Cleavage in American attitudes toward social welfare.

Opinion polls probing both the narrow and broad senses of social welfare among Americans indicate hardly any substantial differences over crucial social sentiments among a variety of groups with at least theoretically divergent interests: rich and pool, men and women, blacks and whites, a variety of ethnic groups, union and nonunion households. The items mainly concern the provision of welfare to the poor through AFDC, now TANF, and Food Stamps but also cover OASDHI. Consistently over more than sixty five years of systematic opinion polling, there is an astonishing consensus, so large in fact that it may undermine any effort to move the American citizenry into a more congregational series of provisions for each Other. In fact, the consensus is antagonistic to the public welfare. Americans by their very actions, opinions, and codified intentions have canceled the notions of class and caste in subverting a generous welfare state.

Key words: attitudes toward social welfare; cleavage; polarization; culture wars

**********

The Englishman William Robson put his finger on the heart of the problem: "Unless people generally reflect the policies and assumptions of the welfare state in their attitudes and actions, it is impossible to fulfill the objectives of the welfare state" (Robson 1976). He might have gone on to point out that it is impossible to sustain any public policy in a democracy with deep divisions-cleavages--among the population.

The issue of cleavages in American attitudes toward social welfare has received surprisingly little attention except as expressed by aggregated data. Anatagonism toward welfare and welfare programs has been widely supported (Dimaggio, Evans and Bryon 1996; Mouw and Sobel 2001; Page and Shapiro 1992; Baggette, Shapiro and Jacobs 1995; Page, Shapiro, and Young [1986]; Shiltz 1970; Erskine 1975; Public Agenda 1995). The few studies suggesting popular support to expand the welfare state (Cook and Barrett 1992; Demos 2002) or documenting a shift in attitudes over the past decades (Teles 1998) have been seriously flawed.

Income class would seem to be one of the most compelling variables in any analysis of decision-making and social attitudes. However, sixty five years of polled attitudes toward social welfare have rarely been disaggregated by income group; Page and Shapiro (1996) is a rare exception but even their treatment is cursory. Attitudes toward welfare are customarily reported by ethnicity, gender, region, and others but undifferentiated by income.

Contemporary disputes over social issues generally--the "culture wars"--and over the source of social sanction for public policy decisions (elites versus masses; class dominance versus pluralism) are sensitive to cleavages in the American polity (Domhoff 1996, 1990, 1967; Domhoff and Dye 1987; Mills 1956; Hunter 1953; Hunter 1991, 1994; Wolfe 1996; Gordon 1994; Downey 2000; Gitlin 1995). Small actual cleavages in American opinions among important political groups--a great consensus over public policy--would reduce these disputes to media events and public entertainments. Large political cleavages would begin to point to the consistent influence of particular groups in determining social policy. So far, the evidence for a general consensus rather than deep cleavages is considerable although again there is hardly any analysis by income class although somewhat more by race and gender (Dimaggio, Evans and Bryon 1996; Mouw and Sobel 2001; Lindaman and Haider-Markel 2002; Brooks and Cheng 2001; Miller and Hoffmann 1999; Hoffman and Miller 1998; Evans 1997; Williams 1997). Presumably the most intense debate--abortion--is distinguished by a split between the pro-choice left and the center but not with the anti-abortion right which appears to very unpopular (Dimaggio, Evans and Bryon 1996). With great support for contemporary social policy or without general support for change or even deep cleavages in attitudes toward current policy, there is little prospect for new policies in a democracy.

Method

The separate polls of the General Social Surveys (1972-1998) (GSS), the National Elections Studies (1948-1998) (NES), the CBS/ New York Times Polls (since 1976) (CBS/NYT) and others were analyzed to describe the cleavage--that is, welfare polarization--in American attitudes to the narrow and broader sense of social welfare and to attempt to place any consensus that may exist within the context of American policy. The narrow sense of welfare is defined as public attitudes toward AFDC, now TANF, and the Food Stamp Program, to a number of specific issues and policies closely allied with those program, mainly the federal role in underwriting the programs and to a associated attitudes especially including those toward blacks. The broader sense of welfare focuses on OASDHI (Old Age, Survivors, Disability, and Health Insurance), what is commonly referred to as Social Security and Medicare, as well as associated attitudes.

Cleavage is explored as the differences between rich and poor whites and blacks, men and women, and to a smaller degree, among ethnic groups and union and nonunion households. The analytic problem is not to find statistically significant differences, since only the tiniest difference will fail to be highly statistically significant with such large samples. Rather, the central task of the research is to interpret whether the differences among the study groups are substantial for purposes of social policy and social welfare. There is no quantitative test of substantiality but rather a number of far more amorphous considerations discussed in the Conclusions. The backup Appendix data tables are available online at www.univ.edu/faculty3/epstein/polls.

The narrow sense of welfare is explored with three types of questions that probe: first, attitudes toward welfare (AFDC, now TANF) and Food Stamps--e.g., whether to increase or decrease spending on them--and their effects, such as whether they decrease work incentives; second, attitudes toward the federal government's role in sustaining these program and toward closely related questions of public responsibility for the poor and needy; and third, attitudes toward blacks and government responsibility to secure their welfare.

The broader sense of welfare focuses first on OASDHI but also explores a variety of adjunctive attitudes that seem to underpin the citizens' sense of general welfare: finances, family, children, education, being cultured, life satisfaction and happiness, the role of government beyond its responsibility for the poor, the trade off between social spending and taxes and so forth.

The Findings section presents highly summarized data. However, it is impossible to array all of the comparisons reported in the paper; and therefore, following current practice, the reader is invited to request specific additional information from the author.

In almost every comparison, cleavage among income groups is virtually absent for the top four quintiles or so. The data are therefore only presented at the ends of the income distributions, purposely searching for the greatest instances of cleavage.

The statistical properties and characteristics of the polls--sampling, question wording and order, representativeness and so forth--can be traced back from their separate code books (Davis and Smith 1996; Miller and Traugott 1989). Both the GSS and the NES conduct face to face interviews, the latter biannually and the former more frequently. The other polls are phone interviews.

Findings

The narrow sense of welfare

The attitude differences toward welfare, that is, cleavage, between rich and poor men, women, blacks and whites, between poor and wealthy union and nonunion households, and among ethnicities are typically small, theoretically insubstantial, or both. The cleavage between poor blacks and wealthy whites is occasionally substantial but this difference has been declining since the 1980s. It is notable that there is rarely any cleavage of note among the top four income quintiles; for all intents and purposes they are indistinguishable. Whatever cleavage exists is most pronounced between the poorest and wealthiest. There were no substantial cleavages among these groups in their attitudes toward welfare increases, personal responsibility, or a range of other associated attitudes. Except occasionally, all groups consistently preferred personal responsibility, limiting welfare payments, the stringent reform measures of 1996, and attitudes hostile toward welfare recipients.

In the 21 polled years between 1973 and 1998, differences between the poorest and wealthiest quintiles were only 23.2 percentage points in reporting to the GSS that "we're spending too much money on welfare" (Table 1). In any polled year differences infrequently exceeded thirty percentage points (Table 2). Yet the income differential between the lowest quintile and the upper quintile is enormous; the upper income threshold of the lower quintile of respondents is barely above the poverty line for a family of three (Appendix Tables 1-6). The differences between the lower decile of respondents, the best off of whom are often well below the poverty line, only adds a few percentage points totaling an average cleavage of 29.5 between them and the wealthiest decile of respondents (Table 1). Differences rarely exceeded 35 percentage points (Tables 2-3, Appendix Tables 4-6). Differentials between whites and blacks were also under thirty percentage points. The only differentials that were larger than fifty percentage points occurred between the poorest blacks and the wealthiest whites and only occasionally (Appendix Tables 4-6). There was no cleavage among ethnic groups (Western Europe, Mid or Central Europe, New World Hispanic, American Indian) and only about 30 percentage points separated blacks from Europeans (GSS tabulations).

In contrast, with five possible responses between independence and government responsibility for the poor, average cleavage was even less between the poor and the wealthy (17.2 percentage points) and blacks and whites (14.5 percentage points) (Tables 4-6, Appendix Tables 7-12). The cleavage increased only slightly in comparing deciles, 21.8 and 18.7 respectively. Whites in particular had substantial preferences for personal responsibility as opposed to government responsibility, one of the areas of consistent but not large disagreement between whites and blacks over the years but also an area of recent convergence (also see Public Agenda 1995).

The consensus preference for personal responsibility and consistently against increasing welfare, is even stronger in light of the fact that much larger percentages of respondents simply agreed rather than strongly agreed that people should care for themselves while about one third of respondents each year stated that welfare payments were adequate as they were. There were only small preferences for government responsibility and increases in welfare even among poor people, indeed, on average thirty-five percent of the poorest white Americans wanted to cut welfare benefits. Again, the top four quintiles provide very similar responses to queries about both the welfare budget and government responsibility, emphasizing the centrality of the natural economic preferences of higher income groups. Continuing the suggestions of Schiltz's (1970) earlier tabulations, the data document America's consistent hostility since the beginning of systematic polling in the 1930s across income classes toward public assistance.

Welfare is perceived increasingly as a local administrative responsibility with enormous majorities of wealthy men, women, and whites preferring state standards and responsibility over federal responsibility for welfare programs (CBS/NY Times April 1995). In fact, these attitudes strongly endorsed the reforms of 1996 (witness the large consensus reported by Public Agenda 1995) and continue to sustain their reauthorization in 2004. Near majorities of poor men women and whites felt the same way. Blacks demurred, but surprisingly poor blacks less than wealthy blacks. In the same poll, wealthy men, women, and whites overwhelmingly wished to limit "the amount of money available for welfare benefits even if this means there might not be enough money to cover all families who qualify." Approximately forty percent of poor men, women and whites agreed. However, cleavages between wealthy and poor were never even twenty percentage points. The same pattern repeats to support the Republicans in Congress as they "completely rebuild the welfare system" along these lines (CBS/NY Times April 1995).

There is a lineage to these types of responses. Only a majority of blacks, and only in 1984, agreed that "families are not getting enough welfare" is a more serious problem than families "getting more welfare benefits than they need" (CBS/NY Times September 1984, January 1988). Only very small percentages of men, women, and whites agreed. One decade later, all groups including blacks were much less sympathetic (CBS/NY Times January 1994). Indeed, while over seventy percent of poor and wealthy males, females, blacks, and whites endorsed government "financial assistance for children raised in low income homes where one parent is missing" (CBS/NY Times July 1977), almost twenty years later, all of these groups except poor blacks cut their support in half for a more leading question: "spending on programs for poor children" (CBS/NY Times December 1994).

The hostility seems aimed toward recipients as much as toward the programs themselves, a difficult distinction to make since the recent social disapproval of racist expression may suppress certain responses. Welfare recipients are obviously considered to be able-bodied and therefore should be independent since Americans consistently agree that "it is the responsibility of the government to take care of people who can't take care of themselves" (e.g., CBS/NY Times January and April 1995). However, very large percentages of rich and poor men, women, blacks, and whites, and often more than fifty percent, agree that "most people who receive money from welfare could get along without it" rather than "most of them really need this help" or "half and half" (CBS/NY Times July 1977, March 1982, January 1994, December 1994). Cleavages were usually less than ten percentage points with even blacks infrequently demurring by much. Moreover, there was hardly any cleavage at all by ethnic descent with only small and intermittent differences, again usually less than ten percentage points, between poor and wealthy Americans who identified themselves as Italian, Slavic, German, Black, Irish, Scandinavian, Latin, British, or "other, American" with (CBS/NY Times September 1976); there was also very little cleavage among the ethnicities themselves (CBS/NY Times September 1976--three questions by ethnicity).

The tenets of unworthiness--tested by the perceived unwillingness to work--cut across almost all groups, rich and poor. Only a majority of blacks and only in December 1994 believed that "most recipients really want to work" (CBS/NY Times January 1994, December 1994, February 1995). Curiously, wealthier respondents customarily endorsed this finding slightly more than the poor perhaps tacitly confessing an ignorance of the unpleasantness of lower paid jobs--but again, hardly any cleavage. Large majorities of poor and wealthy men, women, and whites and a near majority of poor blacks in January 1994 consistently report that "there are jobs available for most welfare recipients who really want to work" (although note that the New York Times may have really wanted this response to endorse their preference for work training or a work program) (CBS/NY Times January 1994, December 1994). But majorities of all these groups, reaching eighty percent for wealthy females report that the jobs do not "pay enough to support a family." Independence from welfare is characteristically preferred over the reduction of poverty--the presumed nobility of work no matter what its consequences.

Enormous majorities believe that "people are so dependent on welfare that they will never get off" (CBS/NY Times January 1994, April 1995) and that unmarried mothers who are under eighteen and "have no way of supporting their children" as well as other welfare recipients should enter work programs and "should stop receiving [welfare] benefits" after a period of time (CBS/NY Times February 1995). Moreover, about twenty percent of all these types of respondents believe that "giving welfare to poor people" increases crime rather than decreases it or has no effect (the dominant response) (CBS/NY Times July 1977); a slightly smaller percentage of all groups believe that "most people are on relief for dishonest reasons" (CBS/NY Times 1995).

Indeed, reported attitudes toward the programs may be proxy for attitudes toward some of the recipients. That is, respondents may make use of the opportunity offered by questions about welfare and welfare recipients to voice their attitudes toward blacks and other minorities or perhaps the poor generally, conflating a sense of moral deficiency with the relief programs themselves.

The NES, sometimes back to the 1960s, and CBS polled for the federal government's responsibility to assist and compensate blacks, for fairness in employment, and for associated attitudes. First it is obvious that little cleavage exists among the various groups and that even the black/white differentials, while consistent, are customarily small (the seemingly large differentials between wealthy blacks and other groups may be artifacts since the group often contained very few respondents and sometimes none at all). Second, recalling that five responses from strong agreement to strong disagreement were offered to the NES questions and that agreement and disagreement customarily contained a larger proportion of responses than the extremes, antagonism to compensation, a federal role, fair treatment for blacks in employment, and others was considerable, perhaps denoting hostility toward blacks themselves and perhaps carrying over to the narrow sense of welfare.

Except for blacks, the nation appears opposed to job preference for blacks even "where there has been job discrimination ... in the past" (CBS/NY Times July 1977, April 1995), believing that "blacks should not have special favors" (NES) (Appendix Tables 13,13a). Again, except for blacks, few strongly agree that "over the past few years blacks have gotten less than they deserve" while many feel that "blacks must try harder" (recalling the confusion created by "must" which may mean that whites believe they should try harder while blacks believe they are forced by racism to carry an extra load) (NES) (Appendix Tables 14, 14a). These antagonisms are even that much greater in light of the socially approved attitude of nondiscrimination and fairness. If indeed the pressures of social comformity suppressed even a modest amount of hostility, then the actual amount of racism and perhaps also hostility to the poor generally--the contemporary notion of an underclass of incorribles, deviants, and malingerers--grows as a daunting impediment to the welfare state.

The broader sense of social welfare

Very large proportions of all study groups between 1984 and 1996 endorsed increases in spending on Social Security; still, there is virtually no cleavage at all (Appendix Tables 15, 15a). As Page and Shapiro (1992) point out, this has become such a standard of America's reported attitudes that it is only infrequently queried. While support for a national health insurance plan seems to have eroded over the past thirty years there is again very little cleavage, on the order of twenty percentage points between wealthy and poor groups (and curiously high nonresponse rates in 1972 and 1984). Cleavage is slightly more but again under thirty percentage points in preferences to "completely rebuild" the American health care system (CBS/NY Times January 1994). At the same time, there are only insubstantial differences, remarkably small considering the income differentials, between poor and wealthy groups in choosing between taxes and spending on "social programs" (GSS 1993). The wording is far more benign than "welfare" but it still did not elicit strong support even among very poor people.

Hardly any group places great trust in the "government in Washington to do the right thing "just about always," even at the height of the Reagan presidency (CBS/NY Times January 1986), which is carried over as a preference for state and local government (CBS/NY Times January 1986). A majority or near majority of all groups responded yes when asked if "there are any groups in America today that are not given a fair chance to succeed economically" (CBS/NY Times 1984); it was surprising that many more did not agree with this near truism. About forty percent of all groups except blacks whose percentages were much higher reported that "government programs created in the 1960s ... made things better" (CBS/NY Times January71986) and majorities, sometimes very large for all groups except wealthy males, agreed that "the federal government should spend money now on a similar effort to try to improve the condition of poor people in this country"--note again the use of "poor" rather than "people on welfare" (CBS/NY Times January7 1986). There was a strong positive response in all groups except wealthy blacks to the proposition "that it is as possible now as when [they] finished school to start out poor in this country, work hard, and become rich" (CBS/NY Times August 1988).

As elsewhere, there is hardly any cleavage at all between the poor and the wealthy relative to a variety of social attitudes: financial security, being cultured, having faith in God, having children, being married, having nice things, being self-sufficient, and having a fulfilling job (GSS 1993). The poor and the rich equally reject nihilism ("life serves no purpose") (GSS 1998). The paradox of satisfaction is even more astonishing; consistently between 1972 and 1998, a cleavage of only about twenty five percentage points separated people well under the poverty line and the wealthy in reporting that they are "very happy ... with the way things are these days" (GSS 1972-1998).

The Case of Organized Labor

Workers presumably form unions in response to the constraints of the labor market, the need to counter the natural tendencies of society to neglect its less well off, and the power of employers relative to individual employees. Moreover, in order to organize, unions presumably develop a greater consciousness of the right for social welfare among their members than would be present among those not in labor unions. However, union households hardly ever differ in their attitudes from non-union households suggesting perhaps that there is no distinct social philosophy underpinning the organization of American labor, only a syndicalist ambition to compel higher wages and benefits. Surprisingly, there are only modest differences between union and nonunion households toward the importance of labor unions themselves, even when their central value is probed (Table 7).

The absence of a distinct social philosophy grounded in the grievances and broader social rights of working people and the general hostility of Americans toward social welfare perhaps explains the decline of organized labor over the past forty years--their absolute decline in numbers, their startling relative decline, and their shift from industrial organization to middle class occupations and the public sector. Today, labor organization is in the process of realizing Kuttner's prophesy of shrinking to craft union size (Kuttner 1985).

Anomalies and differences: blacks vs. whites, broader vs. narrower welfare, ritual vs. operant values

Until recently blacks consistently voiced greater support than whites for public welfare and the role of the federal government, blaming those in need less, and wishing more to address those needs. At the same time the data also corroborate earlier observations by Shapiro [1986] that the cleavages between blacks and whites are narrowing as are differences among income groups, including the poor and the rich. Indeed, Public Agenda's poll in 1995 reported virtually no difference at all between blacks and whites in their attitudes toward welfare and, more surprising, very few differences between them and welfare recipients who are by definition very poor. Schiltz (1970) documents a similar hostility toward public assistance between recipients of public assistance and the general population during the Depression and shortly afterwards.

Yet such as the differences are between blacks and whites, there is no consensus among blacks that suggests the indignation and rage of Malcolm X, Leroy Jones, James Baldwin, or Richard Wright. Indeed, the absence of extreme cleavages and the more recent apparent satisfaction of blacks with social policy may help to explain the decline of black civil disturbances over the past few decades. Voltaire would have been pleased, Marat horrified.

While there is a customary lack of cleavage and a general hostility to the narrow sense of social welfare, there are reported attitudes that would seem to sustain the provision of broader social welfare and some ancillary programs to TANF in contradiction of current social welfare policy. In particular, Americans consistently endorse higher Social Security benefits; many seem to want national health insurance and enhanced job training programs; there is even an enormous agreement among different ethnic groups for the federal government "to see to it that every person who wants to work has a job."

Yet, the direction of federal retirement legislation seems to reward the wealthy who save outside of Social Security and to neglect poorer Social Security recipients. Moreover, approximately forty million Americans are without any health coverage and it is very questionable whether Medicare benefits will improve substantially, even for prescription drug coverage. There is virtually no public sector jobs program and very little job training. The stated support for a broader sense of social welfare beyond and separate from programs for the poor themselves may be illusory. Indeed, much of the apparent support for social welfare in general may be more a shallow hope for good fortune projected upon the federal government but without any strong political will to convert aspirations into enforceable claims. The polls fail to distinguish between real preferences and ritualistic affirmations of America's ceremonial civil religion. So long as American policy making is open and uncoerced, the specific program conditions of public policy, actual policy choices rather than surveyed attitudes, may actually realize the true preferences of the public will.

Conclusions

The bifurcation of social welfare policy in the United States between modest work-related entitlements and inadequate, discretionary assistance for those outside of the labor force has been sustained by broad popular consent. Neither the bifurcation of policy nor the actual insufficiencies of America's social welfare programs appear to be impositions of an elite that is any more predatory than the general citizenship. Rather, the American social welfare state, sustained by the embedded preferences of Americans for market-related social hierarchies and minimal relief of want, institutionalizes the triumph of classical liberalism over welfare state liberalism. Hunter and Gitlin worry with little cause over the ability of America to govern itself; polarization appears restricted to abortion and perhaps a few other "body" issues that in fact have not created much turmoil and that remain peripheral to social welfare. Not coincidentally, the cleavages, such as they may be, relate more to procedural issues of legal right (to abortion or equal protection) and far less to the substantive (financial) issues of equality and poverty. Indeed, the powerful underlying consensus on social welfare both in the narrow sense as well as more broadly defined--TANF and Food Stamps on the one hand and OASDHI on the other--may even be strengthened by displacing social conflict to largely symbolic and procedural issues that preserve more important social values. Inflamed conflicts over abortion and perhaps even the broader feminist agenda, occurring between groups at the political margins, are like the breast thumping among apes, the head butting in goats, and the tail dancing of the stickleback herring that serve vicariously to defuse tensions, select leaders without blood, and reinforce the probity of existing social institutions. There are no culture wars in America apart from the entertainments of the media.

Judged by its social welfare policies, the welfare state in the United States contains a very modest amount of Lowi's (1964) redistributive function, emphasizing regulation with even a tendency toward "distributive" policies (that is, social welfare as group patronage). The popular consent, even if misguided, curiously endorses Domhoff's assertion that "classes and class conflict, along with protest and social disruption, have to be taken seriously to understand pouter in America" but only in the sense that the absence of turmoil is a measure of deep satisfaction with things as they are (Domhoff 1990 282); America's ruling elite seems to enjoy pervasive permission.

Moreover the programs themselves do not seem to be triumphs of autonomous state benevolence, defying by their actual benefits the rudimentary expectations of Skocpol's demands for broad entitlements and full employment (Skocpol 1995, 2000). Still, in the absence of frankly expressed group and class differences, it is methodologically impossible to discern whether the public has been propagandized into agreement or the leadership fairly represents prior, popular references. Contrary to Domhoff, the American state at least judged by its social welfare provisions, is hardly the product of an elite, let alone a predatory one; nor could it possibly be judged a beneficence of leadership by noble, brave, maternal, and informed heroines who act largely within the permissions and constraints of an enlightened pluralism. The reigning and deep consensus profoundly rejects greater sharing, greater entitlements, greater generosity, and more opportunity secured by public interventions. Before expanding the provisions of the social security act it may be necessary for Skocpol to first consider that the embedded displeasure with the redistributive role of government produces little support for a tutelary state of increased welfare and patrician regard.

The robustness of insubstantial cleavages even at the extremes of income and the huge common consensus across time, different surveys, a variety of groups and many different sorts of questions commands attention to a profound American social pact and one that perhaps explains the failure to achieve Robson's hopes for a generous welfare state. Americans may be very satisfied with things as they are, antagonistic toward both the narrowly focused public assistance programs and a greatly expanded government role in securing the general welfare. This uncivil complacency may well erect an insurmountable barrier to expanded entitlements or greater sharing of any sort.

Generosity and ideological diversity while perhaps goals of a vibrant public discourse in an Enlightenment society are apparently not characteristic of the American social welfare ethos, at least since the 1930s and perhaps for the past few centuries. In consideration of the technical ambiguities of the polls, it may even be the fact that America has forged a characteristic political ethos from the vast ethnic and racial ores of its peoples. Thirty percentage point differences among groups that are very differently situated, while seemingly large, are certainly not large enough to constitute class distinctions or even characteristic group attributes. The expectations of class theory and conflict theory would seem to demand far greater cleavages, perhaps on the order of sixty or seventy percentage points. Differences of this magnitude have separated poor blacks from wealthy whites but usually before 1985; they quite obviously carry along with them distinctions of caste made graphic in the cultural abyss that in fact often separates the two groups.

The American political consensus on social welfare has cemented a position quite a bit to the right of center, ideologically centered on voluntary civic participation and good character--"compassionate conservatism:" communitarianism rather than communalism. Not only are Americans antagonistic to welfare narrowly defined but the antagonism is consistent through almost every political division of the nation. Most notably, the poor and the wealthy, blacks and whites, men and women, union and nonunion households, and the variety of ethnic groups share in the same hostile attitudes. The cleavages between the very wealthiest and the very poorest groups of Americans are insufficient to germinate a sizable constituency for more redistributive and generous social welfare policies. Americans cherish their unbounded markets and self-defeating heroic individualism, apparently willing to impose few restrictions for purposes of minimizing economic insecurity or relieving want.

The attempt to retrieve the public's actual but latent generosity from the meanness of standing policy is built on an imagined distinction between the notion of welfare and the welfare programs themselves as if to argue that Americans are for relief but against the poorly run programs that administer their generosity. However, this argument comes apart in light of the widely shared popularity of a work test (that is, the willingness to take a job, any job), the widespread support for mandatory work, and a stolid refusal to acknowledge frank need. Of course, it is a near newspeak tautology for people to support relief for those who deserve it. However, the actual meaning of policy is conferred by the conditions of deservingness.

American social welfare policy itself reflects this consensus of old liberal and new conservative, the dominance of industrial Republicanism and deep-faith traditionalists. The 1996 welfare reforms, grounded in little more than the nation's meanness of purse and spirit, continue to be extremely popular. Compassionate conservatism is laying the track of public policy. Indeed, the enormous amount of reported support for OASDHI is not a hopeful sign of greater American faith in the welfare state. Instead it may represent the nation's private attitudes authorizing the government's parsimonious public programs. Fully forty percent of OAI retirees, typically the poorest paid workers, rely for at least 80% of their income on their Social Security checks which in 1999 averaged only $804 for all beneficiaries (Social Security Administration 1998; Ways and Means Committee 2000). On the other side, OAI maximums (about $2,650 for a family in 1999) are paid to the relatively wealthy whose government checks represent only a fraction of their incomes. Obviously return on investment, not need, generosity, compassion, or forgiveness, is the abiding criterion of American fairness.

Americans appear to be consistently and historically opposed to social welfare policies for the indigent with little desire for even a generous series of social insurances for workers. Neither vertical nor horizontal redistributional policies are popular. The anomalous attitude that government should secure the general welfare is probably voiced as the vaporous hope that traditional American institutions of the market rather than the demeaning programs of public welfare will provide a fair distribution of American plenty.

Classical theoretical assumptions that apparently different social and material conditions greatly affect the attitudes of different economic groups, races, ethnicities, and genders may have been inoperative in the United States for the past seven decades or so and perhaps for even longer. American processes of socialization may enjoy a remarkable triumph over any material or social reality of caste, class, or gender. All would be well for the very large consensus around social welfare policy but for its cruelty to poor and marginal citizens as well as to lower paid workers in general. More than two thirds of recipients of TANF are poor children who are saddled with the miscreancy that the nation ascribes to their parents. Poor children in foster care are given a pauper's mite. The poor who are permanently disabled are treated as if they willfully perpetuated their disabilities. Hardly anything at all is provided for single adults who can not or do not work while many homeless Americans endure parked cars and uninvited pedestrians in their living rooms. And so on with inadequate health and mental health services for poor people perhaps explained by a puzzling tolerance among the working poor, near poor, and the majority of all blue collar workers for their decades of stagnant and inadequate wages and their isolation from even the frankly inadequate services afforded the very poor.

Yet if many Americans are in fact oppressed, they are unaware or blithely accepting of their oppression. Any strategy to mobilize an opposition to current social welfare policies must confront the near identical dispositions of Americans and their apparently great satisfaction if not complacency with social welfare policy both in its narrow and broader senses.

The insistence among a number of the semi-professions, notably social work, that they are liberating the oppressed--a quaint conceit in light of the obliviousness of the oppressed themselves to their suffering--needs first to find a population that acknowledges its need before it applies a remedy. Nonetheless, the literature of the personal social services, including psychotherapy, has been engaging in a delicious irony of success for years: the liberation of the afflicted from afflictions they have do not know they are suffering. Not only is liberationism an emperor without clothes, but also a parade without an emperor, an audience, or reporters to record events--a total fabrication starting with imagination and building back to history. O'Connor's (2001) poverty knowledge and Epstein's (1997) social efficiency seem to correspond well with the popular ethos.

The reported support for increasing social security, substantial endorsements of a universal federal health insurance of one sort or another, and other preferences for an expanded welfare state documented most recently by Demos (2002) might appear to argue for the popularity of the welfare state. Yet the problems with polls may invalidate the reported preferences for expansion (Epstein forthcoming 2006). More to the point than opinions, there has been very little political activism in support of expanded social insurance. President Clinton's abortive attempt at a national health plan and the near constant inability to increase the generosity of OASDI since the 1970s suggest that the program as it exists may be far more expressive of the American consensus than the reported polls. Indeed, the social security system seems perched on retrenchment not expansion and citizen lobbies seem simply protective of the present program. Still, this line of reasoning--minimizing some reported preferences while accepting others--may seem capricious. Yet in light of the substantial methodological deficiencies of opinion polling and the drift of the nation to the right without much political dissent (indeed, with considerable acquiescence), it may be prudent to ground interpretation of reported attitudes in the facts of live political choices and traditional historical discourse. Whatever the ambiguities assigning the American consensus a point on the political continuum, the numbing consistency of reported preferences seems inescapable: there is very little cleavage in American social welfare attitudes.

Inflecting these general conclusions another note, there may be an active hostility to generosity by perhaps one third of the population that taken together with the oblivious middle has probably undercut any serious progressive policy in the United States. Moreover, the constituency for reform--the modern American liberal and the New Democrat in the style of former president Clinton--unfortunately favors procedural equity rather than substantive equality as epitomized by support for affirmative action over compensation and job training over the provision of public jobs. There is hardly any endorsement of major budgetary initiatives to realize true structural reform. Without deep investments there is also little likelihood of addressing America's social problems.

Table 1

Percent responding "we're spending too much" * on welfare and
average differences between poorest and wealthiest groups by race. GSS
1973-1998

Income group    All     White    Black    Difference white/black

Quintile
  Bottom        35.1    41.1     16.8              24.3
  Top           58.3    59.1     37.2              21.9
  Difference    23.2    18.0     20.4              -2.4
Decile
  Bottom        29.5    35.9     16.6              22.3
  Top           59.0    60.2     36.7              23.5
  Difference    29.5    24.3     20.1               4.2

* Responses = too much, about right, too little

Table 2

Percent of all respondents reporting "we're spending too much money
on welfare." * Approximate lower quintile vs. approximate upper
quintile family income, poverty line, income thresholds of quintile
vs. approximate upper quintile family income, poverty line, income
thresholds of quintiles, cumulative percent. 1972-1998 General Social
Survey.

         Those with     Those with
          lower 20%      upper 20%
           income         income
          brackets       brackets
                                      Percentage
Year     N      %       N      %      difference

1973     96    33.2    236    65.2       32.0
1974     75    28.3    219    52.1       23.8
1975    105    31.4    231    55.4       24.0
1976    167    50.0    188    68.6       18.6
1977    147    48.4    214    76.4       28.0
1978    139    41.2    220    73.3       32.1
1980    130    42.6    217    67.2       24.6
1982     80    23.3    260    56.8       33.5
1983    118    37.0    181    62.4       25.4
1984     28    29.2     46    46.5       17.3
1985     53    34.4     92    55.1       20.7
1986     44    29.3     80    50.0       20.7
1987     31    25.4     65    53.7       28.3
1988     32    25.2     75    50.0       24.8
1989     49    32.9     70    53.0       20.1
1990     46    38.3     47    39.5        1.2
1991     39    25.5     90    48.6       23.1
1993     56    39.2    115    64.2       25.0
1994    140    47.1    232    71.4       24.3
1996    120    46.3    158    64.8       18.5
1998     81    29.9    138    50.5       20.6

* Responses = too much, too little, about right.

Table 3

Percent of all respondents reporting "we're spending too much money
on welfare." * Approximate lower decile vs. approximate upper decile
family income, poverty line, income thresholds of deciles, cumulative
percent. 1972-1998 General Social SurveyThose with lower Those with
upper 10% income brackets 10% income brackets

        Those with     Those with
         lower 20%      upper 20%
          income         income
         brackets       brackets
                                     Percentage
Year    N      %       N      %      difference

1973    66    31.0    123    65.1       34.1
1974    26    19.5    111    52.1       32.6
1975    54    27.6    132    57.1       29.5
1976    91    47.6    105    71.9       24.3
1977    54    40.0    165    81.7       41.7
1978    52    31.9    159    74.6       42.7
1980    57    35.6    217    67.2       31.6
1982    48    21.1    139    57.9       36.8
1983    59    34.5     87    61.3       26.8
1984    14    28.0     21    46.7       22.7
1985    17    25.4     45    55.6       30.2
1986    17    25.3     55    48.7       23.3
1987     8    14.8     39    53.4       38.6
1988    11    16.4     49    59.0       42.6
1989    19    24.7     45    57.0       22.3
1990    23    34.3     35    41.7        7.4
1991    12    16.9     33    43.4       26.5
1993    29    39.7     44    60.3       19.6
1994    52    38.2    100    72.5       34.3
1996    56    43.8    104    65.4       21.6
1998    33    24.8     55    46.6       21.8

* Responses = too much, too little, about right.

Table 4

Percent strongly agreeing "that people should take care of themselves"
rather than the "government should improve the living standards of
all poor Americans" * and average differences between poorest and
wealthiest groups by race. GSS 1975-1998

                                          Difference
Income group    All     White    Black    white/black

Quintile
  Bottom        15.9     21.0      9.7        10.4
  Top           33.1     35.2      9.4        25.8
  Difference    17.2     14.2     -0.3        14.5
Decile
  Bottom        14.9     16.6     10.1         5.5
  Top           36.7     36.8     11.6        25.2
  Difference    21.8     20.2      1.5        18.7

* Responses = five responses from strongly agree with the former to
strongly agree with the latter

Table 5

Percent of all respondents stating strong agreement that "people
should take care of themselves" rather than the "government should
improve the living standards of all poor Americans." Approximate
lower quintile vs. approximate upper quintile family income, poverty
line, income thresholds of quintiles, cumulative percnet. 1972-1998.
General Social Survey.

         Those with lower       Those with upper
       20% income reporting    20% income reporting
        people should take     people should take        Percentage
       care of themselves      care of themselves      point difference
                                                       between top and
Year        N      %                N      %           bottom quintiles

1975        61    17.6             133    30.9               13.2
1983        53    16.3             114    39.7               23.4
1984        41    14.4             103    36.8               22.4
1986        46    15.8             107    31.8               16.0
1987        58    15.5              99    27.5               12.0
1988        27    14.4              70    33.0               18.6
1989        25    12.6              84    32.6               20.0
1990        24    13.4              55    24.8               11.4
1991        24    11.8              71    27.0               15.2
1993        44    19.6              81    32.5               12.9
1994        78    20.4             161    35.2               14.8
1996        60    17.3             129    38.2               10.9
1998        58    17.6             149    39.7               12.1

Table 6

Percent of all respondents stating strong agreement that "people should
take care of themselves" rather than the "government should improve
the living standards of all poor Americans." * Approximate lower decile
vs. approximate upper decile. 1975-1998. General Social Survey.

        Approximate      Approximate
        bottom decile     top decile
                                            Percentage
Year     N      %          N      %      point difference

1975     33    16.1       82    34.9          18.8
1983     29    16.5       59    41.8          25.3
1984     19    11.2      103    36.8          25.6
1986     21    15.8       57    37.0          21.2
1987     35    17.7       65    29.3          21.6
1988     11    11.7       46    35.1          23.4
1989     14    14.0       30    30.6          16.6
1990      7     8.0       28    27.2          19.2
1991     10    10.6       41    35.7          25.1
1993     21    22.3       42    40.4          18.1
1994     31    17.1       71    39.7          22.6
1996     31    16.6       85    41.7          25.1
1998     27    16.5       72    42.6          26.1


* Responses=strongly agree that government should improve living
standards, agree, agree with both, agree that people should take care
of themselves, strongly agree.

Table 7

Percent response to "In general, how good are labor unions for the
country as a whole?" by union and nonunion households. General
Social Surveys 1988-91.

                                  Union household

                        Respondent    Spouse in    Both in    Nonunion
                         in union       union       union     household

                100%          10.8          4.9        0.9         83.4

How good are unions?
          Excellent           13.1          0.0        0.0          1.9
          Very Good           35.7         21.1       14.3         15.1
        Fairly Good           39.3         68.4       71.4         47.1
      Not very good            7.1          5.3        0.0         20.1
    Not good at all            2.4          2.6       14.3          5.9
       Can't choose            2.4          2.6        0.0          9.9
            Total %          100.0        100.0      100.0        100.0

References

Baggette, J. R. Y. Shapiro, and L. R. Jacobs. 1995. "The Polls--Poll Trends: Social Security--An Update" Public Opinion Quarterly 59:420-442.

Brooks, C. and S. Cheng. 2001. "Declining Government Confidence and Policy Preferences in the U. S.: Devolution, Regime Effects, or Symbolic Change?" Social Forces 79(4):1343-1375.

Cook, F. L. and Barrett, E. J. 1992. Support for the American Welfare State New York: Columbia University Press.

Davis, J. A. and T. W. Smith. 1996. General Social Surveys, 1972-1996 Storrs, CT: The Roper Center.

Demos. 2002. New Opportunities: Public Opinion on Poverty, Income Inequality and Public Policy: 1996-2001 http://www.demos-usa.org/Pubs/POReport/

DiMaggio, P., J. Evans, and B. Bryson. 1996. "Have Americans' Social Attitudes Become More Polarized?" American Journal Of Sociology 102(3):690-755.

Domhoff, G. W. 1996. State Autonomy or Class Dominance Hawthorne, New York: Aldine de Gruyter.

Domhoff, G. W. 1990. The Power Elite and the State: How Policy is Made in America Hawthorne, New York: Aldine de Gruyter.

Domhoff, G. W. 1967. Who Rules America? Englewood Cliffs: New Jersey.

Domhoff, G. W. and T. R. Dye. 1987. Political Elites and Organizations Newbury Park, California: Sage Publications.

Downey, D. J. 2000. "Situating Social Attitudes Toward Cultural Pluralism: Between Culture Wars and Contemporary Racism." Social Problems 47(1)-90-111.

Downey, D. J. and M. L. Huffman. "Attitudinal Polarization and Trimodal Distributions: Measurement Problems and Theoretical Implications." Social Science Quarterly 82(3):494-505.

Epstein, W. M. 1997. Welfare in America: How Social Science Fails the Poor Madison, Wisconsin: University of Wisconsin Press.

Epstein, W. M. Forthcoming 2006. "Response Bias in Opinion Polls and American Social Welfare." Social Science Journal.

Erskine, H. 1975. "The Polls: Government Role in Welfare" Public Opinion Quarterly 39:257-274.

Evans, J. H. 1997. "Worldviews or Social Groups as the Source of Moral Value Attitudes: Implications for the Culture Wars Thesis," Sociological Forums 12(3):371-404.

Evans, J. H., B. Bryson, and P. DiMaggio. 2001. "Opinion Polarization: Important Contributions, Necessary Limitations." American Journal of Sociology 106(4):944-959).

Gitlin, T. 1995. The Twilight of Common Dreams New York: Henry Holt.

Gordon, L. 1994. Pitied but not Entitled New York: Free Press.

Gordon, L. 1992. "Social Insurance and Public Assistance: The Influence of Gender in Welfare Thought in the United States, 1890-1930." American Historical Review February.

Hamowy, R. 1987. The Scottish Enlightenment and the Theory of Spontaneous Order Carbondale, Illinois: Southern Illinois University Press.

Hoffman, J. P. and A. S. Miller. 1998. "Denominational influences on Socially Divisive Issues: Polarization or Continuity?" Journal for the Scientific Study of Religion 37(3):528-546.

Hunter, F. 1953. Community Power Structure Chapel Hill: University of North Carolina Press.

Hunter, J. D. 1991. Culture Wars: The Struggle to Define America New York: Basic Books.

Hunter, J. D. 1994. Before the Shooting Begins: Searching for Democracy in America's Culture Wars New York: The Free Press.

Kuttner, R. E. 1985. "The State of the Unions." New Republic 192(13):25-29.

Lindaman, K. and D. P. Haider-Markel. 2002. "Issue Evolution, Political Parties, and the Culture Wars." Political Research Quarterly 55(1):91-110.

Lowi, T. J. 1964. "American Business, Public Policy, Case-Studies, and Political Theory," World Politics 16(4):677-715.

Miller, A. S. and J. P. Hoffman. 1999. "The Growing Divisiveness: Culture Wars or a War of Words?" Social Forces 78(2):721-752.

Miller, W. E. and S. A. Traugott. 1989. American National Election Studies Data Sourcebook 1952-1986 Cambridge, MA: Harvard University Press.

Mills, C. W. 1956. The Power Elite New York: Oxford University Press.

Mouw, T. and M. E. Sobel. 2001. "Culture Wars and Opinion Polarization: The Case of Abortion." American Journal of Sociology 106(4):913-943.

O'Connor, A. 2001. Poverty Knowledge Princeton, New Jersey: University of Princeton Press.

Page, B. I. and R. Y. Shapiro. 1992. The Rational Public: Fifty Years of Trends in Americans' Policy Preferences Chicago: University of Chicago Press.

Public Agenda. 1995. The Values We Live By: What Americans Want from Welfare Reform New York: Public Agenda.

Renshon, S. A. 2000. "Political Leadership as Social Capital: Governing in a Divided National Culture." Political Psychology 21(1):199-226.

Robson, W. A. 1976. Welfare State and Welfare Society Great Britain: George Allen and Unwin.

Schiltz, M. E. 1970. Public Attitudes Toward Social Security: 1935-1965 Washington, DC: Unites States Department of Health, Education, and Welfare.

Skocpol, T. 2000. The Missing Middle New York: W. W. Norton.

Skocpol, T. 1995. Social Policy in the United States: Future Possibilities in Historical Perspective Princeton, New Jersey: Princeton University Press.

Skocpol, T. 1992. Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States Cambridge: the Belknap Press.

Shapiro, R. Y., K. D. Patterson, and J. T. Young. [1986] "Economic Status and Other Influences on Pubic Opinion toward Social Welfare Policies." Annual Meeting of the Northeastern Political Science Association, November 13-15, Boston.

Shapiro, R. Y., K. D. Patterson, and J. T. Young. 1987. "The Polls: Public Assistance." Public Opinion Quarterly 51:120-130.

Social Security Administration. 1998. Income of the Aged Chartbook, 1996 Washington, D. C.: Social Security Administration, Office of Research, Evaluation, and Statistics.

Teles, S. M., 1998. Whose Welfare? AFDC and Elite Politics Lawrence, Kansas: University Press of Kansas.

Ways and Means Committee, United States House of Representatives. 2000. The Green Book Washington, D. C.: Government Printing Office.

Weaver, R. K., R. Y. Shapiro, and L. R. Jacobs. 1995. "The Polls--Trends: Welfare" Public Opinion Quarterly 59:606-627.

Weir, M., S. Orloff, and T. Skocpol. 1988. The Politics of Social Policy in the United States Princeton, New Jersey: Princeton University Press.

Williams, R. 1997. Cultural Wars in American Politics: Critical Reviews of a Popular Myth New York: Aldine de Gruyter.

Wolfe, A. 1998. One Nation, After All New York: Viking.

Wolfe, A. 1996. Marginalized in the Middle Chicago: University of Chicago Press.

WILLIAM M. EPSTEIN

University of Nevada, Las Vegas

School of Social Work