The economic and fiscal impacts of special tax treatments in Value Added Tax (VAT) systems

Many countries rely on Value Added Tax (VAT) systems to raise revenues. On theoretical grounds, a VAT with a uniform rate and no exemptions is equivalent to a sales tax and is an efficient tax instrument. Moreover, from a tax administration perspective, the VAT can be more effective than a retail sales tax due to its self-enforcing properties along the supply chain. These properties are arguably the reason why the VAT has been widely adopted. It is rarely the case, however, that a VAT system features a uniform rate and no exemptions. There is typically a revenue-based threshold rule to exempt small firms from VAT registration. Moreover, governments frequently use VAT exemptions and reductions in VAT rates as policy instruments of redistribution, development, or patronage.

The attractive properties of the VAT may be severely weakened once governments grant special tax treatments. For instance, exemptions generate breaks in the VAT system that can affect the entire supply chain: e.g., they change the tax credits firms can get from exempt vs. non-exempt suppliers, incentivise self-supply, and reduce the self-enforcing property of the VAT. Such policies, therefore, can create distortions and have important indirect fiscal consequences beyond the direct forgone revenue. More generally, tax breaks for specific firms can distort competition and factor allocation with negative effects on growth.

Despite their prevalence, there is limited empirical evidence on the fiscal and economic effects of special tax treatments in VAT systems and policymakers have little guidance when evaluating their costs and benefits. The sparse evidence can be attributed to the difficulty in combining (i) the necessary policy variation to identify causal impacts and (ii) the necessary wealth of data to measure those impacts.

We propose to help filling this gap by exploiting a unique setting (São Paulo, Brazil) that will allow us to overcome this difficulty and estimate the economic and fiscal impacts of various special tax treatments in a VAT system. We will have access not only to anonymised administrative tax data on formal firms, but also to data on formal transactions between formal firms. As a result, we will be able to study the effects of, e.g., a substantial change in the VAT threshold on sales and value added for directly affected firms as well as for competitors, suppliers and buyers (and thus on firms’ networks).

The issue of special tax treatments is seen as a very important topic in Brazil, where states have granted a lot of tax exemptions over the years, which are viewed as one of the main factor behind the current dire situation of state budgets in the country. More generally, policymakers and policy advisers around the world have highlighted the lack of empirical evidence on the costs and benefits of special tax treatments for the VAT. This is relevant for all developing countries that, like Brazil, feature limited tax enforcement, rely extensively on the VAT for a significant share of their tax revenues, and grant many special tax treatments.