The Declining Significance of Race: Revisited & Revised

The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters

Citation Wilson, William Julius. 2011. The Declining Significance of Race: Revisited & Revised. Daedalus 140.2: 55-69.

Published Version http://dx.doi.org/10.1162/DAED_a_00077

Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:8052151

Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA The Declining Signi½cance of Race: Revisited & Revised

William Julius Wilson

I published The Declining Signi½cance of Race: Blacks and Changing American Institutions thirty-two years ago, in 1978.1 Given the furor and controversy over the book immediately following its publication, I did not anticipate that it would go on to become a classic. Indeed, the book’s impact on the ½eld of race and ethnic relations–its arguments have been discussed in nearly eight hundred empirical re- search articles, not to mention the nonempirical studies–lends credence to the idea of productive controversy and to George Bernard Shaw’s famous dictum: “[I]t is better to be criticized and misun- derstood than to be ignored.” My motivation for this essay is to reflect on responses to the book that claim to provide an empirical test of my the- sis. In the process, I indicate the extent to which important ½ndings have influenced my thinking WILLIAM JULIUS WILSON, a since the book’s publication. Fellow of the American Acade- my since 1988, is the Lewis P. and he theoretical framework in The Declining Signi½- Linda L. Geyser University Pro- T fessor at Harvard University. His cance of Race relates racial issues to the economic recent publications include When and political arrangements of society. I argued that Work Disappears: The World of the changes in the system of production and in govern- New Urban Poor (1996), which re- ment policies have affected, over time, black/white ceived the Sidney Hillman Foun- access to rewards and privileges as well as racial dation Award; There Goes the Neigh- antagonisms. I advanced this framework to accom- borhood: Racial, Ethnic, and Class plish two major objectives: (1) to explain histori- Tensions in Four Chicago Neighbor- hoods and Their Meaning for Amer- cal developments in U.S. race relations and (2) to ica (with Richard Taub, 2007); and account for paradoxical changes in the black class More than Just Race: Being Black structure whereby, beginning in the last few de- and Poor in the Inner City (2009). cades of the twentieth century, the social and eco-

© 2011 by the American Academy of Arts & Sciences

55 “The nomic conditions of the black poor of individual African Americans. Several Declining deteriorated while those of the black historical shifts accounted for these devel- Signi½cance of Race”: middle class improved. opments. In the preindustrial and indus- Revisited In an elaboration of this framework, I trial periods, the basis of racial inequality & Revised focused on three periods of American was primarily economic; in most situa- race relations: the preindustrial period tions, the state was merely an instrument of antebellum slavery and the early post- to reinforce patterns of race relations bellum era; the industrial period that be- that grew out of the social relations of gan in the last quarter of the nineteenth production.2 Except for the brief period century and ended at roughly the New of fluid race relations in the North from Deal era; and the modern industrial 1870 to 1890, the state was a major instru- post–World War II era. I pointed out ment of racial oppression. that whether one focuses on the way State intervention designed to promote race relations were structured by the racial equality, together with the recipro- system of production, the polity, or both, cal relationship between the polity and racial oppression–ranging from the ex- the economy, has characterized the mod- ploitation of black labor by the business ern industrial period. Indeed, it is dif½- class (including the plantation elite) to cult to determine which factor has been the elimination of black competition for more important in shaping race relations economic, political, and social resources since World War II. Economic expansion by the white masses–was characteristic facilitated black movement from the ru- of both the preindustrial and industrial ral areas of the South to the industrial periods of American race relations. centers and created job opportunities However, I noted that despite the prev- leading to greater occupational differen- alence of various forms of racial oppres- tiation in the African American commu- sion, the change from a preindustrial nity, as an increasing percentage of blacks to an industrial system of production moved into semiskilled and skilled blue- enabled African Americans to increase collar positions and white-collar posi- their economic and political resources. tions. At the same time, government in- The proliferation of jobs created by in- tervention (in response to the pressures dustrial expansion helped generate and of increased black political resources and sustain the continuous mass migration the civil rights protest movements) re- of blacks from the rural South to urban moved many arti½cial discrimination centers, especially the cities of the North barriers with municipal, state, and feder- and West. As the urban black population al civil rights legislation. Moreover, state grew and became more segregated, insti- intervention contributed to the more lib- tutions and organizations in the African eral racial policies of the nation’s labor American community also developed unions with protective union legislation. alongside a business and professional These combined economic and political class af½liated with these institutions. changes created a pattern of black occu- Nonetheless, it was not until after pational upgrading that resulted, for ex- World War II (the modern industrial ample, in a substantial decline of African period) that black class structure began Americans in low-paying service jobs, to take on some of the characteristics of unskilled labor, and farm jobs. white class structure and that economic Given greater occupational differentia- class gradually became more important tion, some aspects of structural economic than race in determining the life chances change have resulted in a closer associa-

56 Dædalus, the Journal ofthe American Academy of Arts & Sciences tion between black occupational mobili- low-wage sector nor the corporate and William ty and class af½liation. Access to higher- government sectors provide the basis for Julius Wilson paying jobs is increasingly based on edu- the kind of interracial job competition cational criteria–a situation that distin- and conflict that plagued the economic guishes the modern industrial period order in previous years. The absorption from earlier systems of production and of blacks into industrial unions and the that has made the position of the black federal government’s protective union poor more precarious. In other words, legislation effectively negated manage- the rapid growth of the corporate and ment’s ability to undercut the demands government sectors has created a seg- of white workers for higher wages by mented labor market that currently pro- replacing them with black workers. The vides vastly different mobility opportu- traditional racial struggles for power and nities for different segments of the Afri- privilege have largely shifted away from can American population. On the one the economic sector and are now con- hand, poorly trained and educationally centrated in the sociopolitical order, as limited African Americans have seen racial tensions have more to do with ra- their job prospects increasingly limited cial competition for public schools, mu- to low-wage sector jobs, they have faced nicipal political systems, and residential rising rates of unemployment and non- areas than with competition for jobs. labor-force participation, and they have Although these developments within endured slower movements out of pov- the sociopolitical order also affect the erty. On the other hand, trained and edu- ultimate life chances of African Ameri- cated African Americans have experi- cans, their respective impact on social enced increased job opportunities in the mobility opportunities is not as great as corporate and government sectors as a racial competition and antagonisms in result of the expansion of white-collar the economic sector. positions and the pressures of state af- Thus, the original argument, as out- ½rmative action programs. lined in The Declining Signi½cance of Race, Accordingly, the mobility pattern of was not that race is no longer signi½cant blacks is consistent with the view that in or that racial barriers between blacks the modern industrial period, economic and whites have been eliminated. Rather, class has become more important than in comparing the contemporary situation race in predetermining job placement of African Americans to their situation and occupational mobility for African in the past, the diverging experiences of Americans. In the economic realm, the blacks along class lines indicate that race black experience has moved historically is no longer the primary determinant of from economic racial oppression experi- life chances for blacks (in the way it had enced by virtually all African Americans been historically). to the economic subordination of the black poor. As a result, a deepening eco- In a paper reflecting on the critical nomic schism has developed in the Afri- reaction to The Declining Signi½cance of can American community, with the Race immediately following publica- black poor falling further and further tion of the book, the late sociologist behind higher-income blacks. Robin M. Williams, Jr., pointed out: Moreover, the center of racial conflict has shifted from the industrial sector to Despite the author’s explicit quali½cations the sociopolitical order. Neither the and speci½cations, some critics seem to

140 (2) Spring 2011 57 “The miss one of the author’s central points: Of the universe of empirical studies Declining that both racial discrimination and class that claim to respond to The Declining Signi½cance of Race”: position importantly affect life-chances Signi½cance of Race, I would like to high- Revisited and that it is the changing character of the light the high-quality publications that & Revised interaction of the two structural conditions correctly address my thesis–including that is critical for understanding the pres- studies that fundamentally uphold or ent situation. The increasing differentials provide partial support for my arguments within the black population in income, as well as those that challenge my basic education, occupational prestige, and claims. In the process, I will show how power-authority seem clear beyond dis- some of these studies have led me to re- pute. That past-institutionalized racism vise or extend parts of my basic thesis, has powerfully shaped these differentials especially as it pertains to race and in- is equally plain, as is the fact that large terracial relations today. average interracial differentials continue to exist. What Wilson argues is only that In her important book Facing Up to the economic class has become more impor- American Dream, Harvard political scien- tant than race in determining job place- tist Jennifer Hochschild states, “One has ment and occupational mobility, as sig- not really succeeded in America unless nalized by the growth of a black middle one can pass on the chance for success to class concurrently with the crystalliza- one’s children.”6 She highlights research tion of a disproportionately large black on the occupational attainments and underclass.3 mobility of blacks revealing that, as late As Williams indicates, my basic argu- as 1960, there was no evidence to suggest ment, including its underlying thesis– that the effect of economic class position regarding the effect of economic and po- could rival the effect of race in terms of litical changes on the shifting relative im- blacks’ achievements in occupation and portance of race and class in black occu- income. Race, or skin color, was such pational mobility and job placement– a powerful factor in life that it clearly is largely unaddressed in the many hun- trumped class. As Hochschild puts it, dreds of studies responding to The De- blacks “‘experienced a perverse sort of clining Signi½cance of Race. For example, egalitarianism’–neither the disadvan- many of the articles whose titles play on tages of poverty nor the advantages of the phrase “the declining signi½cance of wealth made much difference in what race” focus on issues that do not relate they could achieve or pass on to their to my speci½c arguments–issues such children. Discrimination swamped as environmental racial inequality, skin everything else.” However, beginning in tone differentials, mate selection, Amer- the early 1960s, she argues, class began ica’s drug crisis, capital punishment, and to affect career and generational mobil- psychological well-being and quality of ity for blacks as it had done regularly life.4 Others discuss residential segrega- for whites: “Well-off black men thus tion, school racial composition, and dis- could begin for the ½rst time in Ameri- crimination in public places without re- can history to expect their success to lating reported ½ndings to my arguments persist and accumulate. Since 1973 these trends have continued, although regarding the shift in the concentration 7 of racial antagonisms from the econom- less dramatically.” ic sector to the sociopolitical order.5 The research that Hochschild cites includes an important study by sociolo-

58 Dædalus, the Journal ofthe American Academy of Arts & Sciences gist Michael Hout of the University of region, Sakamoto and Tzeng found that William California, Berkeley.8 Analyzing data on the effect of race was smaller in 1990 Julius Wilson intergenerational and intragenerational than in 1940 for every level of education mobility of black men from the Occupa- and sector attainment investigated. Fur- tional Changes in a Generation surveys thermore, when comparing the impact of 1962 and 1973, Hout found support for of education with that of being black, arguments advanced in The Declining Sig- they found that for the vast majority of ni½cance of Race. More speci½cally, he re- black men in 1940, the racial disadvan- vealed that between 1962 and 1973, class tage was greater in absolute value than signi½cantly affected intragenerational the effect of education was; in 1990, how- mobility for African Americans–a phe- ever, the reverse was true: education was nomenon similar to class effects among a much more signi½cant factor than be- whites. As class differences in intergener- ing black. Finally, class effects–in terms ational mobility increased, African Amer- of relative educational attainment–sub- ican men from the most advantaged stantially increased over this time span socioeconomic backgrounds experi- for black men. “These results,” state enced the greatest upward mobility.9 Sakamoto and Tzeng, “support Wilson’s Although Hout’s ½ndings are impor- thesis of the declining signi½cance of tant, as sociologists Arthur Sakamoto race, and they are consistent with his and Jessie M. Tzeng explain, they “gen- claim that in the modern industrial pe- erally pertain to the period immediately riod after the civil rights movement, before and after the civil rights move- ‘economic class position [is] more im- ment”; therefore, they do not cover the portant than race in determining black wide temporal span of The Declining Sig- chances for occupational mobility.’”13 ni½cance of Race, “which is about changes Nonetheless, this comparison over across broad historical periods.”10 By broad historical periods should not lead analyzing the 1940 and 1990 Public Use us to overlook changes in the relative Microdata Sample (pums) data sets (a importance of race and class within the large, nationally representative sample current modern or postindustrial period. of the occupational attainment of black Here, I would include changes that nar- and white males in all sectors of the la- row or increase the role that either race bor force), Sakamoto and Tzeng were or class plays in black occupational ad- able to test my thesis over a broader time vancement. On this connection, Michael span.11 They found that whereas race was Hout’s signi½cant 1984 ½ndings revealed generally more important than class in that public-sector employment “provid- determining occupational attainment ed more high and middle-class occupa- among blacks during the industrial peri- tions for black men than did the private od of 1940, class was clearly more impor- sector employment” and therefore played tant than race in determining occupation- “an important role in both occupational al attainment among black men during upgrading among blacks and the emer- the modern industrial period of 1990. gence of class cleavages within the black Indeed, their results “indicate that the population.”14 In The Declining Signi½cance net disadvantage of being black is sub- of Race, I did not highlight the relative stantially greater in the industrial period contribution of the government sector than in the modern industrial period.”12 and the corporate sector to black occu- More speci½cally, after controlling for pational gains. Given Hout’s ½ndings labor-force experience, schooling, and (and his subsequent research on this is-

140 (2) Spring 2011 59 “The sue, as discussed below), if I were writing when they entered the labor market. By Declining The Declining Signi½cance of Race today, I contrast, younger educated blacks are Signi½cance of Race”: would not only place greater emphasis now entering, and are encouraged to en- Revisited on black gains in the public sector and ter, ½nance, accounting, management, & Revised the major role of the polity in the crys- chemistry, engineering, and computer tallization of a black class structure, I science–½elds from which they were would also underline the role and impor- deterred previously. I quoted a 1978 paper tance of af½rmative action programs. In by Clifton Wharton, then chancellor of the process, I would discuss the impact the State University of New York, who of a possible contraction in government stated, “[I]n 1966, 45 percent of all black employment as well as waning public college graduates were majoring in edu- support for af½rmative action on the oc- cation, today only 26 percent are. In 1966 cupational mobility of the more advan- only 5 percent of the Blacks were study- taged and educated African Americans, ing business, today 18 percent are.”16 issues to which I now turn. I also stated that prior to the 1970s, African American men more often en- Using data from the Current Population rolled in education programs than in pro- Survey, sociologist Melvin E. Thomas grams that prepare students for higher- demonstrated that “contrary to the as- paying corporate jobs, such as business sumption of the declining signi½cance or accounting. For all these reasons, the of race thesis, blacks with higher levels incomes of older educated black males of education were found to be worse off lag signi½cantly behind the incomes than less educated, lower status blacks of comparable white males, whereas when compared to similar whites.”15 I younger college-educated black males ½nd two shortcomings with Thomas’s had approached income parity with treatment of my thesis. First, Thomas their white counterparts. failed to disaggregate the data to show Recognizing the need to focus on comparisons between younger and older younger educated blacks in the post– educated blacks. Second, he neglected civil rights period to provide “an appro- to mention that in the second edition priate test of the declining signi½cance of The Declining Signi½cance of Race (pub- of race within the black middle class,” lished in 1980), I referred to the signi½- sociologists In Soo Son, Suzanne W. cant income gap between all college- Model, and Gene A. Fisher examined educated African Americans and all col- “interracial differences in the net effect lege-educated whites that still exists, of higher education among young work- noting that this ½nding was largely a ers who entered the labor market after consequence of the substantially lower the mid-1960s.” Analyzing data from incomes of older educated blacks. the Panel Study of Income Dynamics Denied the opportunity to move into (psid) from 1968 to 1981 on the occu- higher paying occupations when they pational mobility and earnings attain- graduated from college, or discouraged ment of young black and white males, from pursuing such careers, older black the authors found “evidence of class college graduates tend to be concentrated polarization among blacks in the era in lower-paying ½elds such as teaching, following the 1960s’ antidiscrimination social welfare, and segregated services; legislation.” In 1974, blacks lacking a they were rarely employed as executives high school diploma earned 57 percent or professionals in large corporations of what black college graduates earned,

60 Dædalus, the Journal ofthe American Academy of Arts & Sciences while the ½gure for comparable whites based on a subset of the data used by William was 65 percent. By 1981, blacks without Hout, he stated: “Opportunities for Julius Wilson a high school diploma earned only 36.6 blacks were best in the public sector percent of what blacks with a college where the observed rate of intrasector degree earned, while the analogous ratio upward mobility was actually higher for the two groups of whites declined for blacks than for whites. . . . Since the only to 58.5 percent.17 public sector offers the most opportu- Moreover, Son, Model, and Fisher nity for black advancement, reductions found not only that African American in government employment are likely to men without a high school degree con- be especially detrimental to blacks.”20 sistently earn the smallest proportion of A 1996 study by A. Silvia Cancio, T. comparably educated whites’ incomes, David Evans, and David J. Maume pre- but they were the only group that expe- sents data suggesting that these concerns rienced a decline in their absolute real were justi½ed. The authors also appropri- dollar earnings, bringing their 1981 earn- ately pointed out that “aggregate black/ ings to only two-thirds that of their white white earnings are invalid because older counterparts. Black high school gradu- Blacks presently earn less than whites ates’ earnings were slightly better in both because of past discrimination practices” absolute and relative terms, with an earn- and concluded that a comparison of the ings gap that increased 7 percent between salaries of young workers would be the 1974 and 1981. By contrast, the progress “most appropriate test of the signi½cance of black college graduates was substan- of race in the modern industrial period.” tially greater, with incomes that changed Using psid data from 1976 and 1985, they from fewer than 6 percent of compara- found that the effect of race, after control- ble whites’ incomes in 1974 to matching ling for other variables, increased during the income of their white counterparts this period, and that the proportion of in 1981. Even more spectacular, “[B]lack the racial gap in hourly wages due to dis- college graduates obtain more prestigious crimination (that is, after racial differ- posts than their white counterparts.”18 ences in measured quali½cations were These ½ndings are consistent with the taken into account) also increased dur- data I presented on the black/white in- ing this time span. Thus, they argued, come gap of younger college graduates “[T]he government’s retreat from anti- in the second edition of The Declining discrimination initiatives in the 1980s re- Signi½cance of Race. sulted in organizational discrimination Despite the progress of educated blacks, against blacks and contributed to the Son, Model, and Fisher warned: “[T]he reversal in the postwar trend toward racial parity achieved by young college- racial parity in earnings.”21 educated blacks in the 1970s will be main- Cancio, Evans, and Maume observed tained only if the government’s commit- that until 1980, my arguments of observ- ment to af½rmative action does not slack- able racial progress are essentially correct. en. Ideological and economic pressures However, they stated: “Wilson gave no to reduce federal spending, coupled with indication that he expected the long run a tighter business environment, could trend toward racial parity in earnings to easily lead to fewer opportunities for reverse in the 1980s. But that is what has blacks.”19 Sociologist Marshall I. Pomer happened to young cohorts.”22 In the reached a similar conclusion. In his 1986 epilogue to my book’s second edition, I article on intragenerational mobility acknowledged that vigorous af½rmative

140 (2) Spring 2011 61 “The action programs may still be needed in different decades, as government policies Declining the immediate future “because it is dif½- on the labor market have changed over Signi½cance of Race”: cult to determine if the gains that younger time. If these policies affect careers at their Revisited educated blacks are experiencing in entry starting points, does their impact last into & Revised level positions will be reflected in promo- mid- and late-careers similarly for Whites tions to higher level jobs in later years.” and Blacks?25 But I went on to say: “[A]t this point For the present essay, in the absence there is also reason to believe that trained of longitudinal studies, I examined co- and educated blacks, like trained and horts of male workers ages twenty-½ve educated whites, will continue to enjoy to twenty-nine at ten-year intervals, the advantages and privileges of their using ½gures from the Current Popula- class status. It appears that the powerful tion Survey comparable to the 1977 ½g- political and social movement against ures originally reported in the second job discrimination will mitigate against edition of The Declining Signi½cance of any effective and systematic movement Race. I found that the black/white earn- to exclude quali½ed blacks.”23 I noted ings ratio for college graduates declined that the real issue is improving the plight signi½cantly from 1977 to 1987 (blacks of the black lower class, whose condi- who graduated from college earned 93 tions have not been addressed by pro- percent as much as their white counter- grams like af½rmative action. parts in 1977, but by 1987, that ratio had The research by Cancio and her col- dropped to 73.2 percent). The ratio in- leagues suggests that my optimism con- creased by 9 percentage points between cerning the movement against job dis- 1987 and 1997 (from 73.2 percent to 82.5), crimination was unfounded: “Events in then decreased by 2.9 percent from 1997 the 1980s proved that African Americans to 2007 (from 82.5 percent to 79.6 per- cannot take for granted the political com- cent). Thus, despite some improvements mitments to af½rmative action and equal during the 1990s, by 2007, the income opportunity legislation. . . .Our results ratio of young black college-educated suggest that a waning devotion to these males was signi½cantly below the ratio ideals negatively affected the earnings of 1977.26 of Black workers.”24 Their research clear- Finally, in 1995, political scientist ly underscores the importance of the Theodore J. Davis presented ½ndings strength and direction of future govern- on the consequences of race and class ment efforts to promote racial equality. interaction for both upward and down- They also point to the need for careful ward mobility. Using data from the 1972 longitudinal studies to understand fully to 1989 Cumulative General Social Sur- the racial differences in career dynamics: vey, Davis found that although there is Blacks and Whites are more likely to be some evidence of a very gradual decline paid equally at the beginning of their ca- in the role of race in influencing occu- reers. Research that observes people at pational attainment in the 1980s, and the beginning of their work and examines although both black and white males race differences as they move through the experienced intergenerational occupa- stages of a career . . . will shed needed light tional upward mobility in the 1980s, on the experiences of Blacks within orga- black males were also more likely than nizations. Moreover, it is important to com- white males to experience downward pare cohorts who started their careers in occupational mobility.27

62 Dædalus, the Journal ofthe American Academy of Arts & Sciences Given the research discussed in this cess to jobs and schools, and even expo- William section, I reiterate a point I made earlier: sure to violence. Using individual-level Julius Wilson if I were writing The Declining Signi½cance data from the geocoded version of the of Race today, I would place greater em- psid for the years 1970, 1980, 1990, and phasis not only on the role of the public 2001 to correspond with the decadal cen- sector in accounting for black occupa- suses, urban planner Lance Freeman tional mobility, but also on the impor- found that higher socioeconomic status tance of sustained public support for anti- among African Americans is generally discrimination programs, including af- associated with greater integration and ½rmative action, to ensure that the gains improved locational outcomes. continue or, at the least, are not reversed. The strength of these relationships, However, I also need to address another however, did not increase between 1970 aspect of “the declining signi½cance of and 2000. “Class does matter,” Freeman race” thesis–namely, class changes with- states. “Higher status Blacks generally in the African American community. live in higher-status neighborhoods and those with more Whites. But the impor- One of the basic arguments of The tance of class has not increased since Declining Signi½cance of Race is that there 1970. The determinants of spatial out- has been a deepening economic schism comes for Blacks have been remarkably as reflected in a widening gap between durable at the end of the twentieth cen- lower-income and higher-income black tury. . . . It appears that Blacks will have families. In light of more recent data, to achieve upward mobility in other do- not only has the family income gap mains, such as education, before achiev- between poorer and better-off African ing widespread access to higher-status Americans continued to widen, but the and White neighborhoods.”31 Reaching situation of the bottom ½fth of black this goal may be more of a challenge for families has deteriorated since 1975 black males than for black females. (see Table 1). Indeed, what has also changed since I In 2007, 45.6 percent of all poor blacks wrote The Declining Signi½cance of Race is had incomes below 50 percent of the that the black class structure increasing- poverty line.28 Overall, poor black fam- ly reflects gender differences, especially ilies fell below the poverty line by an among younger blacks, as males have average of $9,266 in 2007, a depth of fallen behind females on a number of poverty exceeding that of all other racial/ socioeconomic indicators: employment ethnic groups in the United States.29 rates, high school completion rates, and Regardless of the reversal of the relative average income, with some of the sharp- income gains of younger educated blacks est discrepancies at the lower end of the reported in the previous section, the gap income hierarchy.32 Black women have between the haves and have-nots in the also far outpaced black men in college African American population continues completion in recent years. Despite the to grow. fact that the gender gap in college degree Research also indicates that “higher attainment is increasing across all racial socioeconomic status Blacks have more groups, with women generally exceeding White neighbors, fewer poor neighbors, men in rates of college completion, this and live in neighborhoods with higher discrepancy is particularly acute among housing values.”30 This fact is important African Americans. That gap has wid- because one’s neighborhood controls ac- ened steadily over the past twenty-½ve

140 (2) Spring 2011 63 “The Table 1 Declining Average Income of Black Families by Income Group Signi½cance of Race”: Revisited 1975–2007 1975–2007 & Revised (changes in (percent 1975 1985 1995 2005 2007 dollars) change)

Lowest Fifth $8,939 $7,284 $7,463 $7,784 $8,143 -$796 -8.9%

Second Fifth 18,533 17,833 20,073 22,085 23,384 4,851 26.2

Middle Fifth 30,650 30,832 35,022 35,842 40,278 9,628 31.4

Fourth Fifth 46,095 49,396 55,408 61,407 64,573 18,478 40.1

Highest Fifth 78,031 90,902 111,767 129,002 132,565 54,534 69.9

Top 5 Percent 106,908 131,672 183,471 212,818 220,916 114,008 106.6

All ½gures reported in 2007 dollars. Source: U.S. Census Bureau, Current Population Survey, 2008 Annual Social and Economic Supplements, Table F-3, “Mean Income Received by Each Fifth and Top 5 Percent of Families.”

years. In 1979, for every 100 bachelor’s creasing proportion of black men in degrees earned by black men, 144 were higher socioeconomic positions. earned by black women. In 2006 to 2007, for every 100 bachelor’s degrees con- In the epilogue to the second edition of ferred on black men, 196 were conferred The Declining Signi½cance of Race, I argued on black women–nearly a two-to-one that a conclusion one could draw from ratio. To put this gap into a larger con- my book was “that the sole concentra- text, for every 100 bachelor’s degrees tion on policy programs dealing with ra- earned by white men and every 100 cial bias makes it dif½cult for blacks to earned by Hispanic men, white women recognize how their fortune is inextri- earned 130 and Hispanic women earned cably connected with the structure and 158, respectively (see Table 2). The gap the functioning of the modern American widens higher up on the educational economy.”33 In concluding the epilogue, ladder. For every 100 master’s degrees I wrote: “Supporters of basic economic and 100 doctorates earned by black reform can only hope that in the 1980s men, black women earned 255 and 193, the needs and interests of the black poor respectively. These ratios have huge im- (as well as those of the other minority plications for the social organization of poor and the white poor) will no longer the black community. If present trends be underrepresented in serious public continue, future discussion of the black discussions, policies, and programs.”34 class structure will have to include a These statements were influenced by my gender component to show the increas- sense at the time that while race-specif- ing proportion of black women and de- ic programs like af½rmative action had

64 Dædalus, the Journal ofthe American Academy of Arts & Sciences Table 2 William Gender Imbalance in Higher Education: Number of Degrees Earned by Women for Every Julius One Hundred Degrees Earned by Men, Academic Year 2006–2007 Wilson

Non-Hispanic Non-Hispanic Hispanic Asian/Paci½c White Black Islander Associate’s Degrees 158 225 171 141

Bachelor’s Degrees 130 196 158 122

Master’s Degrees 167 255 185 122

Doctoral Degrees 124 193 129 108

Source: U.S. Department of Education, National Center for Education Statistics, Integrated Post- secondary Education Data System, Fall 2007, Completions component. elevated and would continue to improve whose 1985 book, Canarsie: The Jews and the employment prospects of trained Italians of Brooklyn Against Liberalism, dis- and highly educated blacks, they had cusses the racial antagonisms of Jews not enhanced the employment oppor- and Italians against inner-city blacks in tunities of the black poor. I felt there- Brooklyn and relates the conflict to my fore that the focus should shift to more central theme regarding the increasing class-based, race-neutral programs. I centrality of racial conflict that origi- no longer support this view. Recogniz- nates “in the sphere of consumption ing that a detailed discussion of policy rather than of production.” In other options would require far more space words, his ½eld research supported the than that allocated for this article, I idea that “competition between blacks would like to conclude with a brief dis- and whites has moved from the sphere cussion of why both race-speci½c and of jobs to the enjoyment of public goods, race-neutral–including class-based– like schools and entitlements.”35 programs must be strongly emphasized The research discussed in the previous and pursued to combat racial inequality. section suggests that the white backlash As I indicated earlier, many studies against racial entitlements such as af½rma- claim to address or challenge “the de- tive action, which is so clearly described clining signi½cance of race” thesis by in Rieder’s book, contributed to the gov- presenting data on residential segrega- ernment’s retreat from antidiscrimina- tion, racial composition in schools, and tion policies during the 1980s, a retreat discrimination in public places without that may have influenced hiring and pro- relating the ½ndings to my argument motion decisions in the corporate sector that the concentration of racial antago- as well. It should come as no surprise that nisms has shifted from the economic waning support for af½rmative action sector to the sociopolitical order. One programs would have an adverse effect notable exception is Jonathan Rieder, on blacks, especially more advantaged

140 (2) Spring 2011 65 “The blacks. A number of empirical studies kets in which employers are looking for Declining have revealed signi½cant differences in workers rather than workers looking for Signi½cance of Race”: the family and neighborhood environ- employers. Revisited ments of blacks and whites that are un- At the time of this writing, the nation & Revised derstated when standard measures of is plagued with one of the highest unem- socioeconomic status are employed. ployment rates since the Great Depres- Take, for example, the question of fam- sion, affecting all racial and ethnic groups ily background. Even when white par- in the United States. For almost ½ve de- ents and black parents report the same cades, the black/white unemployment average income, white parents have ratio was 2.0 or greater, which means substantially more assets than do that the black unemployment rate was at black parents. least twice that of the white unemploy- Whites with the same amount of ment rate in both good and bad econom- schooling as blacks usually attend better ic times. What is unique about the cur- high schools and colleges. Furthermore, rent economic crisis is that the unem- children’s test scores are affected not only ployment rate has surged for both blacks by the social and economic status of their and whites. Since December 2009, the parents but also by the social and econom- black/white unemployment ratio has ic status of their grandparents, meaning fallen below 2.0. The ratio was 1.87 in that it could take several generations be- October 2010 and 1.88 in November fore adjustments in socioeconomic in- 2010.36 equality produce their full bene½ts. Thus, This scenario presents a dilemma for if we were to rely solely on the standard the Obama administration, which has criteria for college admission, such as publicly acknowledged the need to com- sat scores, even many children from bat racial inequality. Given the upsurge black middle-income families would in unemployment among all racial groups, be denied admission in favor of middle- including whites, it would be politically income whites, who are not weighed prudent for the president to advance pro- down by the accumulation of disadvan- grams that address nationwide jobless- tages that stem from racial restrictions ness. However, a strong case could be and who, therefore, tend to score high- made for introducing programs that are er on the sat and similar conventional designed to combat unemployment in tests. For all these reasons, the success the highest areas of joblessness, includ- of younger educated blacks remains heav- ing a mix of private- and public-sector ily dependent on af½rmative action pro- initiatives. For example, in black inner grams, whereby more flexible criteria cities, where the number of very low- of evaluation are used to gauge poten- skilled individuals vastly exceeds the tial to succeed. number of low-skill jobs, a healthy dose The policy implications are obvious. of public-sector job creation is needed. Race-speci½c policies like af½rmative ac- This approach would also apply, say, in tion will be required for the foreseeable white and Hispanic areas that feature future to ensure the continued mobility high rates of joblessness. of educated blacks. But af½rmative action The point is that a continuous struggle programs are not designed to address the is needed to address the problems of ra- problems of poor blacks, which require cial inequality–some calling for race- greater emphasis on demand-side solu- based solutions, like af½rmative action, tions, such as creating tight labor mar- others calling for class-based solutions,

66 Dædalus, the Journal ofthe American Academy of Arts & Sciences such as programs to increase employ- much greater emphasis on the need to William ment in areas with the highest rates of strongly and continuously embrace, as well Julius Wilson joblessness. Accordingly, if I were writ- as advance, both race- and class-based ing The Declining Signi½cance of Race to- solutions to address life chances for day, I would provide more balance in people of color. my policy recommendations by placing

endnotes 1 William Julius Wilson, The Declining Signi½cance of Race: Blacks and Changing American Insti- tutions (1978; 2nd ed., Chicago: University of Chicago Press, 1980). I would like to thank Anmol Chaddha for his help in reviewing the literature on The Declining Signi½cance of Race and for his thoughtful comments on a previous draft of this manuscript. 2 One notable exception was the pattern of black political subjection imposed by the ur- ban political machines in the early twentieth century. However, although the racial devel- opments in the municipal political system had little or no direct or indirect implications for racial interaction in the private industrial sector, one could argue that the systematic exclusion of African Americans from meaningful political participation was a response to the racial antagonisms generated from the social relations of production. Even if one is willing to concede this argument, it could hardly be said that race relations in the ur- ban political system in turn influenced race relations in the private industrial sector. 3 Robin M. Williams, Jr., “Structure and Process in Ethnic Relations: Increased Knowledge and Unanswered Questions,” paper presented at the Annual Meeting of the American Sociological Association, Boston, August 30, 1979. 4 See, for example, Michael Hughes and Bradley Hertel, “The Signi½cance of Color Remains: A Study of Life Chances, Mate Selection, and Ethnic Consciousness among Black Ameri- cans,” Social Forces 68 (4) (1990): 1105–1120; Clarence Lusane, “In Perpetual Motion: The Continuing Signi½cance of Race and America’s Drug Crisis” (The University of Chicago Legal Forum, 1994); Michael Hughes and Melvin Thomas, “The Continuing Signi½cance of Race Revisited: A Study of Race, Class, and Quality of Life in America, 1972 to 1996,” American Sociological Review 63 (6) (1998): 785–795; Aaron Gullickson, “The Signi½cance of Color Declines: A Re-Analysis of Skin Tone Differentials in Post-Civil Rights America,” Social Forces 84 (1) (2005): 157–180; and James Unnever and Francis Cullen, “Reassessing the Racial Divide in Support for Capital Punishment: The Continuing Signi½cance of Race,” Journal of Research in Crime and Delinquency 44 (1) (2007): 124–158. 5 See Joe T. Darden, “The Signi½cance of Race and Class in Residential Segregation,” Journal of Urban Affairs 8 (1) (1986): 49–56; Carl Grant, “The Persistent Signi½cance of Race in Schooling,” The Elementary School Journal 88 (5) (1988): 561–569; Joe Feagin, “The Contin- uing Signi½cance of Race: Antiblack Discrimination in Public Places,” American Sociological Review 56 (1) (1991): 101–116; and Marvin P. Dawkins and Jomills Henry Braddock, “The Continuing Signi½cance of Desegregation: School Racial Composition and African Ameri- can Inclusion in American Society,” The Journal of Negro Education 63 (3) (1994): 394–405. 6 Jennifer L. Hochschild, Facing Up to the American Dream: Race, Class, and the Soul of the Nation (Princeton, N.J.: Princeton University Press, 1995), 44. 7 Ibid. 8 Michael Hout, “Occupational Mobility of Black Men: 1962 to 1973,” American Sociological Review 49 (3) (1984): 308–322. See also Peter Blau and Otis Dudley Duncan, The American Occupational Structure (New York: John Wiley, 1967); and David Featherman and Robert

140 (2) Spring 2011 67 “The Hauser, “Changes in the Socioeconomic Strati½cation of the Races, 1962–1973,” American Declining Journal of Sociology 82 (3) (1976): 621–651. Signi½cance of Race”: 9 A study by Lee Wolfle reached similar conclusions. Wolfle noted that previous research Revisited found that social background variables were more important determinants of educational & Revised attainment among whites than among African Americans. However, his study, using data from the National Longitudinal Study of the High School Class of 1972, controlled for es- timated measurement error structures and found that “social background plays a similar role for whites and blacks. Increments in background social status variables lead to similar increases in educational attainment for whites and blacks. Moreover, the effects of person- al characteristics variables (ability, curriculum, grades) of whites and blacks as they influ- ence educational attainment are also similar for both groups”; Lee Wolfle, “Postsecondary Educational Attainment among Whites and Blacks,” American Educational Research Journal 22 (4) (1985): 501–525, quote at 501. Also, analyzing a subset of the data used by Michael Hout, Marshall Pomer reported that in contrast to higher-status black men, black men in low-paying occupations were signi½cantly less likely than comparable white men to achieve upward mobility; Marshall I. Pomer, “Labor Market Structure, Intragenerational Mobility, and Discrimination: Black Male Advancement Out of Low-Paying Occupations, 1962–1973,” American Sociological Review 51 (1986): 650–659. 10 Arthur Sakamoto and Jessie M. Tzeng, “A Fifty-Year Perspective on the Declining Signi½cance of Race in the Occupational Attainment of White and Black Men,” Sociological Perspectives 42 (1999): 160, 174. 11 Ibid., 161. The authors state: “We restrict the analyses to native-born, noninstitutionalized white and black men aged 25 to 64 who were not enrolled in school and who participated in the labor force at the time of the census. Because most women did not work in the paid labor market in 1940 (Bianchi and Spain, 1986: 141 [Suzanne M. Bianchi and Daphne Spain, American Women in Transition (New York: Russell Sage, 1986)]) and because Wil- son’s (1980) discussion of labor market trends focuses on men, we do not include women in our analyses. The 1940 pums provides systematic empirical evidence about the net ra- cial disadvantage during the industrial period while the 1990 pums provides systematic empirical evidence about the net racial disadvantage during the modern industrial period.” 12 Ibid., 174. 13 Ibid., 174–175. 14 Hout, “Occupational Mobility of Black Men,” 308. 15 Melvin E. Thomas, “Race, Class, and Personal Income: An Empirical Test of the Declining Signi½cance of Race Thesis, 1968–1988,” Social Problems 40 (3) (1993): 328. 16 Clifton R. Wharton, Jr., “Education and Black Americans: Yesterday, Today and Tomor- row,” paper presented at the New York State Black and Puerto Rican Legislative Caucus, Inc., and New York State Conference of Branches, naacp, February 19, 1978. 17 In Soo Son, Suzanne W. Model, and Gene A. Fisher, “Polarization and Progress in the Black Community: Earnings and Status Gains for Young Black Males in the Era of Af½rmative Action,” Sociological Forum 4 (3) (1989): 309, 311, 324. 18 Ibid., 323. 19 Ibid., 325. 20 Pomer, “Labor Market Structure, Intragenerational Mobility, and Discrimination,” 657. 21 A. Silvia Cancio, T. David Evans, and David J. Maume, Jr., “Reconsidering the Declining Signi½cance of Race: Racial Differences in Early Career Wages,” American Sociological Review 61 (4) (1996): 541, 543. 22 Ibid., 554. 23 Wilson, The Declining Signi½cance of Race, 2nd ed., 178.

68 Dædalus, the Journal ofthe American Academy of Arts & Sciences 24 Cancio, Evans, and Maume, “Reconsidering the Declining Signi½cance of Race,” 554. William 25 Julius Ibid. Wilson 26 The income ratios reported here are based on an analysis of Current Population Survey microdata. To draw comparisons with the data I reported in the second edition of The Declining Signi½cance of Race, I used the “white/black” designation, instead of “non-Hispanic white and non-Hispanic black” designation. There are only slight differences between the percentages for the two different designations; thus, the trends reported and conclusions reached would not have changed. 27 Theodore J. Davis, “The Occupational Mobility of Black Males Revisited: Does Race Mat- ter?” The Social Science Journal 32 (2) (1995): 121–135. 28 U.S. Census Bureau, Current Population Survey, 2008 Annual Social and Economic Supple- ment, Table pov28, “Income and De½cit or Surplus of Families and Unrelated Individuals by Poverty Status: 2007.” 29 By comparison, poor non-Hispanic white families fell below the poverty line by an aver- age of $7,957; poor Hispanic families by $8,611; and poor Asian families by $8,959; U.S. Census Bureau, Income, Poverty, and Health Insurance Coverage in the United States: 2007, Cur- rent Population Reports P60-235, Table 4, “People with Income Below Speci½ed Ratios of Their Poverty Thresholds by Selected Characteristics: 2007.” 30 Lance Freeman, “Is Class Becoming a More Important Determinant of Neighborhood Attainment for African-Americans?” Urban Affairs Review 44 (1) (2008): 3. 31 Ibid., 24. 32 William Julius Wilson, More than Just Race: Being Black and Poor in the Inner City (New York: W.W. Norton, 2009). 33 Wilson, The Declining Signi½cance of Race, 2nd ed., 179. 34 Ibid., 182. 35 Jonathan Rieder, Canarsie: The Jews and Italians of Brooklyn Against Liberalism (Cambridge, Mass.: Harvard University Press, 1985), 101. 36 Ratio calculations based on data from the U.S. Bureau of Labor Statistics, Table A-2, “Employment Status of the Civilian Population by Race, Sex, and Age,” December 3, 2010.

140 (2) Spring 2011 69