How Much Should We Tax CEOs?
August 17, 2016
Many politicians have called for raising tax rates on top earners to combat income inequality. Both Sen. Bernie Sanders (I-VT) and Democratic presidential nominee Hillary Clinton have in particular expressed concerns about executive compensation, repeating a commonly cited but misleading statistic about CEO pay. This poses an interesting question: How much should we tax CEOs? In a new American Economic Review paper, Carnegie Mellon economists find that depending on a variety of assumptions, the optimal tax rate on CEOs varies from as low as 13.4% to as high as 40.2%. These results are substantially smaller than what has previously been reported in the optimal tax literature, and suggest that increasing taxes on corporate executives may be unwise.
When economists discuss “optimal taxation,” they are referring to taxes that raise revenues while reducing economic distortions in the economy to a minimum. Taxes generate “deadweight losses,” which are productive economic activities that would have occurred had the tax not be imposed. Some economists, such as Peter Diamond and Emmanuel Saez, have found small deadweight losses from taxing top earners, such as CEOs, and suggest that their theoretical model implies that the optimal tax rate on these earners could be as high as 73%. The current federal marginal income tax rate for top earners is 39.6%, which approaches the mid-40s when accounting for state marginal income tax rates.
However, these Carnegie Mellon economists explain that the Diamond-Saez model ignores how talented CEOs generate positive spillovers for others. For instance, an exceptionally talented CEO will increase the profits of the firm, some of which will be distributed to shareholders and some of which will be distributed to workers. Of course, it is also possible that corporate executives engage in rent-seeking. They may exert their bargaining power to extract resources from the firm to the detriment of workers and shareholders. The authors argue that their assumption of positive spillovers is more consistent with the economic literature than this Piketty-Saez model, which assumes CEOs generate only negative spillovers.
Executive compensation is not only a factor of managerial talent, but also the size of the firm. The authors assume that some firms are larger than others for reasons other than the talent of its managers. Some firms for example may require a large pool of assets in order to provide their products or services. The size of the firm will therefore partly determine the structure of compensation packages that are offered to corporate executives, which can include salaries, bonuses, and stock options.
The authors restrict their analysis to CEOs with reported incomes greater than $500,000. They find that depending on the “marginal social value” the government assigns to firm profits, the optimal tax rate on CE0s varies considerably. According to the model, a government that is moderately concerned about firm profits because of the positive spillovers should set marginal tax rates between 30.1% and 40.2%, roughly within the range of where current rates are for top earners. While further research is needed in this area, this paper provides convincing evidence that we should be cautious about raising tax rates on corporate executives beyond current levels.