The competition to enact a "middle-class tax cut" is only the latest pursuit by the two parties of what the Democratic Leadership Council has called the Holy Grail of American Politics. But who is really middle class? Speaker Gingrich includes families with incomes up to $200,000 a year, or 99 percent of American households. His supporters assert that those unwilling to recognize that lofty income as middle class are pursuing class warfare. President Clinton's tax cut would reach incomes up to $75,000, still almost 90 percent of families. Clinton's educational proposal provides tax deductions for families making as much as $120,000, or about 95 percent of families.
This improbable inclusiveness reflects the great American refusal to confront class. There is a pervasive belief among political analysts and in popular mythology that America is a deeply middle-class country. This belief tends to confuse aspiration with condition. Political opinion oscillates between holding that "classes don't exist in the U.S." and that "almost all Americans feel that they are middle class"--a peculiar way of denying the importance of class by using the idea of class.
Assumptions about middle-class America are at the very center of the Democratic Party's feckless search for ideology and strategy. According to President Clinton's influential pollster, Stanley Greenberg, 90 percent of Americans think of themselves as middle class and have attitudes that reflect middle-class thinking. If that is truly the case, perhaps a tax cut for 90 (or 99) percent of Americans makes sense. But if 90 percent believe they are middle class, that includes many in the lowest quintile, whose average income is under $7,000. It also includes families above the 90th percentile, with incomes of at least $113,000 for a family of four. Do these widely divergent Americans truly have the same view of their economic situation? It seems unlikely.
These misleading claims about class affect both the public and politicians. If indeed most Americans believe they are middle class and share converging perceptions of how well they are doing, politicians will feel forced to operate within a narrow band of policy options. Both parties will compete to offer variations on essentially similar policies. But if a great many Americans actually consider themselves working class, vulnerable, and downwardly mobile, a middle-class political rhetoric may reinforce their feeling that politics excludes them.
Persistence of Class
The idea that almost all Americans think they are middle class is reminiscent of a similar claim more than a half century ago. In 1940, Fortune reported a finding from a Roper poll that the overwhelming majority of American males regarded themselves as middle class. This conclusion became the received wisdom, only a few years after Franklin Roosevelt declared that one-third of Americans were ill fed, ill housed, and ill clothed, at a time when the unemployment rate was still more than 14 percent.
Surprised by the celebration of the Fortune report, the social psychologist Richard Centers analyzed polls conducted in 1945 and 1946 by the Office of Public Opinion Research at Princeton University. In his book, The Psychology of Social Classes, Centers reported that when men were given a choice of lower class, working class, middle class, or upper class, 51 percent chose working class and 43 percent said that they were middle class. Three percent said upper class and only 1 percent identified themselves as lower class. The Roper poll had given respondents an open choice in describing themselves; it was the poll's analysts who had grouped them into the categories upper class, middle class, and lower class. But when Centers added working class it turned out to be the self-identification of half of American men.
Centers also demonstrated that class identification made a difference in people's views of a variety of economic and political issues. For example, Centers asked whether the role of government was "to make it certain that there are good opportunities for people to get ahead on their own" or "to guarantee every person a decent and steady job and standard of living." Middle-class identifiers tended to have a more restrictive view of government than did working- class identifiers.
Nonetheless, the notion of the United States as quintessentially middle class and homogenous in outlook worked its way into popular writing and scholarship. This bias only deepened with the postwar boom, as even poor Americans seemed to share in economic prosperity. But in fact, twenty years after the postwar boom ended, Americans continue to identify themselves according to class much the same way they did in 1940. The General Social Survey of the National Opinion Research Center (NORC) in Chicago, a comprehensive storehouse of annual or biannual data on American attitudes, provides a test of class identification. It asks male (and now female) respondents to choose among upper class, middle class, working class, or lower class. Results sharply contradict the assertion of a big middle- class tent.
As can be seen in the figure, less than half of Americans in 1993 identified themselves as middle class (45.3 percent), while working-class identifiers amounted to 44.9 percent. (Polls conducted by the Survey Research Center of the University of Michigan show that somewhat more Americans call themselves working class than upper middle class when they are given only a choice of these two categories.) The lower two categories are more numerous than the two higher ones (51.6 percent to 48.4 percent). Women, not surprisingly, are more likely to choose lower class and less frequently upper class than are men.
NORC studies of class identification in the 1970s and '80s demonstrated the same general profile of Americans' class identification. While we shouldn't put too much credence on changes over one or two years, differences in NORC findings between 1991 and 1993 are also worth noting. In 1991 only 5.4 percent identified as lower class, while in 1993, 6.7 percent did; middle-class identifiers declined from 49.3 percent to 45.3 percent. Upper-class identifiers, however, increased from 2.3 percent to 3.1 percent. More people felt poor and more people felt rich, but fewer felt middle.
Who identifies as working class? As we would expect and as in Centers's day, those with lower incomes are much more likely to call themselves working or lower class. It is only when incomes are 20 percent above median family income that an overwhelming majority say they are middle or upper class. Only those with a bachelor's degree or more schooling had a large majority choose the middle class label. In that amorphous category of "some college," only 1 percent more identified as middle class than working class. Occupation followed a similar pattern: A majority of those in blue-collar (70 percent), service (60 percent), and clerical jobs (just over 50 percent) put themselves in the lower or working class. In sum, the selection of a working-class identification does not appear capricious. It conforms largely to what we would deduce from the objective income, educational, or occupational characteristics of respondents.
In 1993 nearly half of NORC respondents who said that their financial situation had worsened over the past year identified as working class, while somewhat more than a third of those whose situation had worsened called themselves middle class. Con versely, more than twice as many middle-class as working-class identifiers report satisfaction with their financial situation.
As in Centers's study, attitudes on the government's role in improving economic well-being differ by class identification. Fifty percent more working-class than middle-class identifiers believe the U.S. spends too little on Social Security, while almost two-thirds of the middle class feels the U.S. spends too much on Social Security. The differences on welfare are not as great but still noticeable: 47 percent of working-class compared to 35 percent of middle-class identifiers believe that the U.S. spends too little.
In the political arena, slightly more working-class identifiers say they are Democrats or Democrat- leaning than Republicans or Republican-leaning; among middle-class identifiers 10 percent more declare that they are Republicans or Republican-leaning. Working and lower-class identifiers comprise over half of the large majority with weak partisan affiliations. If one is looking to increase the numbers who vote Democratic and support activist government, the hunt should be among those who identify as working or lower class.
A Class Act?
The NORC picture of American self-descriptions about class has profound political implications. It helps explain why Gingrich trumps conservative Democrats in a politics that is mainly about the affections of the actual middle and upper class.
This country, it turns out, is not so homogeneous in either outlook and or situation. Nor is it old-fashioned to use a term like working class, for many Ameri cans describe themselves just that way. And, as the NORC data suggest, not only are there large numbers of people who feel they are not doing well economically; these people also tend to want an activist government.
Gingrich's characterization of the middle class plays to Repub lican strength. It implies that people struggling at the median income of $37,000 a year or less face the same situation as people with five times that income. But among Democrats, the politics of false inclusion promotes uncertainty and confusion. It suggests to at least half the party's potential base that Democrats fail to sympathize with their perceived condition and have little to offer.
Thus if Democrats promote a tax cut for a broadly defined middle class out of the false belief that nearly everyone feels middle class, they deplete government revenues that would provide tangible benefits for a substantial (and widely ignored) working class that actually supports activist government. Because working-class people have incomes too low to pay much tax, they also get less benefit from the "middle-class tax cut" approach.
The Democratic Leadership Council (DLC) and its offshoot, the Progressive Policy Institute (PPI), narrow and mystify who or what is their middle class. Various DLC papers and speeches urge the Democratic Party to concentrate on "the heart of the middle class," or on "the forgotten middle class," or on "the broad undifferentiated middle class."
The income range of the middle class varies among DLC-PPI writers. Alvin From, president of the DLC, believes that in New Jersey the income range of the middle class is between $50,000 and $100,000, which would exclude the bottom 60 percent of families and include some who are quite affluent. Other PPI writers such as William Galston and Elaine Kamarck use somewhat different income limits. But what is clear in all of these accounts is that there is little room for the party's true traditional base of working-class identifiers. In Washington opinion-leader circles, $50,000 may feel like a minimally decent middle-class income, but it is actually more than what two-thirds of American families earn.
The NORC findings about class do not mean that working- and middle-class identifiers clash on every issue. But they hardly suggest converging attitudes either. The DLC-PPI story is profoundly misleading, especially for Democratic Party strategy, when it suggests that murkily defined "average income" Americans produce homogeneous "social values and moral sensibilities" that the party should emphasize in its broad themes.
Assertions that we are almost all middle class, or that only a limited middle class deserves electoral attention, lead Democrats to embrace essentially Republican views of government. The "middle class" strategy also has the unfortunate effect of further dividing what was once a coalition of working class and poor, and creating a false affinity between working class and affluent. DLC-style middle-class rhetoric suggests policies from which all except the demonized welfare poor will gain. This approach reinforces the false imagery of common interests and positions among working, middle, and upper class. When actual policies do not produce equal gains, and those with lower incomes typically lose out, anger grows and political support evaporates.
The Vanishing Middle
A middle, of course, implies a higher and a lower. In the U.S., high-income groups are becoming even wealthier, while average family income is stagnating and declining. Partly as a result of this capture of economic gains by the upper brackets, the cost of what seems a normal American standard of living is increasing. For example, 20 years ago an average-priced car required a third of an average family's income; today, it takes more than half.
By broadly defining almost everyone as part of an elusive and mythical middle class, both Demo crats and Republicans are ignoring the fact that not only do many Americans define themselves as working class, but also that many who are reported as saying they are middle class or aspiring to it are falling further behind.
Another barrier to a realistic view of income classes in the U.S. is the official poverty line, which is pegged at too low a level. [See John E. Schwarz and Thomas J. Volgy, "Social Support for Self-Reliance: The Politics of Making Work Pay," TAP, Spring 1992, No. 9.] Even if not officially defined as poor, people between the poverty line and 80 percent of the median income derive most of their income from work, not from transfers, and are definitely not able to reach middle-class levels of living.
At the other end, those families with incomes above $75,000, two times the median, but below, say $200,000, are well-to-do or upper middle class. Between the poor and lower-income groups, on one side, and the $75,000-and-above classes, on the other, stand a variety of middling categories, many of whom are in tight economic circumstances or suffering considerable income insecurity. These groups have markedly differing views of their own condition and of politics.
It is not unexpected that the Republican Party would blur these distinctions. It is odd to find Democrats doing the same. One can debate the precise income cutoffs and other variables needed to understand today's class distinctions. Calling (almost) all of us middle class while denigrating the welfare poor neglects the needs of the half of the population who continue to regard themselves as working or lower class, or only insecurely middle.