The impact of taxes on the use of deferred executive compensation

Aspen Gorry, Kevin Hassett, Glenn Hubbard, Aparna Mathur

19 October 2015



Executive compensation has grown dramatically since the 1980s. Along with this rise in total compensation, there have also been dramatic changes in the form of compensation for executives, with increasing reliance upon long-term bonus payments and stock options (Frydman and Saks 2010). Changes in the size of total executive compensation and the form that that compensation takes are of interest not only to the media but also to policymakers. In Gorry et al. (2015), we study whether and to what extent executive compensation responds to changes in marginal tax rates faced by the executive. In particular, while a large literature on the taxable income elasticity (summarised in Saez et al. 2012) studies whether reported taxable incomes respond to tax changes, we focus on whether executives choose to defer taxation on their compensation by choosing to receive compensation as stock options (and stock grants) that are only taxed at the time of realisation. Our empirical results suggest a significant impact of taxes on the composition of executive compensation, with higher tax rates generating an increase in deferral activity.

Tax treatment of deferred income

From a broad perspective, deferring income can generate tax benefits in a number of ways. First, when tax rates vary over time, a stock of deferred income will have an option value since income can be realised in years when tax rates fluctuate downward. Also, graduated income tax brackets provide an additional benefit, allowing workers to realise income in lower tax brackets if income has fluctuated downward. Finally, deferral allows individuals to earn returns on the pre-tax value of their savings.

We show that the tax benefits of deferral depend on the level of both individual, corporate, and capital gains tax rates and depend critically on the corporate deductibility of salary income. Following the analysis in Hall and Leibman (2000), we calculate the tax advantage of options compared to cash compensation. In the case where all compensation is deductible by the firm, the tax gain is generated from avoiding capital gains taxes. The gain is increasing in the capital gains tax rate and decreasing in the personal and corporate rates. Using US tax rates, we find that there is a modest tax benefit of deferral relative to cash of $3.67 per $100 of compensation.

However, in 1993, a US$1 million restriction was placed on the deductibility of executive compensation that is not incentive-based pay. Option payments and bonuses are excluded from the cap. This change dramatically increased the tax benefit to deferred income, making the static gain to deferral $51.86 per $100 of compensation. Moreover, deferred compensation delays when the firm can claim a deduction and potentially allows the firm to generate a larger deduction due to the $1 million limit on deductibility of non-incentive-based pay. That is, a capped deduction can be spread out over future years if income is deferred, to undo the effects of the cap.

With this change in the tax treatment of executive compensation, our study assesses the implications of taxes on the form of compensation. Previous work studied periods where the $1 million rule did not exist. For example, Hall and Liebman’s (2000) sample ends in 1994, only one year after Section 162(m) was enacted. This fact may explain why previous studies only found a limited impact of tax policy on compensation.

Empirical findings

We use data from Execucomp to study how the use of deferred income as a form of executive compensation was influenced by tax policy between 1992 and 2005. We define deferred income as the sum of option awards and restricted stock grants. Since both executive compensation and the use of stock options grew rapidly during our sample period, we study deferred income as a share of total compensation.

We find that higher taxes generate a significant increase in the share of options in executive compensation and a significant reduction in the share of restricted stock grants. This difference arises partly because restricted stock grants are not treated as incentive pay for tax purposes and so are subject to the $1 million rule on corporate deductibility. Moreover, restricted stock grants allow the executive to decide to be taxed immediately or when the stock vests, so it is unclear if taxation is actually deferred.

We find strong evidence that the use of stock options is responsive to the $1 million restriction on executive salaries that was enacted in 1993 (Section 162(m) of the Internal Revenue Code). Because options are classified as incentive pay and are therefore exempt from the rule, theory would predict that there is a tax advantage to taking pay in excess of $1 million in stock options. In line with the theory, we find that the share of income above one million dollars is a strong predictor of the use of options.

Our work contrasts with previous papers that do not find a strong relationship between taxes and the form of executive compensation. Hall and Liebman (2000) study the period from 1980-1994 and conclude that taxes have only a modest impact on the use of options, while changes in corporate governance such as the role of institutional investors and managerial incentives play a much larger role. Similarly, Frydman and Molloy (2014) study how tax policy affected the level of executive compensation between 1946 and 2005, with special focus on the years from 1946-1972. In particular, they look at how changes in labour income taxes influence the use of salary and bonus, stock options, and bonuses after retirement. Using ex ante versus ex post comparisons over tax changes, they find little effect of taxes on the level of compensation.

Implications for government revenue

To conclude, we evaluate the significance of our findings by calculating the effect of the increase in the use of options on the change in government revenue for an increase in the personal tax rate from 35 to 40%. A simple way to calculate the effect on government revenues is to use the static formulas developed in Hall and Leibman (2000). Again, the results vary dramatically depending on corporate deductibility.  Our estimates suggest that the change in revenue from the tax change would be 1.6% lower than a static score would suggest, assuming that all income is deductible. However, when cash compensation is not deductible, the change in the use of options as a result of the tax increase would imply that the government would lose 23.9% of the expected increase in revenue. We find that taxes have important implications for the use of stock options as a form of compensation and that in the presence of the $1 million restriction on corporate deductibility of executive compensation, this change can have economically significant implications for government revenue.


Saez, E, J Slemrod, and S H Giertz (2012), “The Elasticity of Taxable Income with Respect to Marginal Tax Rates: A Critical Review”, Journal of Economic Literature 50(1): 3-50.

Hall, B J, and J B Liebman (2000), “The Taxation of Executive Compensation”, in J Poterba (ed.) Tax Policy and the Economy 14, Cambridge, MA: MIT Press.

Frydman, C, and R Molloy (2011), “Does Tax Policy Affect Executive Compensation? Evidence from Postwar Tax Reforms”, Journal of Public Economics 95: 1425-37.

Frydman, C, and R Saks (2010), “Executive Compensation: A New View from a Long-Term Perspective, 1936-2005”, Review of Financial Studies 23: 2099-2138.



Topics:  Labour markets

Tags:  taxation, CEOs, CEO

Assistant Professor of Economics and Finance, Utah State University

State Farm James Q. Wilson Chair in American Politics and Culture and Director of Economic Policy Studies, American Enterprise Institute

Dean and Russell L. Carson Professor of Finance and Economics at the Graduate School of Business, Columbia University

Resident Scholar in Economic Policy, American Enterprise Institute