General Government Final Consumption Expenditure and Nominal GDP Per Capita in Five East African Countries: A Panel Association Analysis
Wycliffe Mugun *
Department of Economics, Kaimosi Friends University, Kaimosi, Kenya.
*Author to whom correspondence should be addressed.
Abstract
This study examines within-country associations between general government final consumption expenditure and nominal gross domestic product (GDP) per capita in Kenya, Uganda, Tanzania, Burundi, and Rwanda over 2005–2024 using annual World Development Indicators data. GDP per capita in current US dollars is the outcome, and general government final consumption expenditure as a percentage of GDP is the principal explanatory variable; the latter measures government current consumption rather than total public expenditure. Foreign direct investment, population growth, inflation, the official exchange rate, and official development assistance are included as controls. Descriptive and correlation analyses use 100 country-year observations, whereas the log specification is estimated on 84 observations because natural logarithms are undefined for nonpositive values. Country fixed effects are emphasised because persistent country-specific heterogeneity is plausibly correlated with fiscal and macroeconomic covariates. The Hausman comparison is reported, but its non-positive-definite covariance-difference warning means it is not treated as a decisive model-selection test. In the conventional fixed-effects estimates, the coefficient on government final consumption is -0.5811; foreign direct investment and the official exchange rate have positive coefficients, population growth and official development assistance have negative coefficients, and the inflation coefficient is close to zero. These estimates use conventional covariance estimates. Heteroskedasticity is detected, the panel contains only five countries, some variables display nonstationary behaviour, and supplementary cointegration output does not establish a long-run equilibrium relationship. The results are therefore interpreted as exploratory within-sample associations rather than causal, structural, or long-run effects. Particular caution is required because the outcome is nominal and US-dollar denominated and because the fiscal regressor is expressed as a share of GDP.
Keywords: GDP per capita, government final consumption expenditure, panel data, East Africa, fixed effects, fiscal policy