African economic growth reconsidered: Measurement and performance in east-central Africa, 1965-1995.

摘要

Lack of economic growth has overwhelmingly been the focus of studies of the economic history of post-colonial Africa. Ironically, this has diverted attention from explaining the process of economic growth. Explaining African economic growth as it happened, with attention to episodes of growth and changes in incentive structures, is much more demanding of the African growth evidence. There are serious validity and reliability issues with the Africa data. This stands in contrast with the widespread use of the data as functional evidence for economic analysis. The thesis sheds new light on both methodological and substantive issues through a comparative study of the national accounting methodologies in Botswana, Kenya, Tanzania and Zambia. It is found that baseline estimates and growth estimation methodologies are different across countries, and that these to an extent determine differences in measured growth, and therefore might influence conclusions in the literature. The main sources of growth evidence are compared with the national accounts data. It is shown that these different sources do not cohere. These data quality issues are serious enough to compromise research on post-colonial African economic history unless proper care is taken. The final part of the thesis analyses the growth experiences of these four countries on the basis of the national accounts data. At face value the stylised facts about averaged growth rates match the idealised typologies of African economies based on their policy and institutional frameworks. It is shown, however, that when we examine the changes in economic growth rates during the period, and the sources of those changes, the explanations from the case studies do not cohere with the orthodox narrative. While there are clear differences in the growth performance of the countries, these differences in growth rates were determined by events over which the policy makers and the institutional framework could have only limited influence. The case studies underline the importance of looking beyond the averaged aggregate growth rates, because of, rather than despite, the issues of data quality.