Endogenous Growth, Productivity and Economic Policy: A Progress Report

THE NEW “ENDOGENOUS” GROWTH theory that was spawned by the contributions of Romer (1986) and Lucas (1988) has now largely displaced the old neoclassical growth theory of Solow (1956) and Swan (1956) from the frontiers of academic research. But endogenous growth theory vintage 2004 is different in many respects from vintage 1988. These differences are the result of a fruitful interaction that has taken place between theoretical developments and empirical findings, an interaction reminiscent of the manner in which the basic building blocks of Keynesian macroeconomics were refined in the 1950s and 1960s. Attempts to test the theory have in some cases confirmed its predictions and led to new unsuspected applications. In other cases tests have uncovered important weaknesses, which theorists have remedied by introducing elements of reality that were missing from the original theory. My purpose here is to describe how the theory has been changed in the past decade and a half in response to new empirical findings, to point out which of the basic ideas underlying the original models have survived this refining process and which have been cast aside, to survey some of the empirical successes of the theory in its current state and to point out some of its major policy implications. R&D, innovation, productivity and endogenous growth Endogenous growth theory starts by accepting one of the main implications of neoclassical theory, namely that in the long run the main underlying determinant of economic growth is the long-run growth rate of total factor productivity (TFP), which in turn depends mainly on the rate of technological progress. Where it d i f f e r s f r o m n e o c l a s s i c a l t h e o r y i s i n maintaining that the rate of technological progress depends on economic forces, and can b e i n f l u e n c e d b y e c o n o m i c p o l i c y. Technological progress comes from innovations – new products, new processes and new markets that allow us to satisfy our material wants in ways that had never been thought of. Some innovations come from the application of fundamental science, and in that sense they depend on the rate of scientific progress. But many are also the result of economic activity and economic decisions. For example, firms learn how to produce more efficiently as a result of experience, and they also learn how to design b e t t e r p r o d u c t s f r o m t h e i r c u s t om e r s ’ experience. A higher level of economic activity in the economy will give firms and customers more experience in producing and us ing products, which will thus lead to a faster rate of