Conservative economic policy has one central idea: just create a bigger pie, and everyone will have a bigger slice. In fact, conservatives predict that if we cut the rich a bigger piece by lowering their tax rates, the resulting growth will enlarge everyone else's slice, too. This was the core idea of Reagan's tax cuts, and it is central to such current conservative goals as lower capital gains taxes.
Unfortunately, since the 1980s the great majority of Americans have not been getting bigger slices from a growing pie. As many people have noted, median family income has failed to grow. The picture is even more stark for gains in wealth than for gains in income. New research, based on data from federal surveys, shows that between 1983 and 1989 the top 20 percent of wealth holders received 99 percent of the total gain in marketable wealth, while the bottom 80 percent of the population got only 1 percent. America produced a lot of new wealth in the '80s--indeed, the stock market boomed--but almost none of it filtered down.
Few people realize how extraordinarily concentrated the gains in wealth have been. Between 1983 and 1989 the top 1 percent of income recipients received about a third of the total increase in real income. But the richest 1 percent received an even bigger slice--62 percent--of the new wealth that was created (see figure at right).
The most recent data suggest these trends have continued. My preliminary estimates indicate that between 1989 and 1992, 68 percent of the increase in total household wealth went to the richest 1 percent--an even larger share of wealth gain than between 1983 and 1989. As a result, the concentration of wealth reached a postwar high in 1992, the latest year for which data are available. If these trends continue, the super rich will pull ahead of other Americans at an even faster pace in the 1990s than they did in the '80s.
Growing inequality in the distribution of wealth has serious implications for the kind of society we live in. Today, the average American family's wealth adds up to a comparatively meager $52,200, typically tied up in a home and some small investments. While Forbes magazine each year keeps listing record numbers of billionaires--in 1994 Forbes counted 65 of them in the U.S.--homeownership has been slipping since the mid-1970s. The percentage of Americans with private pensions has also been dropping. And, with their real incomes squeezed, middle-income families have not been putting savings aside for retirement. The number of young Americans going to college has also begun to decline, another indirect sign of the same underlying phenomenon. In fact, international data now indicate that wealth is more unequally distributed in the U.S. than in other developed countries, including that old symbol of class privilege, Great Britain.
Economic worries may be at the root of much of the political anger in America today, but there is almost no public debate about the growth in wealth inequality, much less the steps needed to reverse current trends. The debate needs to start with an understanding of how and why America's pie is getting sliced so unequally.
The increasing concentration of wealth in the past 15 years represents a reversal of the trend that had prevailed from the mid-1960s through the late 1970s. The share of total wealth owned by the rich depends, to a large extent, on asset values and therefore swings sharply with the stock market, but some trends stand out (see figure 2 above). During the twenty years after World War II, the richest 1 percent of Americans (the "super rich") generally held about a third of the nation's wealth. After hitting a postwar high of 37 percent in 1965, their share dropped to 22 percent as late as 1979. Since then, the share owned by the super rich has surged--almost doubling to 42 percent of the nation's wealth in 1992, according to my estimates.
Two statistics--median and mean family wealth--help to tell the story of growing wealth inequality in America. A median is the middle of a distribution, the point at which there are an equal number of cases above and below. Median family wealth represents the holdings of the average family. Mean family wealth is the average in a different sense: total wealth divided by the total number of families. If a few families account for a large bulk of the nation's wealth, the mean will exceed the median. The changing ratio between the mean and median is one measure of changes in wealth inequality.
Data from the 1983, 1989, and 1992 Survey of Consumer Finances conducted by the Department of Commerce show that mean wealth has indeed been much higher than the median: $220,000 versus $52,000 in 1992. And mean wealth has grown more rapidly--by 23 percent from 1983 to 1989, and by another 12 percent in the following three years, while median wealth increased by only 8 percent between 1983 and 1989 and then barely moved up at all from 1989 to 1992 (see the figure "Median and Mean Wealth and Income,"). As a result, between 1983 and 1989 the ratio of mean to median wealth jumped from 3.4 to 3.8.
Data for 1992 are as yet incomplete. However, the figures available indicate that the ratio of mean to median wealth saw another steep rise between 1989 and 1992, from 3.8 to 4.2. On the basis of a regression analysis of the historical relation between this ratio and the share of wealth held by the top 1 percent of families, I estimated their share at 42 percent in 1992.
Income inequality also increased over the same period. From 1983 to 1989, the share of the top 20 percent increased from 52 to 56 percent, while that of the remaining 80 percent decreased from 48.1 to 44.5 percent. The preliminary evidence from the 1992 Survey of Consumer Finances suggests a further rise in income inequality between 1989 and 1992 (the ratio of mean to median income increased from 1.57 to 1.61), though this change is much more muted than from 1983 to 1989 (when the ratio moved upward from 1.42 to 1.57).
By the 1980s the U.S. had become the most unequal industrialized country in terms of wealth. The top 1 percent of wealth holders controlled 39 percent of total household wealth in the United States in 1989, compared to 26 percent in France in 1986, about 25 percent in Canada in 1984, 18 percent in Great Britain, and 16 percent in Sweden in 1986. This is a marked turnaround from the early part of this century when the distribution of wealth was considerably more unequal in Europe (a 59 percent share of the top 1 percent in Britain in 1923 versus a 37 percent share in the U.S. in 1922).
The concept of wealth used here is marketable wealth--assets that can be sold on the market. It does not include consumer durables such as automobiles, televisions, furniture, and household appliances; these items are not easily resold, or their resale value typically does not reflect the value of their consumption to the household. Also excluded are pensions and the value of future Social Security benefits a family may receive.
Some critics of my work, such as the columnist Robert Samuelson of the Washington Post, argue that a broader definition of wealth shows less concentration. To be sure, including consumer durables, pensions, and entitlements to Social Security reduces the level of measured inequality. The value of consumer durables amounted to about 10 percent of marketable wealth in 1989; including them in the total reduces the share of the top 1 percent of wealth holders from 39 percent to 36 percent. Adding pensions and Social Security "wealth," which together totaled about two-thirds of marketable wealth, has a more pronounced effect, reducing the share of the top 1 percent from 36 percent to 22 percent. However, even though pensions and Social Security are a source of future income to families, they are not in their direct control and cannot be marketed. Social Security "wealth" depends on the commitment of future generations and Congresses to maintain benefit levels; it is not wealth in the ordinary meaning of the term.
Moreover, the inclusion of consumer durables, pensions, and Social Security does not affect trends in inequality. With these assets included, the share of the richest 1 percent reached its lowest level in 1976, at 13 percent, and nearly doubled by 1989 to 22 percent. Nor does it affect international comparisons. For example, using this broader concept of wealth, we still find that the share of the top 1 percent in the U.S. is almost double that of Britain in 1989--22 percent versus 12 percent.
WHY THE RICH GOT RICHER
Part of the explanation for growing wealth concentration lies in what has happened to the different kinds of assets that the rich and the middle class hold. Broadly speaking, wealth comes in four forms:
- liquid assets, including cash, bank deposits, money market funds, and savings in insurance and pension plans
- investment real estate and unincorporated businesses
- corporate stock, financial securities, and personal trusts.
Middle-class families have more than two-thirds of their wealth invested in their own home, which is probably responsible for the common misperception that housing is the major form of family wealth in America. Those families have another 17 percent in monetary savings of one form or another, with only a small amount in businesses, investment real estate, and stocks. The ratio of debt to assets is very high, at 59 percent.
In contrast, the super rich invest over 80 percent of their savings in investment real estate, unincorporated businesses, corporate stock, and financial securities. Housing accounts for only 7 percent of their wealth, and monetary savings another 11 percent. Their ratio of debt to assets is under 5 percent.
Viewed differently, more than 46 percent of all outstanding stock, over half of financial securities, trusts, and unincorporated businesses, and 40 percent of investment real estate belong to the super rich. The top 10 percent of families as a group account for about 90 percent of stock shares, bonds, trusts, and business equity, and 80 percent of non-home real estate. The bottom 90 percent are responsible for 70 percent of the indebtedness of American households (See figures).
Thus, for most middle-class families, wealth is closely tied to the value of their homes, their ability to save money in monetary accounts, and the debt burden they face. But the wealth of the super rich has a lot more to do with their ability to convert existing wealth--in the form of stocks, investment real estate, or securities--into even more wealth, that is, to produce "capital gains."
Sure enough, we find that when wealth inequality was on the rise, so was the relative importance of capital gains. Between 1962 and 1969, conventional savings--the difference between household income and expenditures-- accounted for 38 percent of the growth of wealth, but from 1983 to 1989 conventional savings accounted for just 30 percent of increased wealth. Conversely, capital gains became a bigger factor in the '80s. The immediate causes lay in a falling savings rate and more rapid growth in the value of stocks than in the value of homes. In addition, the homeownership rate (the percentage of families owning their own home), which had risen 1 percent during the 1960s, fell during the 1980s, by 1.7 percent. The extension of homeownership would suggest a widening diffusion of assets to the middle class. Alas, the homeownership rate peaked in 1980 at 65.6 percent and has been falling ever since.
Widening income inequality was clearly a factor in the growing concentration of wealth. Income inequality did not change much during the 1960s but has risen markedly ever since the early '70s. The wealthy save proportionally more than the middle class. So when the wealthy get a larger share of total income, their share of savings will increase even more. Not surprisingly, between the 1960s and 1980s, the percentage of income saved by the upper third of families has more than doubled, from 9.3 to 22.5 percent. That increase was partly spurred by generous tax cuts during the Reagan years, which were intended precisely to bring about that result.
In contrast, the middle third of the population saved almost 5 percent of its income during the 1960s but by the 1980s saved virtually nothing. This drop in savings reflected the growing squeeze on the middle class from stagnating incomes and rising expenses; the Reagan tax cuts did not produce their predicted effects on this group. The bottom third has historically saved none of its income, and the Reagan years did not turn them into savers and investors as Jack Kemp's happy vision suggested.
According to my estimates, the chief source of growing wealth concentration during the 1980s was capital gains. The rapid increase in stock prices relative to house prices accounted for about 50 percent of the increased wealth concentration; the growing importance of capital gains relative to savings explained another 10 percent. Increased income inequality during the decade added another 18 percent, as did the increased savings propensity of the rich relative to the middle class. The declining homeownership rate accounted for the remaining 5 percent or so.
In short, wealth went to those who held wealth to begin with. For those who didn't have it, savings alone were not sufficient to amass wealth of great significance.
ARE THERE REMEDIES?
These trends raise troubling questions. Will the increasing concentration of wealth further exacerbate the tilt of political power toward the rich? Might it ultimately set off an extremist political explosion? Is it compatible with renewed economic growth?
At least the surface evidence suggests that equality and growth are complementary. The high growth rates of the 1950s and 1960s occurred during a period of low inequality. The slowdown in growth that began in the 1970s was accompanied by rising inequality in both income and wealth. High levels of inequality put better training and education out of the reach of more workers and may breed resentment in the workplace. Analyses of historical data on the U.S. as well as comparative international studies confirm a positive association between equality and growth.
Diffusing wealth more broadly will not be easy. Some of the causes of growing income and wealth inequality lie in changes in the global economy for which no one has any ready policy response. However, the experiences of European countries as well as our neighbor Canada, which are subject to the same market forces, suggest that shifting the tax burden toward the wealthy would spread wealth more widely. In the U.S., marginal income tax rates, particularly on the rich and very rich, fell sharply during the 1980s. Although Congress raised marginal rates on the very rich in 1993, those rates are still considerably lower than they were at the beginning of the 1980s. And they are much lower than in western European countries with more equal distributions of income and wealth.
Another strategy to consider is direct taxation of wealth. Almost a dozen European countries, including Denmark, Germany, the Netherlands, Sweden, and Switzerland, have taxes on wealth. A very modest tax on wealth (with marginal tax rates running from 0.05 to 0.3 percent, exempting the first $100,000 in assets) could raise $50 billion in revenue and have a minimal impact on the tax bills of 90 percent of American families. Even with an exemption of $250,000, such a tax would raise $48 billion.
On the other side of the ledger, the financial well-being of the poor and lower-middle class would be much improved by social transfers similar to Canada's, including child support assurance, a rising minimum wage, and extension of the earned income tax credit.
None of these measures appears politically feasible today. Instead, the current majority leader of the House is promoting a flat tax that would cut in half income tax rates for the rich and entirely eliminate taxes on capital gains. Many prominent Republicans have embraced the flat tax and argued that it should be central to the 1996 election. The rush to give the rich an even bigger piece of the pie ought to be stimulus enough to start a national conversation about where America's wealth is going.
Figure 1: Winners and Losers in the 1980s
Percent of real wealth and income growth accruing to the top 1, next 19, and bottom 80 percent of families, 1983-1989.
Top 1 percent: 61.6%
Next 19: 37.2%
Bottom 80: 1.2%
Top 1 percent: 37.4%
Next 19: 38.9%
Bottom 80: 23.7%
Figure 2: Concentration of Wealth
Share of wealth owned by the top 1 percent of families, 1945-1992 (rounded to the nearest percent)
Figure 3: Mean and Median Wealth and Income
Sources: author's computations from the 1983 and 1989 Survey of Consumer Finances; Arthur B. Kennickell and Martha Starr-McCluer, "Changes in Family Finances from 1989 to 1992: Evidence from the Survey of Consumer Finances," Federal Reserve Bulletin 80 (October 1994), 861-882.
Mean and median wealth and income, 1983-1992 (Wealth: to the nearest $5,000; Income: to the nearest $1,000). Mean to median ratio in parentheses.
Wealth: Year Mean Median 1983 $155,000 $45,000 1989 $190,000 $45,000 1992 $215,000 $45,000
Income: Year Mean Median 1983 $38,000 $28,000 1989 $45,000 $29,000 1992 $43,000 $28,000
Figure 4: ...And the Rich Get Richer
Sources: author's computations from the 1983 and 1989 Survey of Consumer Finances; Kennickell and Starr-McCluer, 1994.
Changing shares of wealth and income, 1983 and 1989
Wealth, 1983 Wealth, 1989
Top 1 percent: 33.7% Top 1 percent: 38.9% Next 19: 47.6 Next 19: 45.7 Bottom 80: 18.7 Bottom 80: 15.4 Income, 1983 Income, 1989 Top 1 percent: 13.4% Top 1 percent: 16.4% Next 19: 38.5 Next 19: 39.1 Bottom 80: 48.1 Bottom 80: 44.5
Figure 5: The Composition of Household Wealth, 1989
Source: authors computations from the 1989 Survey of Consumer Finances
Homes refers to owner-occupied housing; Deposits to liquid assets (cash, bank deposits, money market funds, cash surrender value of insurance and pension plans); Real Estate/Business to investment real estate and unincorporated businesses; Stock to corporate stock, financial securities, personal trusts, and other assets.
The Super Rich*
Homes: 6.6% Deposits: 11.0% Real Estate/Business: 45.1% Stock: 37.3%
Homes: 68.6% Deposits: 17.0% Real Estate/Business: 7.5% Stock: 7.0%
*Defined as families in the top 1 percent of the wealth distribution, with a net worth of $2.35 million or more in 1989.
**Defined as families in the middle quintile, with incomes between $21,200 and $34,300 in 1989.
Source: Author's computations from the 1989 Survey of Consumer Finances.
Families are classified into wealth class on the basis of their net worth. In the top 1 percent of the wealth distribution (the "Super Rich") are families with a net worth of $2.35 million or more in 1989; in the next 9 percent (the "Rich") are families with a net worth greater than or equal to $346,400 but less than $2.35 million; in the bottom 90 percent (Everybody Else) are families with a net worth less than $346,400.
Super Rich: 46.2% Rich: 43.1% Everybody Else: 10.7%
Super Rich: 54.2% Rich: 34.3% Everybody Else: 11.5%
Super Rich: 56.3% Rich: 33.7% Everybody Else: 10.0%
Non-Home Real Estate
Super Rich: 40.3% Rich: 39.6% Everybody Else: 20.0%
Super Rich: 53.6% Rich: 35.4% Everybody Else: 11.0%
Figure 7: Assets and Liabilities Held Primarily by the Non-Wealthy
Source: Author's computations from the 1989 Survey of Consumer Finances.
Families are classified into wealth class on the basis of their net worth. In the top 1 percent of the wealth distribution (the "Super Rich"), are families with a net worth of $2.35 million or more in 1989; in the next 9 percent (the "Rich") are families with a net worth greater than or equal to $346,400 but less than $2.35 million; in the bottom 90 percent (the "Rest") are families with a net worth less than $346,400.
("Deposits" includes cash, currency, demand deposits, savings and time deposits, money market funds, certificates of deposits, and IRA and Keogh accounts.)
Super Rich: 7.4% Rich: 26.3% Everybody Else: 66.3%
Super Rich: 16.8% Rich: 27.7% Everybody Else: 55.4%
Super Rich: 21.0% Rich: 37.8% Everybody Else: 41.2%
Super Rich: 10.1% Rich: 19.9% Everybody Else: 70.0%
1. See my monograph, Top Heavy: A Study of Increasing Inequality of Wealth in America (New York: Twentieth Century Fund, 1995) for technical details on the construction of this series.