End of an era? Managerial losses of African American and Latinos in the public sector

End of an era? Managerial losses of African American and Latinos in the public sector

Social Science Research 54 (2015) 36–49 Contents lists available at ScienceDirect Social Science Research journal homepage: www.elsevier.com/locate/...

354KB Sizes 0 Downloads 3 Views

Social Science Research 54 (2015) 36–49

Contents lists available at ScienceDirect

Social Science Research journal homepage: www.elsevier.com/locate/ssresearch

End of an era? Managerial losses of African American and Latinos in the public sector George Wilson a,⇑, Vincent Roscigno b a b

University of Miami, United States Ohio State University, United States

a r t i c l e

i n f o

Article history: Received 21 February 2014 Revised 17 June 2015 Accepted 25 June 2015 Available online 27 June 2015

a b s t r a c t In this article, we examine whether ‘‘new governance’’ reforms in public sector work over the last two decades have generated managerial wage losses for African Americans and Latinos. Findings from Integrated Public Use Micro-Series data across three time points indicate that the new ‘‘business logic’’ encompassing, most notably, increased employer discretion has progressively disadvantaged African American and Latino men and women relative to their White and gender counterparts. Indeed, for both African Americans and Latinos in the managerial ranks, relative parity in wages that were witnessed in the public sector progressively eroded between 2000 and 2010. Qualifications to these findings indicate that levels of inequality become pronounced for African Americans, and more so among men than women. We discuss the historical niche status of public sector work for racial and ethnic minorities in the U.S. and the importance of conducting further analyses of the public sector because of its fluid nature as a locus of racial stratification. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Sociologists have documented that during the post-1965 ‘‘civil rights era’’ the public sector has constituted an important ‘‘occupational niche’’ for racial minorities (Waldinger, 1996; Model, 1985). In this vein, and relative to African Americans and Latinos, the public sector has been the one in which minority men and women have achieved relative parity in access to, and socioeconomic rewards within, for example, managerial occupations (Waldinger, 1996; Wilson, 2006). One recent summary of this pattern by Landry and Marsh (2011: 371) captures the consensus among sociologists: ‘‘in the civil rights era, the government’s hiring of minority managers compensated for the legacy of private sector discrimination and served to build minority middle classes that are stable and economically viable, including providing favorable life chances on an inter-generational basis.’’ This minority niche status in managerial representation and rewards is supported by an impressive amount of evidence. The presidentially appointed Glass Ceiling Commission (1995), for example, found that across the 1970s and 1980s African Americans and Latinos were nearly as likely as White gender counterparts to be promoted into Managerial/Administrative positions while being less than half as likely to be promoted in the private sector. Moreover, during the 1980s and early 1990s, African American and Latino men and women were nearly as successful as White gender counterparts in retaining managerial positions in the public sector. Compare this to the private sector, where, minority managers were three times as likely to experience downward mobility (Jaynes and Williams, 1989; McBrier and Wilson, 2004). Finally, studies have found that from the 1970s through the mid-1990s African Americans and Latinos managers in public sector work closely ⇑ Corresponding author at: Department of Sociology, University of Miami, Merrick Building, Coral Gables, FL 33124, United States. E-mail address: [email protected] (G. Wilson). http://dx.doi.org/10.1016/j.ssresearch.2015.06.018 0049-089X/Ó 2015 Elsevier Inc. All rights reserved.

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

37

resembled White managers when it came to their earnings ‘‘returns’’ (Elliot and Smith, 2005; Farley, 2005; Tomaskovic-Devey and Stainback, 2007). Despite such positive momentum, sociologists have long noted that the occupational niches of racial minorities are dynamic, historically-contingent, and require ‘‘constant re-examination’’ (Model, 1985; Waldinger, 1996). It is for this reason that we analyze in the current article the status of African American and Latino public sector managers over time, and with some suspicion that their historically better status (compared to their private sector peers) is now in jeopardy owing to ‘‘new governance’’ reform which ‘‘marks the end of government work as we know it’’ (Kamarck, 2007: 131). Steadily gaining momentum in the last two decades, new governance alters the nature of employment from a ‘‘public service’’ to a ‘‘business model’’ and replicates stratification-relevant aspects of the workplace that have historically been the basis of minority disadvantage in the private sector. Specifically, rooted in the logic of ‘‘privatization,’’ new governance has sought to make government more efficient and flexible by reducing in size, restricting traditional worker employment rights and de-bureaucratizing employment-based rules and regulations (Kamarck, 2007; Morgan and Cook, 2014). Empirical assessments of the impact of new governance reforms for managers and especially African Americans and Latinos are, unfortunately, sparse. In some of our prior work, we have denoted significant and growing racial disparities in the prevalence of downward mobility (Wilson et al., 2013) and in overall gross wages between African Americans and Whites (Wilson et al., forthcoming). We still know little, however, about whether higher status workers in the managerial and administrative ranks—i.e., where there should be some protections given positioning, experience and credentials—have been similarly impacted by the changes in the structure of public sector work to which we refer above. Our analyses, which draw on the Public Use Micro-Series (IPUMS), center on changes between 1990 (representing the ‘‘pre-reform’’ period), 2000 (representing the ‘‘early reform’’ period), and 2010 (representing the ‘‘later reform’’ period). Overall, findings regarding growing racial and ethnic wage inequalities shed important light on contemporary public sector employment and opportunity and provide much needed corrective if not elaboration surrounding racial stratification in the contemporary American labor market. 2. The evolving public sector and racial stratification Our discussion of evolving employment conditions and the status of managerial/administrative workers draws primarily from recent studies in public administration (e.g. Kamarck, 2007; Bowman and West, 2007; Lawthler, 2003; Morgan and Cook, 2014). Such work has documented the contours and increasing encroachment of the ‘‘new governance movement’’—a movement that has produced a growing convergence with the historically less minority-friendly private sector. Less attention, however, has been explicitly devoted to the racial stratification implications let alone the impact on higher status workers (Kamarck, 2007; Wilson, 2006). To the extent there is a growing convergence in work conditions and status across economic sectors, we would see it as providing support for ‘‘new institutional’’ organization theory (Dimaggio and Powell, 1991), which in recent years has shifted the focus from explaining why organization are so different to why they are so similar (Frumkin and Galaskiewicz, 2006). The idea of ‘‘coercive isomorphism’’ (Stainback and Tomaskovic-Devey, 2012) is probably most central in denoting how political and ideological-based external scrutiny and regulation cause organizations to gravitate toward isomorphic transformation (i.e., to react similarity across organizations) (Meyer and Rowan, 1991). To this end, practices and statuses associated with new governance are premised on creating ‘‘markets’’ inside government as means of isomorphic transformation. 2.1. Pre-new governance Across the first three decades of the post-1965 civil rights era, the ‘‘public service’’ model of government employment— i.e., based on a decades-old notion of government work as contributing to the ‘‘public good’’ (Kamarck, 2007)—has been indispensable for establishing relative racial parity in socioeconomic rewards among managers (Wilson, 2006; Brown and Erie, 1981). Rooted in a ‘‘career system’’ (Bowman and West, 2007) of employment, at its height around 1990 it applied to a majority of the 18.2 million full-time public sector employees at the federal and state levels in the U.S. (Kamarck, 2007). In this system, employees receive an employment package encompassing favorable status and employment protections in exchange for their service. This has included centralized decision-making and formal bureaucratic procedures in determining stratification aspects of work—such as wages (Berman et al., 2006). Stratification-related decisions are consequently made jointly by managers in conjunction with administrators, who are charged with ensuring equity in the treatment of government workers (Kamarck, 2007; Lawthler, 2003). Further, employees are designated as ‘‘classified’’ so they have ‘‘property rights’’ in their jobs, which restricts worker termination to ‘‘just cause’’ reasons and offers opportunities to invoke relatively elaborate equal employment opportunity laws to contest the decisions of employers regarding, for example, hiring, promotion, firing and wages (Bowman and West, 2007). Under this system, we witnessed substantial minority mobility into the ranks of management and relative racial parity in wage returns for white and minority managers (Glass Ceiling Commission, 1995). 2.2. New governance Beginning in the mid 1990s and accelerating since, new governance has altered the very terms/conditions of government work to which we alluded above. The ‘‘reinventing government’’ movement is a manifestation of the rise of neo-liberal

38

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

economic sentiment in national political discourse that emphasizes the importance of open markets and minimal government involvement in regulating economic activity (Bowman and West, 2007; Kamarck, 2007). Along these lines, the public sector is viewed as most appropriately operating like the private sector so as to ‘‘redress the perceived ills associated with a bloated and inefficient government that had outlived its usefulness in effectively delivering public services and protection of basic public rights’’ (Kamarck, 2007: 134). New governance is ultimately predicated on the logic of ‘‘bottom line’’ financial principles and places a premium on increasing performance, efficiency, and results (Bowman and West, 2007; Hays and Sowa, 2006). Accordingly, public sector work became more avowedly incentive-laden, which is perceived as enhancing productivity at the individual- and department-levels, and increasing the flexibility of managers to make efficiency mandated personnel adjustments at a time when a more fluid public sector is perceived as a response to the rapid pace of social change (Bowman and West, 2007; Wilson, 2006). The explicit scope of this civil service reform extends to ‘‘altering aspects of employees’ status, power, and protections, the implementation of which are widely thought to have been detrimental historically to the stratification-based outcomes of minorities in the private sector’’ (Kamarck, 2007: 31). A hallmark of the new business model is the designation of employees as ‘‘declassified.’’ Decentralization here supplants a highly bureaucratized work environment so that on-site discretion of employers increasingly determines stratification aspects of work (Bowman and West, 2007; Lawthler, 2003). Discretion is manifest, for example, in on-site managers’ making stratification-relevant decisions encompassing promotion salary increases and job retention as well as structuring conditions of work such as task/unit assignments that impact on stratification outcomes (Bowman and West, 2007; Dobbin, 2009; Wilson, 2006). In addition, declassified workers lose property rights in their jobs, becoming employees ‘‘at will’’ (Malamud, 1995) who— absent narrowly carved judicial and legislative exceptions—can be terminated for ‘‘any or not reason at all’’ (Villemez and George, 1994), and, they have restricted opportunities to invoke equal opportunity laws to contest employers’ decisions regarding stratification-relevant aspects of their employment (Wilson, 2006; Dobbin, 2009).1 Interestingly, the legislative enactments ushering in new governance at both state and federal levels are implicitly anti-union, indeed viewing collective bargaining as antithetical to flexibility and bottom-line results that are central to the new governance philosophy (Morgan and Cook, 2014).2 At the state level, there has been variation in the legislation enacting aspects of new governance, including whether aspects enacted apply only to future hires or encompasses all employees. Georgia is a good and influential early example (1996) of the former, while Florida offers a good case in point (1998) of the latter (Bowman and West, 2007). Based, respectively, on the Georgia Act 1816 and the Florida Reform Act, reform eliminated the merit system and ushered in a privatization model that altered practices of recruitment and hiring, pay raises and promotion, and downsizing and discipline (Battaglia and Condrey, 2007). First, reform created power to hire immediately and with more flexibility: the legislation no longer required that agencies confer with the central human resources department on matters of recruitment and selection (Bowman and West, 2007). Second, the adoption of at-will employment status eliminated seniority so that employees had no assurance of reassignment during downsizing and can be relocated without recourse and the length of process for discipline and termination has been reduced dramatically (Bowman and West, 2007). Third, pay raises and promotion are no longer based on formal pay tiers accumulated through seniority but rather are based on the recommendations of on-site managers with minimal administrative oversight and appeal rights of employees (Bowman and West, 2007). Table 1 documents the adoption of the three major aspects of new governance, namely the movement toward ‘‘declassified’’ employment status, the expansion of ‘‘at-will’’ employment status, and the reduction of ‘‘grievable’’ employment issues across all states at each of three periods (i.e., ‘‘pre-reform’’ (1990), ‘‘early reform’’ (2000) and ‘‘later reform’’ (2010)). Table 1 reveals that in the pre-reform period (1990), only two states had adopted any of the three aspects of new governance. During the early reform period (2000) 15 states had adopted aspects of new governance. By 2010, the time point representing the later reform period, 38 states had adopted aspects of it. Further, across all three periods the expansion of at-will employment was the most frequently adopted aspect of new governance: both states adopting new governance in the pre-reform period opted only for the expansion of at-will status. In addition, over 75% of states in the latter two periods adopting new governance opted to expand at-will status while less than half the states opted for expanding de-classified status or reducing grievance issues. Finally and noteworthy, by 2008 five of the six largest federal government agencies had implemented aspects of new governance when President George Bush gained congressional approval to use new

1 We recognize—in accordance with sociology research—that bureaucratic or formalization of employment practices is not always synonymous with the elimination or reduction of racial inequality. A developing literature indicates that the stratification-relevant consequences of formalization is not the presence or absence of bureaucratic rules but rather the underlying logic that promotes formalization (Stainback and Tomaskovic-Devey, 2012; Roscigno, 2007). Bureaucratization is fundamentally a system of control and coordination. In most workplaces, the core goals of bureaucratization are to enhance efficiency of the organization or the power of organizational leaders, but not always addressing issues or racial inequality. These are secondary goals stemming from the efforts of some co0mnstituency who have enough influence to effectuate such ends. All-told, the underlying ‘‘cultural’’ logic and long-standing historical commitment to address issues of racial inequity, we believe, translates into developing and implementing bureaucratic and organizational rules toward ‘‘civil rights ends’’ (Stainback and Tomaskovic-Devey, 2012) a first priority. 2 Wholesale and explicit attacks on public sector unions fall under a different range of state level dynamics, most explicitly—and with considerable national notoriety—in states such as Ohio and Wisconsin in recent years where gubernatorial candidates and governors explicitly vowed to restrict union power and influence, with respect to, for example, right to collectively bargain, contributions to pensions and health insurance plans, right to strike, and, opportunities to opt out of unions, with the justification that unions are financially a drain at a time of fiscal crisis that necessitates significant budget reductions.

39

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49 Table 1 Adoption of new governance.

Pre-reform Early reform Later reform

Period

Number of states adopted

Experienced at-will

Expanded declassified status

Reduced grievance issues

1990 2000 2010

2 15 38

2 12 27

0 14 20

0 6 15

governance as a template for all future federal government hires. It has recently been estimated that by 2013 over 65% of full-time workers at the federal level were subject to aspects of new governance (Morgan and Cook, 2014).

3. Inequality consequences Expectations regarding the impact of new governance on managerial racial/ethnic wage gaps are grounded in sociological research on the dynamics of racial inequality in predominantly White owned and managed private sector firms (e.g. Collins, 1997, 1993; Hite, 2007; Yaffee, 1995; Fernandez, 1981; Dobbin, 2009; Roscigno, 2007; Roscigno et al., 2012; Wilson and Roscigno, 2010; Wilson, 2009; Smith, 2005; Elliot and Smith, 2001; Mundra et al., 1995; Tomaskovic-Devey and Stainback, 2007; Tomkowitz and Brenner, 1996). This research, overall, is rooted in the pioneering work of sociologists such as Allport (1954), Pettigrew (1971, 1964), and Blumer (1966) who mapped out the dynamics of ‘‘modern prejudice.’’ Specifically, employers act in a workplace governed by an avowed ideology of meritocracy: stratification-relevant decisions are rooted in practices that are institutional, situational, and ostensibly non-racial in nature. Employers’ decisions are often made in the interest of perceived ‘‘business necessity’’ such as maintaining a stable and productive workforce and a steady customer/client base and within workplace contexts imbued with unconscious bias, including cognitive distortions inherent in ‘‘attribution bias’’ (Pettigrew and Martin, 1987) and ‘‘statistical discrimination’’ have a disproportionately negative impact on minority men. Such research empirically highlights identify a crucial institutional factor that underlies discriminatory results at the managerial level, namely, the tendency of minorities to be susceptible to unfavorable performance evaluations. Specifically, African American and Latino men experience heightened workplace segregation that limits opportunities to demonstrate the informal or ‘‘particularistic’’ characteristics such as loyalty, trustworthiness and perceived leadership potential that underlie employers’ decision making on outcomes such as wages. For example, Pettigrew and Martin (1987) maintain that restricting the exercise of ‘‘span of control’’ (number of subordinates) and ‘‘span of responsibility’’ (range of superordinate functions, e.g., hiring, firing, say over pay and promotion) that African American and Latino managers have to segregated spheres (i.e., over co-racial group members) leaves them prone to ‘‘information bias’’, a form of statistical discrimination in which demonstrated performance are viewed as less credible than those of Whites. Braddock and McPartland (1987) and Wilson (1997) further argue that allocating minorities to racially delineated work/task groups as well as the segregated operation of traineeship and internship program produces susceptibility to attribution bias (i.e., being evaluated on selective bases the reaffirm negative stereotypes about their suitability for, and productivity at, work). Finally, several authors—including Cox and Nkomo (1990) as well as Elliot and Smith (2001)—maintain that when not segregated, the tendency to subordinate African American and Latino managers/administrators to Whites in authority hierarchies hampers opportunities to demonstrate the range of performance-relevant and informal characteristics that may emerge. Taken as a whole, sociological research addressing private sector dynamics in conjunction with our knowledge regarding the substance of new governance reform over the last two decades provides a solid basis for maintaining that, in the context of managerial wages, racial inequality may be on the rise. Increasing employer discretion—in terms of ‘‘bottom line’’ decision-making and practices regulating the organization of work associated with new governance—should result in the progressive widening of racial wage gaps among African Americans and Latinos managers, relative to Whites, so that these gaps increasingly resemble those found in the historically less minority-friendly private sector.

3.1. Qualifications to theory Sociological research offers a basis for maintaining that niche status varies across minority groups. First, pursuant to new governance reform, the gap in managerial wages vis-à-vis Whites should be greater among African Americans than Latinos. This expectation is derived from work indicating that there is a differential impact of employer discretion across minority groups. Latinos are better able to communicate evaluation-relevant performance and personalistic criteria than African Americans (Fernandez, 1981; Smith, 2005). Specifically, Latinos experience lesser workplace segregation from Whites, which translates into more integrated informal and formal job networks than African Americans (Smith, 2005). Further, Latinos have a less difficult prejudicial burden to overcome in communicating evaluation-relevant criteria. In this regard, survey research indicates that Whites hold less invidious job related stereotypes concerning, for example, suitability for work, intelligence, penchant for criminality, and loyalty toward Latinos than African Americans (Bobo and Massagli, 2001; Kluegel and Bobo, 2001).

40

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

Second, prior work also provides a basis for suggesting that within minority groups, niche status and advantage/ disadvantage will likely vary by gender. Irrespective of race, for instance, we know that women suffer from a shared ‘‘penalty’’ tied to both segregation (Stainback and Tomaskovic-Devey, 2012) and the persistence of gender-based stereotypes (e.g., commitment to work and fitness for lengthy tasks. Accordingly, we posit that tmanagerial wage penalties for African American and Latino women will be distinct from co-racial group men. While suffering a race penalty, for instance, African American and Latino women share a gender cost with their White female peers. As such, and while we expect a generally similar pattern of racial disadvantage for men and women, the overall trend and level of racial inequality will be less pronounced for women. 4. Data and measurement A sample of individuals from the Public Use Micro-Series data (and a sub-set of this data, the American Community Survey) across three periods, the ‘‘pre-reform’’ (1990), ‘‘early reform’’ (2000) and ‘‘later reform’’ (2010), are used to examine the impact of new governance reform on managerial wages among African American and Latino men and women relative to their White and gender counterparts. The IPUMS is a project devoted to collecting and harmonizing U.S. Census Data across time periods and the 5% samples that constitute its core are nationally representative across U.S. households. During each period African American, Latino and White workers between the ages of 18 and 60 were included in the sample if they worked full-time in a non-self-employed capacity within the ‘‘Managerial specialty’’ occupational categories in the private or state/federal government levels. We excluded individuals who worked for the ‘‘local’’ government since the social science literature on new governance has focused on its incidence and its operation at the level of federal and state jobs. The application of these criteria resulted in a sample size of 532,302 men and 324,216 women in the pre-reform period, 519,205 men and 313,768 women in the early-reform period, and 523,607 men and 314,006 women in the later reform. 4.1. Dependent variable The dependent variable in this study, hourly wages, is calculated as the wages earned in sample respondents’ ‘‘usual’’ managerial occupation. The effect of inflation on wages was removed by multiplying the measure by the Consumer Price Index, producing an hourly wage variable expressed in constant 2010 dollars. 4.2. Race and sector variables We code race/ethnicity with the use of two dichotomous indicators of African American and Latinos, with Whites constituting the reference category. Sector is coded as 1 = public, 0 = private. 4.3. Control variables 4.3.1. Human capital credentials The influence of several human capital credentials is controlled. The first is level of educational attainment, represented by two dummy variables: ‘‘college degree,’’ and ‘‘post-college degree.’’ Respondents with a high school degree or less serve as the reference category. Second, because a worker’s physical capacity is treated as a human capital characteristic (Becker, 1964), respondents who said they had ‘‘health problems that limited their capacity to work’’ were coded 1 and all others were coded 0. Third, we measure job commitment as the number of absences from work in the previous year. We reverse coded this indicator so that higher scores reflect greater job commitment. Fourth, job tenure is the amount of time spent with present employer. 4.3.2. Job/labor market characteristics Several job/labor-market characteristics are included as controls. First, union status of job is measured by 1 = yes, 0 = no. Second, to most directly capture the impact of new governance reform we include on a race-specific basis—up to the point of time examined—whether respondent is working in a state that had adopted at least two of three aspects (expansion of at-will status, expansion of declassified status, and reduced grievance procedures) of new governance (‘‘<1 aspect adopted’’) or one aspect of new governance adopted (‘‘1 aspect adopted’’). Specifically, two or greater aspects adopted and one aspect adopted are dummies (coded 1) with no adoption of new governance being the reference category (coded 0). 4.3.3. Family/household Several family/household variables are controlled. First, we control or marital status (1 = married, 0 = unmarried) as marriage has a documented positive impact on earnings. Second, controlled is the number of children in a household (coded as number of children) as their presence tends to be positively related to earnings (Ahituv and Tienda, 2004; Caucutt et al., 2002).

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

41

4.3.4. Sociodemographic Based on its unique—and relatively negative—historic track record regarding inter-group relations (Jaynes and Williams, 1989; Farley and Allen, 1987) and the distribution of socioeconomic resources across racial groups (Farley, 2005), we enter the South as a dummy variable, relative to other regions in the U.S. In addition, age (years) is included as a control.

5. Analytic strategy and results We use ordinary least squares (OLS) regression as our multivariate technique of choice. For each of the three time periods examined, we run separate equations for the two minority groups relative to their White gender counterparts. In assessing racial gaps in wages during each period, for each equation, the coefficient of the race dummy variable is the estimate of the race gap in the private sector. Thus, it constitutes the penalty for being African American and Latino in the private sector. The race gap in the public sector, our main focus, is equal to the sum of the race coefficient and the coefficient of the interaction term between race and sector. This interaction term, thus, represents the ‘‘boost’’ that African American and Latino managers receive for working in the public sector, relative to the private sector.3 We disaggregate the more general findings by gender and, importantly, employ sampling weights for the production of point estimates of population parameters. These weights serve to ensure representativeness across cohorts as well as across sectors, thereby precluding the possibility that findings were driven by compositional differences in the IPUMS sample. Finally, to be as rigorous as possible, we ran additional analyses with a ‘‘distributions changes’’ methodology which assesses the extent to which evolving racial wage gaps are a product of changes in distributional inequality among each group of managers, that is, whether growing inequalities in the public sector are a product not so much of declines in minorities wages but rather changes in Whites’ wages brought about by, for example, differential rates of income growth or income polarization across the managerial income distribution. Secondly, and for comparability sake, we perform a regression decomposition across the three time periods, isolating the changing contribution of discriminatory versus non-discriminatory factors in explaining the wages of both African Americans and Latinos. This allows for assessment of relative impact over time. We also supplement our core findings by rerunning the analyses using quintile regression (Koencker and Barrett, 1978; Powell, 1986). Breaking down the sample at equal quarters of the income distribution, as this analysis does, provides information on whether the racial wage gap across each of the three periods varies across the income distribution.

5.1. Assessing managerial wage gaps by group Table 2 reports mean differences in hourly wages between African American and Latino managers, relative to Whites in the public and private sector during the pre-reform, early reform and later reform periods with wages being presented in raw dollar amounts (the Appendix reports descriptive statistics for variables in the statistical model).4 Results indicate that vis-à-vis White gender counterparts, the relative parity in hourly wages achieved by African American and Latino managers in the public sector—when compared to the private sector—during the pre-reform period, was progressively reduced in the early reform and later reform periods, primarily because of the widening racial inequality in the public sector. Among men, for instance, the gap in hourly wages favoring Whites over African Americans in the public sector is $2.09 in the pre-reform period, expands to $2.70 in the early reform period, and then expands further to $5.27 in the later reform period. The hourly wage gap favoring Whites over African Americans in the private sector, in contrast, is robust but stable across all three periods. Among Latino managers, a similar race-based wage gap emerges across time in the public sector, although the pattern of racial inequality is somewhat less pronounced. Indeed, there is no statistically significant difference during the pre-reform period, but the gap in managerial hourly wages widens to $2.14 in the early reform period and expands to $3.50 in the later reform period. The pattern is largely similar among women, although as predicted the racial wage gaps are less pronounced. Specifically, the gap in hourly wages favoring White compared to African American female managers is $0.83 (not significant) in the pre-reform period, expands to $1.45 (P < .05) in the early reform period and expands further to $2.21 (P < .01) in the later reform period. As was the case for men, racial inequalities in wages in the private sector are moderate but stable across all three periods. Among Latina relative to White female managers, there is no statistically significant difference during the pre-reform and early reform periods but the gap in hourly wages widens to $1.53 (P < .05) in the later reform period. 3 We undertook efforts to address sample selectivity bias. In particular, we were concerned with the possibility that civil rights laws and more equitable treatment have caused highly skilled African Americans and Latinos to gravitate toward the private sector during the early reform and later reform periods. Based on additional analyses there appears to be no evidence of this kind of selection bias build into the sample. In particular, t-tests for differences between African Americans and Whites as well as between Latinos and Whites in skill levels—indexed by level of educational attainment—were not significant in either the pre-reform, early reform, or later reform periods. In addition, the proportion of African Americans and Latinos with ‘‘post-college’’ education (approximately 55%) was virtually identical across all three periods. 4 All analyses were re-estimated without using the IPUMS weights. Results were similar to those reported in this study. Further, a parallel set of all analyses were performed in which the dependent variable—wages—assumed a natural logarithmic transformation. In all cases, results were consistent with those reported in this study.

42

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

Table 2 Hourly wage advantages for whites over African Americans and Latinos (In Real Dollars). Pre-reform

* ** ***

Early reform

Late reform

Afr. Am.

Latino

Afr. Am.

Latino

Afr. Am.

Latino

Public Men (Diff.) Women (Diff.)

$2.09* $0.83

$1.27 $0.65

$2.70** $1.45*

$2.14* $0.59

$5.27*** $2.21**

$3.50** $1.53*

Private Men (Diff.) Women (Diff.)

$5.05*** $2.25**

$3.17** $2.16**

$4.86*** $2.71**

$3.57** $2.22**

$5.39*** $2.82***

$3.62** $2.35**

p < 0.05. p < 0.01. p < 0.0.

5.2. Racial/ethnic managerial wage gaps across time for men Table 3 reports the results of our OLS wage-gap models for men, with controls and at the three relevant time points. Results indicate that the relative parity in wages achieved by African American managers in the public sector, compared to the private sector, during the pre-reform period was progressively reduced across the early reform and later reform periods because of the widening gaps in wages in the public sector.5 Specifically, in the pre-reform period, African American men, relative to White, were highly disadvantaged in the private sector in earnings attainment (b = $3.14, P < .001), yet received a robust public sector ‘‘boost’’ in earnings (b = 2.34, P < .001). This results in a statistically insignificant managerial disadvantage in earnings in the public sector (b = $0.80). In the early reform period, African Americans—similar to the pre-reform period—were highly disadvantaged in the private sector in earnings attainment (b = $3.03, P < .001). The diminishing public sector modest ‘‘boost’’ (b = $1.60, P < .01) during this period results in a moderate level of disadvantage in the public sector (b = $2.23, P < .01). Finally, in the later reform period—similar to earlier periods—African Americans were highly disadvantaged in the private sector in wage attainment (b = $3.16, P < .001). However, in this period they experienced a further deepening of wage inequality in the public sector. They received a statistically non-significant boost in the public sector, resulting in a robust level of disadvantage in the public sector (b = $2.56, P < .001). Results from Table 3 also indicate that similar inequalities emerge over time for Latino male managers though, as expected, the adverse effects of public sector reform on them are lesser than those seen for African Americans. In the pre-reform period, Latino male managers were moderately disadvantaged compared to their white peers in earnings in the private sector (b = $2.01, P < .01). The fact that they experienced a moderate public sector ‘‘boost’’ in earnings, relative to Whites (b = $1.95, P < .01) results in a statistically insignificant difference in public sector wage inequalities. As was the case with African American male managers, the public sector boost erodes over time, so that by the later reform period, the public sector advantage is all but eliminated and the level of disadvantage in the public sector becomes pronounced (b = $1.85, P < .01).6 5.3. Racial/ethnic managerial wage gaps across time for women Table 4 reports the results for OLS wage models, but now for women managers. As we suggested earlier, the pattern of increased public sector wage inequality mirrors that of men although the racial/ethnic gaps are less pronounced. Specifically, in the pre-reform period, African American women managers were moderately disadvantaged in the private sector in earnings attainment (b = $2.17, P < .01) while receiving a moderate public sector ‘‘boost’’ (b = 2.01, P < .01). The result is a statistically insignificant disadvantage in earnings compared to White female managers (b = $0.59). In the early reform period, we find a heightening of public sector earnings inequality, and a modest level of disadvantage (b = $1.24, P < .05). Such wage inequalities between African American and White female managers becomes even more pronounced (b = $2.27, P < .01) in the later reform period. 5 The findings reached here appear to be—at most—indirectly related to union dynamics in the public sector. Along these lines, we were concerned that declining union membership associated with new governance, particularly among African Americans and Latinos, relative to Whites, could explain their declining relative fortunes regarding hourly wages. However, rates of unionization among all three racial groups were altered in a roughly equivalent manner across time periods (about 7% in the early reform period and 12% by the end of the later reform period) so that progressively widening racial gaps in wages in the public sector are not explained, for example, by Whites’ increasing monopolization of union jobs. If a union effect is operating, we suspect it is because of the disproportionate consequence of new governance across racial groups along one of its tenets—the availability of grievance procedures employees can utilize. Through collective bargaining, public sector unions have a ‘‘say’’ in determining the number of available grievance procedures: pursuant to the adoption of new governance, the reduction of opportunities to resort to grievance procedures may have a particularly negative impact on minorities who are more likely to resort to them than Whites (Stainback and Tomaskovic-Devey, 2012). 6 Additional multivariate analyses were performed to further document trends in the impact of public sector reform on trends in the wage gap between African Americans and Latinos vis-à-vis Whites. Specifically, analyses at three years intervals—2003, 2006, and 2009—revealed increasingly significant statistical differences in the racial gaps in hourly wages between both African Americans and Latinos, relative to, Whites. This finding is consistent with the notion that the racial gap in wages favoring Whites in the public sector is accelerating as a greater proportion of public sector workers became subject to new governance.

43

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49 Table 3 OLS Regressions for Hourly Wages Among Men: African Americans and Latinos Relative to Whites in Management. Pre-reform

Early reform

Afr. Am.

* ** ***

Latino

Later reform

Afr. Am

Latino

Afr. Am.

Latino

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

Ascriptive Race

$3.14***

.093

$2.01**

.083

$3.03***

.093

$2.40**

.088

$3.16***

.089

$2.22**

.094

Sector Public

$0.80

.021

$0.72

.048

$2.23**

.021

$1.20*

.056

$2.56***

.078

$1.85**

.063

Interaction Race ⁄ Sector

$2.34***

.077

$1.95**

.074

$0.61

.077

$1.20*

.057

$0.60

.044

$0.37

.031

Human capital Post College College Health limitation

$0.99 $0.37 $083

.057 .028 .046

$1.33* $0.83 $0.43

.062 .046 .031

$1.18 $0.74 $0.98

.060 .044 .053

$1.20* $0.69 $0.84

.055 .049 .049

$1.22* $0.76 $0.23

.053 .045 .016

$0.81 $0.40 $0.44

.046 .026 .030

Family/household Married # of Children

$.079 $0.77

.046 .031

$.066 $0.79

.046 .052

$1.30* $1.22*

.053 .057

$.035 $1.22*

.024 .057

$0.32 $1.01

.020 .060

$.024 $0.33

.013 .021

Job labor market Union Reform >2 aspects Reform 1 aspect

$.073 $2.53** $1.66*

.051 .072 .070

$.087 $1.76* $1.61*

.053 .075 .068

$1.45* $2.44** $2.22**

.060 .080 .079

$0.92 $2.23** $1.87*

.063 .071 .082

$1.77* $3.38*** $2.91**

.074 .055 .90

$1.56* $2.45** $2.06*

.072 .073 .085

Region and Sociod. Region Age

$0.57 $1.31*

.034 .056

$1.28* $0.21

.061 .016

$0.35 $1.45*

.022 .063

$1.32* $0.33

.061 .024

$1.32* $0.75

.063 .045

$1.21* $0.12

.059 .008

Adj. R2

33.2

34.6

37.4

37.2

33.5

35.3

p < 0.05. p < 0.01. p < 0.001.

Results from Table 4 also indicate that a similar, yet less pronounced, pattern of inequality intensification for Latina managers. In the pre-reform period, Latina managers were moderately disadvantaged in the private sector in earnings attainment (b = $2.10, P < .01) while receiving a moderate public sector ‘‘boost’’ in earnings relative to their White counterparts (b = $1.97, P < .01). In fact, here we find a statistically insignificant difference in wages in public sector work. By the later reform period, in contrast, gains related to public sector employment decline for Latina managers (b = $ 1.34, P < .05) in the public sector, resulting in a modest level of disadvantage and statistically significant level of inequality relative to White female managers (b = $1.68, P < .05). 5.4. Decomposing growing public sector inequalities and managerial wage gaps We also considered the extent to which the racial/ethnic wage inequalities across sectors, thus far reported, might be related to the attributes people bring to the labor market and/or potential variations in how such attributes are evaluated. It is for this reason that we assess here, with regression decomposition, the relative contribution of evaluations (intercepts and slopes) and characteristics (means) to the race/ethnic inequalities in hourly wages. Specifically, using the regression slopes from Table 2 and the means from the Appendix, the gaps in hourly wages are decomposed based on the following formula (Jones and Kelley, 1984: 330):

Y W  Y aa ¼ ðaw  amg Þ þ

X X X ðbw  bmg ÞX w þ ðbw  bmg ÞðX w  X mg Þ þ ðX w  X mg ÞBmg

where the subscripts w and mg refer to White and minority group (either African American or Latino), respectively, Y is the mean of logged earnings, a is the y-intercept, b is the slope, and X is the mean of a predictor. The first term on the right side of the equation is the portion of the wage gap due to ‘‘group membership.’’ The second term on the right side of the equation is the portion of the wage gap due to differences in rates of return. The third term on the right side of the equation represents the interaction term between worker characteristics and their evaluation. The fourth term on the right side of the equation is the portion of the wage gap due to ‘‘nondiscriminatory’’ differences in worker characteristics brought to the labor market. Overall, combined are the group membership term and the returns term into a ‘‘discrimination’’ component, which assumes that the portion of the wage gap due to group membership reflects discriminatory factors. The size of the fourth or ‘‘means’’ component designates the expected gap between African Americans and White wages if African Americans entered the labor market with characteristics similar to those of Whites. Table 5 reports the results of this earnings decomposition across economic sectors during the three periods. The findings are generally consistent with those reported in our prior analyses: Benefits to managerial employment for African Americans

44

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

Table 4 OLS regressions for hourly wages among women: African Americans and Latinos relative to whites in management. Pre-reform

Early reform

Afr. Am.

* ** ***

Latino

Later reform

Afr. Am

Latino

Afr. Am.

Latino

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

(b)

(se)

Ascriptive Race

$3.14***

.093

$2.01**

.083

$3.03***

.093

$2.40**

.088

$3.16***

.089

$2.22**

.094

Sector Public

$0.80

.021

$0.72

.048

$2.23**

.021

$1.20*

.056

$2.56***

.078

$1.85**

.063

Interaction Race ⁄ Sector

$2.34***

.077

$1.95**

.074

$0.61

.077

$1.20*

.057

$0.60

.044

$0.37

.031

Human capital Post College College Health limitation

$0.99 $0.37 $083

.057 .028 .046

$1.33* $0.83 $0.43

.062 .046 .031

$1.18 $0.74 $0.98

.060 .044 .053

$1.20* $0.69 $0.84

.055 .049 .049

$1.22* $0.76 $0.23

.053 .045 .016

$0.81 $0.40 $0.44

.046 .026 .030

Family/household Married # of Children

$.079 $0.77

.046 .031

$.066 $0.79

.046 .052

$1.30* $1.22*

.053 .057

$.035 $1.22*

.024 .057

$0.32 $1.01

.020 .060

$.024 $0.33

.013 .021

Job labor market Union Reform >2 aspects Reform 1 aspect

$.073 $2.53** $1.66*

.051 .072 .070

$.087 $1.76* $1.61*

.053 .075 .068

$1.45* $2.44** $2.22**

.060 .080 .079

$0.92 $2.23** $1.87*

.063 .071 .082

$1.77* $3.38*** $2.91**

.074 .055 .90

$1.56* $2.45** $2.06*

.072 .073 .085

Region and Sociod. Region Age

$0.57 $1.31*

.034 .056

$1.28* $0.21

.061 .016

$0.35 $1.45*

.022 .063

$1.32* $0.33

.061 .024

$1.32* $0.75

.063 .045

$1.21* $0.12

.059 .008

Adj. R2

33.2

34.6

37.4

37.2

33.5

35.3

p < 0.05. p < 0.01. p < 0.001.

and Latinos in the public sector appear to increasingly erode, the link between returns and human capital increasingly fractures, and the possibility that discrimination is playing a growing part in wage returns for minorities relative to Whites intensifies over time. This is in contrast to private sector employment, where managerial wage inequalities and the possibility that discrimination is playing a part remain relatively stable across time. We briefly elaborate on these interpretations. The reader will note that the proportion of the wage gap for African American men due to potential discrimination was 32% but increased to 40% and 52%, respectively during the early and late reform periods. Compare this to the private sector, where the change went from 49 to 55. For Latino public sector managers, discrimination as a potential explanatory process rose from 28% to 40%, compared to a rise of 33% to 40% in the private sector. Although racial/ethnic inequalities are certainly less pronounced, we find a similar pattern among women, especially African American women managers. Discrimination as a potential cause rises for African American female public sector managers from 27%, in the pre-reform period, to approximately 37% in the later period. Compare this to the private sector, where the change observed barely moves (from 38% to 40% over time), Similarly, for Latina public sector managers, the change is less pronounced than for men. What especially stands out, however, is that the possibility that discrimination is playing a more explicit role intensifies over time, mostly notable in public sector work (25–32% change, compared to a change from 37% to 39% in the private sector). 5.5. Supplemental analyses and distributional changes We performed additional analyses that are designed to more fully explain the causes of unfolding racial gaps in wages across the three periods. Specifically, we adopted a ‘‘distributional changes’’ approach that focuses on changing relative earnings distributions of minority group managers, relative to their White and gender counterparts. Quantile regression helps assess whether racial gaps among public sector managers differ across levels of the income distribution. Specifically, among African American and Latino men and women we assessed relative to their White at equal fourths (quartiles) of the income distribution. Findings in these regards, not reported but available upon request, indicate evolving racial/ethic inequalities in all of these regressions were almost completely similar across levels of the income distribution. As such, our confidence is bolstered in reporting the overall median racial/ethnic wage gaps as we have done in Tables 3 and 4. Specifically, among men and at all quartiles of wage earners, wage inequalities between African Americans and Whites are non-significant during the pre-reform period, moderately (P < .01) significant during the early reform period, and highly

45

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49 Table 5 Decomposition of race differences in log of hourly wages (All Versus White). Pre-reform

Men early reform

Public

Private

Public

Later reform Private

Public

Private

Afr. Am.

Latino

Afr. Am.

Latino

Afr. Am.

Latino

Afr. Am.

Latino

Afr. Am.

Latino

Afr. Am.

Latino

Men % Discrimination % Non-discrimination Interaction

32 60 8

28 65 7

49 43 8

33 60 7

40 51 9

35 58 7

54 41 5

35 58 7

52 40 8

38 55 7

55 40 5

40 54 6

Women % Discrimination % Non-discrimination Interaction

27 65 8

25 65 10

39 52 9

37 55 8

30 62 8

26 63 11

39 51 16

38 55 9

38 53 11

32 58 10

40 51 9

37 55 8

(P < .001) significant during the later reform period. Similarly, wage inequalities experienced by Latino mangers are non-significant during the pre-reform period, modestly (P < .05) significant during the early reform period, and moderately significant during the later reform period. For women, and at all quartiles of wage earners, the wage gaps between African American and Latina managers relative to their White female counterparts are non-significant during the pre-reform period, modestly significant during the early reform period, and moderately significant during the later reform period. We extended this further by directly examining the extent to which changes in distributional inequality among each group of managers contributed to evolving increases in the racial wage gap in the public sector. That is, we analyzed whether growing inequalities in the public sector are a product not so much of declines in African Americans’ and Latinos wages but rather changes in Whites’ managerial wages brought about by, for example, differential rates of income growth or income polarization across the managerial income distribution. The concept of ‘‘relative distribution’’ is central to such considerations. This distribution is a ratio and compares the density of African American, Latinos and Whites earnings at each level of the earnings distribution. Whites form the reference earnings distribution with African Americans and Latinos constituting the comparison distribution. The relative distribution is the ratio of comparison group earners to reference group earners at each level of the earnings distribution, divided in separate analyses into quartiles (fourths). It takes the value of 1 when the numerator and denominator are equal, which occurs when the fraction of earners at that level of the earnings distribution is the same for both groups. When the fraction of comparison earners is higher or lower than that of the references groups, the value taken by the relative distribution is greater or less than one, respectively. Taking into account relative distributions, we first chart changes in minority group earnings relative to White and gender counterparts between 1990 (pre-reform period) to 2010 (later reform period). Second, we perform an additive regression decomposition to isolate the effects of: (1) change in median ratio over time (2) change in shape of Whites earnings, (3) change in shape of African American and Latinos earnings and, (4) an interaction term. Findings from these analyses, reported in Table 6, indicate that between 1990 and 2010 and at all levels of the earnings distribution, factor 1 contributes more than factors 2 and 3 together in explaining racial wage gaps. By 2010, and among the lower two quartiles of earners, changes in median wages from 1990 explained approximately two-thirds of wage gaps. At the upper two quartiles of earners, it explained approximately three-quarters of the wage gaps. The proportion explained at all levels tended to be slightly higher among African Americans than Latinos relative to their White and gender counterparts. The consistency of these findings with our core modeling in Tables 3 and 4 only bolsters our prior interpretations and lends significant confidence in highlighting the median earnings approach we have employed.

6. Discussion The findings offered in this article, which draw on analyses of IPUMS data, signal the mitigation or disappearance of the public sector as the long-standing economic niche for African Americans and Latinos in managerial occupations. New governance-induced employer discretion—discretion that has been associated with the restriction of worker employment rights and the de-bureaucratization of employment rules and regulations—seems, by most indications, to have progressively disadvantaged minority managers. Similar to long-standing practices in the private sector, discretion is associated with allocation of minority managers into segregated work tasks/roles and their subordination to Whites in authority hierarchies. This seems to have chipped away and relatively effectively at the relative wage parity that minority managers once experienced in public sector employment. Our findings reveal the parallel patterning of rising inequality for African American and Latino public sector managers, although important qualifications emerge. In terms of niche status in the public sector, it is clear that the growing disadvantages reported are more pronounced for Americans than Latinos. Specifically, wage inequalities with Whites more appreciably widened—and, hence, came to more clearly resemble the wage gap in the private sector—for African Americans. We attribute this to the potentially distinct impact of employer discretion across minority groups. As suggested previously,

46

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

Table 6 Changes in distribution of earnings, 1990–2010. Minority group % in white quartiles

Contribution to change

1990

2010

Change

Median change

Shape shift of whites

Shape shift of African Americans

Men African American Highest Quartile Second Quartile Third Quartile Lowest Quartile

22.5 16.7 10.3 10.5

15.8 15.3 12.6 13.3

6.7 1.4 2.3 2.8

75.2 75.0 65.9 64.8

12.6 12.4 17.0 18.3

12.2 12.7 16.1 16.6

Latino Highest Quartile Second Quartile Third Quartile Lowest Quartile

26.4 18.9 13.3 12.8

18.9 17.4 13.6 14.2

7.5 1.5 .3 1.4

72.7 71.5 64.3 62.8

12.1 11.8 14.3 14.1

12.0 12.2 8.9 9.3

Women African American Highest Quartile Second Quartile Third Quartile Lowest Quartile

12.1 16.2 11.7 11.8

16.3 15.7 12.9 13.0

5.8 0.5 1.2 1.2

73.3 72.9 63.4 62.6

11.6 11.4 10.8 11.0

12.1 11.8 14.5 14.7

Latina Highest Quartile Second Quartile Third Quartile Lowest Quartile

13.4 17.5 11.5 10.9

17.4 16.2 13.1 13.2

6.0 1.3 1.6 2.3

71.3 70.7 64.4 62.7

12.6 12.8 11.8 13.7

9.3 9.9 13.2 14.0

Latinos are better able to communicate evaluation-relevant criteria than African Americans and are experience more moderate levels of workplace segregation, which likely translates into more integrated informal and formal job networks. Secondly, women appear to be less disadvantaged in terms of niche status than men. We attribute this finding to: (1) the ‘‘double disadvantage’’ experienced by minority women being partially counteracted by a gender penalty experienced by White women, and (2) the unique position of minority men being penalized on the basis of race in their gender group, which serves to widen inequities. Overall, the findings from this study do not bode well for either African American or Latino managers given that the coercive isomorphism of new governance movement is marching forward. In fact, calls to reform government have received support across political parties in the last decade (Kamarck, 2007; Ingraham et al., 2000). Noteworthy is that in recent years, Democratic policy makers have been prime movers in the ‘‘reinventing government’’ movement which seeks to enact reforms that make government more efficient through downsizing, decentralizing authority, and focusing more on results and less on rules (Kamarck, 2007). Rapidly accelerating new governance and its concomitant loss of parity in earnings likely translate into a series of long-term disadvantages among African Americans and Latinos in managerial positions that may have already commenced. In this vein, minority managers are handicapped because they have relatively few alternative locations in the American labor market to achieve more favorable earnings, and, their relatively negligible accumulation of wealth makes it uniquely difficult to compensate for wage-based hardship (Conley, 1999; Oliver and Shapiro, 1995). Further, declining access to resources wrought by the implementation of new governance threatens the decades-old ability of African American and Latino managers in the public sector to accumulate the financial means to blunt discrimination and provide for the orderly transmission of privileged economic status on an inter-generational basis, a stratification phenomenon that has become a hallmark of the public sector for minorities during the civil rights era (Landry and Marsh, 2011; Hout, 1984).

7. Conclusion If new governance reform continues unabated, it is vital to implement protections that counteract cumulative effects on African Americans and Latinos in managerial occupations. It is for this reason that we call for policy that balances the thrust of new governance and its ‘‘business model’’ orientation with the need to achieve racial/ethnic equity in socioeconomic rewards, such as wages. Important in reaching this compromise, we believe, is providing minority workers with opportunities to contest employer discretion in the workplace. Indeed, sociological research documents that the negative race-based impact of discretionary practices are most effectively blunted if minority workers can resort to a formal grievance procedure—with formal counsel—as part of due process in contesting employer practices (Dobbin, 2009; Kalev and Dobbin, 2006). Incorporating procedural protections in the workplace, mandating that there be more flexible work-group boundaries that offer minorities and their work product increasing exposure, and to a greater range of decision-makers and co-workers,

47

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

and facilitating integrated formal and informal social networks (Kalev, 2009) will be, in our view, especially important safeguards. Finally, firms should make a commitment to placing minority managers in integrated work settings (see Kalev, 2009; Tomaskovic-Devey and Stainback, 2007). Overall, findings from this study should serve to re-orient sociologists about the status of the public sector as a locus of racial/ethnic stratification. Indeed, similar to the private sector, inequality at the managerial level, even in government work, is and should be conceived of as fluid and volatile rather than static and monolithic. Additional analyses should undoubtedly be undertaken, drawing from the arsenal of theoretical/conceptual tools, such as ‘‘social closure’’, and ‘‘queuing’’ theory (Lieberson, 1980), that have shed light on the mechanisms that drive racial stratification in the private sector. Indeed, we find it noteworthy that Weber (1968) maintained as a ‘‘fundamental scenario’’ a factual situation similar to that experienced by African American and Latino managers in the ‘‘new public sector,’’ namely the devaluation of well-rewarded positions. Considerably more research is needed, to be sure. A broader range of socioeconomic outcomes, for instance, merit attention. For instance, worthwhile future analyses might include attention to promotion into managerial positions, supervisory responsibility and the ability to retain privileged status (, i.e., downward mobility). Furthermore, it is important to more directly link discretionary decision-making and race/ethnic inequalities in the new governance era. We believe this can best be accomplished with a ‘‘mixed methods’’ approach that includes case studies of particular workplaces so that purported causal factors can by observed first-hand. In this regard, research should be broad-based and examine workers across sectors and workers in the public sector who vary in the extent to which their jobs have been influenced by new governance. In sum, we hope the findings from this study of wages lead sociologists to undertake this research. We look forward to it, as it will enhance our understanding of a critical issue in racial stratification research—the viability of a niche location among African Americans and Latinos in a privileged occupational category in the U.S. labor market. Appendix Descriptive statistics for IPUMS sample African American

Latino

X

X

SD

White SD

X

SD

Pre-reform Human capital College Post-College Health problems

79% 52% 25%

77% 50% 24%

80% 52% 24%

Job/labor market Unionized Fed. Government <1 reform aspect adopted 1 aspect adopted

25% 27% 6% 7%

23% 21% 5% 7%

33% 20% 7% 7%

Family/household Married Children

41% 2.5

1.1

47% 2.3

1.2

47% 2.4

1.1

Sociodemographic South Age

20% 36.4

4.2

18% 37.1

3.9

18% 37.3

4.0

Early reform Human capital College Post-College Health problems

78% 53% 25%

75% 52% 24%

81% 54% 24%

Job/labor market Unionized Fed. Government <1 reform aspect adopted 1 reform aspect adopted

22% 23% 16% 17%

21% 19% 18% 17%

22% 20% 17% 17% (continued on next page)

48

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

Appendix (continued)

African American

Latino

X

SD

X

SD

White X

SD

Family/household Married Children

44% 2.4

1.2

45% 2.3

1.3

47% 2.4

1.2

Sociodemographic South Age

21% 37.3

4.1

18% 37.2

3.8

16% 37.1

4.0

Later reform Human capital College Post-College Health problems

78% 54% 27%

77% 52% 25%

80% 54% 22%

Job/labor market Unionized Fed. Government <1 reform aspect adopted 1 reform aspect adopted

21% 20% 37% 34%

21% 17% 37% 35%

20% 16% 39% 35%

Family/household Married Children

43% 2.5

1.3

47% 2.4

1.2

48% 2.4

1.2

Sociodemographic South Age

21% 37.2

4.4

19% 37.4

3.7

21% 37.3

3.8

References Ahituv, Avner, Tienda, Marta, 2004. Employment, motherhood, and school continuation decisions of young White, Black, and Hispanic women. J. Labor Econ. 22, 115–158. Allport, Gordon, 1954. The Nature of Prejudice. Addison-Wesley, Reading. Battaglia, R. Paul, Condrey, Stephen, 2007. Framing civil service innovations: assessing state and local government reform. In: American Public Service: Radical Reform and the Merit System. Taylor and Francis, New York, pp. 25–46. Becker, Gary, 1964. Human Capital. National Bureau of Economic Research, New York. Berman, Evan, Bowman, James, West, John, War, Van, Montgomery, 2006. Human Resource Management in the Public Service. Sage, Thousand Oaks. Blumer, Herbert, 1966. Industrialization and race relations. In: Hunter, Guy (Ed.), Industrialization and Race Relations: A Symposium. Oxford University Press, London. Bobo, Lawrence, Massagli, Michael, 2001. Stereotyping and urban inequality. In: OConnor, Alice, Tilly, Chris, Bobo, Lawrence (Eds.), Urban Inequality: Evidence from Four Cities. Russell Sage, New York, pp. 89–162. Bowman, James, West, John, 2007. American Public Service: Radical Reform and the Merit System. Taylor and Francis, New York. Braddock, Jomills, McPartland, James, 1987. How minorities continue to be excluded from employment opportunities: research on labor market and institutional barriers. J. Soc. Issues 43, 5–39. Brown, Michael, Erie, Paul, 1981. Blacks and the legacy of the great society: the economic and political impact of Federal Social Policy. Public Policy 29, 299– 330. Caucutt, Elizabeth, Nezih, Gurner, Knowles, James, 2002. Fertility decisions and gender differences in labor turnover, employment, and wages. Rev. Econ. Dyn. 5, 856–891. Collins, Sharon, 1993. Blacks on the bubble: the vulnerability of black executives in white corporations. Sociol. Quart. 34, 429–447. Collins, Sharon, 1997. Black Corporate Executives: The Making and Breaking of a Black Middle Class. Temple University Press, Philadelphia. Conley, Dalton, 1999. Being Black but Living in the Red. University of California Press, Berkeley. Cox, Thomas, Nkomo, Stella, 1990. Invisible men and women: a status report on race as a variable in organizational behavior research. J. Org. Behav. 11, 459–477. Dimaggio, Paul, Powell, Walter, 1991. The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. In: Powell, Walter, Dimaggio, Paul (Eds.), The New Institutionalism in Organizational Analyses. University of Chicago Press, Chicago, pp. 63–82. Dobbin, Frank F., 2009. Inventing Equal Opportunity. Princeton University Press, Princeton, NJ. Elliot, James, Smith, Ryan, 2001. Ethnic matching of supervisors to subordinate work teams: exploratory findings on bottom-up ascription. Soc. Probl. 48, 258–276. Elliot, James, Smith, Ryan, 2005. Race, gender and workplace power. Am. Sociol. Rev. 69, 365–386. Farley, Reynolds, 2005. The American People. Russell Sage, New York. Farley, Reynolds, Allen, Walter, 1987. The Color Line and The Quality of Life in America. Russell Sage, New York. Fernandez, John, 1981. Racism and Sexism in White Corporations. John Wiley Press, New York. Frumkin, Peter, Galaskiewicz, Joseph, 2006. Institutional isomorphism and public sector organizations. J. Public Admin. Theory 22, 35–72. Glass Ceiling Commission, 1995. Good for Business: Making Full Use of the Nation’s Human Capital Government Printing Office. Washington, D.C.

G. Wilson, V. Roscigno / Social Science Research 54 (2015) 36–49

49

Hays, Steve, Sowa, Jessica, 2006. Changes in state civil service systems. Rev. Public Pers. Admin. 26, 102–117. Hite, Larry, 2007. Hispanic managers and professionals: reflections on life and work. Gender Work Org. 14, 20–36. Hout, Michael, 1984. Occupational mobility of black men: 1962–1973. Am. Sociol. Rev. 49, 308–323. Ingraham, Patricia, Seldon, Sally, Donald, Moynihan, 2000. People and performance: challenges for the future public service. Public Admin. Rev. 60 (1), 54– 60. Jaynes, Gerald, Williams, Robin, 1989. A Common Destiny: Blacks and American Society. National Academy Press, Washington. Jones, Frank, Kelley, Jonathan, 1984. Decomposing differences between groups: a cautionary note on measuring discrimination. Sociol. Methods Res. 12, 323–343. Kalev, Alexandra, 2009. Cracking the glass cage. Am. J. Sociol. 98, 1123–1156. Kalev, Alexandra, Dobbin, Frank, 2006. Enforcement of civil rights law in private workplaces: the effects of compliance reviews and lawsuits over time. Law Soc. Inquiry 31, 855–903. Kamarck, Elaine, 2007. The End of Government as We Know It: Making Public Policy Work. Lynne Reinner, Boulder. Kluegel, James, Bobo, Lawrence, 2001. Prejudice. In: Inequality, Urban (Ed.), O’Connor, Alice, Tilly, Chris and Lawrence Bobo. Russell Sage, New York, pp. 105–152. Koencker, James, Barrett, Gerald, 1978. Regression quantiles. Econometrika 46, 330–350. Landry, Bart, Marsh, Kristin, 2011. The evolution of the black middle class. Annu. Rev. Sociol. 38, 102–121. Lawthler, William, 2003. Privatizing personnel: outsourcing public service functions. In: Hays, Steven, Kearney, Robert (Eds.), Public Personnel Administration: Problems and Prospect. Prentice-Hall Saddle River, pp. 113–151. Lieberson, Stanley, 1980. A Piece of the Pie. University of California Press, Berkeley. Malamud, Deborah, 1995. The last minuet—disparate treatment after hicks. Mich. Law Rev. 1995, 2229–2324. McBrier, Debra, Wilson, George, 2004. Going down? race and downward mobility in the 1990’s. Work Occup. 31, 283–321. Meyer, John, Rowan, Brian, 1991. Institutionalized organizations: formal structure as myth and ceremony. In: Powell, Walter, Dimaggio, Paul (Eds.), The New Institutionalism in Organizational Analyses. University of Chicago Press, Chicago, pp. 143–164. Model, Suzanne, 1985. Economic niches among immigrants. In: Katz, Steven (Ed.), The Historical Origins of the American Underclass. Harvard University Press, Cambridge, pp. 85–135. Morgan, Douglas, Cook, Brian, 2014. New Public Governance. M.E. Sharpe, Armonk. Mundra, Kenneth, Moellner, Anthony, Loez-Aqueres, Wiliam, 1995. Investigating Hispanic underrepresentation in managerial and professional occupations. Hispanic J. Behav. Sci. 25, 513–529. Oliver, Melvin, Shapiro, Thomas, 1995. Black Wealth/White Wealth. Routledge, New York. Pettigrew, Thomas, 1964. A Profile of the Negro American. Van Nostrand Press, Princeton. Pettigrew, Thomas, 1971. Racially Separate or Together. Mcgraw-Hill, New York. Pettigrew, Thomas, Martin, Joanne, 1987. Shaping the organizational context for African American inclusion. J. Soc. Issues 43, 41–78. Powell, James, 1986. Censored regression quantiles. J. Econom. 32, 143–155. Roscigno, Vincent, 2007. The Face of Discrimination. Rowman and Littlefield, New York. Roscigno, Vincent, Williams, Lisa, Byron, Reginald, 2012. Workplace racial discrimination and middle class vulnerability. Am. Behav. Sci. 56, 696–710. Smith, Ryan, 2005. Do the determinants of promotion differ for white men versus women and minorities? an exploration of intersectionalism through sponsored and contest mobility processes. Am. Behav. Sci. 48, 1157–1181. Stainback, Kevin, Tomaskovic-Devey, Donald, 2012. Documenting Desegregation: Racial and Gender Segregation in Private Sector Employment Since the Civil Rights Act. Russell Sage, New York. Tomaskovic-Devey, Donald, Stainback, Kevin, 2007. Discrimination and desegregation: equal opportunity progress in U.S. Private Sector workplaces since the Civil Rights Act. Ann. Am. Acad. Polit. Soc. Sci. 609, 49–84. Tomkowitz, James, Brenner, Orlando, 1996. The relationship between Hispanic stereotypes. J. Soc. Behav. Person. 11, 511–520. Villemez, Wayne, George, Bridges, 1994. The Employment Contract. ILR, Ithaca. Waldinger, Roger, 1996. Still the Promised City: African Americans and Immigrants in Post-Industrial New York. Harvard University Press, Cambridge. Weber, Max, 1968. Economy and Society. Basic Books, New York. Wilson, George, 1997. Pathways to power: racial differences in the determinants of job authority. Soc. Probl. 44, 38–54. Wilson, George, 2006. The rise of at-will employment and racial inequality in the public sector. Rev. Public Pers. Admin. 26, 178–187. Wilson, George, 2009. Downward mobility of women from white collar employment: determinants and timing by race. Sociol. Forum 24, 382–401. Wilson, George, Roscigno, Vincent, 2010. Race and the dynamics of downward mobility from upper-tier occupations across the early work-career. Soc. Sci. Res. 39, 67–77. Wilson, George, Roscigno, Vincent, Huffman, Matt, 2013. Public sector transformation, racial inequality, and downward mobility. Soc. Forces 91, 975–1006. Wilson, George, Roscigno, Vincent, Huffman, Matt, 2015. Public Sector Reform and Racial Inequality. Social Problems 62, 162–185. Yaffee, James, 1995. Latina managers in employment—perceptions of organizational discrimination. Hispanic J. Behav. Sci. 17, 334–346.