We study earnings management (EM) in private firms, where financial reporting mainly serves private communication with stakeholders rather than capital markets. While prior research documents incentives for EM, less is known about whether managers perceive it as socially undesirable, and whether this depends on the underlying motive. In a large-scale online survey experiment, we randomly assign managers to a direct question, an indirect question, or a list experiment to elicit self-reported EM. Around 20 percent report accrual-based EM, with similar rates under direct questioning and the list experiment, suggesting that EM is widely viewed as a routine practice. However, managers are less likely to admit opportunistic motives—such as obtaining better credit terms or influencing business partners—when asked directly. Disclosure of these motives rises when responses are masked in the list experiment. No such gap appears for tax-related motives. Overall, tax-motivated EM seems more socially acceptable, while opportunistic motives trigger under-reporting.

Corporate philanthropy has been increasing in Western democracies in recent decades, a rise often explained by the development of tax policies offering substantial incentives to donate to charities. Yet, corporate philanthropy is also increasingly perceived as a means to influence politics. In this paper, we estimate the tax price elasticity of corporate donations, and investigate how it differs depending on the recipients’ purposes and the donors’ characteristics. To do so, we use an exhaustive administrative panel dataset on firms’ tax returns in France from 2013 to 2022, including the identity of the charities that benefit from the donations, from which we recover their purpose using state-of-the-art NLP methods. Exploiting two reforms that affect the price of donations for firms, we document significant bunching around major regulatory thresholds. We then show that the tax-price elasticity varies significantly depending on the type of beneficiaries.

How do corporate tax systems shape firm boundaries? This paper shows that nonlinear corporate income taxation (CIT) can distort firms’ organizational structures by inducing tax‐motivated firm splitting. I use administrative data on corporations and their owners and exploit two reforms that altered the tax benefits and costs of dividing a firm into multiple entities. I find that a temporary increase in the tax advantage of splitting reduces the share of firms filing jointly for CIT purposes. Once the benefit is perceived as permanent and minimum capital requirements for new firms are abolished, the number of firms per entrepreneur rises significantly and persistently. Reorganizations are primarily driven by tax motives, as I find no effect on firms' total assets, employment, or industry diversification. These findings highlight extensive-margin responses of business organization to corporate taxation, with relevant implications for the understanding of firm dynamics and for tax design.

Zombie firms are firms that continue to operate despite prolonged periods of financial distress. When part of a group, such firms may serve as tax planning instruments by facilitating profit shifting or loss offsets. This paper studies the role of zombie firms in group-level tax planning by examining their response to DAC6, an EU directive mandating the disclosure of aggressive tax planning arrangements. Using firm-level data from ORBIS and a differences-in-differences design, I find that following DAC6, group zombie firms in the EU are more likely to exit the market than their non-EU counterparts. Among surviving firms, the probability of remaining a zombie declines for EU group firms. Heterogeneity analyses show that both firm location and group headquarters location shape these responses, highlighting how zombie firms are used within groups for tax planning purposes.