ABSTRACT

This paper shows that governments can successfully affect the pass through of Value-Added Taxes (VATs) to prices via mandates. To do so, we use high-frequency retail scanner data from Argentina, along with a temporary 21 percentage point VAT cut on essential goods whose pass-through to prices was mandated by the government to be 100% for the VAT cut and no more than 33% for the VAT increase. We implement a difference-in-differences empirical design comparing goods that are subject to the VAT cut and/or to the pass-through mandates to those that are not. We find that, while compliance is not perfect, pass-through is substantially affected---often by a factor of two---by government mandates. Furthermore, and using additional variation, we show that pass-through mandates are four times more effective when prices are monitored by the government. Our findings imply that tax incidence is not an immutable and structural parameter but instead can be affected by governments.

Summary for a broader audience: FocoEcónomico

ABSTRACT

Tax amnesties are a controversial and widely-used policy tool to encourage evaders to disclose income or wealth in exchange for reduced penalties and no prosecution. We investigate the effectiveness of tax amnesties leveraging variation from Argentina, where left- and right-wing governments have implemented different schemes with varying success. For example, after the 2013-15 scheme revealed 0.5% of GDP in hidden assets, a new scheme in 2016 disclosed a remarkable 21% of GDP, dubbed the world's most successful tax amnesty. Using administrative tax data, we document the effects of these amnesties, their externalities, and the factors leading to their success (or failure).

ABSTRACT

We set up a framework to carry out field experiments---whereby units are grouped into mutually exclusive clusters---that allows for multiple treatments and general forms of intracluster correlation. We improve upon existing methods by allowing for cluster size heterogeneity, which is typically ignored when designing experiments. Based on data from several existing experiments, we show that ignoring cluster size heterogeneity can severely overestimate power and underestimate minimum detectable effects. We derive formulas for optimal group-level assignment probabilities as well as the power function used to calculate power, sample size, and minimum detectable effects for each treatment effect. We apply our results to the design of partial population experiments for estimating spillover effects and we run a large-scale randomized tax communication campaign in a municipality of Argentina to estimate total and neighborhood spillover effects on property tax compliance and e-billing sign up. We find evidence of spillover effects for clusters with a high proportion of treated units.

ABSTRACT

We study the implications of delegating tax collection duties on firms. We exploit a major reform to the withholding regime of the Turnover Tax in the City of Buenos Aires, where a substantial number of large firms were appointed to act as Collection Agents (CAs) based on a predetermined revenue threshold. Combining rich firm-to-firm administrative data along with quasi-experimental methods, we show that: (i) firms newly appointed as CAs do not experience any effects on their reported business activity; (ii) firms with pre-existing commercial ties to CAs increase their self-reported income; (iii) the government collects more tax revenue. Additionally, the analysis of a subsequent reform that reduced third-party collection of the tax shows that firms respond symmetrically by reducing their reported income. These results are in line with other papers suggesting that reforms to the tax administration can have a considerable impact when it comes to raising revenue and building tax capacity. In particular, our findings can provide guidance to other middle- and low-income countries on ways to determine who the right tax collector is as a function of the level of development.

Summary for a broader audience: WB Development Impact Blog

ABSTRACT

We exploit a large, quasi-randomized, 2.5-year-long income tax holiday to identify intertemporal labor responses of high-wage earners to net wage changes. In August 2013, the Argentine government exempted a group of wage earners from the income tax for 2.5 years while leaving in place the tax on other high-wage earners. Eligibility was based on whether past wage earnings were below a fixed threshold, thus levying sharply different marginal and average tax rates—effectively 0% for workers below the threshold. Using rich population-wide administrative data and a regression discontinuity design, we estimate a precise and very small wage earnings elasticity of 0.017 for this large, salient, and temporary income tax change. Responses are larger for more flexible outcomes (overtime hours) and for more elastic groups (job switchers and managers). We also find avoidance responses from new entrants who faced no tax if their first monthly wage was below the fixed threshold. This strategic entry below the threshold to dodge taxes required coordination with employers. Our findings suggest the presence of rigidities in the labor market, which imply that wage earners' responses to tax changes depend largely on substantial coordination with employers.

ABSTRACT

We explore whether the way in which tax credits are disbursed affects the gross wage of workers. We exploit an unusual reform in Argentina that shifted the disbursement responsibility of child benefits from employers to a government agency in staggered fashion, from 2003 to 2010. Using population-wide administrative data and an event-study approach based on firms' switching date set by the government, we show that the way tax credits are disbursed matters for the final economic incidence. Our evidence suggests that employers capture about 6-14 percent of the transfers through lower wages when they mediate the payments. We argue that in the firm-based system, transfers were likely understood as part of the starting compensation package and employers exploited this confusion to extract rents. Our findings therefore accord with the hypothesis that transfers are not entirely captured dollar for dollar by workers. More generally, this paper suggests that relying on firms as mediators in the tax-benefit system could have unintended consequences; as less salient schemes may lead to rent capture.

ABSTRACT

We disentangle the extent of imperfect competition in product and labor markets using plant-level data. We derive a formula for the ratio between markups and markdowns assuming cost-minimizing firms that face upward-sloping labor supply and downward-sloping product demand curves. We then separate this combined measure of market power by estimating firm-level labor supply elasticities instrumenting wages with a different set of instruments including the use of intermediate inputs, input price shocks, and TFP shocks. Our results suggest that both markets exhibit imperfect competition, but the variation is mainly driven by markups. We also estimate the relative gains of removing market power dispersion on allocative efficiency, finding that markups are more important on TFP than markdowns.

ABSTRACT

We estimate the response of firms and self-employed workers to two revenue taxes—Monotributo and Gross Receipts Tax—across the turnover distribution using rich administrative data from Argentina. We exploit several revenue-dependent discontinuities that provide incentives to underreport taxable income combined with a bunching design to estimate sales elasticities. We also explore heterogeneities by firm size, type of activities, and type of taxpayers. In the case of small firms, we find sizeable bunching below the thresholds that is stronger for higher tax incentives. The response is stronger in sectors that have more space for manipulation, such as service-based activities. In the case of medium and large firms, bunching is more muted but it suggests that even large firms are able to underreport their gross sales to avoid facing higher tax rates. Firms also seem to find more costly the indirect administrative cost of becoming a collection agent than the direct fiscal cost of the Gross Receipts Tax. We cannot rule out, however, that large firms adjust other margins (or taxes) to compensate for the higher tax pressure.

Two understudied areas in Public Finance are the role of tax professionals and the role of networks in tax administration and enforcement (Slemrod, 2018). In this project I seek to answer two questions: First, whether it is more cost-effective to communicate tax preparers or taxpayers to improve tax compliance; Second, whether there are spillover effects from targeted to non-targeted taxpayers that form part of the same network (those sharing the same accountant). To that end, I run a large scale randomized communication experiment where I send deterrence emails to taxpayers and/or accountants (about 100,000 taxpayers in the treatment group and 900,000 in the control group).

We study the relationship between tax filings and tax withholdings for the Argentine gross receipts tax on businesses and we shed light on the role of tax preparers in that relationship. Firms have to file a gross receipt tax return every month through a centralized online platform and can (i) claim a refund if the amount withheld in advance (third-party reported by suppliers and clients) exceeds the tax liability, or (ii) remit the difference if the tax liability is higher than withholdings. Using rich administrative data we first document sharp bunching exactly at the point where tax liabilities (reported by the firm) equal tax withholdings (third-party reported) and argue that this pattern is due to strategic under-reporting of income so that the tax balance matches the withholding shown by the online platform. Second, by linking firms to tax preparers through shared contact information we study the role of accountants on this behavior. We analyze how correlated are the behaviors of firms who share the same accountant (network). We find a positive relationship between accountants and the bunching behavior observed on their clients. We also exploit information on past audits and show that, within a network, non-audited firms respond to their peer's audit, suggesting that tax preparers might act as diffusers of information across firms.

We study the efficiency and distributional effects of price controls. Despite the controversy surrounding the effects of this regulation, surprisingly little has been done in practice to test the contrasting predictions. In this project, we aim to fill this gap by exploiting scanner data and quasi-experimental variation from price regulations introduced in the Argentine retail sector in 2014. We first study the effect of price controls on the supply of targeted products, spillovers on similar products, and explore how it differs by the level of concentration in the product's industry. Guided by these results, in the second part of the paper we estimate markups at the product level and explore whether price controls can affect the level of concentration in the industry. In the third part, we study the distributional and welfare effects of price controls.