Taking Stock

What have we learned? The ma jor point of this chapter has not been the development of new theory. Our major objective in this chapter has been to see whether we could use the Solow model to have a more informed interpretation of cross-country differences and also use data in order to understand the strengths and shortcomings of the Solow growth model.

At the end of this brief journey, the message is somewhat mixed. On the positive side, despite its simplicity, the Solow model has enough substance that we can take it to data in various different forms, including TFP accounting, regression analysis and calibration. Moreover, each of these different methods gives us some idea about the sources of economic growth over time and of income differences across countries.

On the negative side, however, no single approach is entirely convincing. Each relies on a range of stringent auxiliary assumptions. Consequently, no firm conclusions can be drawn. The simplest applications of the Solow accounting framework suggest that technology is the main source of economic growth over time. However, this conclusion is disputed by those who point out that sufficient adjustments to the quality of physical and human capital substantially reduce or perhaps even totally eliminate residual TFP growth. The same debate is seen in the context of cross-country income differences; while some believe that accounting for differences in physical and human capital across countries leaves little need for technology differences, others show that, with reasonable models, most of the cross-country differences are due to technology.

While complete agreement is not possible, it is safe to say that the consensus in the literature today favors the interpretation that cross-country differences in income per capita cannot be understood solely on the basis of differences in physical and human capital; in other words there are technology differences across countries and these technology differences are likely to be at the heart of cross-country income and growth differences.

Hence one important potential lesson from this data detour is that technological progress is not only important in generating economic growth in the basic Solow model, but also likely to be a ma jor factor in cross-country differences in prosperity. A detailed study of technological progress and technology adoption decisions of households and firms is there­fore necessary. This motivates the detailed analysis of technological progress and technology adoption later in the book. It is also useful to emphasize once again that differences in TFP are not necessarily due to technology in the narrow sense. If two countries have access to the same technology but make use of these techniques in different ways with different degrees of efficiency or if they are subject to different degrees of market or organizational failures, these differences will show up as TFP differences. One indication that TFP differences arising from market or organizational failures are important comes from episodes of severe crises. When countries have large drops in their income, due to civil wars, political instability, financial crises or other reasons, this is almost always associated with a corresponding decline in TFP (and little change in the capital stock and a much smaller change in labor inputs). Naturally, these drops in TFP are not caused by “technological regress,” but result from the breakdown of the market or increases in some other sources of inefficiency. Therefore, technology dif­ferences should always be construed rather broadly and we should pay special attention to cross-country differences in the efficiency of production. By implication, to understand TFP differences across countries, we must study not only differences in the techniques that they use but the way they organize markets and firms and how to incentivize different agents in the economy. This again shapes our agenda for the rest of the book, especially paving the way for our investigation of endogenous technological change in Part 4 and of differences in technology and productive efficiency across countries in Parts 6 and 7.

There is one more sense in which what we have learned in this chapter is limited. What the Solow model makes us focus on, physical capital, human capital and technology, are proximate causes of economic growth in cross-country differences. It is important to know which of these proximate causes are important and how they affect economic performance both to have a better understanding of the mechanics of economic growth and also to know which classes of models to focus on. But at some level (and exaggerating somewhat) to say that a country is poor because it has insufficient physical capital, human capital and inefficient technology is like saying that a person is poor because he does not have money. There are, in turn, other reasons making some countries more abundant in physical capital,

human capital and technology, in the same way as there are factors that make a person have more money than another. I have referred to these as the fundamental causes of differences in prosperity, contrasting them with the proximate causes. A satisfactory understanding of economic growth and of differences in prosperity across countries requires both an analysis of proximate causes and of fundamental causes of economic growth. The former is essential for the study of the mechanics of economic growth and to develop the appropriate formal models incorporating these insights. The latter is important for understanding why some societies make choices that lead them to low physical capital, low human capital and inefficient technology and thus to relative poverty. This is the issue I turn to in the next chapter.

3.8.