Tax authorities implement various strategies to combat tax evasion, tax audits being one of the most important enforcement tools. Previous literature has separately studied the effects of an increase in audit probability, and the impacts of becoming audited, on income and cost reporting behavior of firms and individuals, but their joint effects have obtained less attention. I utilize a natural experiment setting in a highly developed country and study the effects of a tax audit campaign, implemented by the Finnish Tax Administration, on self-reported income and cost items using firm-level quarterly VAT records.

The results suggest that an audit campaign does not necessarily increase voluntary compliance in general, even though finding and taxing hidden income may result in additional revenue for the state. Possible explanations include low prevalence of tax evasion, poor salience of the public announcement about the campaign, and insufficient incentives for firms to increase their compliance.

We analyze firm responses to risk-based tax audits – a central tool in regular tax enforcement – using full-population data on tax audits and tax returns in Finland. We find an immediate and persistent increase in reported profits by the audited firms after being audited compared to matched non-audited firms with a similar development in key outcomes before the audit. This is an indication of significant non-compliance in the baseline. We also examine the anatomy of non-compliance and find that both revenue and labor costs increase after audits, suggesting that some firms may follow a strategy of under-reporting their overall scale of operation. We use novel data on bankruptcy petitions and court decisions to investigate whether stricter tax enforcement has implications for real economic activity. We find a large increase in the likelihood of bankruptcy after audits among non-compliant firms, but no increase in bankruptcies for compliant firms.

Simplified tax regimes are designed to encourage businesses to formalize, yet their effect on tax evasion is not well-documented. These regimes allow firms to report less information to tax authorities, lowering the cost of concealing their income. The study utilizes data from electronic billing systems to calculate a proxy for operational expenses. Additionally, it examines the distribution of firms' income to assess the inconvenience, or "hassle," cost associated with adhering to the conventional tax regime instead of the simplified approach. By integrating operational and hassle cost estimates, we construct a proxy for total cost. When comparing firms with similar cost structures, those operating under the standard tax system report significantly less revenue—between 35% and 50% less—than those under the simplified regime, highlighting the simplified tax regime's potential to reduce tax evasion and alter revenue declaration behaviors.

In this paper, we study formal workers receiving part of their off-the-books, which we refer to as "payments under the table'' (PUT). We conducted the first extensive survey among Brazilian formal workers, finding that PUTs are sizeable and proportionally larger for higher-income workers, despite the widespread belief that third-party reported income exhibits low evasion. Back-of-the-envelope calculations suggest that the government loses 6.8% of income tax revenues due to PUT. We open the black box of how this evasion works in practice.

We combine unique data on labor lawsuits, matched employer-employee, and ownership registries with quasi-experimental variation to shed light on the underlying incentives to engage in collusive tax evasion. We find that employees respond by reducing PUT when benefits for reporting increase. Moreover, employers reduce PUT when perceived risks of being sued are higher.