ABSTRACT
Experimental national income estimates in British colonies in the 1940s and 1950s paved the way for the global expansion of national income accounting. This chapter focuses on early national income accounting in British Africa. National income accounts are "statistical" in both senses of the word: they are state-produced figures and they routinely rely on statistical inference from surveyed samples. One logical solution to the problem of untangling household from business in the rural African communities Phyllis Deane studied was to regard them as very large households, with no "business" at all, thus excluding rural villages from the measured economy entirely. Prest and Stewart provoked equally strong responses from Nigerian economists. Herbert Frankel was a South African of German Jewish descent with a distinguished career as an empirical economist and as a theorist. Dudley Seers saw in the Gold Coast not a homogeneous national economy but a fragile assembly of distinct sectors.
