|
What happens if we put Ocasio-Cortez’s proposal to the test, incorporating behavioral responses in our revenue estimations? Using the Open Source Policy Center’s Tax-Calculator and Behavioral-Responses models developed at AEI, we did just that. A longer version of this discussion can be found here. But the simple answer is that a 70 percent tax on income over $10 million generates less revenue when accounting for a behavioral response greater than zero. Much less revenue, in fact.
Economists call these behavioral responses the elasticity of taxable income (ETI). The ETI captures the change in reported taxable income with respect to changing tax rates, and credible estimates have put the ETI between 0.12 and 0.4. Yet, there is evidence that it is much higher among high-income earners, since these individuals have more access to tax avoidance opportunities, either through sophisticated tax planning, timing decisions, or through the use of deductions and credits.
The long answer: When we apply a 70 percent tax on all ordinary income (wages, salaries, interest and business income) above $10 million, merely accounting for behavioral responses reduces the revenue gain of this proposal by 27.8 to 66.8 percent of that of the static estimation, depending upon the magnitude of the elasticity assumption.
Below we present the revenue generated from a tax on ordinary income under a static assumption (ETI equals 0), under an average ETI estimate of 0.25, and an ETI that is likely closer to capturing the true behavior of those at the very top of the income distribution.
Source: Authors’ estimations using OSPC’s Tax-Calculator version 0.24.0.
Our estimations, which are far lower than the $700 billion per decade published by the Washington Post, are in line with those of other economists. The Tax Foundation finds this proposal to generate $291 billion from 2019 to 2028 under a static model, while only $189 billion if accounting for an ETI of 0.25. In other words, there is 35 percent less revenue gained when incorporating a (conservative) behavioral response. This is right in line with our calculations that use the same ETI. Economists using the Penn Wharton Budget Model similarly underscore the major implications of behavioral responses to tax changes. They compare a static estimation of the revenue gain to a dynamic estimation arising from a specific income-shifting response: the reorganization of business owners (who make up a large share of those earning above $10 million) from pass-through businesses to C corporations. This incentive to decrease one’s tax liability arises since pass-through businesses would be subject to the 70 percent marginal rate, while C-Corporations would still be subject to the corporate income rate of 21 percent. A major conclusion is that such behavioral responses may decrease the revenue otherwise gained from this proposal by 57 percent. Again, while slightly different methodologies are used in these studies, the main point holds: behavioral responses to tax increases matter. A lot.
Several alternatives, such as a carbon tax, would generate more revenue than the proposed surtax on the wealthy, and could be used to fund programs that specifically benefit low-income households, such as the Earned Income Tax Credit program. Plus, a carbon tax relates directly to the Green New Deal’s fundamental goal of clean energy. When leveraging tax increases to fund any social welfare program, particularly tax increases on those with the largest incentives and means to alter their tax liability, it is essential to account for behavioral responses. Neglect of such responses may be a keen political design, but it is certainly not an advisable economic strategy.
What happens if we put Ocasio-Cortez’s proposal to the test, incorporating behavioral responses in our revenue estimations? The simple answer is that a 70 percent tax on income over $10 million generates less revenue when accounting for a behavioral response greater than zero. Much less revenue, in fact.
|
