Social Capital, Household Welfare and Poverty in Burkina Faso

This paper investigates empirically the importance of socialcapital, in the form of local associations and networks, forthe welfare of rural households in Burkina Faso. It draws ona unique database combining standard information on householdwelfare with multidimensional measures of social capital. Theanalysis finds that higher levels of social capital are associatedwith higher household per capita expenditures and better accessto credit. The distribution of social capital was found to bemore equal than that of other assets. Poor households and thosewho own little land obtain a higher return from social capitalthan other households.     

THE CONCEPTUAL BASIS OF MEASURES OF HOUSEHOLD WELFARE AND THEIR IMPLIED SURVEY DATA REQUIREMENTS*

This paper outlines a conceptual basis for the measurement and analysis of levels of welfare. It reflects the thinking that has been ongoing in the World Bank's Living Standards Measurement Study. Three alternative approaches to the measurement of welfare for the purpose of ranking households are surveyed, and the data requirements and analytical techniques for each highlighted. Various issues are discussed regarding the causal analysis of welfare levels and the changes in them. It is argued that the consideration of several dynamic aspects of welfare is significant for the identification of the poor and the potentially poor and for more accurate measurement of levels of living between socioeconomic groups.     

Correcting for Sampling Bias in the Measurement of Welfare and Poverty in the Cote d'Ivoire Living Standards Survey

The sampling aspects of a household data set are important toanalysts. The early years of the Côte d'Ivoire LivingStandards Survey (CILSS) had a sampling bias, which seriouslyaffected estimates of population statistics such as householdsize. The bias arose from sampling procedures that overrepresentedlarger dwellings. Assuming that samples drawn in later yearswere unbiased, a correction procedure is applied that uses weightsbased on household size. Results from the weighted data arethen compared with the unweighted findings to assess the seriousnessof the bias. Estimates of household expenditure per capita inthe early years of the survey are found to be significantlyunderestimated, resulting in an overestimation of poverty. Thesampling bias also resulted in an underestimation of the upwardtrend in poverty during 1985–88. The CILSS has been apopular and fruitful data set for policy analysis. These findings,however, cast doubt on the robustness of earlier work. Thus,the effort to trace sampling information is particularly worthwhilefor policy-oriented applied research.