Disaggregated income components and welfare analysis among households in Uganda

Abstract

Welfare is a long-time key goal for most economies in formulation and implementation of policies. Evidence reveals tremendous decline in welfare in Sub-Saharan Africa manifested by low consumption expenditure and high poverty levels. This study’s general objective was to analyze the effect of household income on welfare among households across districts in Uganda. Specifically, the study sought to establish the effect of farm income, non-farm income, remittance income and diversified income on household welfare in Uganda. The study was grounded on Pareto Optimality theory to test hypothesized relationship between household income and welfare. A positivistic research orientation was used. The study adopted an explanatory research design. Panel data of Uganda National Panel Survey for the period 2013 to 2020 was used. Secondary data obtained from Uganda Bureau of Statistics for 128 districts was analyzed using static panel, Logit and Probit models. Household consumption expenditure and household poverty status were used as proxies for household welfare in analysis. The study employed the instrumental variables fixed effects and random effects as well as instrumental variables probit estimation techniques in addressing endogeneity bias. The study tested for unit roots (Fisher type test), normality (Jaque Bera test), multicollinearity (VIF), FE-RE choice (Hausman test) and Endogeneity (Wald test). Results from the study confirmed that household income could assist to explain household welfare. Using consumption expenditure as a measure of household welfare, results indicated that coefficient estimates of farm income (𝛽 = 0.014527, 𝑝 = 0.082) was not significant at 5% level while non-farm income (𝛽 = 0.262181, 𝑝 = 0.000); remittance income (𝛽 = 0.011652, 𝑝 = 0.000) and diversified income (𝛽 = 0.071074, 𝑝 = 0.000) had a significant and positive effect at 5% level on household welfare in Uganda. Utilizing poverty status as a measure of household welfare, results indicated that coefficient estimates of farm income(𝛽 = −0.021325, 𝑝 = 0.076) was not significant at 5% level while non-farm income(𝛽 = −0.246568, 𝑝 = 0.000), remittance income (𝛽 = −0.016760, 𝑝 = 0.000) and diversified income (𝛽 = −0.200253, 𝑝 = 0.000) had significant and positive effect at 5% level on household welfare in Uganda. The study concluded that non-farm income, remittance income and diversified income are key determinants influencing household welfare. In addition, the study came to the conclusion that, of all the different types of income of the household, non-farm income had the most significant marginal influence on household welfare. In order to improve the welfare of households, it is necessary for both households and the government to develop methods that increase non-farm, remittance, and diversified income. Income from non-farm sources, income from remittances, and income from diverse sources should be the primary focus of households, governments, policymakers, and other stakeholders who are interested in alleviating poverty. The cuurrent study concentrated on measuring the welfare impact over a relatively short period of time, whereas future studies should focus on estimating the welfare effects over longer periods of time using more advanced models such as ARDL and GMM.