Research

Do perceptions about government spending affect willingness to pay taxes? We test this hypothesis with a natural field experiment that focuses on the allocation of property taxes to public schools. Our results show that taxpayers often misperceive the destination of their tax dollars. By introducing shocks to households’ perceptions via an information-provision experiment, we find that perceptions of how tax dollars are used significantly affect the probability of filing a tax appeal. Moreover, the effects are consistent with reciprocal motivations: individuals are more willing to pay taxes if they believe that the government services funded by those taxes will provide greater personal benefit.

The canonical model of Allingham and Sandmo (1972) predicts that firms evade taxes by optimally trading off between the costs and benefits of evasion. However, there is no direct evidence that firms react to audits in this way. We conducted a large-scale field experiment in collaboration with a tax authority to address this question. We sent letters to 20,440 small- and medium-sized firms that collectively paid more than 200 million dollars in taxes per year. We find that providing information about audits significantly affected tax compliance but in a manner that was inconsistent with Allingham and Sandmo (1972).

Based on detailed administrative tax records, we implement a bunching design to explore how individual taxpayers respond to personal income taxation in Uruguay. We estimate a very modest elasticity of taxable income at the first kink point (0.06) driven by a combination of gross labour income and deduction responses. Taxpayers use personal deductions more intensively close to the kink point and underreport income to the tax authority. Our results suggest that the efficiency costs of taxation are not necessarily large in contexts characterised by limited deduction opportunities. Policy efforts should be directed at broadening the tax base and improving enforcement capacity.

Using an experimental-questionnaire method combined with randomized information treatments, this paper analyzes the drivers of individual inequality aversion. We elicit in- equality aversion by asking a sample of more than 1800 Uruguayan students to choose a society for a hypothetical grandchild. Participants make a sequence of choices between imagined societies characterized by varying levels of average income and income inequal- ity. In addition, we prime competing narratives regarding the sources of inequality in so- ciety. The main findings are that (1) the prevalence of inequality aversion is high: Most participants’ choices revealed inequality-averse preferences; (2) inequality aversion is in- creasing in the position of the hypothetical grandchild in the income distribution, like a normal good; (3) participants are more likely to accept inequality when it results from effort rather than luck independently of their grandchild’s position; (4) the effect of so- cial mobility on inequality aversion depends on the grandchild’s income position: Mobility opportunities reduce (increase) inequality aversion if participants expect their grandchild’s income to increase (fall). The latter result is consistent with the idea that mobility may impact the desire for more or less redistribution through rational expectation and risk aversion.

For some entities, the utility-maximizing evasion rate depends substantially on tax audit features, such as audit probabilities and penalty rates. Bergolo et al. (2017) document large misperceptions about these features. In this paper, we expand the analysis of survey data to explore potential sources of these misperceptions. Of all the channels that we explore, recent contact with audits best explains differences in misperceptions.

In this paper, we analyze how top income earners (TIEs) respond to changes in personal income taxation. Using an unprecedented combination of administrative records from the Tax and Social Security Agencies that covers most sources of personal income, we exploit a unique reform to Uruguay’s progressive labor income tax schedule that generated quasi-random variation in the marginal tax rates affecting labor income earners in the top 1% of the distribution. Using a difference-in-differences design, we estimate the elasticities on the intensive, extensive, and income shifting margins to changes in the labor income tax rates. Our preferred specification estimates an intensive margin elasticity of 0.577, which is partially explained by a real labor supply adjustment through fewer hours worked. Responses on the extensive margin are larger. Our preferred estimates indicate an extensive margin semi-elasticity of 2.479, which is mostly driven by shifts from the labor to the corporate income tax base (with a semi-elasticity of -1.967). Based on a simple model that allows individuals to choose between different tax bases, we estimate that the efficiency costs of the reform are, at most, 31.3% of the projected tax revenue.