CRS Reports on Capital Gains Taxes

Gravelle presents arguments for and against capital gains tax rate reductions. The positives include a reduction of the lock-in effect and increased economic growth and entrepreneurship. Among the negative aspects of a rate reduction is that such a tax break would benefit mostly high-income taxpayers and encourage tax shelters and complexity, she notes.

Jane G. Gravelle Senior Specialist in Economic Policy Government and Finance Division

Summary

[1] The Taxpayer Relief Act of 1997 reduced capital gains taxes on several types of assets, imposing a 20 maximum tax rate on long-term gains. There is also an exclusion of $500,000 ($250,000 for single returns) for gains tax on home sales. The capital gains tax has been a tax cut target since the 1986 Tax Reform Act treated capital gains as ordinary income. An argument for lower capital gains taxes is reduction of the lock-in effect. Some also believe that lower capital gains taxes will cost little compared to the benefits they bring and that lower taxes induce additional economic growth, although the magnitude of these potential effects is in some dispute. Others criticize lower capital gains taxes as benefitting higher income individuals and express concerns about the budget effects; particularly in future years. Another criticism of lower rates is the possible role of a larger capital gains tax differential in encouraging tax sheltering activities and adding complexity to the tax law.

What Are Capital Gains and How Are They Taxed?

[2] Capital gain arises when an asset is sold and is the difference between the basis (normally the acquisition price) and the sales price. Corporate stock accounts for 20 to 80% of taxable gains, depending on stock market performance. Real estate is the remaining major source of capital gains, although gain arises from other assets (e.g., timber sales and collectibles). The appreciation in value can be real or reflect inflation. Corporate stock appreciates both because the firm's assets increase with reinvested earnings and because general price levels are rising. Appreciation in the value of property may simply reflect inflation. For depreciable assets, some of the gain may reflect the possibility that the property was depreciated too quickly.

[3] If the return to capital gains were to be effectively taxed at the statutory tax rate in the manner of other income, real gains would have to be taxed in the year they accrue. Current practice departs from this approach. Gains are not taxed until realized, benefitting from the deferral of taxes. (Taxes on interest income are due as the interest is accrued). Gains on an asset held until death may be passed on to heirs with the tax forgiven; if the asset is then sold, the gain is sales price less market value at the time of death, a treatment referred to as a "step-up in basis."

[4] Under current law, there is a maximum tax of 20% on capital gains held for a year, although the ordinary income tax rates reach as high as 39.6%. Where ordinary tax rates are 15 percent, the capital gains tax is 10 percent. Gain from assets held five years and acquired after 2000 will be subject to a maximum rate of 18 percent. For gains in the 15 percent bracket, an 8 percent rate applies to gain on assets held for five years and sold after 2000, with no required acquisition date. Gain arising from prior depreciation deductions is taxed at ordinary rates, but there is a 25 percent ceiling rate on the gain from attributable to prior straight-line depreciation on real property.

[5] Under prior law, several rules permitted avoidance or deferral of the tax on gain on owner-occupied housing, including a provision allowing deferral of gain until a subsequent house is sold (rollover treatment) and a provision allowing a one-time exclusion of $125,000 of gain for those 55 and over. These provisions were replaced with a general $500,000 exclusion ($250,000 for a single individual), which cost only slightly more in revenue.

[6] In contrast to these provisions that benefit capital gains, capital gains are penalized because many of the gains that are subject to tax arise from inflation and therefore do not reflect real income.

A Brief History

[7] Capital gains were taxed when the income tax (with rates up to 7%) was imposed in 1913. An alternative rate of 12.5% was allowed in 1921 (the regular top rate was 73%). Tax rates were cut several times during the 1920s. Capital gain exclusions based on holding period were enacted in 1924, and modified in 1938, to deal with bunching of gains in one year. In 1942 a 50% exclusion was adopted, with an alternative rate of 25%. Over time, the top rate on ordinary income varied, rising to 94% in the mid-1940s, then dropping to 70% after 1964. In 1969 a new minimum tax increased the gains tax for some; the 25% alternative tax was repealed.

[8] In 1978 the minimum tax on capital gains was repealed and the exclusion increased to 60% with a maximum rate of 28% (0.4 times 0.7). The top rate on ordinary income was reduced to 50% in 1981, reducing the capital gains rate to 20% (0.4 times 0.5). The Tax Reform Act of 1986 reduced tax rates further, but, in order to maintain distributional neutrality, eliminated some tax preferences, including the exclusion for capital gains. This treatment brough the rate for high income individuals in line with the rate on ordinary income -- 28%.

[9] In 1989, President Bush proposed a top rate of 15%, halving top rates. The Ways and Means Committee considered two proposals: Chairman Rostenkowski proposed to index capital gains, and Representatives Jenkins, Flippo, and Archer proposed a 30% capital gains exclusion through 1991 followed by inflation indexation. This latter measure was approved by the Committee, but was not enacted.

[10] In 1990, the President proposed a 30% exclusion, setting the rate at 19.6% for high income individuals. The House also passed a 50% exclusion with a lifetime maximum ceiling and a $1,000 annual exclusion, but this provision was not enacted into law. When rates on high income individuals were set at 31%, however, the capital gains rate was capped at 28%.

[11] In 1991, the President again proposed a 30% exclusion, but no action was taken. In 1992, the President proposed a 45% exclusion. The House adopted a proposal for indexation for inflation for newly acquired assets: the Senate passed a separate set of graduated rates on capital gains that tended to benefit more moderate income individuals. This latter provision was included in a bill (H.R. 4210) containing many other tax provisions that was vetoed by the President.

[12] No changes were proposed by President Clinton or adopted in 1993 and 1994 with the exception of a narrowly targeted benefit for small business stock adopted in 1993. The value of the tax cap was increased, however, in 1993 when new brackets of 36% and 39.6% were added for ordinary income.

[13] In 1994, the "Contract With America" proposed a 50% exclusion for capital gains, and indexing the basis for all subsequent inflation, while eliminating the 28% cap; this exclusion would be about a 40% reduction on average from current rates. The Ways and Means Committee reported out H.R. 1215 which restricted inflation indexing to newly acquired assets (individuals could "mark to market" -- pay tax on the difference between fair market value and basis as if the property were sold to qualify for indexation), did not allow indexation to create losses and provided a flat 25% tax rate for corporations). The 1995 reconciliation bill (H.R. 2491) that was vetoed by the President, included these revisions but delayed the indexation provision until 2002. During the 1996 presidential election, Mr. Dole proposed a slightly larger capital gains cut, and both candidates supported elimination of capital gains taxes on virtually all gains from home sales.

[14] In 1997, the President and Congress agreed to a tax cut as part of the reconciliation. The Administration tax cut proposal included the change in tax treatment of owner occupied housing. The House bill included a reduction in the 15% and 28% rates to 10% and 20%, about a 30% cut. Capital gains would also be indexed for assets acquired after 2000 and held for 3 years; mark-to-market would also be allowed. The Senate and the final bill did not include indexing. The capital gains issue was briefly revisited in 1998, when the holding period for long term gains was moved back from the 18 months set in 1997 to the one year period that has typically applied.

Revenue Effects

[15] Over the past several years, a debate has ensued regarding the revenue cost of cutting capital gains taxes (see The Revenue Cost of Cutting Capital Gains Tax Rates, CRS Report 97-559 E for further discussion). For example, in 1990 when the President proposed a 30% exclusion, Treasury estimates showed a $12 billion gain in revenue over the first 5 years, while the Joint Committee on Taxation found a revenue loss of approximately equal size.

[16] Although the estimates seemed quite different, they both incorporated significant expected increases in the amount of gains realized as a result of the tax cut. For example, the Treasury would have estimated a revenue loss of $80 billion over 5 years with no behavioral response, and the Joint Tax Committee a loss of $100 billion. (The gap between these static estimates arose from differences in projections of expected capital gains, a volatile series that is quite difficult to estimate).

[17] Empirical evidence on capital gains realizations does not clearly point to a specific response and revenue cost. Recent research suggests long-run responses may be more modest than those suggested by the economics literature during the 1990 debate, but the short-run response is still difficult to ascertain. (For a survey of this literature, see Jane G. Gravelle, The Economic Effects of Taxing Capital Income, Washington, D.C.: MIT Press, 1984, pp. 143-151.)

[18] Any revenue feedback effect will be smaller the larger the tax reduction. When the tax reduction is large, although there will be a larger response, any induced revenues will be taxed at the new lower rates. Thus, a 50% exclusion will not have as large a feedback effect relative to the static estimate as a 30% exclusion. Moreover, allowing a prospective tax cut that depends on selling and acquiring a new asset to qualify (or marking to market) as is the case for the reduction from 20 percent to 18 percent for five year property causes a gain in the short run as well.

[19] Arguments have also been made that a capital gains tax cut will induce additional savings, also resulting in a feedback effect as taxes are imposed on new income. This effect is uncertain, as it is not clear that an increase in the rate of return will increase savings (savings can decrease if the income effect is more powerful than the substitution effect) and what the magnitude of the response might be. (See Congressional Budget Office Memorandum, An Analysis of the Potential Macroeconomic Effects of the Economic Growth Act of 1998). There is also a debate about the effect of the capital gains tax on growth through its effect on innovation. Regardless of these empirical uncertainties, any effect of savings on taxable income in the short run is likely to be quite small due to the slow rate of capital accumulation. (Net savings are typically only about 2% to 3% of the capital stock, so that even a 10% increase in the savings rate would result in only a 2/10 to 3/10 of a percent first-year increase in the capital stock.) A related argument is that the tax cut will increase asset values; such an effect is only temporary, however, and will, if it occurs, only shift revenues from the future to the present. (For a discussion of savings and asset valuations, see testimony of Jane G. Gravelle, Congressional Research Service before the Senate Finance Committee, February 15, 1995 and the House Ways and Means Committee, March 19, 1997.)

Impact on Effective Tax Burdens

[20] The tax burden on an investment is influenced by both the tax rate and any benefits allowed or penalties imposed. One way to measure this tax burden is to calculate a marginal effective tax rate that captures in a single number all of the factors that affect tax burden. It is the percentage difference between the before- and after-tax return to investment, or the estimated statutory rate that would be applied to economic income to give the taxpayer the same burden as the combination of tax benefits and penalties.

[21] The effective tax rate on capital gains can be either higher or lower than the statutory rate, depending on the inflation rate relative to the real appreciation rate and the holding period. Also, assets held until death are not subject to tax. For example, under prior law, assuming a 20% statutory tax rate, the gain on a growth stock (paying no dividends) with a real appreciation rate of 7% and an inflation rate of 3% would, if held for one year, 7 years, 20 years, and until death, be subject respectively to tax rates of 27%, 22%, 14%, and 0%. In the future, the rate on 7- and 20-year assets would be 18 percent, for effective tax rates of 19% and 12%). (With no inflation and a 20 percent rate, the rates would be 19%, 16%, 12% and 0%; inflation penalizes assets held a shorter period more heavily than assets held for a longer period). Since less than half of gains that are accrued are realized, the effective tax rate is probably lower than the statutory tax rate. These benefits are larger for individuals who in the highest tax brackets (31%, 36%, and 39.6%) because the capital gains tax rate is capped at 20%.

Issues: Efficiency, Growth, Distribution, and Complexity

[22] One argument in favor of reducing the capital gains tax is the lock-in effect. If this effect is large, the tax introduces significant distortions in behavior with relatively little revenue gain. Another way to reduce lock-in is accrual taxation (i.e., tax gains on a current basis as accrued), but this approach is only feasible when assets can be easily valued (e.g., publicly traded corporate stock). Another way to reduce lock-in is to tax gains passed on at death. These solutions may face technical problems and taxation of gains at death has been unpopular.

[23] In the case of owner-occupied housing, although there were many ways to avoid the tax under current law, the rules may have resulted in individuals remaining in houses that are too large if economic circumstances have declined (for example, through job loss), retirement has brought a preference for a smaller home, or there is relocation to a lower-cost area.

[24] A case might be made for lower capital gains taxes on corporate stock because corporate equity capital is subject to heavy taxation. This heavy taxation encourages corporations to take on too much debt and directs too much capital to the noncorporate sector. On the other hand, lower capital gains taxes increase the relative penalty that applies to dividends and introduce tax distortions in the decisions of the firm to retain earnings.

[25] Arguments have also been made that lower gains taxes will increase economic growth and entrepreneurship. While evidence on the effect of tax cuts on savings rates and, thus, economic growth is difficult to obtain, most evidence does not indicate a large response of savings to an increase in the rate of return. Indeed, not all studies find a positive response, because a higher rate of return may allow individuals to save less while reaching their desired goal. (See Gravelle, The Economics of Taxing Capital Income, MIT Press, pp. 24-28, for a survey; see.) A more effective route to increasing savings may be to take revenues that might otherwise finance a tax cut and reduce the debt.

[26] Although arguments are made that lower gains taxes stimulate innovation and entrepreneurship, there is little evidence in history to connect periods of technical advance with lower taxes or even high rates of return. The extent to which entrepreneurs take tax considerations into account is unclear; however, there is some reason to doubt that capital gains taxes are important in obtaining large amounts of venture capital, in part because much of this capital is supplied by those not subject to the capital gains tax (i.e., pension funds, foreign investors). (See CRS Report RL30040, Capital Gains Taxes: Innovation and Growth, for a further discussion of capital gains taxes and venture capital.) Moreover, there is no evidence that longer corporate stock holding periods lead to more investments in long-term assets, including R&D, a rationale for lowering rates for assets with longer holding periods. (Note that indexing, initially proposed, and then dropped, favors assets with shorter holding periods).

[27] A major complaint made by some about lower gains rates cut is that they primarily benefit very high income individuals. Capital gains are concentrated among higher income individuals because these individuals tend to own capital and because they are likely to own capital that generates capital gains. For example, the Treasury indicated in 1990 that 54% of the proposed capital gains tax cut would go to individuals with incomes over $200,000 and 74% would go to individuals with incomes over $100,000 (using a 5-year average expanded income definition to classify taxpayers). Individuals with $200,000 of income account for about 1% of taxpayers and individuals with incomes over $100,000 account for less than 5%. The distributional effects of the capital gains relief for homes is somewhat less concentrated at the higher end (although lower income individuals are much less likely to own homes). The revisions may also enhance horizontal equity by treating taxpayer in different circumstances more evenly.

[28] Critics of lower capital gains taxes cite the contribution of preferential capital gains treatment to tax sheltering activities and complexity. For example, individuals may borrow (while deducting interest in full) to make investments that are eligible for lower capital gains tax rates, thereby earning high rates of return. These effects are, however, constrained in some cases (i.e., a passive loss restriction limits the deductions allowed in real estate ventures). Capital gains differentials complicate the tax law, especially as applied to depreciable assets where capital gains treatment can create incentives for churning assets unless recapture provisions are adopted. Indexation of capital gains for inflation is more complicated than a simple exclusion, since different basis adjustments must apply to different vintages of assets and proper indexation of gains on depreciable property is especially difficult. Thus, foregoing indexing probably kept the tax law simpler.