Population Concentration in Less Developed Countries: New Evidence

Economic theory associates the increase in population concentration, i.e., the proportion of national population residing in the core economic region, with scale and agglomeration economies. Wheaton and Shishido (1981) estimated that these persist until real per capita national income reaches US$5,000 (1985 rates). After this point in a country's economic development, population redistribution towards the core region will cease and the proportion of national population residing in the core region will commence to decline. The experience of developed countries (DCs) in the 1970s and 1980s broadly conformed to this pattern, albeit with exceptions. Evidence from less developed countries (LDCs) through the 1980 round of censuses led Vining (1986) to propose a weakened version of the US$5,000 rule, in which this point is characterized only by a slowing rate of population re-distribution towards the core, not by an outright reversal. This paper updates previously reported trends in population redistribution in LDCs and reports on many new countries. Taken as a whole, post-war data reinforce the need for caution of the sort expressed by Vining. While there is a weak negative correlation between the rate of net migration into the core region and per capita income, the share of the population residing in the core region may continue to rise even when per capita income has grown to well beyond US$5,000. Men, thinly scattered, make a shift, but a bad shift, without many things. A smith is ten miles off: they do without a nail or staple. A taylor is far from them: they'll botch their clothes. It is being concentrated which produces high convenience (Boswell 1786, p.169)