Thanks to international agencies such as the World Bank and private nongovernmental organizations, we now have respectable quantitative measurements for economic and even political developments for many countries for the past 20 to 30 years. Such data permit the application of statistical techniques to test various theories about some age-old questions. Barro's short book, based on his Lionel Robbins lectures at the London School of Economics, is an excellent example of this now-fashionable approach to social scientific analysis. The author addresses two important questions: What factors are conducive to national economic growth? and Under what conditions will democracy flourish?
The results on growth are drawn from over 80 countries since the 1960s. High growth in per capita income is invariably associated with high rates of investment. But what determines these? The author focuses on more fundamental factors, and his results confirm what has now become conventional wisdom: high levels of secondary schooling, high life expectancy (reflecting good health), and a greater degree of rule of law are all conducive to growth. And of course it helps to have a persistent rise in one's export prices. High fertility, high government expenditure (other than on education and defense), and very high inflation rates all have a negative impact on growth. Some democracy seems to help growth, but too much slows it down, in Barro's analysis. Democracy, as has been hypothesized for many years, seems to require some degree of economic well-being and good health to thrive. Primary education of women, but not secondary education of men, also has a positive effect. Income inequality, at least according to the incomplete measures available, does not seem to affect it one way or the other. The advantage of this large-numbers approach to traditional questions is that it can get beyond anecdotes and theories based on a few examples. But the analysis accounts for only about half the variation in growth rates across countries. There are many exceptions to Barro's general conclusions, and much remains to be explained.