Semi-autonomous revenue authorities (SARAs) have been at the centre of tax administration reform in Sub-Saharan Africa for the last 30 years. Nevertheless, the revenue effect of this reform remains highly debated (Ahlerup et al., 2015; Ebeke et al., 2016; Fjeldstad and Moore, 2009; Sarr, 2016; Von Haldenwang et al., 2014). This paper adds to the debate by controlling for the dynamics in tax revenue, which otherwise confound the effect of SARAs on tax revenue. Using a panel dataset of 46 countries over the period 1980-2012 and accounting for revenue dynamics, we show that, in contrast to previous findings, there is no robust evidence that SARAs have increased revenue performance in Sub-Saharan Africa. These findings are supported by an instrumental variable estimation which relies on donor influence. When broadening our scope, we fail to find any effect from SARAs on tax effort, revenue volatility and corruption. We, thus, conclude that there is little statistical support for a systematic relationship between semi-autonomous revenue authorities and tax capacity in Sub-Saharan-Africa.
semi-autonomous revenue authority Sub-Saharan Africa tax administration tax reform