker, K.M. Murphy, and R.F. Tamura, "Human Capital, Fertility, and Economic Growth," NBER Working Paper No. 3414, August 1990.

(8) R.G. King and S.T. Rebelo, "Transitional Dynamics and Economic Growth in the Neoclassical Model," NBER working Paper No. 3185, November 1989, and "Public Policy and Economic Growth: Developing Neoclassical Implications," NBER Working Paper No. 3338, April 1990.

(9) M.A. King and M. Robson, "Endegenous Growth and the Role of History," NBER Working Paper No. 3151, October 1989.

(10) N.L. Stokey, "Human Capital, Production Quality, and Growth," NBER Working Paper No. 3413, August 1990.

(11) J. Benhabib and B. Jovanovic, "Esternalities Growth Accounting," NBER Working Paper No. 3190, November 1989.

(12) K.M. Murphy, A. Shleifer, and R.W. Vishny, "Increasing Returns, Durables, and Economic Fluctuations," NBER Working Paper No. 3014, June 1989.

(13) R.J. Caballero and R.K. Lyons, "The Role of External Economies in U.S. Manufacturing," NBER Working Paper No. 3033, July 1989.

(14) P.M. ROmer, "Are Noncovexities Important for Understanding Growth?" NBER Working Paper No. 3271, February 1990.

(15) J. Greenwood and B. Jovanovic, "Financial Development, Growth, and the Distribution of Income," NBER Working Paper No. 3189, December 1989.

(16) P.M. Romer, "Endogenous Technological Change," NBER Working Paper No. 3210, December 1989.

(17) G. M. Grossman and E. Helpman, "Comparative Advantage and Long-Run Growth," NBER Working Paper No. 2809, January 1989; "Endogenous Product Cycles." NBER Working Paper No. 2913, March 1989; and "Growth and welfare in a Small Open economy," NBER Working Paper No. 2970, May 1989.

(18) P. Aghion and P. Howitt, "A Model of Growth Trhough Creative Destruction," NBER Working PAper No 3225 January 1990.

(19) G.M. Grossman and E. Helpman, "Quality Ladders and the Theory of Growth," NBER Working Paper No. 3099, September 1989, and "Quality Ladders and Product Cycles." NBER Working PAPers No 3201, December 1689.

(20) A.O. Krueger "goverment Failures in Development," NBER Working Paper No. 3340, April 1990.

(21) W. Easterly, "Policy distortions, Size of Government,and Growtgh," NBER Working No. 3214, December 1989.

(22) R.J. Barro and X. sala-i-Martin, Public Finance in Models of Economic Growth," NBER Working Paper No. 3362, May 1990.

(23) Anyone interested in obtaining documentation and floppy disks containing the cross-century data collected and analyzed by Barro can write to Holger Wolf at the NBER in Cambridge. Also, in 1991, floppy disks containing the Penn World table can be obtained by writing to the Publications Department of the NBER in Cambridge. In the future, we plan to make available a dataset containing personal income by state for the United States from 1880 to the present.
COPYRIGHT 1990 National Bureau of Economic Research, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:National Bureau of Economic Research Inc.
Author:Barro, Robert J.; Romer, Paul M.
Publication:NBER Reporter
Date:Sep 22, 1990
Previous Article:Developing Country Debt and Economic Performance, vol. 2, Country Studies - Argentina, Bolivia, Brazil, and Mexico.
Next Article:U.S.-Japanese corporate finance.

Related Articles
State economy should slow in 1999.
U.S. Economic Expansion Longest on Record: Can the "New Economy" Take the Credit?
Slow growth ahead, economists predict.
Montana avoids recession and terrorist attack impacts. (The Montana Outlook).
Is Indiana's economy improving? Catching fire now--business investments and exports.
Street smart economics: know when to hold 'em and when to fold 'em.
Military modernization in an era of uncertainty.
Montana's economy: all sectors contribute to recent growth.
The Montana economy zooms along.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters