One of the functions that people routinely expect governments to perform is to reduce inequality
and poverty. This goal sits somewhat uncomfortably beside the more traditional concerns among
economists for economic efficiency, including the provision of public goods. But it is important
politically and socially, perhaps more so than issues of economic efficiency. Even the most neo-classical
policymaker must heed a policy’s consequences for the poor.
In Africa, a generation of new, nationally representative household surveys have shown that the
distribution of resources is surprisingly unequal. While the Kuznets hypothesis would suggest that
Africa’s relatively poor economies would have less inequality than middle-income ones, many African
Economies are in fact among the most unequal in the world (Table 1). At a first glance then, the need for
equalizing policies appears very important on the continent. Conceptually, government could achieve
this either with progressive taxation (i.e. taxes that fall disproportionately on the rich) or with progressive
expenditures (i.e. programs or services that go disproportionately to the poor). In this paper, we describe
the extent to which different kinds of taxes and expenditures succeed in transferring resources to the poor
in Africa. While we consider a broad range of taxes and expenditures, the list is far from comprehensive.
On the tax side, because our data come from household surveys, we cannot say anything about
corporation taxation, and our results on some important types of taxes, most notably import duties,
depend on strong assumptions. For expenditures, we are even more limited. By the very nature of public
goods such as defense, public order, and the judiciary, it is impossible to identify their beneficiaries, so
we cannot comment on large parts of the budget. Further, the transfer payments schemes that account for
much of the government’s redistributive policies in richer economies are almost nonexistent in Africa.
Many of the benefits of social services, especially health and education, however, accrue directly to
individuals and thus are identifiable. Fortunately, these are also the expenditures that people most
commonly expect to have a redistributive impact in Africa and they are generally covered in household
surveys. Thus, our coverage of expenditure incidence will concentrate on benefits of publicly provided
health and education services.