Raymond Richman - Jesse Richman - Howard Richman
Richmans' Trade and Taxes Blog
Solving the Capital Gains Riddle
Under the U.S. personal and corporate income tax codes, realized capital gains and losses are considered income. However, when calculating national income, capital gains and losses are excluded. Why the difference in treatment? No serious economist would argue that changes in the value of assets during a period should be added in calculating the national income for that period. Nor even that realized capital gains and losses should be added in calculating national income. Yet most appear to believe that realized capital gains at least should be counted as income for purposes of personal and corporate income taxation. Some argue that to exclude realized capital gains would be unfair to taxpayers whose income comes from other sources. In this view, even unrealized capital gains (referred to in the economic literature as accrued capital gains) are income but need to be excluded for administrative reasons, i.e., one should not be forced to sell some of his capital to pay the tax liability. Herein we argue that accrued capital gains are not income, that realized capital gains are income if they are consumed but not income if they are re-invested.
As distinguished an economist as the late Herbert Stein, former chairman of the Council of Economic Advisors, in an op-ed piece in the Wall Street Journal, "Don't Cry for Capital Gains Taxpayers" (WSJ, July 5, 1995), wrote:
"Because of the value of time, the tax on capital gains is lower than it seems--and lower than the tax on many other [sic!] kinds of capital income. That is due to the deferral of the tax on the gain from the time the gain accrues to the time the gain is realized by sale or exchange."
When he writes that an accrued gain is a form of capital income, he is confusing capital and income.
American economists have been confusing capital and income for decades. When he writes that income tax is Adeferred@ from the time it accrues to the time it is realized, he means that the capital gain is income as soon as the value of a capital asset increases, a view he shares with nearly all American economists. Most accrued gains in common stocks are the result of increased corporate earnings or the expectation of increased corporate earnings. All the while the investor owns those shares, the corporate earnings are subject to corporate income taxation and the investor, in addition, pays taxes on the corporation=s dividends. Many economists consider this to be double taxation of corporate income. No doubt Prof. Stein had in mind the fact that the retained earnings of corporations are not subject to personal income taxation and their reinvestment can be expected to increase future earnings and result in the appreciation of the company=s shares. There was a time when the difference between corporate and personal marginal income tax rates was very great which would seem to justify Prof. Stein=s statement. The reverse relationship is true today. The corporate rate of income tax on income over $75,000 is 34 to 39 percent depending on the bracket and is higher than the current maximum rate on long-term capital gains of 20 percent.
Any view that considers accrued capital gains as income confuses capital and income. The value of a capital asset is simply the capitalized value of the income it is expected to yield. Taxing capital gains as they accrue would result in a double tax on income, a tax on the capital gain followed by annual taxes on the future yields which are economically equivalent to the accrued capital gain. Indeed, if you taxed the accrued capital gain, the taxpayer=s future income would be lower by what the tax paid would have yielded. For this reason, the British historically never considered capital gains as income for income tax purposes. They have avoided double taxation of corporate earnings, too, by giving individual investors a full tax credit for the taxes paid by corporations.
A capital gains tax is equivalent to a tax on the expected future income that produced the gain in the first place. Taxing accrued but unrealized capital gains is no more justified than taxing a medical doctor upon completion of his training on the capital value of his future income stream. Suppose a doctor can reasonably anticipate earning an income of $200,000 after graduation, an income stream that has a capital value of $2 million. The cost of his education, let us say including income foregone, was $500,000. Using a 10 percent capitalization rate, he has an accrued capital gain of $1.5 million. No one would argue that he should pay a tax on the accrued gain. He=ll pay tax on his annual income of $200,000 as he earns it. Owners of stocks and bonds and real estate similarly will pay tax on their earnings from those sources as they earn it.
Capital is a source of income and the source needs to be distinguished from its yields that accrue to its owner in the form of rent, interest, profits, and indeed, as in the above example, wages and salaries. If you tax accrued capital gains which are the capitalized value of an expected increase in future yields and you tax the yields as they are received, you are taxing the same thing twice.
In his classic work, Value and Capital, Prof. J. R. Hicks defined income as the maximum value that a person can consume during a period of time and still be as well off at the end of the period as he was at the beginning. American economists, as Prof. Stein did, thought this meant that accrued gains were income. Obviously a person is better off if the value of his investment has appreciated. In this case, it is because his future earnings from that investment are expected to increase. The difficulty as he points out is with measuring how much of his capital stock he can consume and still be as well off at the end of the period as at the beginning. Should capital be valued in money terms or real terms? Should one count changes in value resulting from changes in the interest rate used in capitalizing (i.e., calculating the value of) the future income stream? These are all questions that are relevant to the appropriate tax treatment of capital gains. But if a person consumes all of the increase in the value of the capital stock, whatever the definition, there is no possibility of income growth, of larger consumption in the future. To quote Prof. Hicks,
"The income which is relevant to conduct must always exclude windfall gains: if they occur, they have to be thought of as raising income for future weeks (by the interest on them) rather than as entering into any effective sort of income for the current week. Theoretical confusion between income ex post and ex ante corresponds to practical confusion between income and capital." (2nd Ed., 1946, p.179)
Unfortunately most American economists ignored Prof. Hicks=s warning and are guilty of confusing capital and income.
But wait, there is more. This same logic can be used to justify the taxation of realized capital gains when they are consumed and to exclude realized capital gains that are reinvested. The identity of capital and its yields, that one is the equivalent of the other, suggests that realized capital gains that are consumed should be taxed as income because the taxpayer is in effect consuming the yields that holding the asset could be expected to produce. Taxing the consumption of realized gains as income does not confuse property and income. When one sells a capital asset, one is realizing the capitalized value of its future income stream. If the sales proceeds are reinvested, the seller is simply exchanging one equivalent income stream for another. But when one consumes the proceeds, one is consuming the future yields which would have been subject to tax had the asset not been sold.
Realized capital gains are as spendable as any other income. The fact that they may be spent is one of the reasons why economists argued that realized capital gains should be taxed as income. However, capital gains are not necessarily consumed. If they are reinvested, the investor=s capital stock remains the same and any tax on the gain is a tax on capital. To allow taxpayers who realize capital gains to roll over the proceeds to other capital assets without incurring tax liability would give capital gains the same treatment that is presently accorded the capital gains of homeowners. But it is also an argument for taxing realized capital gains only if they are consumed. If a taxpayer chooses to consume his gains, he is choosing to treat the gain as spendable income.
The law recognizes certain exchanges as tax free. For example, home-owners have the privilege of avoiding capital gains tax if the householder buys another home of equal or higher value. Such a transaction is called in technical jargon a "roll-over". However, the privilege should extend only to realized capital gains that are reinvested, not consumed, i.e., genuinely rolled-over. Just as accrued gains represent a future stream of income, realized gains represent a future stream of income and when the gain is consumed the taxpayer in effect consumes that future income stream. The investor who purchases that asset is investing; the consumer of realized gains is dis-investing, converting the savings of the buyer to consumption. Realized capital gains can be consumed or saved just as wages, rent, interest, and profits can be consumed or saved. So we err by taxing reinvested gains and the British err by their failure to tax the consumption of realized capital gains as income.
Another virtue of permitting the roll-over of gains without taxing the gain is that it would make irrelevant all the arguments for special treatment of capital gains and simplify tax administration enormously. Many capital gains are illusory due to inflation and many are due to changes in interest rates (the rate used in calculating present value). No distinction would need to be made between illusory gains and Areal@ gains and between short- and long-term gains. There would be no need to tax gifts and inheritances. When the ownership of an asset is transferred by inheritance, the heirs take as their basis the value of the asset at the time of the decedent=s death. Thus, one way to avoid a tax on capital gains is to hold the capital asset until death. Economists have long advocated closing this loophole. However, the remedy is not to close this loophole but to tax inheritances as one should tax capital gains, that is, if and when any part is consumed, it will be taxed as income. The roll-over should be extended to the recipients who would pay taxes at regular income tax rates if and when they consume the assets they receive. Those who press for lower estate taxes argue that payment of the tax forces the sale of family businesses and farms to pay the estate taxes. Inheritances could be treated as roll-overs. When and if the donees and legatees sell the businesses or farms and consume the proceeds, they will be taxed. They won=t be taxed at all if they do not consume their inheritances. That is really tax simplification, equity, and sound economically.
Compare the simplicity of this approach with the other problems and suggested solutions to the inequity of the current treatment of capital gains. Because capital gains may accrue over a number of years, counting them as income in the year they are realized may push the taxpayer into a higher tax bracket than would have been the case had the gain been taxed each year as it accrued. This fact is often advanced by those advocating lower taxes on so-called long-term capital gains. The suggested solution to this inequity is tax averaging which complicates the income tax. But it isn=t a problem at all if you tax consumed gains and un-tax reinvested gains. The investor does his own averaging by choosing not to consume his gains is a single time period. No need for any distinction between so-called long-term and short-term gains and losses or by arbitrarily excluding 50 percent or some other percent of the capital gain. If and when Congress restores averaging, fairness requires that it should apply to all forms of income, not only to capital gains.
Another problem in the current treatment of capital gains is that they are often illusory. Indexing to adjust for inflation has been suggested as a solution to the problem of illusory gains. The rationale for such indexing is clear; illusory income isn't income at all. And real capital gains, gains remaining after adjusting for inflation, should receive the same tax treatment as income from other sources. The question of illusory gains doesn=t arise if rolling-over capital gains without creating a tax liability becomes the policy.
The argument is sometimes made that other forms of income are affected similarly by inflation. That is not the case at all. A $5.00 per hour wage in 1980 and a $10.00 wage in 1990 may have the same purchasing power measured in 1980 dollars. But there is a real income in both years of $5.00 measured in 1980 dollars. In the case of an asset worth $10,000 in 1980 and $20,000 in 1990, the asset has a value measured in 1980 dollars exactly equal to what was paid for it. The apparent $10,000 gain is an illusion. The $5.00 real wage income in both years is no illusion. Once again, this is no problem if we tax gains that are consumed. There is nothing to index. The consumption is current just as the consumption of wage income is current.
There is no justification for a cap on the rate applicable to capital gains because it discriminates against capital gains taxpayers in lower income brackets as well as those taxpayers who derive their income from wages, dividends, etc. There is no justification for giving wealthy taxpayers a tax break denied taxpayers who are less wealthy. The former get a capital gains rate cap under the present law while those in lower brackets get little or no preferential treatment at all.
The politicians of both political parties have been playing Russian roulette with capital gains and shooting themselves in the foot since the early 1920s, rolling the barrel with every change of administration. The policy that endured the longest and which was in effect at the time of the 1986 "reform" was to count 50 percent of long-term gains as income with a maximum rate of 50 percent, in effect a cap of 25 percent. It had the virtue at least of giving all capital gains recipients a tax benefit. Currently all of the capital gain is counted as income but the maximum rate of 20 percent benefits only taxpayers in the top income brackets. Middle class taxpayers derive no benefit whatsoever from the present treatment.
Capital losses under the present treatment may be used to offset capital gains but the excess of losses over gains is limited to $3,000. This provision alone should alert economists that realized capital gains may not be income at all since you cannot treat capital losses the same way as gains. Once again taxing only consumed gains as income requires no provision of losses at all. They are in effect Arolled over@; it is a change in the value of capital and has no relevance to income tax.
Income taxes have two opposite effects, a substitution effect and an income effect. The substitution effect discourages working and saving which are taxed and encourages leisure and consumption which are not taxed. The income effect results from the fact that taxes leave you poorer and you therefore have to work harder and save more to maintain or reach your desired life style and provide for your retirement. Leisure isn't worth much without money to spend -- you're just unemployed. Lower rates of tax on capital gains are probably one cause, perhaps even the major cause, of the rich saving a smaller proportion of their income than they did historically. Coupled with the unprecedented bull market of the last decade, favorable treatment of capital gains has made it much easier for wealthy investors and corporate insiders to consume more and save less. Taxing the consumption of capital gains and permitting their roll-over would be neutral insofar as incentives to save or consume are concerned.
A basic reform that would postpone taxation on income until it is consumed is the essence of the Unlimited Savings Act sponsored by Senator Nunn, a Democrat, and Senator Domenici, a Republican several years ago. Excluding savings whatever the source from income taxation would have the effect of converting the progressive income tax to a progressive tax on consumption, in the process simplifying the tax system enormously. People would be taxed not on what they put into the economy but on what they take out of it. There would be no need for depreciation and depletion allowances, gift taxes, inheritance and estate taxes, retirement plans, and, of course, special treatment of capital gains. One would simply report the money received from any source and subtract money that is invested. We then would have a tax system that stimulates investment and promotes economic growth.
We know what an income tax ought to be. It ought to be fair, economical to administer and to comply with, economically neutral, and progressive. Fairness requires that persons in equal economic circumstances pay the same amount of tax. Economical to administer means that it should not be costly to collect and enforce and should not impose unnecessary burdens on the taxpayer in the form of time and money. Since the income tax diverts resources from households to the government, the use that the government makes of those resources at the margin should contribute at least as much to the country's well-being as those funds would contribute if left in the private sector. It should be neutral between economic activities that have equal social value. The income tax is a progressive tax taking a higher percentage of income from the rich than from the poor thus reducing after-tax inequality. While opinions may differ as to how progressive the income tax should be, few disagree with the proposition that ability to pay increases faster than the growth of income. Income like any other good is subject to the law of diminishing marginal value. How progressive should the tax be? The limit to progressivity is the trade-off between income redistribution and economic growth. Taxing capital gains that are consumed as income while permitting the roll-over of capital gains that are reinvested accords with these sound principles.
Economy - Long Term
Economy - Short Term
Last 100 Years
Real Estate Taxation
Journal of Economic Literature:
Atlantic Economic Journal: