Changes from Above

We begin with a brief account of actions taken by the international union. Unions operate within a federated structure with “locals” representing geographically based collective bargaining units and an “international” that serves as the central governing council for what are usually U.S.-based, and sometimes Canadian-based, locals. The degree of autonomy and self-governance among locals vis-à-vis their parent internationals can be very high, although an opposite trend of greater consolidation and centralization at the international level has been on the rise among a number of unions within the U.S. labor movement (Milkman Reference Milkman2005). This has been the case with Unite Here. With regard to the questions motivating our case study, Unite Here's International has adopted an explicit commitment to diversity issues that has filtered down to the local level with considerable, albeit not uniform, success.

Unite Here's adoption of diversity as an explicit union issue stems from necessity. Industries characterized by jobs with relatively low skill-barriers to entry, such as the hotel and food service industries represented by Unite Here, were poised to absorb the dynamic demographic changes wrought by increasing immigration flows over the last several decades. As the demographics of Unite Here's membership changed in most urban markets across the United States, the union was forced to address a new set of issues, concerns, and conflicts that arose from its increasingly immigrant membership. Yet in many cities, native-born workers, especially African Americans, remained employed in hotels and restaurants represented by Unite Here. Diversity had found the union, whether the union wanted it or not. What the union chose to do given these circumstances, however, is the story we partly describe in this article.

Unite Here was formed by a 2004 merger of two independent service-sector unions—UNITE (Union of Needletrades, Industrial, and Textile Employees) and HERE (Hotel Employees and Restaurant Employees). Although the merger yielded some short-term rocky internal politics, leadership and union practices within the hospitality division—the HERE side of the union—have been relatively stable. A strong commitment to diversity issues within the hospitality division extends from before the merger to the current time period under study, a period under the consistent leadership of the current president of Unite Here, John Wilhelm.

Wilhelm began formulating and implementing an organizational and political program around diversity shortly after he was first elected president of the former HERE in 1998 (he later became the hospitality industry president of Unite Here in 2004 and president in 2009). Internally, HERE's membership ranks had come to be dominated by immigrant workers experiencing greater vulnerability on the job due to their immigration status, undocumented or not. These workers connected their employment experiences to a broad array of immigration concerns affecting not only themselves as workers, but their families and communities as well. Wilhelm and other union leaders recognized that addressing immigration issues was fundamental to both representing these workers on the job as well as responding to their issues as members of a democratic organization.

Yet increased attention to immigration also brought to the fore the threat of displacement faced by African Americans in the industry—what one union staffer called “the white elephant in the room.”Footnote ³ This staffer went on to report:

African American members still constituted a disparate share of the union's membership—35% of Local 1's ranks (Warren Reference Warren2005)—and were not hesitant in voicing their concerns about their declining numbers in the hotel industry. In tackling immigration issues, the union risked alienating one of their most politically active and engaged member constituencies.

In order to forestall divisiveness, union leaders made a deliberate decision to link the issues of immigration reform and African American employment under the rubric of diversity. Wilhelm laid out a formal program to expand the union's work on diversity at HERE's International Union Convention in 1999. Most visibly, this moment initiated a strong push forward on immigration-related activities, especially increased political mobilization calling for federal immigration reform. Yet, as one union staffer described, Wilhelm's initiative was guided by a simultaneous commitment to addressing “both the immigration crisis and the displacement of African Americans by immigrants” (Interview, Laura Hausen, May 2011). The Immigrant Workers Freedom Rides (an immigration reform advocacy project) and the African American Hiring Initiative were two of the first projects to emerge from the new diversity program. We discuss both in the following section.

Nearly a decade later, with Wilhelm's election to the presidency of Unite Here in 2009, the union instituted a number of organizational changes and reforms at the national level aimed at further bolstering Unite Here's diversity efforts. Chief among these was the creation of a new executive position, the General Vice President for Immigration, Civil Rights, and Diversity, one of the union's five general officers. Additionally, three executive board members were assigned to form an official “Diversity Committee.” As Unite Here's own press copy announces, these organizational changes demonstrate “an institutionalized commitment to diversity” (Unite Here 2009). Significantly, these steps represent the codification and evolution of previous and ongoing activities at both the national and local level.

We now turn to the practices and policies engaged by Unite Here Local 1, the Chicago affiliate. It is important to note that, although an autonomous body, Local 1 shares Wilhelm's political commitment to diversity and has willingly engaged in many of the international's diversity initiatives. Demographically, its workforce is among the most diverse of the Unite Here locals measured in terms of African American and immigrant representation. As a case, Local 1's practices and policies provide a strategic research site at which to observe how workers “manage” diversity at the worksite through their own forms of organization. Our case also reveals the ways in which experiences with diversity drawn from different locales get translated into new strategies in new places. In our case, the international union serves an important heuristic function as a clearinghouse for local efforts—efforts that are then honed and refined over time and adapted to new local contexts.

Union Diversity Practices

How unions and other democratic organizations structure themselves, particularly in an environment of members with diverse, and potentially conflicting, identities based on race, ethnicity, nativity, and gender is directly related to the organization's effectiveness, collective identity, and specific outcomes. These internal institutional rules, norms, practices, and structures can either enable or constrain workers' ability to challenge power relations, especially the racial and ethnic division of labor at the workplace. While some organizational structures and practices can lead to competition and conflict among a diverse membership, others can build solidarity between groups while ensuring that each subgroup's unique concerns are adequately represented. One specific outcome in the union context that illustrates whether there is conflict or solidarity among African American and immigrant workers specifically is a collective bargaining agreement between a union and employer specifying wages, benefits, and working conditions. In our case study, Unite Here Local 1 won unprecedented diversity language in its contract covering over 7,000 hotel workers in the city of Chicago, representing the specific concerns of its diverse membership. But how the union got to that outcome is a result of its unique and deliberate organizational structure, one in which the specific racial, ethnic, nativity, gender, and sexual identities of its workers are taken into account together with workers' job classifications and class identities.

One staff member explains the underlying organization's philosophy around the explicit recognition of racial, ethnic, and gender differences in this way, “The union really ought to be representative and it really ought to be widely representative of the real backbone of our union anyway, which is housekeeping, which is overwhelmingly female and overwhelmingly female of color” (Interview, Julie Smith, December 2002). For the union's leadership, marginalized women-of-color housekeepers, the lowest paid and most disadvantaged members of the union, are seen as a source of strength, not weakness (“the real backbone”). Union staff consciously creates internal structures intended to represent, lift up, and empower these workers and their multiple identities.

Unite Here Local 1 has not always been organized in such a way that recognizes explicitly the multiple identities of its members. In order to create a more democratic and representative union, since 1999 union staff have sought to “raise expectations” of members, “develop new leaders” through “building the committee,” and show employers, politicians, community members—and themselves—that they are willing to “take it to the streets” and do whatever it takes to win better working conditions and standards in their industry and in local politics. Specific constraints—from the union's own history, to the broader political and economic context, employer opposition, and worker attitudes—are ubiquitous, even if their shape or contour differs depending on industry, union, and overall context. How political actors negotiate these organizational constraints, while redesigning existing institutional structures, is important in assessing how successful they will be at representing all of their members and empowering them to contest the racial and ethnic division of labor.

With the arrival of new union leadership in late 1999, staff members engaged in organizational redesign as a method of restructuring the union to empower its diverse members and their diverse interests. This new leadership team first took control of what was a moribund and corrupt union with a broad mandate to change.Footnote ⁴ The new leadership rooted out the corruption by firing the old staff and allowing others to leave voluntarily, and brought in outside and diverse leaders, staff, and organizers. One example of this change is that over half of the full-time staff speaks Spanish, compared to none before. In addition, all of the union's membership meetings and publications are translated into Spanish, Bosnian, Polish, and Chinese to better accommodate and engage their multiethnic members in the life of the union.

Besides translating publications and membership meetings, another practice instituted by the new staff was to survey the membership to discover their specific concerns and issues—something the union had never done before. Staff then began to organize rank-and-file members and build a representative internal committee of worker-leaders to be active participants in the union. “Rank-and-file organizing committees” are internal leadership structures that some unions use to advance their many goals. For instance, instead of relying just on staff or elected leaders to talk to workers about their concerns and construct a collective identity, the most prominent method of most unions, some like Unite Here Local 1 train, and then rely on, an ever-growing volunteer “committee” of rank-and-file leaders to do these tasks. This has several effects: 1) workers are more likely to be honest with and trust co-workers compared to union staff; 2) it helps to develop new leaders and teach them political skills and capital which then empowers more members to engage in collective action in the workplace as well as other union functions such as political action; 3) it helps unions shift staff resources over to organizing and recruiting new workers. By training rank-and-file workplace leaders to handle shop-floor problems and issues themselves, it frees up valuable staff time to devote to organizing new members, and not “servicing” existing members; and 4) Finally, by encouraging high norms of participation and decision-making, these internally created leadership structures provide another mechanism of democratic accountability within organizations in addition to formal electoral rules.

Part of the goal of surveying the membership was to get information and assess the concerns of members, a key aspect in the process of building the organizing committee. The surveys were also used as an organizing tool to organize workers and recruit leaders. If workers could complete ten surveys of their co-workers, they passed a good test of having the raw materials of being leaders (Interview, Sam Donaldson, December 2002). In addition, this process of surveying what was ultimately a fifth of the membership (2500 completed surveys) also helped in mapping out the social and political networks of members. The survey included questions about church and political organizational membership. The resulting 250 churches listed later became a key basis of the union's outreach efforts to garner support from the religious and wider community in its various contract campaigns. The outcome was that the union took advantage of the internal resources its members bring to the workplace and union hall: the social, community, and political networks in which workers are embedded outside of the workplace (Kurtz Reference Kurtz2002; Needleman Reference Needleman2003). Insofar as these networks are highly structured by patterns of racial and ethnic residential segregation, the ability of the union to engage in a process to discover and then utilize these significantly diverse networks is unusual.

The focus on developing a strong and large internal rank-and-file committee of leaders to take ownership of and some control over the union is an organizing model that Unite Here has been developing, implementing, and perfecting for some time. And for the most part, it is an unusual model within the broader labor movement. Although the union did not achieve its goal of having a committee that represents 10% of the 14,000 workers (a leadership committee of 1400), during the height of its campaigns, it often recruits a rank-and-file organizing committee numbered in the hundreds (400–700). But it is one thing to build an internal rank-and-file organizing committee of worker-leaders, and quite another to deliberately build a representative committee of this magnitude. This approach to organizational redesign increases the likelihood that issues affecting a range of workers will get on the table and that the bargaining unit as a whole will actually win those issues at the contract table, and most important in this case, it increases the ties between African American and immigrant workers who are then empowered to contest the racial division of labor at work.

How exactly did the staff and leadership of the union build a representative—and by representative we mean by hotel, specific job classification, race, ethnicity, immigration status, gender, and sexual orientation—internal rank-and-file committee of over 500 workers from scratch in three years (1999–2002)? First, they planned a series of committee and member actions with several goals in mind: to educate the public about workers' plight, to educate members about their own conditions, and to begin to put pressure on the employers as an expression of the union's increasing strength. These actions, in addition to helping increase the organizational capacity of the union, also served to transform and raise the expectations of members about what they could possibly achieve. Second, and related to the building of an over-500-member representative rank-and-file leadership committee, was the establishment of a sixty-five-member rank-and-file negotiating committee among the “die-hards.” Again, this committee was representative by job classification, race, ethnicity, nationality, gender, and sexual orientation. Upon its creation in 2000, this rank-and-file member negotiating committee—generally rare among unions in the American labor movement—marked the first time that rank-and-file workers had been involved in contract negotiations in this local union's history.

The volunteer 500-rank-and-file-member organizing committee highlights the role of organizational redesign. Under certain conditions, political organizations can be constructed in such a way as to recognize explicitly the multiple identities of members, structure the organization around these identities in a federated arrangement, take advantage of the existing social networks and relationships these subgroups of diverse workers have, and mobilize these groups of workers around a collective good, undercutting the potential problem of undermobilization (Kurtz Reference Kurtz2002). This particular strategy of organizational redesign is not necessarily based on high-road political principles of democratic and equal representation, important and ideal as those are. Instead, it is based more on the pragmatic need to ensure every significant subgroup of workers is included and involved in the union, lest employers exploit the already existing racial and ethnic division of labor inherent in any heterogeneous workforce and union membership. To do this, the union had to “make it possible to have as many different people involved as possible” (Interview, Sam Donaldson, December 2002) which was the primary justification for language translation to facilitate the involvement of immigrant workers and the subsequent demands around ensuring that African American workers maintain a presence in the industry.

Marshall Ganz (Reference Ganz2000) and Morris and Staggenborg (Reference Morris and Staggenborg2002) argue that the most successful social-movement organizations are those with leadership teams that have what they call both “insiders” and “outsiders.” The interaction of organizational outsiders who bring political skills and capital with insiders (e.g., members) with a different set of resources expands the organization's capacity. A staff member, John Williams, described exactly how this interaction between inside and outside leaders worked as rank-and-file member-leaders learned to organize and mobilize their co-workers:

This strategy is nothing new; it can be found in contemporary community organizing and in previous eras of union militancy (Ganz Reference Ganz2000; Sen Reference Sen2003; Smock Reference Smock2003; Warren Reference Warren2010). As the staff member, John Williams, explained, “The model has been around a long time. When you look at the thirties … the CIO really had it. When unions stopped organizing, it went away” (Interview, December 2002). Linking the political skills leaders from outside the labor movement bring to union organizing, one of Voss and Sherman's (Reference Voss and Sherman2000) respondents described how he learned to build committees from his experience as a community organizer:

By taking a page from one of the most successful models of unionism in terms of the inclusion of workers of color and women (the CIO), and from community organizing models which rely more on the internal resources of their members, Unite Here has revived and adapted the model for use in a different yet contemporary context of a racialized and gendered global economy. When union staff member John Williams was asked by the authors about how the union deals with the potential problems that come with having a diverse and occupationally segregated workforce, he responded, “If we do our committee-building right, it takes care of itself. If we don't, then we are in deep shit in many ways” (Interview, December 2002).

The effects of building these large internal representative rank-and-file organizing committees go beyond workplace-specific contract campaigns or internal union politics. The bonds of solidarity and political capital that this diverse group of worker-leaders learn—talking to, organizing, and mobilizing their co-workers, neighbors, politicians, community, and religious leaders around issues of social justice—is transferable to other contexts and situations. For example, under its “member-avoidance program” before the union's transformation, the local never engaged rank-and-file members in political activity. Instead, relying on the routine tactics characteristic of the majority of U.S. unions, Local 1 contributed campaign funds and lobbied its political friends for favors. But the local's shift to a more engaged membership has had direct effects on Chicago politics, especially in Black and Latino neighborhoods. One key instance of this is that roughly forty Black union members get paid time off of work to conduct voter mobilization in African American wards during elections. An important political result of this was in 2004 when Unite Here African-American members contributed directly to increasing turnout rates in several Black city council wards, contributing to the election of Barack Obama to the U.S. Senate.

Another example is the creation of solidaristic bonds between native-born African American and immigrant workers through internal political education like that conducted on the 2003 Immigrant Workers Freedom Ride (IWFR) buses. Foreshadowing the recent book by Isabel Wilkerson (Reference Wilkerson2010), the union helped build bonds of solidarity between subgroups of its members by linking the migration histories of Black and immigrant workers. Through popular education activities implemented on the IWFR bus ride and the subsequent “Migration Oral History Project,” African American and immigrant workers were encouraged to focus on their own family narratives of migration, uplifting the “push” and “pull” factors compelling their migration history.

Part of the process of organizational redesign and the empowerment of different subgroups of workers based on their race, ethnicity, nativity, and gender is the construction of a collective identity that workers then use to contest the racial and ethnic division of labor at work and at home. An important aspect of this political process is the use of particular frames to educate and convince members to take action. The fact that a union represents its multiracial, multiethnic workers and their multiple identities does not automatically translate into the framing of issues or even a broader collective identity in any one particular way. Unions and political organizations use a variety of strategies to build a collective identity and frame their demands, and the results of any particular strategy depend on a number of factors. Different strategies can unite or divide the membership; prevent or enable some issues to get onto the agenda; and mobilize or undermobilize specific internal and external resources (Kurtz Reference Kurtz2002).

Throughout American political history, especially among social reform movements, activists have most often avoided explicit framing or discussion of issues deemed controversial such as race, class, gender, or sexuality, leaving these structures of inequality uncontested (Allen and Allen, Reference Allen and Allen1974; Cohen Reference Cohen1999; Iton Reference Iton2000). In the case of organized labor, labor historian Bruce Nelson (Reference Nelson2000) describes how historians and organizers have seen “identity” issues as impediments to labor solidarity. He writes:

This dilemma—of figuring out under what conditions identity and class or worker identities can strengthen rather than undermine each other—poses a significant challenge for organizers to which there are no easy panaceas or answers. It is the case that within diverse constituencies, the ways in which particular issues are constructed, especially around race and ethnicity, can, under some conditions, be divisive. It is also the case that when organizations ignore certain issues affecting the most marginalized out of fear of their divisive impact, those populations often fail to be adequately represented and those structures of inequality are reinforced instead of contested (Cohen Reference Cohen1999; Strolovitch Reference Strolovitch2007). We also know that in the union context, employers have historically sought to emphasize divisive issues in efforts to divide the workforce by race and ethnicity and prevent collective action.

In the case of Local 1, the dominant frame used by union leaders to construct a collective identity and create solidaristic bonds among its members and across their differences is that of a singular “union” identity on the surface, but one which is constituted and laden with deeper meanings of the multiple identities and concerns of union members. Local 1 partly uses John Rawls' (Reference Rawls1971) difference principle in efforts to build broad solidarity while also protecting its most vulnerable members. The thrust behind the difference principle is that a community emphasizes the least advantaged (or most disadvantaged) when agreeing to principles of justice. Time and time again, the leaders of Local 1 emphasize the need to fight for the least advantaged, linking their fate to the fate of the entire union. For example, the local's President would often say to rank-and-file organizing committee members in articulating the union's agenda, “If the most vulnerable aren't protected, the boss wins and we all lose” (Warren Reference Warren2002). This is the most common frame used by leaders, staff, and members alike when advocating for issues affecting the union's most disadvantaged workers. We have heard this frame used time and again to advance the issues of immigrant workers, gay and lesbian workers, African American workers, and housekeepers who are overwhelming women of color (Warren 2001–Reference Warren2011). This strategy of constructing frames that emphasize difference-within-unity and the need to maintain solidarity across a diverse membership experiencing multiple inequalities is an example of the internal political process used to contest the racial and ethnic division of labor.

Bargaining Diversity

In 2006, Unite Here Local 1 in Chicago won unprecedented diversity language in its collective bargaining contracts, ensuring that hotel employers hire and retain African Americans. Albeit new, this language added to the broad and expansive nondiscrimination language the union won in previous contracts, as well as the broad scope of equity provisions the union had secured for its diverse workforce, such as domestic-partner benefits for its lesbian, gay, and transgendered members, gender equity in health-care coverage (e.g., contraceptive equity), and protections for its immigrant workforce (Warren Reference Warren2005). Pragmatically, contract provisions targeted at African American workers became a concrete answer to what one union official at Unite Here Local 1 described as its most pressing diversity challenge: “How do we move forward on immigration issues without isolating our African American members?” (Interview, Amy Hall, 2011).

The diversity language comprises two parts: a commitment to a diverse workforce by the employer and a set of “affirmative steps” by which the employer will demonstrate a good-faith effort to fulfilling this commitment. Section 16, Part 1, titled “Commitment,” from Unite Here Local 1's 2006 contract with hotel employers reads as follows:

As our informants described, the employers were by and large open to the inclusion of this language, but the concrete nature of the commitment was highly significant as both a policy and a political victory. The contractual commitment represents the formal adoption of union diversity goals by management and provides the union with a mechanism by which to hold employers accountable. Additionally, the language provides a means by which to hold employers accountable in their dealings with workers on a day-to-day basis, outside of formal grievance dealings and contract negotiations. In short, signing such a commitment shifts the tone and tenor of future interactions around race, immigration, and diversity, on the shop floor and beyond. Of course, by securing the diversity language, the union demonstrated that it could win on an issue of great concern to many of its members. By so doing, the union also provided its members with a means by which to hold the union accountable to the substantive realization of its diversity commitments and practices.

Section 16, Part 2, “Affirmative Steps,” outlines a program of action centered on community outreach as the primary means by which to actualize the diversity commitment. The 2006 collective bargaining agreement states:

Specifically, the contract mandates that the union and the employer jointly develop “an annual strategic action-oriented outreach program” designed to accomplish the following:

1. 1. Inform and educate members of underrepresented community about job and career opportunities with the Employer;

2. 2. Establish contacts with diverse community groups and schools that serve underrepresented communities and seek to develop partnerships with them to enhance their knowledge of the Employer and jobs and career opportunities for community members with the Employer.

Lastly, the contract stipulates that results must be tracked and reported annually.Footnote ⁶

The impetus behind this diversity contract language stems back to the African American Hiring Initiative created by local union leaders and the Executive Board of HERE in 2000 as part of Wilhelm's larger diversity program. The initial focus of the initiative was on targeted training programs to connect African Americans to hotel employment. Las Vegas provided a successful model. A program undertaken in the early 2000s by the Culinary Training Academy, a labor-management partnership that provides training for the hospitality industry, and the hotel union (Unite Here Local 226) had some success in bringing more African Americans into the hospitality industry by using high schools to recruit students, particularly young African American men, into training programs at the Culinary Training Academy.Footnote ⁷ Efforts replicating the Las Vegas model moved forward in Boston and Los Angeles. In Boston, an entire floor of the union hall, outfitted with a training kitchen and hotel rooms, was dedicated to providing training in partnership with hotel management.

In each city targeted by the African American Hiring Initiative, community outreach was key. In particular, the union recruited African-American ministers (sometimes working through the National Association for the Advancement of Colored People (NAACP) and the Urban League) to inform them about the hiring initiative, training opportunities, and the pay and benefit levels of union hospitality jobs. Laura Hausen, an international staff member who worked on the African American Hiring Initiative for several years, relayed in a 2011 interview that much of her outreach work involved educating community members about the quality of union hotel jobs. “Hotel jobs are not viewed as good jobs in the Black community … so we needed to explain that these jobs could be good jobs.” But community leaders were receptive and eager to move forward. This union staffer described finding “a tremendous hunger for these kinds of programs [in the Black community].” In an effort to further institutionalize and build upon these early efforts, local union leaders in a number of cities committed to fight for the inclusion of diversity language in their next contracts, all up for negotiation between 2005 and 2007. Such a step would help to further the union's efforts to attract and retain African American workers by formalizing the participatory efforts of hotel management as well as signaling a heightened level of commitment to its members. In addition to Chicago, the Boston and Los Angeles locals were successful in winning some form of diversity language in their 2006 contracts.

While the contracts are not identical, each secures a commitment to diversity from management and outlines affirmative steps centered on community outreach and targeted training. All contracts stipulate that a strategic program must be developed; they differ, however, in specifying how and by whom. The Chicago contract states simply that the union and the employer, “through the mutual commitment of time and resources” (Section 16.2.b), will design an annual strategic plan together. By contrast, the Boston contract specifies the creation of a “Citywide Diversity in the Hospitality Industry Taskforce” made up of hotel, union, and community representatives, the latter appointed by the mayor.Footnote ⁸ An Ombudsman is designated to facilitate the “consensus building process” of the Taskforce (Article 48, Section B.2) and is charged with additional tasks ranging from discussing “complaints by employment applicants about the Employer's hiring practices or decisions” to coordinating the analysis of hiring, promotion, and recruitment data provided by the hotels (Article 48, Section 7). The contract charges the taskforce to meet at least quarterly.

The Boston contract reveals the union's grasp of the complexity of the problem of increasing the employment of underrepresented groups, such as African Americans, as well as the complexity of labor market processes in general. Although the Boston contract pays heed to the supply side of the labor market—how to recruit and train eligible applicants—it goes the furthest of all contracts in addressing the most critical moment in shaping the racial and ethnic division of labor: hiring. Although the final hiring decision remains the right and privilege of the employer, the Boston contract's diversity provisions give union and community representatives a voice in the hiring process. As members of the taskforce, union and community representatives are able to “review and make recommendations to the Hotels regarding suggested amendments to the application and hiring procedures that may present obstacles to members of the African-American and broader diverse community members” (Article 48.5.b). While not guaranteed, this provision gives union and community representatives the opportunity to influence hiring procedures for the purpose of increasing diversity.

Perhaps most significantly, this mandate increases transparency around hiring practices and procedures. In order to comply with the “review and recommendation” mandate of the taskforce, employers must identify, regularize, and report their procedures. Doing so helps to formalize the hiring process. Additionally, the task force institutionalizes a monitoring process, further reinforcing formalization. Past research indicates that transparency and formalization greatly reduce discriminatory practices (intentional or not) and are of the greatest benefit to African American applicants (Dobbin Reference Dobbin2009; Holzer and Neumark, Reference Holzer and Neumark2006). While the advisory nature of the taskforce protects employers' autonomy in hiring, its mere existence can potentially influence hiring procedures in ways that are beneficial to underrepresented applicants by simply opening up hiring practices and procedures to review.

In addition to the citywide task force, the Boston contract establishes a Hotel Diversity Committee at each hotel, with equal representation by labor and management, “to assess the success of the Hotel's hiring, promotion, and recruitment practices when compared with the benchmarks and recommendations of the Citywide Taskforce” (Article 48, Section 6.f). These local hotel committees locate monitoring closest to the point of hire, at specific worksites, while the taskforce serves to set benchmarks for the industry as a whole and coordinates outreach across the entire local labor market. Such a strategy guards against unevenness and laggard employers, even as it generates efficiencies of scale and mutuality—no one employer is expected to undertake community outreach or facility training on its own. But each hotel must attend to its own hiring practices. The Hotel Diversity Committee provides the union with a means to monitor these local practices, and the contract secures a pledge from each hotel “to negotiate in good faith with the union to correct any failures to follow the recommendations and benchmarks described by the Citywide Taskforce” (Article 48, Section 6.f). As research has unequivocally demonstrated, labor and employment violations are greatly reduced when workers have a voice in labor-management relations (Fine and Gordon, Reference Fine and Gordon2010). Workers, when enfranchised within a “backward-mapping” implementation process, provide invaluable on-the-ground knowledge necessary for successful monitoring and problem solving (Barenberg Reference Barenberg, Bercusson and Estlund2008; Elmore 1979–Reference Elmore1980). Through the creation of these Hotel Diversity Committees, the Boston diversity language goes furthest in institutionalizing worker participation in the hiring process.

Notably, the Boston contract formalizes the participation of the city and the community in addressing the issue of employment diversity, especially the hiring and retention of African American workers. Institutionalizing the involvement of multiple stakeholders serves both political and pragmatic ends. Bringing outsiders into the issue expands the scope of conflict in an effort to build widespread support for the union's diversity goals that stretches beyond the shop floor (Schattschneider Reference Schattschneider1960). City and community participation helps to connect both the union and hotel management to a diverse set of resources and outreach venues. For example, the contract charges the taskforce to “work with existing Community job development and training programs that will assist Employers in identifying potential job applicants” (Article 48.5.c). Lastly, involving community representatives at the start helps to solidify “buy-in” through mutuality. Community outreach, then, becomes a bilateral process with the community rather than one driven at the community. Further, the requirement of community participation externalizes recruitment and outreach to actors most likely to identify underrepresented applicants.

Differences in the specific content of the diversity language among the Chicago, Boston, and Los Angeles contracts reflects differences in union density and negotiating power, past practices, and urban political regimes. Although we do not take these differences up as the analytical question of interest in this paper, we point to differences in contract language regarding training between Boston and Chicago as a further illustration of locally contingent outcomes. In Boston, the union had already garnered management's participation in a labor-management training program—the Greater Boston Hotel Employees/Local 26 Education/Training Program—that the contract stipulates should be utilized “to further the efforts of the Taskforce” (Article 48.5.d). This training program also receives workforce development funding.

By contrast, no such training program exists in Chicago. Yet the Chicago local did win training language in its 2006 contract that serves as “placeholder language,” as one Chicago union official described (Interview, Amy Hall, May 2011). The steps outlined are dependent upon a commitment from the Mayor's workforce development office to support a hospitality-industry training program. Section 67, “Training Fund,” of the contract stipulates union and hotel participation in the creation and funding of such a program:

Although Chicago Mayor Richard M. Daley once toured the Culinary Training Academy in Las Vegas and proclaimed it a model program, he never moved forward with plans to create a similar program in Chicago through a joint city-labor-management partnership.

Far from symbolic, the diversity language—even that which serves as placeholder language—included in these Unite Here contracts was hard-won. Negotiations were tense in a number of cities. The Boston Globe (2006) reported that “contract negotiations in Boston erupted in a heated exchange” (p. A12) on the issue of African American representation, specifically over the numbers of African Americans employed in the city's Starwood-operated hotels. As one union official told us, Chicago employers were initially uncomfortable with the diversity demands, but all agreed to the idea of diversity in principle (Interview, Amy Hall, May 2011). In Boston, as reported in the Boston Globe (2006), the hotel industry negotiator publicly stated that the Boston hotels “‘are prepared to do whatever outreach is necessary’” (p. A12). At root, the diversity language makes hotel employers formally accountable to carry out this necessary outreach.

Unite Here's contract diversity language represents a significant way in which workers, beyond the confines of their own social networks, can shape and influence the racial and ethnic division of labor. By institutionalizing outreach efforts and the monitoring and review of hiring practices, the diversity provisions provide workers with some formal influence on hiring. Even without these provisions, workers can do much to shape the applicant pool and affect application and hiring procedures through word-of-mouth recruitment, coaching of acquaintances through the application process, and vouching for certain applicants to the boss (Waldinger and Lichter, Reference Waldinger and Lichter2003). Yet Unite Here's contract requirements formalize these processes in an effort to lessen information asymmetries that can exclude particular groups from gaining entry into employment. Significantly, the diversity language attempts to open the black box of hiring to workers through their union representation. In the case of Boston, the community also gains inclusion in the process. Ultimately, the contract diversity language provides workers with a formal accountability mechanism. While workers cannot demand specific hires, they can demand that employers demonstrate good-faith efforts toward achieving diversity. Such leverage differs strikingly from workers' informal influence on the composition of the workforce achieved through their recruitment networks and other forms of social capital. This influence derives from workers' political capital, achieved through their explicit political mobilization as union members.

Recognizing Unite Here's contract diversity provisions as an outcome of politics reveals job competition as a socially contingent and politically mediated process. As a democratic organization, Unite Here must manage the multiple, seemingly competing, demands of its demographically diverse membership—one composed of significant numbers of African Americans and immigrants. A Local 1 official put it this way, “The question for the union is how do we defend one another's causes. How do we bridge the gap between these two [African American and immigrant] groups of workers? The diversity language was one way to demonstrate commitment to both” (Interview, Amy Hall, May 2011). In order to pursue protections for its immigrant members, the union needed to secure provisions for its African American workforce, and vice versa. Under the rubric of “diversity,” the union sought to build common cause between two workforces with highly group-specific demands. This linkage of immigration and African American concerns requires constant negotiation and mobilization, but it reveals that competition is not a given among different groups dependent upon the same economic resources, such as jobs. It is politics, however, that makes the difference.