Skip Navigation
| Skip Navigation |
|
|
|
|
|
SUMMARY
I. INTRODUCTION
II. ACTUAL CHANGES IN THE DISTRIBUTION OF TAXES AND INCOME, 1980-1983
III. STATIC EFFECTS OF THE TAX CUTS IN ERTA
IV. BEHAVIORAL RESPONSES TO TAX POLICY CHANGES: THE EVIDENCE FROM 1980-1983
APPENDIXES
A. Other Analyses of Distributional Effects of the 1981 Tax Cuts
B. Definition of Expanded Adjusted Gross Income
C. Derivation of Equations in Chapter
D. Changes in AGI/EAGI, Taxable Income/AGI, and Taxes/Taxable Income
E. The Gini Coefficient and the Suits Index
F. The CBO Capital Gains Equation
| TABLES | |
| I.1 | Steps in Construction of Baseline and in the Measurement of Static and Feedback Effects |
| I.2 | Expanded Adjusted Gross Income Groups |
| II.1 | Distribution of Income Taxes: 1980-1983 |
| II.2 | Overview: Percentage Growth in Income Taxes, 1980-1983 |
| II.3 | Percentage Growth of Income Taxes: 1980-1983 |
| II.4 | Weighted Growth of Components of EAGI: 1980-1983 |
| II.5 | Distribution of Taxes Paid, Before-Tax EAGI and After-Tax EAGI |
| II.6 | Summary Measures of Inequality of Income Distribution and Tax Progressivity |
| II.7 | Real Expanded Adjusted Gross Income Per Return |
| III.1 | Percent Changes in Statutory Marginal Tax Rates; Joint Returns: 1980-1983 |
| III.2 | Percent Changes in Tax Liability; Selected Taxable Income Levels; Joint Returns: 1980-1983 |
| III.3 | Static Estimates: Tax Liabilities in 1983 Simulated Under 1980 and 1983 Law |
| III.4 | 1983 Deductions for Contributions to IRAs and for Two-Earner Married Couples |
| III.5 | Share of Tax Liabilities in 1983: Simulated Under 1980 and 1983 Law |
| IV.1 | Effects of Tax Policy Changes: Comparison of Projected and Actual Taxes Paid by Income Groups |
| IV.2 | Comparison of Actual and Projected Income and Deductions Per Return: Top Percentile of Returns |
| IV.3 | Growth in Capital Gains and Changes in Marginal Tax Rates on Gains: 1980-1983 |
| IV.4 | Estimated Effects of Different Factors on Capital Gains Realizations: 1980-1983 |
| IV.5 | Estimated Effects of Additional Capital Gains Realizations on Taxes Paid by Top Percentile of Returns: 1983 |
| IV.6 | Growth in Wages and Salaries and Changes in Marginal Tax Rates on Wages and Salaries: 1980-1983 |
| A.1 | Rates of Growth of Components of Income, from National Income and Product Accounts |
| D.1 | Explanation of Percentage Change in Ratio of Adjusted Gross Income (AGI) to Expanded Adjusted Gross Income (EAGI): 1980-1983 |
| D.2 | Explanation of Percentage Change in Ratio of Taxable Income (TY) to Adjusted Gross Income (AGI): 1980-1983 |
| D.3 | Explanation of Percentage Change in Itemized Deductions: 1980-1983 |
| D.4 | Annual Growth of Tax/Taxable Income: 1980-1983 |
In the Economic Recovery Tax Act of 1981 (ERTA), the Congress lowered the top marginal tax rate on individual income from 70 to 50 percent, reduced other marginal tax rates by 23 percent over a three-year period, and enacted a number of other provisions that reduced individual tax payments and lowered taxes on the business income of both individuals and corporations. The corporate tax reductions in ERTA were partially offset later by the provisions of the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA).
ERTA also provided for indexation of personal exemptions, the zero bracket amount (ZBA), and the width of tax brackets to changes in the Consumer Price Index beginning in 1985. Because indexing did not take effect immediately, however, the real value of personal exemptions, the ZBA, and bracket widths continued to decline between 1980 and 1984.
This paper examines the effects of ERTA and TEFRA on changes in the
distributions of individual income tax payments and after-tax incomes between
1980 and 1983. The total change in the distribution of tax payments is
separated into a "static" component attributable only to the tax changes,
and a component labelled "feedback and other" that is attributable to changes
in the distribution of pretax income. The latter changes reflect effects
of the tax changes on the percentage of income received in taxable forms
in different income classes, and the effects of changes in economic conditions.
Particular attention is directed toward behavioral responses, especially
those for taxpayers in the upper 1 percent of the income distribution.
OVERALL DISTRIBUTIONAL CHANGES
Between 1980 and 1983, the share of individual income taxes paid by taxpayers in the top 1 percent of the income distribution increased from 19.1 percent to 20.6 percent. This increase occurred even though the group experienced the largest reduction in average tax rates. Other taxpayers in the top half of the income distribution paid a lower share of taxes over this period, notably those between the 2nd and 25th percentiles of the income distribution, whose share fell from 54.1 percent to 52.7 percent. The share of taxes paid by taxpayers in the next highest quartile fell slightly, while the share of taxes paid by taxpayers in the bottom half of the income distribution increased slightly from 6.9 percent to 7.0 percent.
The principal reason why those in the top percentile paid an increased share of taxes was that their incomes grew faster. Income for this group increased by 42.4 percent between 1980 and 1983, compared to a 24.5 percent growth for income averaged over all returns. A major component of this relatively greater income growth was realized capital gains. For the top percentile, realized capital gains increased by 89 percent between 1980 and 1983 and were responsible for more than the entire difference between the growth in income in the top percentile and the growth averaged over all returns.
The tax system was less progressive in 1983 than in 1980, despite the increased share of taxes paid by the top percentile. Summary measures based on the distribution of after-tax income, arguably the best way to determine progressivity, show that the distribution of after-tax income was less equal in 1983 and that the tax system had a smaller effect in reducing inequality.
It is important to note that an increase in the share of after-tax income received by high-income groups does not necessarily mean that other groups are becoming worse off in absolute terms. Tax reductions that raise the income share and tax payments of upper income groups can also increase the after-tax incomes of lower income groups if (1) increased saving or work effort by those in the top bracket, by adding to the capital stock or the availability of skilled labor, increases real wages for all groups over time or (2) if higher tax payments by upper-income groups allow for larger tax reductions for lower-income groups. The first of these effects would be expected to appear only in the longer term, while the second would occur only as a result of subsequent legislative action.
Such effects cannot be detected in the 1980-1983 data. During this period, the real after-tax income per return in the bottom half of the income distribution declined by almost 3 percent and remained virtually constant for returns in the next highest 25 percent of the income distribution. For the top percentile of returns, the increase in real after-tax income per return was almost 23 percent.
Individual income tax revenue in 1983 was about $40 billion below the level that would have resulted if all incomes and deductible expenses had grown at the same rate as average personal income per capita between 1980 and 1983, and if average tax rates had been held constant. As mentioned above, the estimated reduction in revenue can be decomposed into two parts: (1) a "static" component attributable to the direct effect of the tax cuts measured at a constant 1983 level and distribution of income and (2) a component labelled "feedback and other" attributable only to changes in the distribution, but not the level, of income between 1980 and 1983.
The component labelled "feedback and other" includes changes in taxable
income that may be tax-induced, but also could be independent of changes
in the law. These changes include increases in realized capital gains and
earnings, shifts from tax-exempt or tax-deferred income to taxable income,
and reductions in tax-deductible expenditures. They may have been induced
by changes in the tax law, but may also have been caused by other economic
events, such as the 1981-1982 recession, the stock market boom in 1982-1983,
changes in market interest rates and deregulation of financial institutions.
Because total income growth in the no tax change case is assumed to be
the same as actual growth, the question of the effect of the tax cuts on
total economic growth is not addressed.
STATIC EFFECTS
The static analysis of ERTA shows that the direct benefit from tax changes between 1980 and 1983 was proportionately greater in the highest income groups. At a 1983 level and distribution of income, tax payments in the top percentile of tax returns were about 15 percent less than they would have been if 1980 tax law had remained in effect but had been indexed to average growth in per capita personal income. In contrast, tax payments in the second quartile of returns were 9 percent lower than they would have been under indexed 1980 law while tax payments in the bottom half of the income distribution were 3 percent higher than they would have been under indexed 1980 law.
This pattern of tax changes resulted from a combination of factors. The benefits from the reduction in marginal tax rates were distributed proportionately across income classes. The benefits from the major changes in the definition of the tax base went primarily to taxpayers in the upper part of the income distribution. The reduction of the value of personal exemptions, the ZBA, and bracket widths relative to income raised taxes proportionately the most for taxpayers in the bottom half of the income distribution. Consequently, when measured at the 1983 level and distribution of income, the relative position of taxpayers in the bottom half of the income distribution worsened between 1980 and 1983, while the position of taxpayers in the top 1 percent of the income distribution improved the most.
The static effect is the best measure of the initial impact of tax policy
changes on the well-being of taxpayers. But such changes can also affect
the economic decisions of taxpayers, who can increase their benefits or
reduce their losses by changing their behavior. These behavioral changes
may increase or decrease taxes paid, but will always make taxpayers better
off than they were immediately after the change in tax policy. For example,
if high-income taxpayers choose to sell assets so as to realize more capital
gains when tax rates are lowered, their tax payments will increase but
this will be more than offset by the increase in the value to them of higher
realized income. Because the total change in the distribution of tax payments
may include some induced behavioral response, it does not accurately reflect
the distribution of benefits from tax changes.
FEEDBACK AND OTHER EFFECTS
The study finds no evidence that behavioral responses to the tax cuts resulted in any overall revenue feedback effects for the vast majority of the taxpaying population. In the aggregate, changes in the distribution of income, the use of deductions, and the realization of taxable income relative to total personal income are estimated to have had virtually no effect on total taxes paid. The estimated total revenue loss of $39.5 billion was slightly greater than the estimated static loss of $38 billion.
For the top percentile, however, "feedback and other" effects offset over half (about 60 percent) of the static revenue reduction. Virtually all of this was accounted for by increased realizations of capital gains and increased wage and salary income. It is plausible that the higher incomes may have resulted from changes in behavior induced by the tax rate changes, which increased marginal after-tax income per dollar of pretax capital gains and wages by a greater percentage for people in this group than for other taxpayers.
The increase in capital gains in the top percentile between 1980 and 1983 can be explained in part by the reduction in marginal rates on capital gains and in part by the fact that over time realized capital gains have been rising proportionately faster than other income and also stock prices. The 1983 increase in capital gains realizations, however, was much larger than predicted by historical relationships, suggesting that tax rates were not responsible for all of the increase. The growth in wage and salary income in the top percentile may also be attributed in part to behavioral responses. However, in light of estimates from studies of labor supply behavior, the increase in wage and salary income exceeded the expected response to lower marginal tax rates. Another reason for the relatively larger wage and salary growth in high-income groups may have been the increase in unemployment between 1980 and 1983, because high unemployment disproportionately reduces earnings growth for low-income groups.
Once more, it is important to emphasize that this study does not incorporate all possible types of behavioral and feedback effects and in particular assumes that the overall growth rate of income was unaffected by the tax changes. Estimation of the effect of the tax changes on the overall growth rate during the 1980-1983 period is very difficult. It can be said, however, that the effect would have had to be very large to contradict the study's finding that the tax cut of 1981, as modified by TEFRA, reduced tax revenues. The overall growth rate was probably somewhat affected by behavioral responses, however, and such responses may have more effect in the longer run.
The study does provide insight into the differences among behavioral responses of specific income groups to changes in tax rates and suggests that future researchers should carefully examine such responses by the highest income class, especially the effect on capital gains realizations.
This document is available in its entirety in PDF.