Tax audits are a necessary component of the tax system, but policymakers and others have expressed concerns about their potentially adverse real effects. Understanding the causal effects of tax audits has been hampered by lack of data and because typically tax audits are not randomly assigned. We use administrative data from random tax audits of small businesses to examine the real effects of being subject to a tax audit. We find that audited firms are more likely to go out of business following the audit. The effect is concentrated in firms that underreport their taxes, although we find some evidence that the administrative costs of an audit also negatively affect firm survival. Among firms that continue as going concerns, we find evidence that audits have adverse effects on future revenues but no effect on future wages, employment, or investment. Finally, we find that tax audits have side benefits, causing firms to make changes to improve their tax efficiency.

Internal and external parties meaningfully shape corporate tax outcomes. However, we lack a holistic understanding of the major parties involved and their comparative effects. Using proprietary IRS data for public and private firms, we identify the top executives, corporate accountants, external accounting firms, and individual tax preparers and examine the comparative importance of these parties on corporate tax outcomes. We find that external individual tax preparers matter much more than the accounting firms that employ them. Internal actors (top accountants and executives) explain more of the variation in corporate tax outcomes than external actors (individual tax preparers and accounting firms). We also find some evidence that individuals’ characteristics are associated with the tax behavior of the corporations they serve. Overall, we conclude that some of the actors who are unobservable in public data play a greater role in corporate tax outcomes than parties that are a focus of prior research.