Capital gains taxes are a perennial issue in US tax-reform debates. Some people maintain that preferential rates on capital gains encourage entrepreneurship and capital formation, while others question whether these benefits are worth the costs.
What are those costs, exactly? It’s clear in terms of direct fairness costs: the wealthiest 1 percent of US households accounted for two-thirds of capital gains realizations in the Federal Reserve’s 2019 Survey of Consumer Finances. However, the fiscal costs, which are estimated by the Joint Committee on Taxation, are far less clear. In the parlance of policy makers, the JCT is considered the official “scorekeeper” that decides how tax legislation “scores” if implemented. The prevailing wisdom in the taxation-scorekeeping community appears to be that the revenue-maximizing rate for capital gains is about 30 percent, which is well below both current top marginal tax rates on other income and top rates currently under debate. But in a simple exercise, we estimate that increasing capital gains rates to match the ordinary income level could raise more than $1 trillion over a decade. This illustrates the need to rethink scorekeeping in the debate.
The prototypical example of a capital gain is a share of corporate stock. An individual who bought an $18 share of Amazon when it went public could sell that share today and pay taxes on more than $3,100 of appreciation.
If the revenue-maximizing rate is 30 percent, setting a rate too far above this level will actually reduce the total amount of revenue collected, as the gains expected will fail to materialize because the dynamic response of taxpayers will dramatically shrink the tax base.
Such a response could take the form of an investor retiming a stock sale to avoid realizing a capital gain event. This certainly happens, but we suspect that in most instances the investor doesn’t avoid paying taxes on that gain entirely, just immediately. The tax is simply postponed, in which case these behavioral effects are overstated, resulting in a potentially severe underestimate of the revenue at play.
The current realization elasticity used by the JCT and others in the scorekeeping community is approximately -0.7, determined on the basis of both historical scores and more recent academic research. A crude application of this elasticity implies that if capital gains tax rates doubled (to match top ordinary income levels), only 53 percent of gains would be realized. In concrete terms, roughly $1 trillion of annual realizations would shrink to around $500 billion, and the rate hike would be scored as raising no new revenue.
However, this style of calculation neglects a material offsetting factor: medium-term retiming of realizations would offset lost revenues in the short run. Suppose that doubling capital gains rates from 20 percent to 40 percent causes realizations to occur half as often: instead of realizing gains every year, individuals realize them every two years. If assets grow at 10 percent annually, then in the low-tax regime, $100 of assets yields realizations of $10 in Year 1 and $10.80 in Year 2 (after the individual pays $2 of tax in Year 1). In the high-tax regime, $100 of assets yields realizations of $0 in Year 1 and $21 in Year 2. Despite the appearance in Year 1 of a large elasticity of realizations in response to the tax increase, total revenues over both years increase from $4.16 in the low-tax regime to $8.40 in the high-tax regime. In this simple example, without other behavioral responses, the short-term revenue score is zero, and the medium-term revenue score is double the baseline. Clearly, the latter score is more relevant for policy purposes.
Reform capital gains taxation, and you also reduce wasteful effort by taxpayers and their planners to devote resources to circumventing tax liabilities by exploiting preferential capital gains rates and sheltering opportunities.
There are also other indications that conventional elasticities may be overstated. For example, the composition of capital gains has shifted in recent years, so that nearly half of capital gains now accrue through pass-through and mutual-fund distributions outside of the direct control of taxpayers. It is unclear whether scorekeeping models account for compositional changes and dynamic weights.
This matters because it’s one thing to time a stock sale, but it’s harder to time pass-through gains, such as those produced by the growth of carried-interest compensation offered to general partners of hedge funds and venture-capital and private-equity firms. Partnership agreements typically require funds to be returned within 10–12 years of the initial commitment. Investors in these structures cannot time realization decisions around favorable tax environments, nor can they typically defer their gains indefinitely. It seems plausible that 30–50 percent of capital gains cannot be easily timed in response to tax changes, and if that’s the case, no matter how large the easily timed elasticity is, doubling rates to top ordinary income levels will still raise substantial revenues.
Consider, also, a few examples of how changes in the capital gains tax might affect other tax bases. Preferential tax treatment encourages avoidance in the form of misclassification of wage income for fund managers through the carried-interest loophole. Similarly, the tax code favors employee stock options, which, when held for long enough, qualify for capital gains taxation. The different treatment of capital gains and dividends affects the relative attractiveness of distributing corporate profits via share buybacks versus dividends. And capital gains tax preferences can affect the allocation of capital across industries and locations, due to sheltering opportunities such as like-kind exchanges in real estate and oil and gas, investments in Opportunity Zones, and incomplete recapture of depreciation deductions following asset sales. Reform capital gains taxation, and you thus also reduce wasteful effort by taxpayers and their planners to devote resources to circumventing tax liabilities by exploiting preferential capital gains rates and sheltering opportunities.
Some have argued that lower capital gains rates promote investment, but this case appears overstated. Among other reasons, it is hard to imagine entrepreneurs making decisions about investment and risk on the basis of the capital gains tax regime: Mark Zuckerberg was not focusing on the capital gains tax when he was in his dorm room coding up Facebook.
Given the magnitudes at stake, scorekeeping procedures employed in evaluating capital gains should be made more transparent and the subject of external professional debate and review. This transparency would facilitate discussion between professional scorekeepers and outside experts about the extent to which models can be improved and new data collected. It would also facilitate the comparison of estimates across a broader set of proposals and help ensure that consistent scorekeeping practices are applied.
Transparency is a double-edged sword. Given the importance of official scores to legislative decision-making, releasing the assumptions underlying scorekeepers’ estimations to the public would invite greater lobbying around those assumptions by supporters and critics of different reforms. Our proposal is not to open the floodgates with respect to scorekeeping writ large. A natural structure is in place: the Congressional Budget Office already has a panel of advisors who provide input on economic issues. This group or a related subgroup of experts can be convened to advise the JCT, as well as the CBO and the Treasury’s Office of Tax Analysis. It would be important for diverse views to be represented in this body, and it would be valuable to work with the full set of scorekeepers to select a panel of people who are thoughtful and likely to be taken seriously by the revenue-estimating community. Short of such a formal gathering, promoting informal conversations and collaborations between scorekeepers and academics would facilitate advancing the research frontier in the most useful directions.
Our call to action is born from a position of enormous respect and admiration for the integrity and seriousness of the scorekeepers. The ultimate goal is to continue to advance our understanding of taxpayer behavior and the revenue potential of capital gains and other tax-reform efforts to inform the policy-making process.
Natasha Sarin is assistant professor of law at the University of Pennsylvania.
Lawrence H. Summers is the Charles W. Eliot University Professor and president emeritus at Harvard University, and served as the 71st secretary of the US Treasury and the director of the National Economic Council.
Owen M. Zidar is professor of economics and public affairs at Princeton.
Eric Zwick is associate professor of finance and a Fama Faculty Fellow at Chicago Booth.