Value added taxes increase a country’s revenue

The remarkable rise of the VAT has attracted relatively little analytical or empirical attention. This paper aims to help fill this gap, exploring both the causes and consequences of the spread of VAT. Using a panel of 143 countries observed over 25 years, we estimate a recursive two-equation system characterizing both the decision to adopt a VAT and the impact of the VAT on revenue once adopted. The latter is of interest not only in itself but also, it has been shown, as an indicator of whether or not the VAT has proved an especially effective form of taxation. There emerges a rich set of determinants of VAT adoption. Income per capita, interestingly, is not amongst them: the VAT is no longer the preserve of high income countries. A country is less likely to adopt a VAT, however, the larger is its agricultural sector (this being a hard-to-tax group, and perhaps harder under the VAT than under other taxes), and the lower is its past revenue ratio (suggesting a greater need to enhance revenues). The IMF also appears to have played an independent and significant role in the spread of the VAT: participation in a non-crisis IMF program in itself significantly increases the likelihood of adopting a VAT. Yardstick competition also seems to have been significant, with a country more likely to adopt a VAT the greater the proportion of others in the region who have already done so. Most strikingly, and intriguingly, the results indicate strongly that, all else equal, more open economies are less likely to adopt a VAT. A central question in assessing the consequences of the VAT is simply: Has it enabled countries to increase their tax revenues? This is important both in itself–put another way, it relates to the truth of the presumption underlying the ‘money machine’ concerns raised prominently in recent discussions of possible VAT adoption in the United States–and because, in the simple framework analyzed above, such a revenue gain proves to be necessary and sufficient for the VAT to have reduced the marginal cost of public funds. The simplest empirical framing of this question gives a clear-cut answer: a dummy for the presence of a VAT, added to a standard tax effort equation, proves to be significantly positive but modest in size: adoption of the VAT is associated with a long run increase in the overall revenue-to-GDP ratio of about 4.5%. Allowing the impact of the VAT to vary with country circumstances, however, a more complex picture emerges. The shift effect of the VAT now becomes negative, but acting in the opposite direction are gains that tend to be greater in higher income and in more open economies. The former may plausibly reflect a greater ability to administer and comply with the tax. The latter seems likely to reflect the relative ease with which VAT can be collected on imports (while the lower likelihood of adoption in such circumstances, mentioned above, may reflect the ease of then relying on tariff revenue rather than undertaking the fundamental modernization of the tax administration that is often needed to implement a VAT). These opposing effects imply that while some countries may have gained (or would gain) revenue from adopting the VAT, others have (or would) not. This ambiguity has been explored by calculating predicted revenue effects for each country in the sample (Table 3). While these calculations can be no more than suggestive, the impression that emerges is that adoption has increased revenue and improved effectiveness in the large majority of countries that have implemented a VAT, and would do so too for most of those that have not. Only in Sub-Saharan Africa is the weight of evidence mixed. What this suggests, however, may be not that the VAT in these countries is intrinsically inappropriate. It could be that the VAT there has significantly improved the overall efficiency of the tax system, but with a greater degree of offset through reduced use of other taxes (notably, perhaps, tariffs). It does suggest, however–given clear signs of associated revenue and effectiveness gains in some countries of sub-Saharan Africa–that significant work remains to be done in improving the functioning of the VAT in much of the region.