The aim of this study is to examine the effects of tax reforms and civil conflicts on tax performance in Burundi. The results from a regression analysis on a tax performance equation indicate that civil conflicts did not significantly affect total tax revenue, international trade taxes or income tax. One possible explanation for that finding is that, apart from the chaotic period 1993–1995, for the rest of the civil war period, conflicts affected mostly rural areas and had far less of an effect on the capital city Bujumbura, home to most industries and services. Consequently, after 1995, the civil war affected the agricultural sector to a much greater extent than the industrial and services sectors, which have the most taxpayers. However, civil conflicts have had significant negative effects on goods and services taxes. This is due to the negative effect of conflict on economic activity. The results of the effect on tax reforms suggest that tax reforms do not have a significant effect on total tax revenue or the tax categories. The reasons why tax reforms may not have had an effect on tax revenue performance include the prevalence of fiscal corruption, the negative effect of conflicts on the economy, abusive tax exemptions, and failure to focus on widening the tax base. Further results from the estimation of tax buoyancy and elasticity indicate that international trade tax is the strong point of the tax system in the short run, while tax on goods and services is the strong point in the long run. In addition, a high tax effort is estimated, which can be explained by the narrowness of the tax base comprised mainly of a few big taxpayers, coupled with a very large informal sector. However, it should be noted that while the tax system in Burundi is characterized by over exploited taxable capacities, it still relies on foreign aid and grants to finance much of its expenditure requirements. There is a need to rethink the implementation of tax reforms in Burundi to enhance their effectiveness.
