The United States entered the COVID-19 crisis with an unusually large budget deficit for an economy at or close to full employment. Even if employment, output, and growth were to recover quickly to where they were at the end of 2019 (something that is far from certain), the deficit, under current law, will remain large.

The good news is that interest rates are likely to remain well below the rate of GDP growth for the foreseeable future, as they have since the beginning of the century. As long as that remains the case, there is no danger of an “exploding debt” scenario in which a large but constant federal deficit causes debt to grow without limit as a share of GDP. Still, the debt and deficit will remain high by historical standards. Accordingly, as the recovery proceeds, we can expect to hear it argued on both economic and political grounds that the deficit should be reduced. 

At that point, the search will be on for ways to close the budget gap. Although politicians on both side of the aisle will undoubtedly roll out their favorite targets for spending cuts, much of the heavy lifting is going to have to come on the revenue side of the budget. As former Trump adviser Gary Cohn put it recently, talking to CNN’s Fareed Zakaria

Our next Congress, the Congress that sits down in 2021, almost has to sit down and look at our spending and our revenue side. … How we spend money? There are a lot of places where we could cut back. In addition to that, I think they have to look at our tax system and think of ways that we raise revenue

One area of the tax code that is ripe for reform is the preferential treatment given to capital gains. While incomes from wages and salaries face a maximum tax rate of 37 percent, capital gains on assets held for more than a year, in most cases, are taxed at a maximum rate of just 20 percent. 

The benefits of the capital gains tax preference flow predominantly to the rich. Some 70 percent of the benefits go to taxpayers with annual incomes of $1 million or more, who enjoy annual benefits of $145,000 each. Benefits for households with incomes of $50,000 or less average about $10. For years, backers have tried to find broader justifications for this tax break, contending that benefits to the few somehow trickle down to the rest, but their efforts are less than convincing. 

Here are some of the issues raised by the capital gains tax preference, and some of the reasons why many economists put its elimination near the top of the list in the search for additional sources of federal revenue.

Are capital gains really income?

One argument for lenient tax treatment is that even if income is to be taxed, capital gains are not income at all, but only a reflection of a transfer of ownership of an existing asset. If we want assets to come under the control of those who can put them to best use, say defenders of the preference, we should not erect barriers to their purchase and sale. Just as we tax labor income when it is paid out to workers, we should tax capital income only when it is paid out as interest, dividends, royalties, or whatever, but not when ownership of an asset is transferred from one party to another.

That might make sense if it were possible to distinguish cleanly between income and capital gains, but it is not. Instead, it is all too easy to transform income from almost any source into something that looks like a capital gain.

Consider a fanciful example: I make a contract with my wife to supply 2,000 hours of labor over the coming year for a payment of $2,000. Under the terms of the contract, she can, if she wants, call on me to use those hours in any way she wants. She could ask me to mow the lawn and make lasagna for dinner, but she could also sell the local university an option that would allow it to buy my services at the same $2,000 she paid for them. The university willingly pays her $95,000 for the option, which is a good deal for them since they would otherwise have had to pay me $100,000 to teach my quota of courses. When we file our joint tax return come April 15, we will be taxed at the capital gains rate on my wife’s $95,000 income from the option sale, while I report ordinary income of $2,000. Our before-tax income is a little lower than it otherwise would be, but our total taxes are a lot lower. 

Although this example seems absurd, the real world is full of situations where businesses can do much the same – structure a transaction to make it look like a capital gain instead of ordinary income. In fact, one law firm advertises an options-based strategy exactly like that described above, except that it is used to shelter income from real estate rather than from labor. Choices like organizing a firm as a partnership instead of a corporation, paying executives with stock options instead of a salary, or imaginatively structuring a real estate deal can also convert ordinary income to capital gains. The strategies are endless.

One of the most controversial ploys is the carried interest rule, which allows hedge fund and private-equity managers to structure the income they receive for their services in a form that qualifies for taxation as capital gains. Even some commentators who are otherwise enthusiastic about lenient taxation of capital gains draw the line at carried interest. For example, David Frum, who thinks a lower tax rate for capital gains is good policy, agrees that the rule is “utterly unjustifiable. If you’re investing with other people’s money,” he says, “What you are earning is income — and it should be taxed as such.” The 2017 Tax Cuts and Jobs Act, which slashed the corporate tax rate, was supposed to close the carried interest loophole, but it ended up making only minor modifications. For the most part, say tax analysts, the rule lives on. 

Tax avoidance strategies that convert ordinary income to capital gains would be a problem even if they did nothing but generate inequities and reduce federal revenues, but that is not the whole story. Such strategies may require more than just waving an accountant’s wand over something a firm would do anyway. Instead, structuring transactions to take advantage of specific tax rules often requires changing actual business practices, such as the choice of financing methods, the timing of investments, even the choice of one’s whole line of business. Often, the changes would be unprofitable except for their tax advantages. Although proponents of the capital gains preference claim it supercharges growth and efficiency, tax-induced changes in business practices are a significant drag on the real economy.

Do we need a capital gains preference to correct for inflation?

A second argument used by supporters is that we need low tax rates on capital gains to avoid taxing “phantom” gains produced by inflation. The argument is superficially plausible. When there is inflation, asset owners may be taxed on nominal gains even when real asset values do not increase. 

For example, suppose you buy some shares of stock at $100 and sell them for $120 a couple years later. Inflation has meanwhile pushed up the cost of living by 10 percent. That leaves you with an inflation-adjusted pretax gain of just $10. If you pay 37 percent tax as ordinary income on the $20 nominal gain, your tax rate on the $10 real gain is 74 percent. Cutting the rate on nominal capital gains to 20 percent reduces the real rate to 40 percent – not quite enough even to fully level the playing field, but a move in the right direction, it would seem.

But there are two flaws in that argument.

First, any arbitrary rule, such as a fixed lower tax rate or an exclusion of a portion of capital gains, can only crudely approximate the necessary adjustment for inflation. The 20 percent rate that is close to right in the example we just gave becomes too low if inflation slows (as it has in recent years). If inflation instead accelerated, it would become too high. Furthermore, the rate that just levels the playing field for a person in one tax bracket could be too high or too low for those in other brackets.

A more nuanced approach would be to index the basis on which capital gains are calculated to reflect actual inflation between purchase and sale. That would avoid the taxation of phantom capital gains, but not a second, equally serious problem: Other forms of investment income, too, are subject to phantom taxation when there is inflation. 

Suppose, for example, that in a zero-inflation world, borrowers would offer a 5 percent coupon rate on top-rated corporate bonds. If the rate of inflation rises to 5 percent, borrowers would be willing to offer a 10 percent nominal coupon rate on the bond, since they know they will be able to pay future interest and principle in less valuable dollars. The 10 percent nominal rate leaves your real return and their real interest cost at 5 percent. So far, so good. But suppose now that you are subject to a 20 percent tax on your interest income. In the zero-inflation case, your interest income after tax is 4 percent. In the inflationary case, you have to pay tax on the whole 10 percent nominal rate, leaving you an 8 percent nominal return after taxes. When you subtract 5 percent inflation, that 8 percent nominal return becomes just 3 percent. In short, even if borrowers adjust nominal interest rates to fully reflect inflation, inflation increases the effective tax rate on bondholders.

The situation would be similar for income from the common stock of a firm that has constant real profit, from which it pays a constant fraction in dividends. Faster inflation would increase the real effective rate of taxation on the dividends.

A helpful 1990 paper from the Congressional Budget Office explores the problem in detail. The paper confirms that faster inflation raises the effective tax rate on investment income, but it points out that the effect is inherently smaller for capital gains than for dividend or interest income. Attacking the problem of phantom capital gains in isolation by whatever means — a preferential capital gains rate, an exclusion, or indexation – only widens the gap between the way inflation affects capital gains and the way it affects interest and dividends. Doing so increases the attractiveness of tax avoidance strategies that involve inefficient business practices.

The ideal solution to distortions caused by inflation would be to index the entire tax system. Indexation would have to cover not only all forms of investment income, but also taxation of ordinary income, real estate, inheritance, and everything else. But trying to remove the effect of inflation on capital gains taxes separately is likely to make things worse, not better. 

Do we need the preference to avoid double taxation of corporate profits?

“Double taxation” of corporate profits is a third common argument in defense of lenient tax treatment of capital gains. The idea is that corporate profits are taxed once at the business level and then again at the individual level when they are paid out as management bonuses, dividends, or capital gains. 

It is true that a preferential rate on capital gains would be one way to attack the distortion — one way, but a bad one. A much better way would be to fix the flaws in corporate taxes that are the source of the problem rather than apply a Band-Aid to capital gains.

Actually, part of the job was done in the 2017 Tax Cuts and Jobs Act, which lowered corporate tax rates across the board. A further step would be to get rid of the numerous loopholes in the corporate tax system that allow a big chunk of corporate profit to escape taxation altogether while those unlucky enough not to qualify pay much more. 

But the 2017 corporate tax cut left a key part of the job unfinished. If we want to enjoy the potential efficiency benefits of corporate tax reform, those taxes should not just be reduced; at the same time they should be shifted to the individual incomes of the managers and shareholders who are the ultimate recipients of corporate profits. To do that would require eliminating the capital gains preference. A regime that had no corporate income tax and full taxation of profits, whether earned by shareholders as capital gains, dividends, or in any other form, would eliminate double taxation once and for all without an unfair redistribution of the overall burden of taxation.

The lock-in effect

The “lock-in” effect is a final problem with capital gains taxation. As the tax is currently administered, people do not have to pay capital gains on an asset until it is sold. As a result, the after-tax return on an appreciating asset increases the longer it is held.

Compare a bond that pays a steady interest income every year to a stock that increases in value by the same amount each year. Over time, the effective tax rate on the stock would be lower even if the statutory tax rate were the same on both interest income and capital gains. The reason is that bondholders have to pay their tax year by year, while stockholders can defer payment of the tax until they sell their shares, possibly many years later. 

Moreover, if people still have not sold their stock or other assets when they die, their heirs never have to pay taxes on the capital gains at all. Instead, the value of the assets is “stepped up” to the market value at the time of the original purchaser’s death. In the meantime, the original owners can live quite well by borrowing against the value of the assets. When they die, their heirs can sell enough shares to pay off the loans, without paying capital gains taxes either on those shares or the ones they keep.

The lock-in effect, then, creates an artificial incentive for owners to hold on to stocks or real assets longer than they otherwise would. In many cases, that means that assets do not move smoothly from the hands of those who own them to those of new owners who could make better use of them. Furthermore, the lock-in effect greatly reduces the revenue that the government realizes from capital gains taxes. 

Unfortunately, removing the tax preference and taxing capital gains at the same rate as ordinary income would, by itself, make the lock-in problem worse. Assets would move from hand to hand even more slowly than they do now. As a result, the increase in revenue from lifting the capital gains preference would be disappointingly small.

Fortunately, there are ways to overcome the lock-in problem. A recent study from Brookings by Grace Enda and William G. Gale discusses several possible reforms.

Two of the simplest reforms attack the so-called “Angel of Death” loophole that allows heirs to fully or partially escape capital gains taxes. One version would eliminate the valuation step-up at death. Suppose John Sr. buys an asset for $1,000 in 1990 and dies in 2010 when the asset is worth $2,000. John Jr., the heir, finally sells the asset in 2020 when it is worth $3,000. Under the current regime, John Jr. pays no capital gains tax until 2020, and then only on the $1,000 gain that has occurred since the time of inheritance. Without the step-up, John Jr. would pay tax in 2020 on the full $2,000 gain that had taken place since the original purchase. An even stronger version of the same reform would require John Jr. to pay tax on the first $1,000 of the gain in 2010, at the time of inheritance, and pay the tax on the remaining $1,000 when the asset is sold in 2020.

A still more far-reaching reform would be to tax capital gains annually on an accrual or mark-to-market basis. Suppose that your tax rate is 30 percent, and in year 0, you buy 100 shares of stock at $50 a share. They rise in price to $100 in Year 1, fall to $75 in Year 2, and rise again to $150 in Year 3, at which point you sell them. Under mark-to-market taxation, in Year 1 you pay tax of $1,500 on a capital gain of $5,000; in Year 2, you have a negative tax liability of $750 on a capital loss of $2,500, which you can carry forward; and in Year 3, you have a capital gain of $7,500 on which you owe tax of $2,250, $750 of which you cover with the carried-forward loss. Your total tax over the 3 years is $3,000, the same as it would be if you were taxed only at the time of sale, but because there is no deferral of taxes, there is no lock-in effect.

Mark-to-market taxation would be easy to implement for assets like stocks and bonds that were actively traded, but not so easy for hard-to-value assets like real estate or private business interests. An approach called retrospective taxation could be used in such cases. Under that system, capital gains would be taxed at the time of sale, but the tax rate would be higher the longer the asset had been held. The rate of increase in the tax rate would depend on market interest rates. If it were done right, the cost of the rising tax rate would exactly balance out the benefit of deferring taxes until sale. In short, retrospective taxation would eliminate the lock-in effect without encountering the problem of annual evaluation of hard-to-value assets.

Any of these reforms, alone or in combination, would mitigate the lock-in effect. As a result, capital markets would operate more efficiently, since assets would move more freely into the hands of the owners who could put them to the best use. It would also increase the revenue gain from elimination of the capital gains tax preference.

The bottom line

Those who defend the preferential rate on capital gains are right when they argue that all taxes affect business decisions. But it is a non sequitur to say that because they affect business decisions, capital gains taxes should be as low as possible. The proper conclusion, instead, is that we should consider capital gains taxes in the context of the tax system as a whole:

  • Taxing capital gains at lower rates than other forms of investment income does little to encourage investment in general. However, it does a lot to encourage the structuring of investment in ways that avoid taxes, even if they are inherently less efficient.
  • Theoretically, an ideal tax system would be fully indexed for inflation, but singling out capital gains for special treatment while other forms of capital income are not adjusted actually increases the degree to which inflation undermines the equity and efficiency of the tax system.
  • A case can be made against double taxation of corporate profits, but the proper reform would be to tax capital gains and dividends as ordinary income at the same time that profit taxes at the corporate level were reduced or eliminated.
  • The lock-in effect is real, but the appropriate way to mitigate it would be through elimination of the angel-of-death loophole while taxing all capital gains on a mark-to-market or retrospective basis.

In short, as we look ahead to the likely need for additional federal revenues as the U.S. economy fully recovers from the COVID-19 downturn, a thorough reform of our system for taxing capital gains should be a high priority.