1. This work was first published in Great Britain and Tanzania and later (1974) in the United States. All citations herein are to the American edition.

2. This section makes no attempt to cover all the problems involved in using national per capita income for intercountry comparisons of well-being. Some of the problems not dealt with here are the following: There is great variation among African countries in terms of their national income accounting procedures and the way they define national income. The extent to which certain kinds of economic activity (such as the work of women done in the home) is included can vary greatly, as can the way in which the production of foreign enterprises is treated. Another problem is that to make the sort of international comparisons Rodney makes, one must convert national income figures into a common currency, such as dollars, franks, yen, or marks. In many countries, however, there are multiple exchange rates, and it is impossible to choose the exchange rate. Beyond that, however, even if there is only one exchange rate, exchange rates don't necessarily reflect the internal purchasing powers of their respective currencies correctly, hence a per capita national income of $100 or $150 per capita—reported for many African countries—may suggest that countries are much poorer, or “underdeveloped” in some way, than they really are, because they conceal the fact that $100 will purchase many more goods and services than is the case in the reader's country. Still another problem is the fact that national income figures at their best only say something about the amount of economic activity; they say nothing about the needs of the population involved. Needs vary enormously from country to country, not only because of such things as genetic and cultural differences, but also because of climate. Tropical countries clearly need to invest less of their resources in protecting themselves from the cold of winter than does the rest of the world. More detail on these and still other problems can be found in Robson, P. and Lury, D.A. (eds.), The Economics of Africa (London, 1969), pp. 24–28 Google Scholar and the sources cited therein. See also Morgenstern below.

3. A margin of error of ±10 percent for the national income of the United States emerges from the work of Simon Kuznetz, who won the Nobel Prize for his work on national income accounting ( Morgenstern, Oskar, On the Accuracy of Economic Observation, Princeton, 1963, p. 255 Google Scholar).

4. Even for the handful of African countries in which some of the dietary staples are marketed through centralized government agencies (e.g., the maize marketing board of Kenya), black markets are commonly used, so that only a fraction of a crop may enter the government marketing channels.

5. See, for example, Miracle, M.P., Maize in Tropical Africa (Madison, 1966), pp. 241-42, 248.Google Scholar

6. Charron, K.C., The Welfare of the African Labourer in Tanganyika (Dar es Salaam, 1946), p. 43 Google Scholar; Food and Agriculture Organization, Food Composition Tables for Use in Africa (Rome, 1968), p. 34.Google Scholar

7. Platt, B.S., Table of Representative Values of Food Commonly Consumed in Tropical Countries (London, 1945), pp. 8, 9, 12, 13 Google Scholar; Food and Agriculture Organization, Tables, p. 13.Google Scholar

8. Platt, , Tables, pp. 8–13 Google Scholar; Food and Agriculture Organization, Tables, p. 21.Google Scholar

9. The most readable treatise on this subject is Bennett, M.K., The World's Food (New York, 1954), pp. 94–114.Google Scholar

10. Farnsworth, Helen C., “Defects, Uses, and Abuses of National Food Supply and Consumption Data,” Food Research Institute Studies (November, 1961).Google Scholar

11. Ibid., p. 188.

12. Poleman, Thomas T., “The Food Economies of Urban Africa: The Case of Ghana,” Food Research Institute Studies 2 (May, 1961): 131–35.Google Scholar