What Have We Learned?

It may be useful to first summarize the most important aspects and takeaway lessons of our analysis. My summary would be as follows.

(1) Growth as the source of current income differences: at an empirical level, the in­vestigation of economic growth is important not only for understanding the growth process, but also because the analysis of the sources of cross-country income differ­ences today requires us to understand why some countries have grown rapidly over the past 200 years while others have not (Chapter 1).

(2) The role of physical capital, human capital, and technology: cross-country differ­ences in economic performance and growth over time are related to physical capital, human capital and technology. Part of our analysis has focused on theoretical and empirical investigation of the contributions of these factors to production and growth (Chapters 2 and 3), and the remaining, larger part, has focused on understanding physical capital accumulation, human capital accumulation, and technology cre­ation and adoption decisions (Chapters 8-15 and 18-19). One conclusion that has emerged concerns the importance of technology in understanding both cross-country and over-time differences in economic performance. Here, technology refers both to advances in techniques of production, thus to the accumulation of knowledge and blueprints for more efficient machinery, and also to the general efficiency of the or­ganization of production, which will be affected by the incentives that a society (and its government) provides to firms and workers, by its contracting institutions, and by the types of market failures that prevent the development of more productive economic relationships (Section 18.5 in Chapter 18 and Chapter 21).

(3) Endogenous investment decisions: while we can make considerable progress by un­derstanding the role of physical capital and human capital, and using cross-country data on differences in investments in machinery and in education to account for the process of economic growth and development, we also need to endogenize these in­vestment decisions.

(4) Endogenous technology: likewise, technology should be thought of as endogenous, not as mana from heaven. There are good empirical and theoretical reasons for thinking that new technologies are created by profit-seeking individuals and firms, via research, development and tinkering. In addition, decisions to adopt new tech­nologies are likely to be highly responsive to profit incentives. Since technology appears to be a prime driver of economic growth over time and a major factor in cross-country differences in economic performance, we must understand how tech­nology responds to factor endowments, market structures, and rewards. Developing a conceptual framework that emphasizes the endogeneity of technology has been one of the major objectives of this book. The modeling of endogenous technology necessitates ideas and tools that are somewhat different from those involved in the modeling of physical and human capital investments. Three factors are particularly important. First, fixed costs of creating new technologies combined with the non­rival nature of technology necessitates the use of models where innovators have ex post (after innovation) monopoly power. The same might apply, though perhaps to a lesser degree, to firms that adopt new technologies. The presence of monop­oly power changes the welfare properties of decentralized equilibria and creates a range of new interactions and externalities (Chapters 12 and 13 and Section 21.5 in Chapter 21). Second, the process of innovation is implicitly one of competition and creative destruction.

(5) Linkages across societies and balanced growth at the world level: while endogenous technology and endogenous growth are major ingredients in our thinking about the process of economic growth in general and the history of world economic growth in particular, it is also important to recognize that most economies do not invent their own technologies, but adopt them from the world technology frontier or adapt them from existing technologies (Chapter 18). In fact, the process of technology transfer across nations might be one of the reasons why after the initial phase of industrialization, countries that have been part of the global economy have grown at broadly similar rates (Chapter 1). Therefore, the modeling of cross-country income differences and the process of economic growth for a large part of the world necessi­tates a detailed analysis of technology adoption, technology diffusion and technology transfer. Two topics deserve special attention in thinking about technological link­ages across countries and technology adoption decisions. The first is the contracting institutions supporting contracts between upstream and downstream firms, between firms and workers, and between firms and financial institutions.

(6) Takeoffs and failures: the last 200 years of world economic growth stand in stark contrast to the thousands of years before. Despite intermittent growth in certain parts of the world during certain epochs, the world economy was largely stagnant until the end of the 18th century. This stagnation had multiple aspects. These included low productivity, high volatility in aggregate and individual outcomes, a Malthusian-type configuration where increases in output were often accompanied by increases in population does only having a limited effect on per capita income, and a largely rural and agricultural economy. Another major aspect of stagnation has been the failed growth attempts; many societies grew for certain periods of time and then lapsed back into depressions and stagnation. This changed in the 18th century. We owe our prosperity today to the takeoff in economic activity, closely related to industrialization, that started in Britain and Western Europe, and spread to certain other parts of the world, most notably to West European offshoots, such as the United States and Canada.

(7) Structural changes and transformations: modern economic growth and development are accompanied by a set of sweeping structural changes and transformations. These include changes in the composition of production and consumption (the shift from agriculture to industry and from industry to services), urbanization, financial devel­opment, changes in inequality of income and inequality of opportunity, the trans­formation of social and living arrangements, changes in the internal organization of firms, and the demographic transition. While the process of economic development is multifaceted, much of its essence lies in the structural transformation of the econ­omy and society at large (Section 17.6 in Chapter 17, Chapters 20 and 21). Many of these transformations are interesting to study for their own sake. Moreover, they are also important ingredients for sustained growth. Lack of structural transformation is not only a symptom of stagnation, but often also one of its causes. Societies may fail to takeoff and benefit from the available technology and investment opportu­nities in the world partly because they have not managed to undergo the requisite structural transformations, so they do not have the right type of financial relations, the appropriate skills, or the types of firms that would be necessary as conduits of new technologies.

(8) Policy, institutions and political economy: the reward structures faced by firms and individuals play a central role in shaping whether they undertake the investments in new technology and in human capital necessary for takeoff, industrialization, and economic growth.

(9) Endogenous political institutions: policies and institutions are central to understand­ing the growth process over time and cross-country differences in economic perfor­mance, but these social choices are in turn determined within society’s political in­stitutions. Democracies and dictatorships are likely to make different policy choices and create different types of reward structures. But political institutions themselves are not exogenous in the long-run equilibrium of a society. They are both deter­mined in equilibrium and change along the equilibrium path as a result of their own dynamics and as a result of stimuli coming from changes in technology, trading opportunities, and factor endowments (Chapter 23). We have already seen some simple models that provide various useful insights, but much remains to be done. Towards a more complete understanding of world economic growth and the income

differences today, we therefore need to study the following: (1) how political institu­tions affect policies and economic institutions, thus shaping incentives for firms and workers; (2) how political institutions themselves change, especially interacting with economic outcomes and technology; (3) why political institutions and the associated economic institutions did not lead to sustained economic growth throughout history, why they enabled economic takeoff 200 years ago, and why in some countries, they blocked the adoption and use of superior technologies and derailed the process of economic growth.

In this brief summary, I focused on the ideas most relevant for thinking about the process of world economic growth and cross-country income differences we observe today. The reader will recall that the focus in the book has been not only on ideas, but also on careful math­ematical modeling of these ideas and mechanisms so that coherent and rigorous theoretical approaches to these core issues can be developed. Nevertheless, to avoid repetition I will not mention the theoretical foundations of these ideas, which range from basic consumer, producer and general equilibrium theory to dynamic models of accumulation, models of mo­nopolistic competition, and dynamic models of political economy. But I wish to emphasize again that a thorough study of the theoretical foundations of these ideas is necessary both to develop a satisfactory understanding of the main issues and also to find the best way of making them empirically operational.

24.2.