Criminal Penalties on Officers Can Deter Corporate Tax Avoidance

There is a worldwide debate over how to prevent overly aggressive corporate tax avoidance. In 2015, the Australian government doubled civil penalties on large multinational corporations that engage in the practice. Similarly, recent legislation in the United States, like the Stop Tax Haven Abuse Act, would increase penalties only on corporations, while the UK treasury also plans to increase corporate penalties. However, the former UK Liberal Democrat chief secretary, Danny Alexander, and others suggest that the treasury should consider criminal penalties on corporate officers as well.

Our forthcoming study empirically examines whether high penalties deter corporate tax avoidance, a practice that can reduce social welfare and place a disproportionate tax burden on individual tax payers. In the U.S., criminal penalties vary by state and range from prison sentences of one year (e.g., Florida) to a sentence of 20 years (e.g., New York). Civil penalties, on the other hand, are levied on corporations as a percentage of the amount of tax underpayment. Some states’ civil penalties could be as high as 200 percent of the underpayment amount (e.g., Massachusetts), while other states have a maximum penalty of 25 percent or less (e.g., Connecticut).

We find that the extent of state-level penalties on officers affects the degree of deterrence for state tax avoidance, while the extent of penalties on corporations and their shareholders does not. Specifically, when the maximum possible length of prison sentence for officers in a state is longer, a firm avoids a lower amount of state corporate income taxes. Such maximum prison sentence is intended to reflect not only the perceived threat of penalty but also the state’s attitude toward imposing penalties directly on the officer. Similarly, we find that corporations avoid a lower amount of taxes in states where the maximum penalty for overly aggressive tax reporting is a felony rather than a misdemeanor. These findings are consistent with our expectations, because corporate officers rather than shareholders are the direct decision makers in corporate tax reporting. They are consistent with the agency theory that officers’ overly aggressive tax reporting decisions do not incorporate the extent of penalties imposed on shareholders.

We further use the increase in the penalty on officers in New York in 2009 as a plausibly exogenous event to test the impact of legal penalties on state tax avoidance. In 2009, New York increased the penalty on an officer for overly aggressive corporate tax reporting from four years to 25 years in prison. Using a difference-in-differences research design, we find no difference in state effective tax rates between firms with at least one subsidiary in New York and other firms prior to 2009. However, after 2009, the state effective tax rates of New York subsidiary firms are significantly greater. Second, we identify a sample of firms that do not have subsidiaries in New York prior to 2009 but moved into New York after 2009. These firms do not have different tax avoidance behavior before 2009. Also, we do not find evidence that a firm’s tax avoidance behavior predicts its decision to move into New York after 2009.

From a policy perspective, the implication is that if the costs of implementing the two penalties are equal, authorities should favor criminal penalties on officers over civil penalties on the corporation. In other words, both federal and state governments should impose greater legal penalties on corporate officers for aggressive corporate tax avoidance.

This post comes to us from professors Mark (Shuai) Ma at the University of Pittsburgh’s Katz Graduate School of Business and Wayne B. Thomas at the University of Oklahoma’s Price College of Business. It is based on their recent paper, “Legal Environment and Corporate Tax Avoidance: Evidence from State Tax Codes,” forthcoming in The Journal of the American Taxation Association and available here.

Leave a Reply

Your email address will not be published. Required fields are marked *