Determinants of value added tax revenue in Kenya
Wawire, N. H. W.
MetadataAfficher la notice complète
Past studies that have been undertaken on the responsiveness of tax revenues to changes in GDP in Kenya have found a positive relationship between tax revenues and GDP. However, the studies omit some key determinants of tax revenues, such as the nature of the tax system and institutional, demographic and structural features of the economy. Due to this omission, the estimated income elasticities of Value Added (VAT) revenues are unreliable for planning purposes, a situation that might be responsible for the recurring budget deficits. The specific objectives of this study were to establish the determinants of VAT revenue and assess the response of VAT structure to changes in the in its tax bases. The study is important because its results can be used to design pro-growth tax policies and implement tax changes that are equity enhancing. The paper uses Paul Samuelson's (1955) fundamental general equilibrium analysis of the public sector to derive its main results. In the framework, the demand function for the public good was derived from a constrained model of utility-maximization. In the same vein, tax revenues were taken as functions of household incomes, which paved the way for the estimation of Engel curves for public goods. The study finds that growth elasticities for VAT are all greater than one. The estimation results show that total GDP elasticity of VAT revenues is less than the elasticities with respect to monetary GDP, suggesting the existence of an underground economy in Kenya over the period of analysis. It is found that VAT revenues respond with substantial lags to changes in its determinants and that VAT revenues are sensitive to unusual circumstances. The study concludes that Kenya’s VAT revenue is very responsive to changes in their determinants especially international trade. There is therefore the challenge of creating a stable VAT system so that tax revenues can increase rapidly as the economy grows.