The Tax Cut So Regressive Even Mitt Romney Has Doubts

October 1, 2019

Imagine a world in which the most pressing issue is to slash taxes for the rich and only the rich, costing the US government hundreds of billions of dollars and doing little to spur economic growth. Imagine a policy so unequal that even Mitt Romney has his doubts. Reader, I give you the capital gains tax cut.


In recent weeks, President Trump has hinted that he will bypass Congress to lower capital gains taxes by indexing the gains to inflation. Under current law, households owe taxes on the nominal value of capital gains. For example, for a stock bought for $100 and sold for $110, the household would face taxes on $10 of income.

Trump’s proposal would index asset values to inflation, thus taxing only real gains and not nominal gains. In the above example, if average prices of all goods had increased by 10 percent between the period the stock was bought and sold, there would be no taxation on gains.

On the surface level, there are a number of reasons this proposal is a bad idea. There is little evidence that lowering capital gains tax rates will raise savings or investment that would lead to economic growth, especially since the bulk of tax cuts in the medium term will apply retroactively to investments made in the past.[1] The 2017 corporate tax cut, which resulted in a bonanza of stock buybacks and little evidence of increased investment, should dispel the notion that lower tax rates will boost economic growth.[2]

Moreover, it is unlikely that the president even has the legal authority to make the changes.[3]

Dig a little deeper, and it becomes clear that the benefits of this policy overwhelmingly flow to a miniscule group of Americans. In the analysis below, I show that capital gains are the most unequally distributed form of income there is—much more unequally distributed than wealth, total income, or wages. Cutting taxes on this source of income is thus the most regressive tax policy possible.

To analyze which groups will gain the most from the law, I use data from the Distributional National Accounts, a data source that has information on income and capital gains tax payments of a representative sample of individuals in the US.[4] We can thus measure which groups will gain from a cut in capital gains taxes.[5] We will focus on the richest 10 percent by total income (the ‘top 10 percent’) and the richest 1 percent by total income (‘top 1 percent’).

Figure 1 shows that the benefits of the tax cut overwhelmingly accrue to the richest percentages of the population. The top 1 percent would receive 74 percent of the gains, with the top 10 percent receiving 92 percent.  The story is even worse the further up you go. The richest top 0.1 percent would receive 53 percent of the gains, with the richest 0.01 percent receiving 32 percent of the gains.

 

 

There are three primary reasons behind the concentration of the benefits. First, overall wealth is highly concentrated, with the top 1 percent owning 35.2 percent of aggregate wealth in 2016. Second, wealth is particularly concentrated in asset classes that tend to generate the largest amount of capital gains: the top 1 percent owns 65 percent of corporate equities.[6] Finally, the bottom part of the distribution tends to own assets in which capital gains are not taxable, such as assets in defined contribution plans or residential housing.

These are not small numbers we are talking about; capital gains are a very large and growing source of income in the US. It is common for total capital gains income to be 10 to 25 percent of GDP in a given year.[7] Capital gains have been particularly large since 1980, due to increasing profits, share buybacks, and lower interest rates. The Penn Wharton Budget Model estimates that Trump’s proposed tax change would cost the Treasury $100 to $200 billion over 10 years,[8] money that is desperately needed for infrastructure, Social Security, health care, and climate change remedies. The best use of $200 billion is not funneling it to the top.


[1] See Huang and Marr (2012).
[2] See Furman, “Not Much”: What macroeconomic data says about the impact of the Tax Cuts and Jobs Act. HTTP://WWW.AEI.ORG/PUBLICATION/NOT-MUCH-WHAT-MACROECONOMIC-DATA-SAY-ABOUT-THE-IMPACT-OF-THE-TAX-CUTS-AND-JOBS-ACT/
[3] See Hemel and Kamin (2019).  HTTP://YALEJREG.COM/ARTICLEPDFS/36-JREG-693-HEMEL.PDF. Under George HW Bush, the White House Office of Legal Counsel concluded that the Treasury Department could not index gains without Congress. See HTTPS://WWW.JUSTICE.GOV/FILE/20536/DOWNLOAD.
[4] Data is available at HTTP://GABRIEL-ZUCMAN.EU/USDINA/.
[5] We assume, following the CBO, that the entire economic incidence of personal income taxes falls on the income earner. Furthermore, we assume that individuals across the income distribution have the same holding periods for assets. To the extent that richer individuals hold assets for longer (shorter) time periods, these estimates will tend to underestimate (overestimate) the share of the benefits accruing to top income groups. The Penn Wharton Center microsimulation (HTTPS://BUDGETMODEL.WHARTON.UPENN.EDU/ISSUES/2018/3/23/INDEXING-CAPITAL-GAINS-TO-INFLATION) estimates that the top 1 percent will receive 86 percent of the benefits, and the top 20 percent will receive 99 percent of the benefits.
[6] Corporate equities directly held, not those held in pension plans.
[7] See Robbins (2018)
[8] See Huang and Bryant (2019)

References

Hemel, Daniel, and David Kamin. “The False Promise of Presidential Indexation.” Yale J. on Reg. 36 (2019): 693.

Huang, Chye-Ching, and Chuck Marr. “Raising Today’s Low Capital Gains Tax Rates Could Promote Economic Efficiency and Fairness, While Helping Reduce Deficits.” Center on Budget and Policy Priorities (2012): 14.

Huang, Chye-Ching, and Kathleen Bryant. “Indexing Capital Gains for Inflation Would Worsen Fiscal Challenges, Give Another Tax Cut to the Top.” Center on Budget and Policy Priorities (2019).

Robbins, Jacob A. “Capital Gains and the Distribution of Income in the United States.” (2018). Working Paper. http://jacobarobbins.com/jr_inequ_jmp.pdf