Cleveland, David A. (1991) Migration in West Africa: A Savanna Village Perspective. Africa [London] 61(2):222-246. ARTICLE AS PDF FILE.

ABSTRACT
Local migration in response to population pressure is part of the history of northeast Ghana. First by physical coercion, then by economic coercion, colonialism drastically changed the pattern of migration to one of long-distance movement from northeast Ghana and the northern savannas in general to southern Ghana. Migration in turn affected social organization, agriculture and population dynamics in savanna communities. While colonial policy was not always consistent, one dominant and ultimately effective strategy seems evident: to break up locally self-sufficient economies and societies in order to simulate the temporary migration of labour from largely subsistence agriculture to work in commercial agriculture, mining and public works in the south. These sectors were directly tied to the European economy for the benefit of Britain. Low wages and poor working conditions encouraged most migrants to return to their savanna villages when they were sick, injured or too old to work.

When Ghana gained its political independence from Britain this new pattern of migration had become firmly established and was maintained by changes in the social, economic and transport systems. Data from Zorse and the Upper Region show that migration at any one time takes about 50 per cent of working-age males and 15 per cent of working-age females to southern Ghana for periods of a year or more. Significantly increased dependency ratios mean that as a result of this migration each four remaining working-age adults must support themselves plus four dependents, instead of supporting only three dependents, as would be the case without migration. Since remittances by Zorse migrants are equal to only a small fraction of the value of their lost productive labour, the net effect of migration on the food consumption level of those remaining in the village will be determined by the balance between the increased output required of each remaining working-age adult and the decreased yield required of the total area of arable land. While I do not have all the quantitative data needed to resolve this question, statements by Zorse residents, evidence of chronic under nutrition, a long-term decrease in land productivity due to erosion and lack of organic matter, and serious labour shortages during periods o critical farm activity, suggest that the net effect of migration on Zorse is negative. That is, neither labour productivity nor land productivity is likely to compensate for the higher dependency ratio.