Date: Mon, 27 Nov 1995 15:24:41 -0800
From: Richard Ira Lavine <dlavine@tenet.edu> To: Multiple recipients of list <pen-l@anthrax.ecst.csuchico.edu> Subject: [PEN-L:1593] Income Distribution

Here is an recent analysis of some previously unpublished data from the Congressional Budget Office. The work is by the Center for Budget and Policy Priorities, a liberal think tank.

Original analysis included charts that could not be posted. Please contact Michelle Bazie at CBPP for a hardcopy of the report.

Revised November 22, 1995

UNEQUAL SHARES: RECENT INCOME TRENDS AMONG THE WEALTHY

BY ISAAC SHAPIRO

The budget debate this year has frequently focused on whether high-income families are being treated appropriately by plans to balance the budget over the next seven years while cutting taxes. To help put this discussion into context, this analysis exami nes how recent economic and policy trends have affected high-income and wealthy families. The analysis relies heavily on new data from the Congressional Budget Office, as well as on data from the Census Bureau. It finds:

*Both short-term and long-term trends in income growth have heavily favored high-income and wealthy families. Their incomes have risen at a much faster pace in recent decades than other groups, and they are the only group which has recaptured the ground lost in the recent recession. After adjusting for inflation, the average income of the wealthiest one percent of the population has nearly doubled since 1977, while the incomes of middle-class families have been stagnant and low-income families have expe rienced a significant loss.

*Although federal tax changes since the mid-1980s have improved the progressivity of the tax system, the proportion of income that the wealthy pay in federal taxes remains lower than in 1977.

*In the past two decades, the share of national income received by high-income and wealthy families has risen substantially. The wealthiest one percent of the population (2.5 million people) now has nearly as much after-tax income as the bottom 40 percen t of the population (about 100 million people). The top 20 percent of the population has fully as much after-tax income as the bottom 80 percent of the population. Income inequality is at the widest level on record.

In this analysis, the term "wealthy" generally applies to the richest one percent of families. According to the CBO data, the top one percent of families are projected to have AFTER-tax income averaging $438,000 in 1996; they receive 12 percent of nation al after-tax income. The term "high-income" generally applies to the 20 percent of families with the highest incomes. Their after-tax incomes are projected to average $89,000 in 1996; they receive half of national after-tax income.

In light of the above trends, it would seem inadvisable to carve a path towards a balanced budget under which the incomes of wealthy families are likely to increase further while the incomes of other families -- and especially low- and moderate-income fam ilies -- are likely to decrease substantially. But, as the last part of this analysis finds, that is the current Congressional course.

INCOME TRENDS

Most public information on income trends is from the Census Bureau. Census data, however, do not include information on income trends for the top one percent of the population, nor do they focus on the effect of taxes on incomes. For this reason, this a nalysis also relies heavily on information from the Congressional Budget Office. For trends over time, this analysis uses unpublished CBO data covering the period from 1977 (the first year of this CBO series) to 1992 (the latest year for which actual, as opposed to projected, figures are available).(1) The data show:

*From 1977 to 1992, the average AFTER-tax income of the wealthiest one percent of the population nearly doubled, increasing by 91 percent after adjusting for inflation.(2) The average after-tax income of the top fifth of families rose 28 percent during this period.

*In sharp contrast, the average income of the middle fifth of families was essentially stagnant, ticking up by just one percent over this 15-year span. The bottom fifth of families experienced an average decline in after-tax income of 17 percent. (See F igure 1.)

More recent data for BEFORE-tax income are available from the Census Bureau for 1994. They show that the average income of the bottom 80 percent of the population is well below what it was in 1989 -- the last year before recession set in -- and that only the top fifth of households(3) has recovered fully from the recession.

*The Census data show that the average income of the bottom fifth of households in 1994 was 7.5 percent -- or $629 -- below its level in 1989, after adjusting for inflation. The average income of the middle fifth was 6.3 percent -- or $2,185 -- below its 1989 level.

*By contrast, the average income of the top fifth was about one percent -- or at least $1,000 -- higher in 1994 than it was in 1989. The average income of the top five percent of households rose even more, increasing by at least four percent -- or $7,000 -- over this period. (The Census Bureau data do not include information for the top one percent of households.)

These developments primarily reflect trends in before-tax incomes, including changes in the private economy. Earnings trends have diverged widely for different groups, with those at the top of the wage scale gaining substantially, and the rest of workers receiving wages that have been essentially stagnant or have declined. For the wealthy, capital gains income also has risen substantially since 1977.

TAX TRENDS

The federal tax system is progressive; that is, higher-income families pay a higher proportion of their income in federal taxes than middle- and lower income families do. The progressive nature of the tax system reflects the generally accepted principle that federal tax burdens should, in part, reflect a family's "ability to pay."

But the system is only modestly progressive, as CBO data on projected federal tax burdens for 1996 illustrate. (Projected data for 1996 are used here in order to capture the full effects of tax change enacted in 1993.) The CBO data show that the top one percent of families are projected to have 13.4 percent of the nation's pre-tax income in 1996 and to pay 18.4 percent of the nation's federal taxes. The top 20 percent of families are projected to have 52.1 percent of the nation's income and to pay 61.4 percent of the nation's taxes. It also should be kept in mind that the progressivity of federal taxes is muted by the regressivity of state and local taxes. State and local taxes consume a larger proportion of income of low- and middle-income families than of upper-income families.

Of further interest, while the federal tax system has become more progressive since the mid-1980s, CBO data show that wealthy families still pay a smaller share of their income in federal taxes than they did in 1977. The wealthiest one percent of familie s will pay 32.7 percent of their income in federal taxes in 1996 under current law; they paid 35.4 percent of their income in federal taxes in 1977. If the wealthiest one percent of families paid the same percentage of their income in taxes in 1996 as in 1977, they would owe another $19 billion in taxes next year.

INCOME INEQUALITY

The divergence in the extent to which different groups have benefited from recent economic trends has led to a dramatic growth in income inequality. As the CBO data show:

*In 1992, the wealthiest one percent of families received 12.1 percent of the nation's after-tax income while the bottom 40 percent received 14 percent. That year the richest 2.5 million Americans had nearly as much after-tax income as the 98 million Ame ricans with the lowest incomes. (See Figure 2.)

This is in sharp contrast to 1977, when the total after-tax income of the bottom 40 percent of Americans was MORE THAN DOUBLE the total after-tax income of the richest one percent.

*The top 20 percent of families received 49.8 percent of the nation's after-tax income in 1992, up from 44 percent in 1977. This means the top fifth of the population received as much after-tax income in 1992 as the other 80 percent of the population com bined.

Some of the changes in after-tax income shares in Figure 2 may appear relatively small, such as the decline in the after-tax income share of the middle fifth of the population from 16.3 percent in 1977 to 15.2 percent in 1992. But seemingly small shifts in income shares reflect differences of tens of billions of dollars in the amounts of income going to various groups.

*If the middle fifth of the population had received the same share of total after-tax income in 1992 that it did in 1977, the families in this income bracket would have received $37 billion more in income in 1992. This is the equivalent of $1,790 more pe r family.

*Similarly, if the poorest fifth of the population had the same share of total after-tax income in 1992 that it had in 1977, those families would have received $54 billion more in income in 1992, or $2,670 more per family.

*The story is quite different for the wealthy. If the richest one percent of the population had the same share of total after-tax income in 1992 as in 1977, those families would have had $162 billion LESS in income, or $154,000 less per family. (Stated differently, because of the rise in their share of national after-tax income, the richest one percent of families were $154,000 better off per family than if their share of national after-tax income had remained constant.)

Census data for before-tax income that cover a longer time period are available; they depict the same trend towards greater income inequality. The Census data show that in 1994, the shares of national before-tax income held by the top 20 percent of house holds and by the top five percent of households were the highest ever recorded, with annual data going back to 1947.(4) Meanwhile, the share of national before-tax income held by the bottom 40 percent of households was at the lowest level ever recorded . (Census data also show that the inclusion of noncash or "in-kind" benefits in these calculations have little effect on these figures.)(5)

Also of interest, two major international studies that have been published recently found that income disparities are wider in the United States than in all other Western industrialized nations examined. The first study focused on children and found that affluent children in the United States are better off than their counterparts in all of the other 17 industrialized nations studied.(6) At the same time, the study found that low-income children in the United States are poorer than low-income children in 15 of the 17 other industrialized nations. As a result, the income gap between affluent children and poor children is much wider in the United States than in any of the other nations examined.

The second study examined all individuals.(7) It also found that the gap between the incomes of rich and poor individuals was wider in the United States than in any of the 17 countries examined (all of which are industrialized members of the Organizati on for Economic Cooperation and Development, OECD). Of further interest, this study found that the average level of government income assistance to low-income people in the United States was the second lowest of all the countries examined.

INCOME LEVELS

This section examines the average after-tax income levels of various groups of families. Current CBO projections for 1996 are again used here to reflect the effects of the tax changes in the 1993 budget agreement.

*The average after-tax income of the top one percent of families is projected to equal $438,000 in 1996. For the top 20 percent of families, it will average $89,000.

*In contrast, the middle fifth of families will have average after-tax income of $28,500, one-fifteenth the income of the wealthiest one percent and one-third the income of the top 20 percent.

*The bottom fifth of families will have average after-tax income of $8,230.(8)

THE CURRENT CONGRESSIONAL COURSE

Balancing the budget entails sacrifice on the part of many Americans. How these sacrifices should be apportioned among various income groups constitutes an important question for policymakers.

The income trends discussed here suggest that at a minimum, the sacrifices should be evenly shared. Preferably, those who can best afford some sacrifice and have fared best economically in recent years -- people at the highest income levels -- should be asked to bear something more than a proportionate share of the burden.

Congress, however, is charting a very different path -- one in which the poorest families are asked to sacrifice the most and the wealthiest families, rather than bearing a share of the sacrifice, are to have their after-tax incomes increased further. Va rious Center on Budget and Policy Priorities analyses elaborate on these issues, but a few key findings help illustrate the point.

• The tax provisions of the Congressional reconciliation bill would sharply exacerbate income tax disparities, with wealthy families gaining much and many low-income families actually facing tax increases. Even the child tax credit would be inequitably d istributed. Between now and the year 2000, the wealthiest 20 percent of taxpayers would receive $38 billion more from the credit than the bottom 40 percent of taxpayers, according to a Center analysis of Joint Tax Committee data. Nearly half of all child ren -- 44 percent -- would receive either a partial credit or no credit because they live in families with incomes too low to qualify for a full credit. Almost 24 million children would receive no aid from the credit.

The gains from other provisions of the Congressional tax package -- such as the large reductions in capital gains taxes and the liberalization of IRAs -- are much more skewed towards the wealthy than the child tax credit. As a result, the wealthiest one percent of families would ultimately gain an average of $8,450 apiece from the tax package each year, according to an analysis by the Department of Treasury.(9) The Treasury Department estimates that the top one percent of families would receive 17 per cent of the tax benefits from the Congressional reconciliation bill when it is fully phased in, while the top 20 percent of families would receive 63 percent -- or well over half -- of the benefits from the tax package.(10)

By contrast, as a group low- and moderate income families would not gain significantly from the reconciliation tax package. Indeed, Treasury Department data indicate that 7.7 million households -- including 3.3 million families with children -- would be worse off than under current law under the family income tax provisions of the package. The Treasury analysis examined the full effects of the cuts to the Earned Income Tax Credit -- a tax credit primarily benefiting low- and moderate-income families wit h children -- as well as the benefits from the new child tax credit and the standard deduction increase.(11)

• Programmatic cuts, in turn, are skewed against low-income families. Of the $1 trillion in budget cuts over seven years in the Congressional reconciliation bill, about $400 billion would come from programs targeted on low-income households.(12) The p roportion of the budget cuts that these programs would absorb would be about twice their share of the federal budget.

In fact the reductions in means-tested entitlements reflected in the Congressional budget are so deep that by 2002, even after adjusting for inflation, these reductions will be seven times larger than the total of all means-tested entitlement cuts enacted under President Reagan in 1981 and 1982, when the Reagan-era cuts in low-income programs reached their deepest point.(13)

Further evidence of the disparate impact of the Congressional spending reductions comes from a Clinton Administration analysis of the budget cuts that have moved through Congress. Even though the analysis did not cover all of the proposed budget cuts, it found the bottom fifth of families with children would eventually lose an average of more than $2,000 a year in income and health benefits. By contrast, the top fifth of families -- those who would receive the most generous tax cuts -- would lose an ave rage of less than $200 in income and health benefits from the budget reductions, according to the Administration data.(14)

(1) The CBO data described here were released to selected Congressional staff at the end of June but, until now, have not been generally released to the public.

(2) The CBO figures are based primarily on Census income data but reflect a fuller measure of income than Census data do. CBO also assigns families to different groups in a somewhat different manner than the Census Bureau traditionally does. The Census Bureau groups families strictly by the amount of income they have, so that the 20 percent of families with the lowest incomes become the bottom fifth or "quintile". CBO, on the other hand, accounts for the fact that larger families have greater income n eeds than smaller families. Accordingly, it adjusts each family's income according to the family's size before assigning the family to a quintile. This is a superior method, since a $20,000 income is far more ample for a single-person family than for, s ay, a family of five.

(3) CBO uses the term "families" to describe the income units in its data tables. Its definition of families includes people living alone and thus in many respects corresponds to the Census definition of "households".

(4) While the various Census indicators of income inequality were at record levels in 1994, the one-year changes between 1993 and 1994 were not statistically significant. The Census Bureau's technical measure of income inequality -- the "Gini" index -- was also at its all-time highest level.

(5) The standard Census data used in the text and the CBO data do not account for the value of government in-kind benefits received by households. Since low- and moderate-income households receive more assistance from government in-kind benefits than hi gh-income households, these benefits do moderate income disparities somewhat. The effect, however, is relatively small.

This can be seen by examining experimental measures of income that the Census Bureau uses. These measures -- which provide information back to 1979 -- incorporate both sources of income that make income distribution more unequal (e.g., capital gains inco me) and tax and in-kind benefit policies that lessen income disparities.

Under all these experimental measures, income inequality in 1994 was greater than in any other year in the time series, with the exception of 1986. Even in comparison to 1986, some measures of income inequality were higher in 1994, while others were lowe r. Moreover, the 1986 data are skewed at the top because of a one-time jump in capital gains income. This jump reflected the "cashing in" of capital gains before higher capital gains tax rates took effect in 1987, as a result of the 1986 Tax Reform Act.

Also of note, the share of national resources received by the bottom fifth of households does not increase dramatically when in-kind benefits are considered. Even when an expansive measurement of in-kind benefits that includes some valuation of Medicare and Medicaid benefits is used, the share of national income received by the bottom fifth of households rises by only 0.7 percentage points. Excluding Medicare and Medicaid, their share rises by 0.5 percentage points.

Finally, while GOVERNMENT in-kind benefits tend to moderate income inequality somewhat, PRIVATE in-kind or noncash benefits -- such as health or pension benefits provided through a job -- are of relatively little help to low-income families. Low-wage job s frequently do not provide such benefits. If these private in-kind benefits were included in these calculations, low-income families would be shown to receive a smaller share of national income. (6) Lee Rainwater and Timothy M. Smeeding, Luxembourg Income Study, DOING POORLY: THE REAL INCOME OF AMERICAN CHILDREN IN A COMPARATIVE PERSPECTIVE, August 1995.

(7) Anthony Atkinson, Lee Rainwater, and Timothy Smeeding, WORKING PAPER NO. 120, INCOME DISTRIBUTION IN ADVANCED ECONOMIES: EVIDENCE FROM THE LUXEMBOURG INCOME STUDY, October 1995.

(8) The second poorest fifth is projected to have average after-tax income of $18,367 and the second richest fifth $41,051. Switching to before-tax income, the projected average figures are: bottom fifth $8,665; second poorest fifth $21,578; middle fift h $35,536; second richest fifth $53,020; top fifth $123,749; and top one percent $651,274.

(9) The Citizens for Tax Justice estimates that the top one percent of taxpayers would ultimately gain an average of $10,475 each year under the reconciliation package.

(10) The Treasury data also show that the next-to-top fifth of families would ultimately receive 24 percent of the annual benefits from the Congressional tax package, the middle fifth would receive 11 percent of the benefits, and the next-to-lowest fifth a scant one percent. On average the bottom fifth of families would owe slightly MORE in taxes under the reconciliation bill, because the cuts in the EITC would slightly more than offset the reductions in other taxes.

(11) The Joint Tax Committee data that fully account for the EITC cuts are also of interest. They show that from 1995 to 2000, as a group taxpayers with incomes between $0 and $10,000 would lose $2.7 billion under the reconciliation tax package as compa red to current law; taxpayers with incomes between $10,000 and $20,000 would gain $1 billion; and taxpayers with incomes between $20,000 and $30,000 would gain $700 million (though from 1997 to 1999 this group of taxpayers would be worse off compared to c urrent law). In combination, as a group taxpayers with incomes below $30,000 would be $1 billion worse off between now and the year 2000.

(12) The $1 trillion in cuts required under the budget resolution reflects the dimension of the cuts when they are measured from CBO's "capped baseline." This has been CBO's standard way of measuring budget reductions. On November 16, CBO Director June O'Neill reported to Congress that its budget reconciliation legislation would produce budget cuts of $1.017 trillion relative to CBO's baseline projection.

(13) The figures used here for the size of the means-tested entitlement cuts enacted in 1981 and 1982 are those of the Congressional Budget Office, as published in a 1983 CBO report and as adjusted for inflation to 2002.

(14) The Administration estimated that the House budget plan would yield direct income and health benefit cuts that averaged $2,820 for the bottom fifth of families of children while the Senate budget plan cut these benefits by an average of $2,024 for t hese families. The Administration has not yet released an estimate of the average cuts under the reconciliation bill, but it will fall between the House and Senate figures. The Administration analysis is POTENTIAL POVERTY AND DISTRIBUTIONAL EFFECTS OF WELFARE REFORM BILLS AND BALANCED BUDGET PLANS, November 9, 1995.

For more information, contact:

Michelle Bazie
HN0026
Center on Budget and Policy Priorities
777 N. Capitol St., NE Suite 705
Washington, DC 20002
(202) 408-1080