(1990) has compared allocations to education amongst fiscally distressed (adjusting) and other (non-
adjusting) developing countries and Sahn has explored changes in the share of education of the
government budget before and after the first World Bank adjustment loans. Neither are able establish
systematic effects though Gallagher concludes that reductions in social sector spending are associated with
slower economic growth. Conversely Kakwani, Makonnen and Gaag (1990) and a recent World Bank
study (World Bank 1990) associate adjustment directly with declining resources: these studies suggest the
share of the public budget and GDP allocated to education increased in all country groups after 1980
except those that experienced intense adjustment.
In a later paper Sahn (1992) restates his argument that there is no pattern of increase or decrease in real
levels of social sector spending before and after adjustment. This study is based on unweighted country
statistics of a sample of 22 Sub Saharan African countries in the 1980's; it fails to support predictions of
widespread negative effects of adjustment. It appears to demonstrate that educational expenditures
increased after structural adjustment as a percentage of discretionary government expenditure in 6 cases
(e.g. Ghana increased from 20.5% to 23%, Kenya from 19.3% to 21.7%) and fell in 10 (e.g. Sierra Leone
16.6% to 9.9°/0, Nigeria 9.4% to 7.4%). In about half the cases the changes were very small. Real
expenditures per child may have behaved very differently to this especially where significant devaluation
has occurred (Nigeria, Ghana, Sierra Leone) and where relationships between total government
expenditure and total government discretionary expenditure have varied, to name just two complicating
factors. Sahn (1992) concludes by suggesting that there is no compelling evidence that structural
adjustment of itself worsens provision of social services; rather deterioration is a chronic problem in
failing economies which are those most likely to require adjustment programmes. Noss (1991) also takes
the view that though there is correlation between adjustment programmes and reductions in public
expenditure - and thus social sector provision - causal relationships are poorly understood and analysis is
extremely complex.
Figure 1: Educational Expenditure Before and After Structural Adjustment