Black Class Exceptionalism:
Insights from Direct Democracy on the Race vs Class Debate
Zoltan L. Hajnal, University of CaliforniaSan Diego*
Abstract
What implications does the growing economic divide between poor and middle class blacks have for the political arena? Traditional accounts suggest that increased economic diversity should lead to increased political division as the middle class becomes more conservative. Others maintain that race will continue to trump class because of ongoing racial inequality and widespread racial discrimination. I argue for a third alternative. I suggest that for blacks and possibly for other racial minorities increasing class status reinforces race. Class gains may increase the salience of race because economic success often means working in a predominantly white world and experiencing discrimination more regularly. I test these theories using the vote in direct democracy. I find that middle class blacks are more rather than less likely to support a liberal or black agenda. Class works differently for African Americans than for whites.
* Zoltan L. Hajnal is Assistant Professor of Political Science, University of California, San Diego, La Jolla, CA92093-0521. (). The author wishes to thank Neal Beck, Amy Bridges, Darren Davis, Paul Frymer, Claudine Gay, Andrew Grant Thomas, Jennifer Hochschild, Sam Kernell, Taeku Lee, Chris Parker, John Skrentny, and Barbara Walter for helpful comments on earlier drafts.
A substantial economic gulf has emerged within the African American community. While the proportion of blacks in the middle class has more than doubled, the number of blacks who are not only poor but who are also living in extremely poor inner-city neighborhoods has also doubled.[1] As a result, blacks are in many ways more divided than are whites. The disparity between the top and bottom fifth of the black population in terms of income, education, victimization by violence, and job status is now greater than the disparity between the top and bottom fifth of the white population (Hochschild and Rodgers 1999). In other words, African Americans are increasingly being divided into haves and have nots.
This growing economic diversity raises important questions about the future of black politics. Traditionally, African Americans have formed one of the most cohesive and reliably liberal political groups in the country. Although blacks are not as cohesive as some assume and surveys have revealed real divisions on non-racial policy questions, blacks still seem to be firmly united and Democrat when it comes to the vote (Welch et al 2001, Cohen 1999, Bositis 2000, Dawson 1994, Tate 1994). There are, however, concerns that this unity will not continue in the future. One group of scholars has suggested that economic diversity will lead to political divisions as the black middle class becomes more conservative and less willing to support a liberal, pro-black agenda (Wilson 1980, 1987, Thernstrom and Thernstrom 1997). But others sharply disagree. A second set of scholars has maintained that ongoing racial inequality and continued discrimination mean that race will continue to trump class. A sense of linked fate will leads blacks of all classes to maintain a united political front (Dawson 1994). Or is yet a third alternative more likely? Ironically, more in-depth studies of the black middle class have found that the economic success of the black middle class is bringing them into a largely white world in which they are more apt than other blacks to experience discrimination and thus more apt to feel racialized and potentially more eager to support a black agenda (Hochschild 1995). This racialized world might then lead to increased rather than decreased support for a pro-black agenda.
The degree to which economic heterogeneity alters black political preferences will likely provide critical insight into the meaning of class. But this is more than a theoretical debate. How divided or united the black community is politically has wide-ranging implications not only for how regularly blacks win elected office but also for the ability of African Americans to get a favored policy agenda passed. The willingness of either major party to begin to seriously court black voters and the attractiveness of blacks as a coalition partner in America’s increasingly racially diverse population are also very likely to be tied in different ways to how conservative or cohesive blacks are politically. Thus, how and why class matters is critical to the future of the black community.
To test these three different theories of racial and class politics, I turn to a rarely used measure of public opinion, the vote in direct democracy. Three features of direct democracy - relatively limited partisan influences, real and immediate costs to the outcome, and a wide array of policy questions on the ballot -may mean not only that direct democracy is likely to provide a different perspective from candidate elections or most public opinion surveys but also that it could offer a potentially telling look at class divisions.
Analysis of the direct democracy vote in California over the last three decades indicates that on most concrete policy decisions African Americans are at least somewhat divided. The analysis also shows that predictions of a major class divide in the black community are overblown. The growing economic status of the black middle class has not led to growing conservatism or the abandonment of a traditional black political agenda. Rather, the opposite seems more true. Patterns of voting in direct democracy indicate that middle class blacks are slightly more supportive of a liberal agenda than are less well off blacks. Class does matter but it shapes policy decisions differently for blacks than for Anglos. The racial solidarity of middle class blacks suggests that race is not likely to fade away in the near future, and that further increases in the economic position and integration of blacks may reinforce rather than reduce the significance of race.
The paper proceeds as follows. I begin by detailing three different accounts of the role of class in black politics: one that highlights the potentially important role of economic heterogeneity in creating a class divide, a second that emphasizes the ongoing importance of race for all African Americans, and a third which focuses on the increasingly racialized world of middle class blacks and argues that class works differently for blacks than for whites. I then briefly describe the direct democracy data and list some of the reasons why direct democracy is a potentially telling measure of public opinion. This is followed by a presentation of the analysis of the direct democracy vote in California and finally by a brief discussion of the implications of this research.
Three Views of Class Politics
Middle Class Conservatism?
Should the growing economic divide in the black community affect the cohesiveness of the black community and the direction of black politics? Standard accounts of class politics maintain that political decisions are shaped by material interests and that as a consequence economic heterogeneity should lead to political division (Lipset 1981[1960], Weber 1946). Extended debates have been waged over whether this class effect is declining and on what set of issues class still matters but the basic story remains largely intact.[2] The notion is that as one moves up the economic ladder, one’s material interests change. As class status increases, individuals should be more interested in maintaining their economic advantages, less interested in economic redistribution, and ultimately more supportive of a conservative economic agenda.
Several scholars have argued that this logic should apply to African Americans as well (Thernstrom and Thernstrom 1997, Wilson 1980). For Wilson (1980, 1987) structural changes that have devastated poor black communities and changes in race relations that have allowed middle class blacks to compete for high skilled, high paying jobs have led to a situation in which the interests of middle class blacks are increasingly distinct from those of poor or working class blacks. Especially on issues that redistribute resources from the well off to more disadvantaged groups, the new status and interests of the black middle class should clash with the liberalism of a traditional black agenda and political unity should decline (Hamilton 1982).
Race Trumps Class
A far more common view among scholars of the black community is that despite significant economic gains among some members of the black population, race continues to trump class. According to this view, recent changes have simply not been large enough to overcome centuries of exclusion and discrimination. African Americans still lag far behind whites on most measures of well-being (Blank 2001). Moreover, there are signs that prejudice and racial discrimination persist and that race retains its significance in American life (Kirschenman and Neckerman 1991, Massey and Denton 1993). Thus there is much reason for blacks to remain racially conscious and to believe that their individual fates remain tied to the fate of the whole black community and, as a consequence, little reason to expect significant class division in the political sphere (Dawson 1994).
Class Gains Reinforce Race
More recently, a third view of the black middle class has emerged. This third view agrees that race strongly shapes black life chances. But it differs from earlier views in that it also argues that different segments of the black community experience the effects of race differently. Ironically, in-depth accounts of the black middle class experience have revealed a world in which race is more rather than less important. Despite the growing economic advantages of the black middle class, these studies suggest that because middle class blacks often work or live in predominantly white environments, they are subject to racial misperceptions and racial hostility more regularly than other blacks (Cose 1995, Feagin 1991). For these ‘successful’ blacks, glass ceilings, small slights, and more blatant acts of prejudice all lead to frustration and anger and above all else reinforce rather than erode racial consciousness (Hochschild 1995, Feagin 1991).
Moreover, even in cases where whites do not actively discriminate against middle class blacks, the mere fact that middle class blacks are likely to be one of a small number of black ‘representatives’ means that race is likely to be more salient for this group. Because middle class blacks will almost always be easily identified as black and will be unable to hide or assimilate the way white ethnics have done in the past, race for middle class blacks is likely to be “a consuming experience” (Willie 1978:15). As such, the black middle class experience is likely to be fundamentally different from the traditional middle class experience.
Unfortunately, none of these studies has theorized about how this unique black middle class experience should affect the politics and more specifically the policy preferences of the black community. Nevertheless, I believe the implications are clear. If the black middle class experience does, in fact, lead to heightened racial consciousness then this should offset the traditional effects of class. To the extent that race grows in importance rather than fades in importance, class gains should be associated with increasing support for a set of liberal, pro-black policies. If true, class will matter for blacks but it will work in the opposite direction from its effects on the white community.
Measuring the Effects of Class: Why Direct Democracy?
Conventionally, to evaluate these divisions within the black community, studies have looked at partisanship, voting in candidate elections, or attitudes in public opinion surveys. At least two of these three measures point to relatively high levels of political unity within the black community and all three suggest that class divisions are minimal. The numbers in terms of partisanship are clear. Over the last three decades, African American allegiance to the Democratic Party has never dropped below 80 percent (Tate 1994). Similarly, analysis of the black vote in candidate based elections tends to reveal a community that is solidly Democratic, clearly liberal, and largely united(Hamilton 1982, Parent and Stekler 1985, Bositis 2000). In partisan contests at the state and national level, Democratic candidates usually garner the vast majority of the black vote. Even in nonpartisan contests at the local level, blacks appear to form a cohesive voting bloc favoring left-leaning candidates (Deleon 1991, Lieske and Hillard 1984). This unity is ultimately reflected in the fact that less than one percent of all black elected officials are Republicans (Bositis 1996). Research based on public opinion surveys is more mixed. Many public opinion surveys highlight a number of issues on which blacks are far from united(Welch et al 2001, Dawson 2001, Cohen 1999, Bositis 2000). Blacks, for example, appear to be just as fragmented as whites on women’s rights, immigration, crime, and foreign affairs (Hochschild and Rogers 1999, Bositis 2000, Hajnal and Baldassare 2001). Some even claim that outside a core set of racial policy questions, blacks are not distinct ideologically from whites (Canon 1999). Importantly, however, even the available survey evidence suggests that these divisions are generally unrelated to class (Welch and Combs 1985, Parent and Steckler 1985). Moreover, public opinion studies have found that regardless of class most blacks support a pro-black agenda on a core set of racially relevant issues (Kinder and Sanders 1996, Dawson 2001, Gilliam 1986, Welch and Foster 1987).[3] The overwhelming perception among observers of American politics and the conclusion of most scholars is that, “African American political choices are not significantly influenced by individual economic status or level of education” (Dawson 1994:157).
There are, however, reasons to suspect that these conventional measures may minimize class divisions and overstate black solidarity in the political arena. To further assess how class divisions affect political interests, I turn to a seldom used measure of public opinion, the direct democracy vote. The vote in direct democracy differs from conventional measures in at least two ways that may help to gauge how class differences alter individual political interests.
First, unlike in most candidate elections, in direct democracy political parties usually play a limited role. The relative absence of party labels may make it easier for individual African Americans to oppose the black majority when their economic circumstances differ from the rest of the community. Second, there are real, immediate, and often times considerable consequences to the vote in direct democracy. The passage of one recent proposition in California, for example, committed $12 billion dollars to education in the state. With all of these resources at stake, there is likely to be a strong incentive to vote based on material interests and there may be higher costs to voting out of allegiance to a particular group. Thus, direct democracy may make it more difficult to choose race over class. None of this means that direct democracy will necessarily be more accurate or more informative than other measures of public preferences but it does suggest that direct democracy could offer a unique perspective that may weigh certain considerations like material interests more heavily.
Limitations of Direct Democracy
The vote in direct democracy has potential advantages over traditional measures but it is not without limitations. First, as with other analyses of the vote, in direct democracy the preferences of non-voters are ignored. In some ways this is an advantage in that direct democracy only measures ‘activated’ public opinion (Lee 2002). However, to the extent that the preferences of non-voters differ from those of voters, assessments based on direct democracy will not accurately reflect the views of the entire public. Fortunately, the majority of research on this subject suggests that non-voters do not have substantially different preferences than voters (Verba, Schlozman, and Brady 1995, Wolfinger and Rosenstone1980). As Ellcessor and Leighley note in a recent article, “one of the least contested conclusions in the study of political behavior is that voters’ political attitudes and policy positions are fairly representative of non-voters” (2001:127).[4] The skew created by looking only at voters should be relatively limited.
Another potential problem with using the vote in direct democracy is that voters may not know much about the policies they are voting on (Magleby 1984). While this is clearly the case for some voters, it does not appear to be a problem for most voters. Recent research on voting patterns in initiative elections suggests that individuals do tend to vote rationally - supporting propositions that benefit their own interests and changing their minds over the course of the campaign when important new information is made public (Bowler and Donovan 1998, Lupia 1994). Moreover, it is not at all clear that this problem is worse in direct democracy than it is on conventional measures. Existing evidence suggests that voters in candidate elections and respondents in surveys are often poorly informed and regularly give top-of-the-head responses that may be only distantly related to true preferences (Delli Carpini and Keeter 1996,Zaller 1992, Ferejohn and Kuklinski 1990). Since the vote on statewide initiatives often follows intense and protracted campaigns in which the two sides compete for the hearts and minds of voters, voters in direct democracy may actually be relatively well-informed. Any study of public opinion is likely to be plagued by poorly informed respondents but voters in direct democracy may have at least some advantages over survey respondents and voters in candidate elections.