Insights / Research BriefFeb 05, 2021

Rethinking How We Score Capital Gains Tax Reform

Based on BFI Working Paper 2021-10, “Rethinking How We Score Capital Gains Tax Reform,” by Natasha Sarin, University of Pennsylvania; Lawrence H. Summers, Harvard Kennedy School of Government; Owen M. Zidar, Princeton University; and Eric Zwick, Associate Professor of Finance and Fama Faculty Fellow, Chicago Booth
Key Takeaways
  • Capital gains tax rates have varied over the years and are currently at 20 percent, roughly half of the top income tax rate.
  • Debate over capital gains tax rates has focused on how they affect entrepreneurship and capital formation, and whether such benefits outweigh the equity and fiscal costs of lower rates.
  • While tax scorekeepers disagree on an optimal rate, this work suggests that capital gains tax rates could approach 40 percent, which would add over $1 trillion to US Treasury coffers.
Capital gains tax rates have varied over the years, from well over 30 percent in the 1970s to 15 percent in the early 2000s to 20 percent today, which is roughly half the highest rate on taxable income. The reason capital gains tax rates have varied so much is because of the intense political debate surrounding their impact: Some maintain that low rates encourage entrepreneurship and capital formation, while others doubt that such benefits outweigh the equity and fiscal costs of lower rates.

How can such an important tax rate vary so widely over time? For those with a more technocratic view of the world, like economists and other analysts, this is a quantifiable problem. Adjusting the rate up or down should have measurable effects that, at least theoretically, would account for costs and benefits and point to an optimal rate or range. Such “scorekeeping” does occur and is conducted by the Joint Committee on Taxation (JCT), a nonpartisan, independent committee of the US Congress that “interacts with members of Congress, members of the tax-writing committees, and their staff on a confidential basis and enjoys a high-level of trust from both sides of the political aisle and in both houses of Congress.”1

The JCT doesn’t actually set rates, that is the job of Congress, which can take or leave JCT’s scorekeeping. However, questions have been raised over the years about the nature of JCT’s scorekeeping and whether the committee’s models need updating. In particular, in “Rethinking How We Score Capital Gains Tax Reform,” Natasha Sarin, Lawrence H. Summers, Owen M. Zidar, and the UChicago’s Eric Zwick challenge current assumptions about the revenue potential of raising capital gains tax rates. Indeed, at a time of increasing federal budget deficits, the authors argue that raising such rates could deliver substantial tax revenues to the US Treasury.

Chart
Figure 1: Evolution of Net Capital Gains and C-Corporation Equity Wealth
Notes: These graphs plot net realized capital gains, C-corporation equity wealth, and the maximum long-term capital gains tax rates between 1960 and 2014 as a share of GDP. Data on capital gains and C-corporation wealth from Smith, Zidar and Zwick (2020). Data on tax rates from the U.S. Treasury (https://www.treasury.gov/resource-center/tax-policy/ tax-analysis/Documents/Taxes-Paid-on-Capital-Gains-for-Returns-with-Positive-Net-Capital-Gains.pdf) and the Tax Foundation (https://taxfoundation.org/federal-capital-gains-tax-collections-1954-2009/). GDP data from the World Bank (https://data.worldbank.org/indicator/NY.GDP.MKTP.CD).

How far can you “stretch” a tax?

Intuition tells us that there is a tax rate above which most taxpayers will cease to pay—not through illegal means but by foregoing the activity that generates the revenue. By way of simple example, let’s say a tax of 10 percent on an activity generates $1 million in revenues. You may assume, then, that increasing the tax rate to 11 percent should add one-tenth in revenues, or $1.1 million in total. However, in this case the higher tax rate drives off enough taxpayers that only $50,000 in extra revenue is gained. Following this logic, it is easy to see that a tax rate at too high a rate could reduce economic activity so much that total revenue could actually dip below $1 million. In economic parlance, this taxable income is described as elastic because it responds to changes in tax rates. Clearly, assumptions about elasticity rates matter a great deal when crafting tax policy.

Conventional wisdom among a number of tax scorekeepers holds that the revenue-maximizing capital gains tax rate is 30 percent, or 50 percent higher than today. This means that setting a rate too far below 30 percent leaves tax revenues on the table, while taxing too far above 30 percent could reduce total collected revenue.

Regarding long-term capital gains taxes, the conventional wisdom among a number of tax scorekeepers holds that the revenue-maximizing capital gains tax rate is higher than 20 percent, although these scorekeepers do not agree on a precise number. Regardless, the point here is that setting a rate too far below the optimal rate leaves tax revenues on the table, while taxing too far above that rate could reduce total collected revenue or raise little revenue at significant efficiency cost. 

On the one hand, the dynamic effects of capital gains are more pronounced than for other policies, as retiming a capital gain realization in an investor’s stock portfolio is easier than changing investment strategy for executives seeking to avoid a corporate tax increase or reducing labor supply for workers when income tax rates rise. People, in other words, are not required to take a capital gain at a given time—they can hold it or divert it to, say, a charity or tax-favored bequest to avoid taxation. 

On the other hand, the authors posit that these behavioral effects are overstated, resulting in a potentially severe underestimate of the revenue at play from capital gains tax increases. They offer three reasons to suggest that elasticities on capital gains taxes are overstated, which highlight the need for more scorekeeping transparency:

  • First, most existing research on capital gains taxes is focused on short-term effects of rate changes; that is, what will taxpayers do, say, in the year following a change in rates. The new rates, of course, persist over time, and the current short-run focus misses the medium- and long-run revenue capture that would occur in years two, three, four, and so on. 
  • Second, revenue estimates may understate the substitution between capital gains and other forms of income, such as dividends or wages for finance professionals and executives. 
  • And third, the uncertainty about how scorekeepers incorporate dynamic effects into various capital gains reform proposals complicates apples-to-apples evaluation of alternatives. 
  • Broadly speaking, then, while people certainly can, and do, defer capital gains to avoid taxation, they eventually realize far more of those gains than scorekeepers assume. This raises the question of how many more revenues could be gained by raising the capital gains tax rates to, say, the level of income taxes—or roughly doubling the rate from 20 to 40 percent. The authors calculate that such a change could raise over $1 trillion dollars over time. 

Importantly, while advocating for more transparency in capital gains tax scorekeeping, the authors offer one important caveat: Transparency is a double-edged sword. Given the great interest in taxation levels, by legislators and the public, making the assumptions underlying scorekeepers’ estimation publicly available will invite greater lobbying around those assumptions by supporters and critics of different reforms. Cracking the “confidential basis” that exists between the JCT and Congress may challenge the “high-level of trust from both sides of the political aisle and in both houses of Congress.” 

Conclusion

This work challenges the assumption that substantially raising capital gains tax rates would have a negative effect on tax revenues. Current elasticity assumptions miss the mark and, likewise, raising capital gains tax rates would result in significant additional tax revenues.

CLOSING TAKEAWAY
While people certainly can, and do, defer capital gains to avoid taxation, they eventually realize far more of those gains than scorekeepers assume. This raises the question of how many more revenues could be gained by raising the capital gains tax rates to, say, the level of income taxes—or roughly doubling the rate from 20 to 40 percent? The authors calculate that such a change could raise over $1 trillion over time.

However, beyond this specific policy insight, this paper argues for more transparency in how tax rates are determined. In doing so, the authors question whether current scorekeeping makes use of up-to-date economic research. Applying the latest research and doing so more transparently will result in a capital gains tax rate that will more accurately respond to taxpayer behavior and return optimal revenues. 

Finally, the authors clearly state that they have no animus with JCT or other tax scorekeepers. Indeed, they stress their enormous respect and admiration for the integrity and seriousness of these tax professionals. Rather, the goal of this work is to advance understanding of taxpayer behavior and the revenue potential of capital gains (and other) tax reform efforts to inform the policymaking process. 


1https://www.jct.gov/about-us/overview/. Other groups engage in their own tax policy scoring, including the Tax Policy Center, Tax Foundation, and the Penn Wharton Budget Model.