Regional Welfare And Income Distribution In Kenya: A New Approach To The Measurement Of Wellbeing

Abstract

The income distribution of a region is a summary measure of many social processes and is directly associated with a region's level of welfare. This notwithstanding, regional welfare has been conventionally measured only with regard to a region's income. The primary objective of this study is to construct a theoretically consistent social welfare index that takes into account the level and distribution of income as well as the basic needs requirement of the population of a region. Using data obtained from a Kenyan household survey, we use a probit model to estimate parameters of an abbreviated social welfare function. Since welfare is not observable, it is proxied by a measurable variable, child survival at the household level. Regional child survival probabilities are computed and analyzed. The probit index for child survival turns out to be a good approximation to an abbreviated social welfare index at the household level. This index, once computed, can be used to compare levels of wellbeing enjoyed by households and regions, given a certain income level, the distribution of that income, and the extent to which basic needs in a region are being met. The study also investigates the pattern of income distribution across regions. We use data from an integrated labour force survey to estimate the gini coefficients. To complement this, we use the same data set to compute income shares of different ordinal groups by rural and urban areas. The results indicate that income inequality in Kenya is more pronounced in rural than in urban areas. For the rural areas, Rift Valley Province registers the highest and Coast, the in lowest inequality measure, while for the urban areas, Nairobi has the highest inequality measure. Child survival probabilities are estimated and reported for each of the eight regions. Results show that the urban areas have higher overall survival probabilities, than the rural areas. Further, Nairobi has the highest survival probabilities, and North Eastern, the lowest. Coincidentally, North Eastern also has the lowest income inequality, and Nairobi, the highest urban inequality. However, whenever these two indices are combined to measure welfare, Nairobi ranks the first on the welfare ladder while North Eastern ranks the last. The approach to the measurement of welfare developed in this thesis is new in the sense that the weights of its arguments are determined by household preferences and are consistent with utility theory. These weights are unique because they prevail only when the household welfare is maximum. The empirical results reported in this thesis should help policy makers gain insights into the patterns of regional income distribution in Kenya, and provide guidance on how regional welfare differences may be addressed. With this information, it is possible for the government to design policies that can be implemented to improve welfare of regions that have lagged behind for decades.