Performance of VAT Revenue Collection in Tanzania: An ARDL Approach on selected Macro-Economic Variables

Abstract

Value-Added Tax (VAT) is a cornerstone of revenue mobilization in Tanzania, yet its dynamic relationship with core macroeconomic drivers remains empirically underexplored. This study investigates the short- and long-run determinants of VAT revenue performance in Tanzania from 2006/07 to 2021/22. Utilizing an Autoregressive Distributed Lag (ARDL) bounds testing approach, we analyze the impact of nominal GDP and inflation on VAT collections. While the informal sector size and VAT taxpayer registration were initially considered, diagnostic tests revealed prohibitive multicollinearity, leading to their exclusion from the final model, an instructive finding that highlights critical data and structural challenges for tax policy. The results indicate a significant positive immediate effect of economic growth on VAT revenue. However, the long-run GDP coefficients are negative, suggesting a complex, lagged adjustment mechanism. Contrary to the traditional Tanzi effect, inflation shows a modest but persistent positive influence in the short run. The study concludes that Tanzanian VAT collections are predominantly influenced by current economic activity and exhibit strong internal persistence. We recommend policies focused on sustaining formal economic growth and improving tax base measurement to enhance VAT revenue predictability and performance.