Introduction

A large number of studies of income distribution and development have shown that it is difficult to improve living standards of the poorest groups in LDCs, even in cases where aggregate growth has been rapid (see Bigsten, 1983; Sundrum, 1990; World Bank, 1990). To improve our understanding of the relationship between economic change and the incomes of the poor, analyses of poverty issues must be undertaken at a disaggregated level. In response to this need, new analytical tools have been developed while old ones have been refined. One of the most useful of the new tools is the Social Accounting Matrix (SAM), developed by Richard Stone and others in the Cambridge Growth Project. The SAM has increasingly been used for income distribution analyses in LDCs (see, e.g., Pyatt and Round, 1977). It can be applied in many ways from simple sectoral descriptions to incorporation in full-blown CGE models. The type of SAM analysis provided here represents a half-way house between partial analyses and a full-scale model.

In this chapter we use the Kenyan SAM of 1976 for income distribution analysis. Another SAM has been constructed for 1986, but unfortunately it is not a proper update of the SAM of 1976. In particular, it is lacking in detail with regard to information on household receipts and outlays. The household sector is in fact not disaggregated at all in the 1986 SAM. It is therefore not possible to repeat the analysis for 1976 for the latter year.