Taxing Capital Gains as Ordinary Income
A look-back approach can be used to tax capital gains like ordinary income
Welcome Back to Win-Win Democracy
A key component of the win-win approach to reduce corporate political spending that we discussed last time is taxing investment income as ordinary income. Doing so will accomplish several goals we’ve already discussed:
Slow and reduce the ability of individuals to build oligarch-level wealth by deferring and often never paying capital gains taxes.
Offset tax revenue lost by eliminating corporate income taxes for companies that agree to be bound by constraints on corporate political spending.
This newsletter issue addresses both nuts-and-bolts and policy issues of taxing capital gains as ordinary income. I’m going to assume that you understand the basics of the current system; if not, (re)read the sections Some Taxing Stories and Exploiting Capital Gains for Oligarch Wannabes in the Growing Oligarchs issue of the newsletter.
Thanks for reading Win-Win Democracy! Subscribe for free to receive new posts and support my work.
When you earn a wage, you pay income tax on the wages you earn during the calendar year. When you own an asset that increases in value, you pay income tax on the increase in value only when you realize the gain, i.e., when you sell the asset. This is known as deferral because, while the gain occurred over potentially many years, you didn’t pay taxes on the asset’s increase in value each year, but instead deferred the income taxes until the gain was realized.
As we saw in Exploiting Capital Gains for Oligarch Wannabes, deferral is immensely valuable with high-growth investments held over long time periods. Indeed, deferred taxation of capital gains is a major reason that individuals can amass fabulous wealth. Without deferral, individuals could still become very wealthy, but we would see far less Bezos- or Musk-like astronomical wealth.
Eliminating deferral would also go a long way to place wage income and investment income on an equal footing.
Suppose we wanted to eliminate deferral by collecting tax on an asset’s gain in value every year. To do this, you need to know how much the asset is worth at the end of each year, easy for something like a publicly-traded stock, but difficult for a work of art, other collectible, real estate, or private stock.
The only certain way to value an asset for which there’s no public market is to see what it fetches when you sell it. This is how our current system works — you only pay capital gains tax when you realize the gain, i.e., sell the asset.
You can imagine a system that eliminates deferral on easily-valued assets and taxes hard-to-value assets only when the gain is realized. This would be a mistake, however, because it would create incentives for people to invest in hard-to-value assets instead of in easy-to-value assets.
An important design principle, therefore, is:
The effect of deferral on taxes should be eliminated for all assets.
The phrase “effect of deferral” is important because it would be unreasonable to require that an asset actually be sold at the end of every year just to find the asset’s value or to get the funds to pay taxes on the annual gain.
Look-back Taxation of Capital Gains
Suppose that we only collect tax on capital gains when the gain is realized, as in the current system. We could nevertheless implement the effect of eliminating deferral by computing (or estimating) what the income tax would have been each year if deferral was not allowed.
To compute (or estimate) what the tax would have been in previous years if deferral was not allowed, we need two pieces of information:
The asset’s gain in value each year. This gain becomes additional ordinary income in that year.
The additional income tax that would have been paid each year on that additional income.
As an example, let’s revisit the story of Alice the investor from Some Taxing Stories. Alice bought 1,000 shares of Amazon stock on January 2, 2017 for $795.99 per share. She sold it five years later on December 27, 2021 for $3,372.89 per share. Her realized gain was $2,576,900.
We can look up the price of Amazon stock at the end of each year and compute the gain she made each year, which she would have reported as ordinary income on each year’s tax return:
Of course, Alice couldn’t have reported that additional income during those past years, but the IRS could do this calculation and compute the additional tax Alice would have paid in each of these years based on the ordinary income tax rates then in effect and Alice’s previously reported income for that year.
Alice would then owe those deferred taxes, plus interest (the IRS already sets standard interest rates for payment of deferred taxes every quarter), for the year in which she sold her Amazon stock.
Alice has effectively taken a loan for the deferred taxes from the government. That’s why she pays interest on those deferred taxes. If Alice preferred not to take such a loan from the government (if, for example, she believed that her return on having that money invested is less than the interest rate), she could estimate the deferred taxes and pay them every year, effectively settling up with the government when she sold the stock.
Let’s change the story so that instead of buying Amazon stock, Alice buys the stock of an unlisted startup company.
We can’t know the asset’s yearly gain in value, so we estimate it. The most realistic simple approach is to compute the constant daily rate of return from the investment. Using simple formulas in a spreadsheet, we can do this, and, from the daily rate of return, we can compute the estimated value of the unlisted stock at the end of each year. Here’s the result:
The “gain” column shows the estimated gain in value that year, which Alice would report as additional income for that year.
Other than using estimated values, everything else is the same as with the example using a publicly-listed stock.
What if Alice (in either scenario) lost money on her investment, either overall or just in certain years? Current tax law allows up to $3,000 (for a couple filing jointly) in losses to offset other ordinary income; the rest of the loss can carry forward to future years to offset future gains. This approach, however, is inconsistent with look-back taxation of capital gains because there is no distinction between tax on capital gains and tax on ordinary income.
Instead, we create exactly the same kind of tables as before, using actual values for easy-to-value assets and estimates based on constant daily returns for hard-to-value assets. For years in which the gain was negative, compute how much the investor’s tax for that year would have been reduced and the government would refund the difference with interest.
Treating capital gains as ordinary income implies that step-up-in basis would be gone.
So, for easy-to-value assets, your estate would pay capital gains taxes on your appreciated assets when it files your final tax return, just as if you had been alive to do it. The estate might have to sell your appreciated asset to pay the tax or it might pay the tax from other funds. Regardless, your heirs would receive what is left of the asset as your will directs.
In this situation, your estate might as well sell the asset, pay the tax, and give the heirs cash because there’s no tax advantage to passing the asset on directly.
Hard-to-value assets and illiquid assets are more difficult. Imagine that the asset is a family business. It is unreasonable to force the estate to sell such an asset just to determine its value for tax purposes.
I see two possible approaches.
Accountants have methodologies for valuing businesses and other kinds of illiquid assets. These methodologies are already used for gift and estate tax purposes.
Such estimates could also be used for capital gains tax purposes: The estate would use such an estimate as the value of the asset then compute the capital gains tax as described above for a hard-to-value asset. The estate would pay the capital gains tax out of other funds, the heirs would assume the asset, and would use the estimated value as their cost basis going forward.
The disadvantage of this approach is that the estate must be able to get the funds to pay the capital gains tax due.
An alternative approach could be used to solve this problem: Instead of treating the death as a realization event, carry forward the information needed to be able to do a look-back tax computation when the heirs actually want to sell the assets.
Basically, the information required would be the deceased’s income for every year the deceased owned the asset. Then, when the asset is eventually sold, that information would be used to compute the deferred taxes and interest based on the valuation from the sale.
This approach would require that the IRS be able to identify an asset (e.g., an asset identifying number analogous to a taxpayer id number) and associate the required information with the asset.
Donating or Gifting Appreciated Assets
Under our current system, when one donates an appreciated asset that you’ve held for more than a year to a charity, tax on the capital gain is never paid, neither by the donor nor by the donee.
If we are going to tax capital gains like ordinary income, there would be no special treatment of a donation of an appreciated asset. For a liquid asset, you’d sell your asset and donate money to the charity. Or, for a hard-to-value, illiquid asset, you’d get an appraisal of the asset’s value and pay the capital gains tax due based on the appraised value of the asset.
Similarly, if you gift your asset to someone else, this would be treated as a realization of the gain and you’d pay the capital gains tax on it. Then you could give the asset to your child or other recipient and their cost basis would be the value of the asset on that day.
People are affected when tax policy is changed. For example, many people purchased homes assuming that their property taxes were fully deductible. This suddenly changed when the TCJA was passed in 2017 and many people were unhappy about this sudden change.
We have a similar situation here: People who purchased assets under the current tax policy will be affected by eliminating preferential treatment of capital gains. The vast majority of impact would be on people in high-income households since 95% of the benefit of preferential treatment of capital gains flows to people in the top quintile of household income; indeed, 75% flows to people in the top 1 percent of household income.
This suggests a transition period that would work something like this:
Assets purchased on or after the effective date would be taxed according to the new rules.
Assets purchased prior to the effective date and sold within ten years of the effective date would be taxed under the old rules unless the owner’s household income places the owner in the 1% of household incomes, in which case the new rules would apply. The rationale for this is that people in the top 1% have the resources to adjust to these changes quickly.
Assets purchased prior to the effective date and sold more than ten years after the effective date would be taxed according to the new rules.
Extremely wealthy people whose wealth consists primarily of unrealized capital gains, which includes many billionaires, would owe large tax bills. But they would have ten years to plan for how they would pay those bills.
Of course, there is a lot of room for discussion on this topic.
I did not make this stuff up in a vacuum!
There is academic literature on this topic and related issues. It is widely recognized that our current system of taxing capital gains both favors the wealthy and is full of loopholes.
An approachable overview, written from the perspective of taxing wealth by taxing investment income appears in a paper by Leiserson and McGrew. The Tax Foundation, a conservative-leaning think tank has written an analysis of look-back approaches to eliminate the deferral of capital gains tax. I disagree with some of their framing and analysis, but it is nevertheless worth reading. The Wharton business school has modeled the impact of various approaches on tax revenues.
Finally, Senator Ron Wyden, Chair of the Senate Finance Committee, has proposed a bill that would tax unrealized capital gains of billionaires using similar approaches to what we’ve discussed here. He would not, however, change the overall preferential treatment of capital gains. A form of this proposal has made it into President Biden’s 2023 budget proposal, but seems to have little chance of being enacted.
In the previous newsletter, I proposed a win-win approach to reducing the influence of money in politics that would (1) eliminate corporate income tax for corporations that agree to abide by restrictions on corporate political spending, and (2) revamp the tax on capital gains to both reduce the ability of individuals to obtain Bezos and Musk-level wealth and to offset tax revenue from the loss of corporate income tax.
The discussion in this newsletter issue provides some detail about how the revamping of the capital gains tax could work. I’m not a tax expert and many more details would have to be decided to be able to implement something like this.
The fourth major threat to democracy identified in Taking Stock, which we have not yet discussed, is economic resentment among the tens of millions of Americans who have been left behind in the country’s economic growth. We will take up that topic in the next newsletter issue and we’ll find this issue’s discussion of capital gains taxes quite relevant.