2-Severin


2013 COLUMBIA JOURNAL OF RACE AND LAW 215 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

WE BUILT THIS CITY: THE LEGALITY 

OF COMMUNITY BENEFIT 

AGREEMENTS FOR BIG BOX 

CONSTRUCTION UNDER TITLE VII AND 

THE EQUAL PROTECTION CLAUSE 

CHADÉ SEVERIN* 

Community groups have begun to employ Community Benefit Agreements (CBAs) to combat the legacy 
of discrimination and segregation in the construction industry. The U.S. Government as well as state 
and city governments have implemented various plans since the 1960s to try to eradicate discrimination 
and segregation with a varied pattern of success. In light of the Supreme Court’s decisions in Adarand 
v. Pena and City of Richmond v. J.A. Croson, it is nearly impossible for a governmental entity to 
impose a hiring quota or percentage for minority groups. Therefore, without state intervention, the best 
way for communities to secure adequate opportunities may be to negotiate with the developers directly. 
This is where CBAs come in. CBAs, however, can take a couple of forms: private-private between 
private community groups and private developers or private-public between a private developer and some 
state actor. CBAs in each category may also have either geography-based hiring criteria, or race-based. 
This Note examines seven CBAs that fit under these different categories and measures them against the 
Supreme Court’s Title VII and Equal Protection jurisprudence. This Note ultimately seeks to suggest 
strategies for making CBAs successful against possible legal challenges. 

 

I.  INTRODUCTION…………………………………………………………… 217 

II.  BACKGROUND……………………………………………………………… 218 

A. Unemployment for Minorities in Construction and Service Sectors. 218 

B. Past Legislative and Judicial Efforts to End Discrimination………. 218 

1. Philadelphia Plan…………………………………………….. 218 

2. Hometown Approach………………………………………... 220 

                                                
* J.D. 2013, Columbia Law School. 



216 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

3. Title VII Lawsuits……………………………………………. 220 

C. Changing Tides: Croson  and Adarand……………………………… 221 

III.  A MODERN HOMETOWN APPROACH WITH LEGAL TEETH: 
 COMMUNITY BENEFIT AGREEMENTS……………………………… 224 

A. Private-Private Agreements with First Source Offices, Targeted  
Hiring and Quotas Based on Geography and Income……………... 225 

1.  Bayview CBA…………………………………………………. 226 

2. Staples CBA…………………………………………………... 227 

B. Private-Public with Hiring Quotas Based on Geography………….. 228 

1. Cherokee-Gates CBA………………………………………… 228 

2. LAX Project CBA…………………………………………….. 229 

C. Private-Private and Private-Public with Hiring Based on Race or  
Gender………………………………………………………………… 229 

1. Atlantic Yards………………………………………………… 229 

2. Milwaukee Park East Redevelopment Area………………… 230 

3. Christina Avenue Composting Facility CBA……………….. 231 

IV.  POTENTIAL LEGAL CHALLENGES TO CBAS AND HOW TO  
AVOID THESE CHALLENGES…………………………………………… 232 

A. Title VII Challenges…………………………………………………. 232 

1. What are Affirmative Action Plans and do CBAs Meet  
Their Criteria?........................................................................  232 

2. Voluntarily Negotiated Affirmative Action Plans…………... 233 

3. Disparate Impact Liability…………………………………… 237 

B. Constitutional Challenges to Race-Based CBAs Under the  
Equal Protection Clause……………………………………………… 239 

1. Government as Signer of or Party to the Agreement……….. 239 

2. State Action Doctrine………………………………………… 239 



2013 COLUMBIA JOURNAL OF RACE AND LAW 217 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

3. Analysis of the State Action Doctrine as Applied to Atlantic  
Yards………………………………………………………….. 244 

4. Would a Race-Based CBA Withstand Strict Scrutiny?.......... 246 

V.  CONCLUSION………………………………………………………………. 249 

 

I.  INTRODUCTION 

As of December 2011, African American unemployment was twice that of white Americans.1 
Even before the Great Recession, African American unemployment rates were typically twice that of 
white Americans.2 This disparity cannot be traced to one or even a few main causes. But the 
construction industry in particular has erected some of the most structural barriers to employment for 
minorities. Within the industry, unions, although generally in decline throughout the country, remain a 
crucial source of training, hiring, and bargaining power between contractors and developers. But the 
unions often become exclusive clubs, impenetrable to minorities seeking construction work. 
Furthermore, legislation aimed at reducing the employment disparity has pitted governments and courts, 
acting on behalf of minorities, against the unions and further hardened their intransigence. 

In addition to these employment barriers, minorities face increasing economic inequality within 
their own communities. Urban gentrification can increase the polarization between low-wage and high-
wage jobs within their communities.3 Current scholarship in urban politics and development shows that 
developers are essential to growth and economic success in cities, but also that urban development has 
exacerbated racial, economic, and geographic inequities.4 To resolve this tension, some community 
groups have joined together to create Community Benefits Agreements (“CBAs”)—legally binding 
agreements between a coalition of community-based organizations and developers or governmental 
bodies. The community groups pledge support for the development in return for, inter alia, priority in 
jobs, low-income housing, and living wages.5  

CBAs are relatively new and show incredible promise. However, there is little legal scholarship 
on their legal strengths and weaknesses, especially in light of historical efforts to reduce discrimination 
                                                

1 Statistics are seasonally adjusted. See Table A-2. Employment status of the civilian population by race, sex, and age, 
BUREAU OF LABOR STATISTICS, http://www.bls.gov/news.release/empsit.t02.htm (last visited Mar. 31, 2013). 

2 Christian E. Weller & Jaryn Fields, The Black and White Labor Gap in America, CTR. FOR AMERICAN PROGRESS 

(July 25, 2011), http://www.americanprogress.org/issues/labor/report/2011/07/25/9992/the-black-and-white-labor-
gap-in-america.  

3 Virginia Parks & Dorian Warren, The Politics and Practice of Economic Justice: Community Benefits Agreements as a 
Tactic of the New Accountable Development Movement, 17 J. CMTY. PRACTICE 88, 89 (2009). For the purposes of this Note, I 
use “urban gentrification” to mean an influx of capital into a neighborhood that can improve the quality of residential 
and commercial areas of that neighborhood, but also displace low-income residents that previously resided there. For a 
more comprehensive overview of gentrification, see Maureen Kennedy & Paul Leonard, Dealing with Neighborhood Change: 
A Primer on Gentrification and Policy Choices, THE BROOKINGS INSTITUTE (Apr. 2001), 
http://www.brookings.edu/~/media/research/files/reports/2001/4/metropolitanpolicy/gentrification. 

4 Parks & Warren, supra note 3, at 91.  
5 Id. at 89.  



218 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

against minorities in construction and development.6 This Note will examine some current CBAs in the 
context of the legal history of integration within urban minority communities, particularly within the 
construction trade. Part II will outline historical efforts to integrate the construction industry, starting 
with the Philadelphia Plan in 1969 and ending with the Supreme Court decisions of City of Richmond v. 
J.A. Croson and Adarand Construction v. Pena in 1988 and 1995, respectively. Part III will examine the 
employment provisions of a few CBAs primarily negotiated by groups affiliated with the Partnerships for 
Working Families.7 Part IV will then analyze potential legal challenges to the CBAs and how groups can 
best structure these agreements to withstand or avoid such challenges. 

II.  BACKGROUND 

A. Unemployment for Minorities in Construction and Service Sectors 

Although the construction industry is a particularly salient example of employment 
discrimination against African Americans, it certainly is not the only industry in which such 
discrimination has occurred. CBAs include provisions for permanent hiring in retail and service sector 
positions within the new project area. Discrimination in the service sector, although less historically 
documented and analyzed, is nevertheless real and also relevant to CBAs.8 This Note uses the 
construction industry to contextualize the history and evolution of Title VII and Equal Protection 
analysis, but that analysis extends to other industries as well.  

B. Past Legislative and Judicial Efforts to End Discrimination  

1. Philadelphia Plan 

After President Johnson signed the Civil Rights Act of 1964, the Office of Federal Contract 
Compliance (OFCC) led the effort to reduce discrimination among private companies with government 
contracts. The construction industry was one of the hardest to regulate because the unions wielded so 

                                                
6 There are, however, many articles analyzing different aspects of CBAs. See, e.g., Vicki Been, Community Benefits 

Agreements: A New Local Government Tool or Another Variation on the Exactions Theme?, 77 U. CHI. L. REV. 5 (2010) 
(examining the benefits and drawbacks of using CBAs in the land use approval process); Michael L. Nadler, The 
Constitutionality of Community Benefits Agreements: Addressing the Exactions Problem, 43 URB. LAW. 587 (2011) (analyzing the 
potential of CBAs to violate the Takings Clause); Christine A Fazio & Judith Wallace, Legal and Policy Issues Related to 
Community Benefits Agreements, 21 FORDHAM ENVTL. L. REV. 543 (2010) (broadly analyzing the benefits and drawbacks to 
CBAs, particularly fairness to communities, developers, and taxpayers and the question of whether developers can 
negotiate concessions outside of local law); Patricia E. Salkin & Amy Lavine, Community Benefits Agreements and 
Comprehensive Planning: Balancing Community Empowerment and the Police Power, 18 J.L. & POL’Y 157 (2009) (analyzing the 
effects of CBAs on comprehensive municipal urban planning); Debra Bechtel, Forming Entities to Negotiate Community 
Benefits Agreements, 17 WTR J. AFFORDABLE HOUSING & COMMUNITY DEV. L. 145 (2008) (discussing options for 
creating corporate entities to negotiate CBAs to increase the enforceability and credibility of CBAs). 

7 The CBAs negotiated by groups within the Partnership for Working Families represent only a percentage of 
the total CBAs in the United States, but they are typically the most successful. Parks & Warren refer to these as “strong” 

CBAs, as compared to “weak” CBAs negotiated primarily in New York and New Jersey. Parks & Warren, supra note 3, 
at 91–92.  

8 For a discussion of racial discrimination in the service sector, see generally Devah Pager, Bruce Western, & 
David Pedulla, Employment Discrimination and the Changing Landscape of Low-Wage Labor Markets, 1 U. CHI. LEGAL F. 317 
(2009).  



2013 COLUMBIA JOURNAL OF RACE AND LAW 219 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

much power.9 Each union determined which employees would work on particular projects, and 
selections were usually based on union membership and seniority.10 The unions also limited their 
membership to ensure that all members had sufficient work opportunities and had strict qualification 
requirements that usually precluded African Americans from joining in the first place.11 Without 
sufficient apprenticeship or journeyman experience, African Americans could neither get into unions nor 
obtain sufficient seniority to be contracted out for lucrative construction work. Because of these 
practices, the OFCC sought to enforce antidiscrimination in construction contracts through 
experimental “special area plans” in St. Louis, San Francisco, Cleveland, and Philadelphia throughout 
1966 and 1967.12 The St. Louis and San Francisco plans loosely required contractors to work with unions 
to develop and provide equal opportunities for minority groups in employment and in journeyman and 
apprenticeship programs.13 Though the contractors drafted detailed statements of their plans to 
cooperate with these requirements, the programs never showed actual improvement in minority 
placement rates.14 

After the failures of the St. Louis and San Francisco plans, a group of regional federal officials 
developed the Philadelphia Plan in November 1967. The Plan created a system to review unions’ 
affirmative action programs before the government awarded them contracts. The Plan also established a 
numbers-based model for these programs by compiling basic information on racial population ratios, the 
construction workforce, sources of minority recruitment, and the expected amount of construction in a 
particular area.15 However, the Comptroller General struck down this holistic approach for being too 
vague. He stated that the Plan was defective because the basis for approval or disapproval of the 
affirmative action plans was unclear.16 The Nixon administration revised the Plan in 1969 by giving 

                                                
9 JOHN DAVID SKRENTNY, THE IRONIES OF AFFIRMATIVE ACTION 136 (1996).  
10 Id.  
11 Damon Stetson, Negro Groups Step Up Militancy in Drive to Join Building Unions, N.Y. TIMES, Aug. 28, 1969, at 27 

available at http://select.nytimes.com/gst/abstract.html?res=FA0717FE3B551B7B93CAAB1783D85F4D8685F9. 
Stetson also mentions that when civil rights groups demanded that the building trade unions hire more African 
Americans as journeymen, a union official said, “Would you want your house wired by an amateur?” Id. See also Peter 
Millones, Labor; Building Trades Say, ‘Enter, Negroes,’ N.Y. TIMES, Feb. 18, 1968, at E4, available at 
http://select.nytimes.com/gst/abstract.html?res=F10817FC3F5E1A7B93CAA81789D85F4C8685F9 (describing an 
example of discrimination when a group of African American youths in New York scored extremely high in a test for a 
sheet metal apprenticeship program, but the union nevertheless tossed out their scores and delayed their entry into the 

program); SKRENTNY, supra note 9, at 197–98 (discussing the building trades’ contestation of the allegations of 
discrimination and quoting renowned labor union leader George Meany’s assertion that “the Building Trades [are] being 
singled out as being . . . the last bastion of discrimination. . . . [T]his is an amazing statement, when you figure how small 
participation of Negroes . . . is in, for instance, the banks in this country, the press, on the payroll for newspapers and 
communications media”); David R. Jones, U.S. Aides Will Discuss Bias With Officials of Building Trades, N.Y. TIMES, June 18, 
1967, at 32, available at http://select.nytimes.com/gst/abstract.html?res= 
FB0915F73B5B1A7B93CAA8178DD85F438685F9 (regardless of whether the discrimination was intentional, the 
government sought to increase minority membership in building trade unions first through dialogue and eventually 
through government-mandated action).  

12 SKRENTNY, supra note 9, at 136.  
13 Id. at 136–37.  
14 Id. 
15 Id. at 137.  
16 Id. at 138.  



220 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

contractors specific percentage targets for minority employees.17 This revision sought simultaneously to 
remedy the shortcomings of the ambiguous 1967 Plan and to avoid the argument that the “targets” were 
quotas, which Title VII forbade.18 The Solicitor of Labor even tried to justify the targets by analogizing 
the Plan to permissible school integration plans that also had numerical goals.19 By revising the Plan, 
Nixon emphasized that he specifically intended to target the discriminatory practices of the building 
trade unions, stating, “[W]e cannot have construction unions which deny the right of all Americans to 
have those positions.  America needs more construction workers, and . . . all Americans are entitled to an 
equal right to be a member of a union.”20 Despite vicious opposition to the Plan from the building trade 
unions and other contractors, the Third Circuit upheld the validity of the Plan. The Supreme Court 
denied certiorari, implicitly sanctioning the Third Circuit’s holding.21 

2. Hometown Approach 

Before officially revising the Philadelphia Plan, Nixon also sanctioned another plan, called the 
“hometown approach,” to integrate the construction trade and improve minority recruitment.22 Under 
the hometown approach, the parties voluntarily negotiated agreements with each other. For example, 
through private negotiation, the unions willingly accepted the same racial quotas they had rallied against 
under the Philadelphia Plan. This approach produced more actual integration than the Philadelphia Plan 
because employers had to accept minority apprentices into the unions as “trainees.”23 In addition, the 
hometown approach made sure that minorities were direct signatories on these agreements, which 
increased the likelihood of successful implementation.   

3. Title VII Lawsuits  

                                                
17 The plan required that bidders on any federal or federally assisted construction contract for projects 

exceeding $500,000 must submit an acceptable affirmative action program with specific goals for employing minority 
employees in specific skilled crafts. The Executive Order applied to the five-county Philadelphia area. The Department 
of Labor held public hearings in Philadelphia and determined various employment ranges for minority workers, 
including five to nine percent for ironworkers in 1970, eleven to fifteen percent in 1971, sixteen to twenty percent in 
1972, and twenty-two to twenty-six percent in 1973. Contractors Ass’n of E. Pa. v. Sec’y of Labor, 442 F.2d 159, 163–64 
(3d Cir. 1971) cert. denied, 404 U.S. 854 (1971). 

18 Id. at 172–73. Specifically, if a contractor failed to hire within the range of targeted percentage of minority 
employees, it would have to demonstrate “good faith” that it had tried as hard as possible to achieve that range. If the 
contractor failed to convince the OFCC of its good faith, the contract could be canceled or denied. Id. For a more in-
depth discussion of the political debate surrounding the Philadelphia Plan, see SKRENTNY, supra note 9, at 177–221.  

19 SKRENTNY, supra note 9, at 195.  
20 Id. at 197.  
21 Contractors Ass’n, 442 F.2d at 159. 
22 William B. Gould, The Seattle Building Trades Order: The First Comprehensive Relief Against Employment 

Discrimination in the Construction Industry, 26 STAN. L. REV. 773, 778 (1974). 
23 WILLIAM B. GOULD, BLACK WORKERS IN WHITE UNIONS 302 (1977). Becoming a trainee would place 

minorities on a track to get the coveted journeyman’s card, which would give them relative economic security in the area 
of the union’s jurisdiction. However, unions could satisfy the Plan by hiring non-union minority workers, thereby failing 
to allow minorities to get footholds into seniority within the unions. See also Paul Good, The Bricks and Mortar of Racism, 
N.Y. TIMES, May 12, 1972, at SM24, available at http://select.nytimes.com/gst/abstract.html?res= 
FA0C15FA3F5F117B93C3AB178ED85F468785F9 (quoting Curtis Alexander, a black with a journeyman or union card 
that he obtained because of federally funded job training, who said that without such a card, “[Y]ou get no respect.”).  



2013 COLUMBIA JOURNAL OF RACE AND LAW 221 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

In addition to the Philadelphia Plan and the hometown approach, changes within the courts 
directly affected integration within the unions. First, Federal Rule of Civil Procedure 23 was amended in 
1966, making it easier for lawyers to represent large classes of individual minority construction and trade 
workers.24 This development encouraged civil rights groups to file many more Title VII lawsuits against 
the unions’ discriminatory practices. By 1974, the unions had suffered multiple defeats in federal courts 
and began to fear that more lawsuits would threaten their solvency.25 Fear of future defeats led the 
unions to sign consent decrees with various companies and the federal government to provide black 
workers financial compensation or institute affirmative action programs.26 The increasing number of 
Title VII cases also required more federal judges; the appellate bench increased by forty-three percent 
during the 1960s and thirty-six percent during the 1970s.27 Finally, federal courts—including the 
Supreme Court—expansively interpreted the provisions of Title VII to support affirmative action and 
dismantle seniority systems, job requirements, or entrance examinations ostensibly targeted at 
entrenching the segregated status quo.28   

The Philadelphia Plan, hometown approaches, and Title VII lawsuits were powerful foes to the 
forces of union discrimination. However, their momentum was relatively short-lived. By the 1970s, 
conservatives began to combat all aspects of the civil rights groups’ successes, with powerful results.29  

C. Changing Tides: Croson  and Adarand 

The 1980s and 1990s brought a more conservative judicial bench that significantly constrained 
the integration victories of the prior two decades. The Supreme Court decided to hear two cases—City of 

                                                
24 Class-action litigation represented only a few dozen cases in 1965 and more than a thousand cases a decade 

later. PAUL FRYMER, BLACK AND BLUE: AFRICAN AMERICANS, THE LABOR MOVEMENT AND THE DECLINE OF THE 

DEMOCRATIC PARTY 85 (2008).  
25 Id. at 70–71.  
26 Specifically, the United Steelworkers of America signed a consent decree with nine steel companies and the 

federal government to pay 55,000 black workers who had filed a class action suit against the union. Id. Peter Schoemann, 
president of the Building Trades in 1968, gave a speech supporting affirmative action programs, but only out of fear of 
pattern-or-practice Title VII suits. “‘[W]e carried the fight just about as far as we could.’ But to avoid further lawsuits, 
‘the building trades need a single policy in this area.’” Id. at 70. Another cause of the expansion of consent decrees was a 
change in how the courts used Rule 53, which provided for special masters. Courts let the masters play a more expansive 
role in enforcing consent agreements between civil rights groups and unions. The special masters directly supervised and 
reported on unions’ affirmative action programs and enabled courts to invoke fines against unions that did not comply. 
Through this program, judges effectively replaced the EEOC and DOL as agencies overseeing enforcement of 
integration, which Frymer argues often went beyond legislative intent. Id. at 85–86.  

27 Id. at 86. The Federal Magistrates Act of 1968 allowed federal judges to appoint magistrates whenever 
necessary to help them with their caseload.  

28 Id. at 87–88. See, e.g., Griggs v. Duke Power Co., 401 U.S. 424 (1971); United Steelworkers of Am. AFL-CIO-
CLC v. Weber, 443 U.S. 193 (1979); Contractors Ass’n of E. Pa. v. Sec’y of Labor 442 F.2d 159 (3d Cir. 1971); Quarles 
v. Philip Morris, 279 F. Supp. 505 (E.D. Va. 1968). 

29 FRYMER, supra note 24, at 94–95. Conservatives started their own advocacy-driven law firms, targeted 
changes to the Federal Rules of Civil Procedure, and encouraged Congress to pay more attention to and stay active in 
the process of legal rule making. Congress also restricted attorney fee opportunities for lawyers bringing Title VII 
lawsuits to reduce incentives to bring many cases with large damages. Id. at 94–95.  



222 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Richmond v. J.A. Croson and Adarand Constructors, Inc. v. Pena30—in which statutes required general 
contractors to hire socially disadvantaged subcontractors for a specific monetary percentage of each total 
awarded contract. In Croson, the city of Richmond passed a local plan that required any general 
contractor that received a construction contract from the city to subcontract at least thirty percent of a 
contract’s value to qualifying minority business enterprises.31 The Supreme Court held that the program 
was unconstitutional under the Equal Protection Clause because it awarded those contracts on the basis 
of race without showing a compelling state interest in doing so.32  Although Croson applied only to state 
and local programs,33 Adarand required federal programs that considered race to undergo strict scrutiny 
as well.34 

These decisions had a significant effect on state, local, and federal efforts to reduce 
discrimination in government contracting. By 2009, sixteen state and federal courts had applied strict 
scrutiny to affirmative action programs.35 Twelve had invalidated programs under the Equal Protection 

                                                
30 City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989); Adarand Constructors, Inc. v. Pena, 515 U.S. 200 

(1995).  
31 Croson, 488 U.S. at 469. Richmond’s Plan defined minority group members to be “Blacks, Spanish-speaking, 

Orientals, Indians, Eskimos or Aleuts.” Croson, 488 U.S. at 477–78. Richmond’s Minority Business Utilization Plan was 
similar to the federal Public Works Employment Act of 1977, which was the first federal act to require at least ten 
percent of every federal construction grant to be expended for “minority business enterprises.” Pub. L. No. 95-28, Tit. I, 
§ 103, 91 Stat. 116 (1977) (codified as amended at 42 U.S.C. § 6705 (2006)).  

32 Croson, 488 U.S. at 490–92. The Supreme Court evaluates whether state or federal legislation violates the 
Equal Protection Clause by using different levels of judicial scrutiny.  

At a minimum, a statutory classification must be rationally related to a legitimate government purpose. 
Classifications based on race or national origin, and classifications affecting fundamental rights, are 
given [strict] scrutiny. Between these extremes of rational basis review and strict scrutiny lies a level of 
intermediate scrutiny, which generally has been applied to discriminatory classifications based on sex 
or illegitimacy. 

Clark v. Jeter, 486 U.S. 456, 461 (1988) (internal citations omitted). Before the Court decided Adarand and Croson, it was 
implicitly understood that strict scrutiny only applied to legislation that discriminated on the basis of race, not to legislation 
that aimed to help minorities (also known as “benign racial discrimination”). Adarand and Croson clarified that any 
classification on the basis of race is cause to apply strict scrutiny. See Mary J. Reyburn, Strict Scrutiny Across the Board: The 
Effect of Adarand Constructors, Inc. v. Pena on Race-Based Affirmative Action Programs, 45 CATH. U. L. REV. 1413, 1416 

(1996). For a law to survive under strict scrutiny, the legislating government must show the legislation fulfills a 
“compelling governmental interest,” and the legislation is “narrowly tailored” to that interest. Gratz v. Bollinger, 539 
U.S. 244, 270 (2003).  

33 Croson, 488 U.S. at 490 (“That Congress may identify and redress the effects of society-wide discrimination 
does not mean that, a fortiori, the States and their political subdivisions are free to decide that such remedies are 
appropriate.”). 

34 Adarand, 515 U.S. at 200. Adarand analyzed section 106(c) of the Surface Transportation and Uniform 
Relocation Assistance Act of 1987 as well as the Small Business Act of 1953. The former offered states financial 
assistance with highway construction if the state gave ten percent of the amount to small businesses owned and 
controlled by socially and economically disadvantaged individuals, which was usually defined in terms of race. Lynn 
Ridegeway Zehrt, A Decade Later: Adarand and Croson and the Status of Minority Preferences in Government Contracting, 21 
NAT’L BLACK L.J. 1, 5 (2009).  

35 Zehrt, supra note 34, at 2.  



2013 COLUMBIA JOURNAL OF RACE AND LAW 223 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Clause. Of the four that validated such programs, three focused primarily on federal affirmative action 
statutes in the highway construction industry, not on state or local development.36   

The foregoing history demonstrates a clear pattern since at least the 1960s: integration in the 
construction industry has been cyclical, with booms and busts. The late 1960s saw tremendous 
momentum within the civil rights community and political establishment to integrate the construction 
unions. However, by the 1970s and 80s, political support had waned and those opposed to integration 
had obtained greater political strength. This change led in turn to a decline in enforcement of affirmative 
action programs and a slide back to pre-Plan segregation. A more conservative judicial bench in the last 
twenty-five years has constrained the scope of state and federally mandated affirmative action programs 
with cases like Croson and Adarand. And current labor advocates must contend with big-box retailers like 
Wal-Mart, Target, and Costco. These retailers have a history of discriminating against minorities, offering 
low wages, imposing bad working conditions, and being staunchly anti-union.37 This state of affairs 
leaves a mystifying path for African American construction workers to follow: should they set their 
efforts toward integrating trade unions, which would increase their general bargaining power and wages 
for a larger number of jobs, or should they abstain from joining a union in the hope of obtaining 
nonunion work with large, anti-union employers?38 And, in light of Supreme Court jurisprudence on the 
legality of affirmative action programs both under Title VII and the Equal Protection Clause, what role 
should federal, state, or local governments play in this process?   

The next Section examines how community groups have developed CBAs as an innovative 
response to four major shifts in labor activism and urban development. First, union membership and 
political presence has declined precipitously in the last fifty years.39 Second, a dramatic rise in the service 
sector (as compared to manufacturing) has increased the proportion of non-exportable jobs in urban 
areas. This rise should increase the bargaining power of labor groups and employees; however, a third 
major shift, the simultaneous rise of anti-union big-box retailers has served to undermine rather than 

                                                
36 Id. The only case that validated a program was in Denver. I will explore these cases more in Part IV, infra.  
37 FRYMER, supra note 24, at 1–2. See also Dan Frosch, Immigrants Claim Wal-Mart Fired Them to Provide Jobs for 

Local Residents, N.Y. TIMES (Feb. 8, 2010), http://www.nytimes.com/2010/02/09/us/09walmart.html; Reuters, Wal-
Mart Settles Lawsuit on Hiring, N.Y. TIMES (Feb. 20, 2009), 
http://www.nytimes.com/2009/02/21/business/21walmart.html (Wal-Mart settles lawsuit claiming it discriminated 
against African-Americans in hiring for $17.5 million); Target Corp. to Pay $500,000 for Race Discrimination, EEOC (Dec. 10, 
2007), http://eeoc.gov/eeoc/newsroom/release/12-10-07a.cfm; Target Corp. to Pay $775,000 for Racial Harassment, EEOC 
(Jan. 26, 2007), http://www.eeoc.gov/eeoc/newsroom/release/1-26-07.cfm; Steven Greenhouse, Trying to Overcome 
Embarrassment, Labor Opens a Drive to Organize Wal-Mart, N.Y. TIMES (Nov. 8, 2002), 
http://www.nytimes.com/2002/11/08/us/trying-to-overcome-embarrassment-labor-opens-a-drive-to-organize-wal-
mart.html (stating that not one of Wal-Mart’s one million workers is a union member and that the retailer has crushed 
the only successful effort to organize a group of butchers that lasted a mere two weeks). 

38 Nor are these approaches necessarily mutually exclusive. For example, labor unions and local politicians in 
Chicago got Wal-Mart to agree that in order to allow its stores into the city, they would have to be union-built. Stephanie 
Clifford, Wal-Mart Gains in Its Wooing of Chicago, N.Y. TIMES (June 24, 2010), 
http://www.nytimes.com/2010/06/25/business/25walmart.html. Wal-Mart reached the same deal with the Building 
and Construction Trades Council union in New York City, even though the project has yet to be approved. Elizabeth A. 
Harris, Wal-Mart Skips Council Hearing as Impact of Stores is Assailed, N.Y. TIMES (Feb. 3, 2011), 
http://www.nytimes.com/2011/02/04/nyregion/04walmart.html.  

39 Scott L. Cummings, Law in the Labor Movement’s Challenge to Wal-Mart: A Case Study of the Inglewood Site Fight, 95 
CAL. L. REV. 1927, 1933 (2007). 



224 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

encourage labor activism. Finally, an increasingly conservative federal judiciary and a large number of 
judicial vacancies40 have made litigation a less-tenable option. As a result, CBAs have become a hopeful 
and often successful shift for minorities from targeting union discrimination to targeting non-exportable 
industries and large retailers who seek to build in local areas.41   

III.  A MODERN HOMETOWN APPROACH WITH LEGAL TEETH: COMMUNITY 
BENEFIT AGREEMENTS 

CBAs are contracts negotiated between prospective private developers and individuals or groups 
representing the affected community.42 They can also be contracts between private entities and the 
State.43 By negotiating CBAs, community groups gain a developer’s promises to directly benefit the local 
community through means including affordable housing, employment provisions, public space, 
monitoring provisions, and potential remedies for breach.44 The developer gains the community’s 
support for the development, which in turn helps avoid costly delays or cancellation of the project.45 

Often the most important and politically salient features of CBAs are their employment-related 
provisions, which promise to provide job access and job quality.46 Many CBAs have provisions for 
“targeted hiring programs,” which require that employers in a development promise to hire certain 
individuals through some combination of a tiered priority system, a “first source” office, or local job 
training programs.47 The individuals that usually benefit from the program are those harmed by the 

                                                
40 As of March 31, 2013, there were eighty-four vacancies out of 874 total federal judgeships, representing 

almost ten percent of judgeships. Federal Judgeships, U.S. COURTS, 
http://www.uscourts.gov/JudgesAndJudgeships/FederalJudgeships.aspx (last visited Mar. 31, 2013); Judicial Vacancies, 
U.S. COURTS, http://www.uscourts.gov/JudgesAndJudgeships/JudicialVacancies.aspx (last visited Mar. 31, 2013). 
Increased vacancies mean it is harder to get employment cases into and through the already-backlogged courts. 

41 For a more thorough discussion of how localism is an effective alternative labor strategy to twentieth century 
federal union efforts, see Cummings, supra note 39, at 1942–51. 

42 JULIAN GROSS, GREG LEROY & MADELINE JANIS-APARICIO, COMMUNITY BENEFITS AGREEMENTS: 
MAKING DEVELOPMENT PROJECTS ACCOUNTABLE 9–10 (2002) (Good Jobs First & The California Partnership for 
Working Families 2005), available at http://www.goodjobsfirst.org/sites/default/files/docs/pdf/cba2005final.pdf. Julian 
Gross, the director of the Community Benefits Law Center, also created a precise definition of a CBA: “A CBA is a 
legally binding contract (or set of related contracts), setting forth a range of community benefits regarding a 
development project, and resulting from substantial community involvement.” Julian Gross, Community Benefits 
Agreements: Definitions, Values, and Legal Enforceability, 17 J. AFFORDABLE HOUSING 35, 37 (2008), available at 
http://www.forworkingfamilies.org/sites/pwf/files/publications/CBAs_Definitions_Gross_2008.pdf.  

43 See discussion infra Part III.B. 
44 Barbara Bezdek, Putting Community Equity in Community Development: Resident Equity Participation in Urban 

Redevelopment, in LAW, PROPERTY AND SOCIETY: AFFORDABLE HOUSING AND PUBLIC-PRIVATE PARTNERSHIPS 93, 109 

(Robin Paul Malloy & Nestor M. Davison eds., 2009).  
45 Community Benefits Agreements, GOOD JOBS NEW YORK, http://www.goodjobsny.org/resources-

tools/community-benefits-agreements (last visited May 3, 2013).  
46 Parks & Warren, supra note 3, at 92. Job access means both providing residents relevant training and getting 

residents into jobs, while job quality means that those jobs provide a decent, or even “living wage.”  
47 Gross, supra note 42, at 43. A “first source” office receives notice of job openings from employers, maintains 

contact with a variety of job training organizations to access their applicant pools, and refers qualified workers to 
employers. A first source office can benefit job training organizations and targeted individuals by giving them reliable 
access to information about job openings. It can help the targeted hiring program meet its goals. Id. at 46. For a 

 



2013 COLUMBIA JOURNAL OF RACE AND LAW 225 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

development in the first place: those whose jobs or homes are displaced by the development and 
residents of the neighborhoods around the development. But these programs often also target residents 
of low-income neighborhoods within the entire metropolitan area or individuals referred by community 
job training organizations.48 

CBAs create provisions to provide these individuals with jobs in a variety of ways. Some 
implement referral and hiring processes. These include requiring employers to target prioritized 
individuals on their own by giving notice of job openings through certain channels or in certain 
geographic areas, interviewing only priority candidates for a limited period of time after they notify those 
candidates of openings, interviewing only people referred by designated sources, or meeting percentages 
of priority candidates hired in order to be in compliance with the program.49   

The next few Sections consider the employment provisions of some existing CBAs to highlight a 
few of the forms they can take. The CBAs chosen for this Note met certain criteria.  First, the Note 
examines both historically significant or publicly controversial CBAs. The Staples CBA is an example of 
the former and the Atlantic Yards CBA the latter. Second are CBAs that appeared to have more direct 
involvement by local, state, or federal governments and are therefore more easily subject to 
constitutional challenges. The Cherokee-Gates CBA and LAX CBA met these criteria. Finally are CBAs 
with clear hiring quotas for minority employees that implicate cases involving affirmative action plans. 
The Atlantic Yards and San Francisco CBAs are examples of this type of CBA. The CBAs discussed in 
this Note are neither wholly illustrative of the many kinds of CBAs in the country nor are they meant to 
indicate “important” or “better” CBAs. Rather, their provisions are potential signposts for legal analyses 
of whether different combinations of factors are more or less permissible under Title VII and the 
Fourteenth Amendment.50 

A. Private-Private Agreements with First Source Offices, Targeted Hiring, 
and Quotas Based on Geography and Income 

CBAs between two private actors with hiring practices based on geography or income could be 
subject to disparate impact challenges under Title VII.51 If attributable to state action, however, such 

                                                                                                                                                       
description of a tiered priority system, see First Source Hiring Policy, Section IV: Responsibilities of First Source Referral System, 
PARTNERSHIP FOR WORKING FAMILIES, 
http://www.forworkingfamilies.org/sites/pwf/files/documents/StaplesCBA.pdf. 

48 These organizations are often called first source offices. Gross, supra note 42, at 43.  
49 Id. at 45–46. 
50 For an introduction to the broad range of issues facing CBAs, see Community Benefits, PARTNERSHIP FOR 

WORKING FAMILIES, http://www.forworkingfamilies.org/campaigns/CBA (last visited May 3, 2013); Amy Lavine, 
National Survey on CBAs, COMMUNITY BENEFITS AGREEMENTS (Aug. 12, 2010), 
http://communitybenefits.blogspot.com/2010/08/national-survey-on-cbas.html. 

51 Suits for disparate impact do not require proof of intentional discrimination, but rather proof that an 
employment policy—whether intentional or unintentional—had a disparate impact on a particular minority group.  

Title VII of the Civil Rights Act of 1964 . . . prohibits employment discrimination on the basis of race, 
color, religion, sex, or national origin. Title VII prohibits both intentional discrimination (known as 
‘disparate treatment’) as well as, in some cases, practices that are not intended to discriminate but in 
fact have a disproportionately adverse effect on minorities (known as ‘disparate impact’).  

Ricci v. DeStefano, 557 U.S. 557, 577 (2009). 



226 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

challenges would also receive an equal protection analysis.52 If a potential plaintiff has a case for both 
disparate impact, and it involves some state action, disparate impact alone will not bring a CBA within a 
strict scrutiny analysis without a discriminatory motive.53 Rather, the claimant would have to show that 
the relevant actors had a discriminatory purpose regarding him personally54 or that, given proof of a 
disparate impact, the relevant actors maintained the policy because of, not in spite of, the disparate 
impact.55 In short, it generally would be easier to prove a violation of Title VII’s disparate impact 
provisions than to prove a constitutional violation under the Equal Protection Clause. These factors will 
receive deeper analysis in Part IV. 

1.  Bayview CBA 

In May 2008, a coalition of three San Francisco community groups entered into a CBA with 
Lennar, a national housing developer.56 Under the agreement, Lennar promised to provide low-income 
housing, housing assistance funds, and specific hiring goals under a “first source hiring program.”57 The 
articulated purpose of the hiring program was to “facilitate the employment of targeted job applicants by 
employers in the project . . . through a non-exclusive referral system.”58 The agreement defines 
“employer” as a “non-governmental business or nonprofit corporation that conducts any portion of its 
operations in the Project Site with at least eight (8) regular full time equivalent employees.”59 The 
definition includes contractors but not construction contractors.60 The agreement defines a “targeted job 

                                                
52 See discussion infra Part IV.B. 
53 See, e.g., McClesky v. Kemp, 481 U.S. 279 (1987) (holding that the state of Georgia’s sentencing to death of 

more black defendants and killers of white victims than white defendants and killers of black victims did not manifest a 
discriminatory purpose in violation of the equal protection clause). See also Washington v. Davis, 426 U.S. 229, 246–48 
(1976). 

Under Title VII, Congress provided that when hiring and promotion practices disqualifying 
substantially disproportionate numbers of blacks are challenged, discriminatory purpose need not be 
proved[.] . . . However this process proceeds, it involves a more probing judicial review of, and less 
deference to, the seemingly reasonable acts of administrators and executives than is appropriate under 
the Constitution[.] . . . We are not disposed to adopt this more rigorous standard for the purposes of 
applying the Fifth and the Fourteenth Amendments in cases such as this. 

Id. 
54 McCleskey, 481 U.S. at 292. 
55 Id. at 298–99. (Claimant would have to prove that the “Georgia Legislature enacted or maintained the death 

penalty statute because of an anticipated racially discriminatory effect . . . [or] because of the racially disproportionate 
impact”) (emphasis in original). 

56 The three San Francisco groups were the San Francisco Labor Council, ACORN, and the San Francisco 
Organizing Project (SFOP). San Francisco Labor Council Pens Affordable Housing Agreement with Lennar, 
FOGCITYJOURNAL.COM (May 16, 2008), http://www.fogcityjournal.com/wordpress/455/san-francisco-labor-council-
pens-affordable-housing-agreement-with-lennar.  

57 See Core Community Benefits Agreement: Hunters Point Shipyard/Candlestick Point Integrated Development Project, 
PARTNERSHIP FOR WORKING FAMILIES (May 30, 2008), 
http://www.forworkingfamilies.org/sites/pwf/files/documents/BayviewHuntersPointCBA.pdf.  

58 Id. at 26. 
59 Id. 
60 Id. The exclusion of construction contractors is usually due to the way the industry is structured, which has 

led to low-bid contracting. The Partnership for Working Families seeks to remedy these discrepancies in the 

 



2013 COLUMBIA JOURNAL OF RACE AND LAW 227 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

applicant” as an individual referred to any employer by a “first source referral system,” which is the 
agency designated to implement the first source hiring program.61 The CBA sets three tiers of targeted 
job applicants that an employer must prioritize above other applicants: first, individuals whose residence 
or place of employment will be displaced because of the project; second, low- and moderate-income 
individuals living in the area of the project; and third, low- and moderate-income individuals living in zip 
codes within the city.62 

The agreement defines its goal for covered jobs for a six-month period to be fifty percent 
targeted job applicants for entry-level jobs.63 If an employer meets this goal, it will be considered in 
compliance with the first source hiring program.64 But an employer will also be in compliance with the 
Program if it has observed the plan’s other provisions even if it has not met the goal.65  

2. Staples CBA 

The Staples Center CBA, though not officially the first CBA, is typically considered the 
preeminent and groundbreaking CBA negotiated between community groups and a large private 
developer. In May 2001, a coalition of over thirty labor groups negotiated a comprehensive CBA 
covering the area surrounding the Staples Center in Los Angeles.66 This CBA pioneered the first source 
hiring program to benefit “targeted job applicants” as well as employers by providing a pool of qualified 
job applicants. Many other California community group coalitions modeled their CBAs directly on the 
Staples example.67 

                                                                                                                                                       
construction industry in addition to its other CBA work. See Construction Career Opportunities Project, THE PARTNERSHIP 

FOR WORKING FAMILIES, https://communitybenefits.rdsecure.org/section.php?id=168 (last visited Mar. 31, 2013). 
61 Id. at 27. 
62 Id. The agreement also requires a specific process for covered employers to hire targeted job applicants. First, 

at least six months before initial hiring, each employer must notify the referral system of the best estimate of the 
approximate number and type of jobs that will need to be filled. As new jobs arise, an employer must also notify the 
referral system of available job openings and the skills and qualifications required. For three weeks after notifying the 
referral system of job opportunities, the employer may only hire qualified targeted job applicants as long as operations 
on the project site have not yet started. Once operations have started, the employer can only hire targeted job applicants 
for five days after notifying the referral system of job opportunities. When both exclusive time periods are over, the 
employer must still make good-faith efforts to hire targeted job applicants, but may also hire any applicant recruited or 
referred from any source. Id. at 28.  

63 Id. at 27. 
64 Id. at 28. 
65 Id. 
66 See Staples CBA, supra note 47. 
67 The Bayview CBA was partially based on the Staples CBA. See also North Hollywood Mixed-Use Redevelopment 

Project Community Benefits Program , PARTNERSHIP FOR WORKING FAMILIES (Nov. 2001), 
http://www.forworkingfamilies.org/sites/pwf/files/documents/NoHoCBA.pdf; Marlton Square Redevelopment Project 
Developer Community Benefits Program, PARTNERSHIP FOR WORKING FAMILIES, 
http://www.forworkingfamilies.org/sites/pwf/files/documents/cba_marltonsquare.pdf (last visited May 3, 2013); and 
Hollywood & Vine Mixed-Use Development Project Community Benefits Agreement Gatehouse Hollywood Development, L.P., 
PARTNERSHIP FOR WORKING FAMILIES, https://communitybenefits.rdsecure.org/downloads/CBA%20Gatehouse%20 
FINAL%205-7-04.pdf (last visited May 3, 2013).  



228 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

In the Staples CBA, like the Bayview CBA, targeted job applicants are also separated into a tiered 
priority system with three levels with a more specific stratification of local geography: the first priority 
goes to individuals whose place of residence or employment has been displaced by the Staples Center 
project and to low-income individuals living within a half-mile radius of the project.68 The second 
priority goes to low-income individuals living within a three-mile radius of the project.69 The third 
priority goes to low-income individuals living in any area throughout Los Angeles.70 An employer is also 
required to use the first source referral system when hiring for any jobs located within the project area.71  

B. Private-Public with Hiring Quotas Based on Geography 

As opposed to completely private CBAs, in private-public CBAs the action is already attributable 
to the state as a direct participant and signatory of the CBA. Therefore, legally, both the Equal 
Protection Clause and Title VII will apply. The level of scrutiny under equal protection analysis will be 
subject the previous caveats: proof of a disparate impact maintained precisely because of its impact, or a 
discriminatory implementation with respect to one or more individuals. Without this showing, the CBAs 
would receive a rational basis analysis. 

1. Cherokee-Gates CBA 

The Cherokee-Gates Project in Denver is similar to many other CBAs, providing affordable 
housing and first source hiring, but it also includes an unprecedented agreement to pay prevailing wages 
to every construction worker who engages in the publicly funded construction of site infrastructure and 
maintenance of public spaces and facilities.72 The first source local hiring program promises to maximize 
both job opportunities for disadvantaged residents and the outcome of public investments.73 The biggest 
difference between the Cherokee-Gates CBA and other CBAs is the amount of public funding used for 
the project. The Front Range Economic Strategy Center (FRESC) led a coalition of community groups 
and union organizations to negotiate for $126 million in subsidies from the city of Denver to the 
developer of the project, Cherokee Denver, LLC.74 In return, Cherokee Denver agreed to a variety of 
conditions, including good wages for construction workers and first source hiring for nearby residents.75 
In this case, negotiation between the community groups and the developer proved difficult and often 

                                                
68 Staples CBA, supra note 47, at A-16.  
69 Id. 
70 Id. 
71 Id. 
72 Community Benefits Achievements at the Cherokee-Gates Project, THE CAMPAIGN FOR RESPONSIBLE DEVELOPMENT, 

http://www.forworkingfamilies.org/sites/pwf/files/documents/CherokeeGates.pdf (last visited May 3, 2013) 
[hereinafter Community Benefits Achievements].  

73 First Source Local Hiring: Overview & History of Denver’s First Source Policy, THE FRONT RANGE ECONOMIC 

STRATEGY CENTER, http://www.fresc.org/downloads/First%20Source%20Local%20Hiring%20-
%20Overview%20and%20Denver%20History.pdf (last visited May 3, 2013). 

74 Tory Read, The Gates Cherokee Redevelopment Project: “A huge step forward for low-income people in Denver,” THE 

FRONT RANGE ECONOMIC STRATEGY CENTER (2006), 
http://www.fresc.org/downloads/Gates%20Report%20MCD.pdf (last visited May 3, 2013). The total subsidy included 
$85 million in tax increment financing and $41 million in metro district financing subsidies. The $126 million 
represented about thirteen percent of the total project budget of $1 billion. Id. at 15. 

75 Id. at 2–3. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 229 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

frustrating, so the city stepped in to incentivize Cherokee Denver to agree to the groups’ conditions in 
return for subsidies.76 However, even though the city provided the crucial link, it was not a signatory to 
the agreement. Therefore, including the Gates CBA under the private-public heading is slightly 
misleading. This CBA will not necessarily qualify as state action, but the enormity of the subsidies and 
the city’s involvement in the CBA necessitate a searching legal inquiry into whether this amount of 
governmental support would make this agreement appear more private-public than private-private.77  

2. LAX Project CBA 

The 2004 Los Angeles Airport CBA is the largest CBA to date, encompassing an $11 billion 
modernization plan.78 The CBA was signed by the LAX Coalition for Economic, Environmental and 
Educational Justice and the Los Angeles World Airports (LAWA), the governmental entity that operates 
the airport.79 The employment provisions of the CBA look almost identical to those of the Staples 
Center CBA, the predominant difference being the size and scope of the project. The targeted job 
applicants in the LAX project are divided into only two tiers: first, low-income individuals living in the 
“Project Impact Area”—an area defined by the project’s environmental impact report—and, second, 
low-income individuals residing in the entire city of Los Angeles.80  Unlike the Staples and Bayview 
CBAs, however, the LAX CBA does not include a goal of targeted job applicants to be hired for LAWA 
to be in compliance with the agreement. Rather, LAWA must only hire targeted applicants for a set 
period of time when making initial hires for the commencement of operations and when making hires 
after the commencement of operations.81 Here, the CBA definitely involves state action, but the hiring 
criteria are facially neutral without an obvious discriminatory purpose.  Therefore, this CBA is at most 
subject to a Title VII disparate impact attack based on the demographics of Los Angeles. 

C. Private-Private and Private-Public with Hiring Based on Race or Gender 

Finally, some private-private or private-public CBAs have racial or gender-based quotas. As 
opposed to the geographically-based hiring goals of other CBAs, these CBAs impose affirmative action 
goals on the basis of race and gender. They are, therefore, subject to a disparate treatment analysis under 
Title VII, but may fall within the permissible scope of voluntary affirmative action plans first analyzed in 
United Steelworkers v. Weber.82 The CBAs in this category that are attributable to state action will most 
likely be facially discriminatory and subject to strict scrutiny. Again, they will receive deeper analysis in 
Part IV. 

1. Atlantic Yards 

                                                
76 Id. at 21–23. 
77 See discussion infra Part IV.B. 
78 John M. Broder, Los Angeles Groups Agree to Airport Growth, for a Price, N.Y. TIMES (Dec. 17, 2004), 

http://www.nytimes.com/2004/12/17/national/17lax.html. 
79 Policy & Tools: Community Benefits Agreements & Policies in Effect, PARTNERSHIP FOR WORKING FAMILIES, 

http://www.forworkingfamilies.org/page/policy-tools-community-benefits-agreements-and-policies-effect#cbas (last 
visited Mar. 29, 2013).  

80 Exhibit C to Community Benefits Agreement: First Source Hiring Program for Airport Employers, PARTNERSHIP FOR 

WORKING FAMILIES (2004), http://www.forworkingfamilies.org/sites/pwf/files/documents/LAXFirstSourceFinal12-6-
04.pdf. 

81 Id. at 3. 
82 See United Steelworkers of Am. AFL-CIO-CLC v. Weber, 443 U.S. 193 (1979).  



230 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Negotiated in 2004 and 2005, the Atlantic Yards CBA in Brooklyn, New York concerns the 
Atlantic Yards development, including the new Barclays Center stadium, housing, and retail space.83 
However, due to extensive controversy and lack of adequate enforcement, Atlantic Yards has become 
the paradigm example of a “weak CBA.”84 Despite being modeled on the successful Staples Center 
project and negotiated by prominent Brooklyn and minority community groups, the CBA has failed to 
deliver some of its key promises to minority construction workers.85 Nevertheless, a lawsuit filed by 
seven construction workers illustrates how specific promises within an enforceable CBA can enable 
putative beneficiaries to sue for breach.86  

The employment provisions of Atlantic Yards have simultaneous goals. First, they provide that 
developers would use good faith efforts to employ or cause to be employed at least thirty-five percent 
minority and ten percent women construction workers during construction of the arena and the project 
as a whole.87 Second, the CBA envisions that enrollment priority for the broader employment provisions 
and first source hiring program will extend first to residents of city-owned apartments and second to 
low-income members of the neighboring community, and then will expand outward both geographically 
and to individuals of higher income.88 The Atlantic Yards CBA thereby creates a combination of quotas 
for minority and female employment, particularly during construction, as well as facially neutral 
geographic and income-based requirements.89 

The most directly challengeable aspects of the Atlantic Yards CBA would likely be the quota 
requirements for minority and female hiring. However, the geographic requirements could also be 
challenged if they imposed a disparate impact on non-minority residents seeking jobs in the large 
development. 

2. Milwaukee Park East Redevelopment Area 

                                                
83Community Benefits Agreement, BENEFITING BVHP, 

http://www.benefitingbvhp.org/docs/docs/cba/ATLANTIC_YARDS_CBA.pdf (last visited May 3, 2013) [hereinafter 
Atlantic Yards CBA]. 

84 See supra note 7.  
85 John Marzulli & Erin Durkin, Promise of union jobs a lie by Atlantic Yards, suit by construction workers charge, NY 

DAILY NEWS (Nov. 15, 2011), http://www.nydailynews.com/new-york/promise-union-jobs-a-lie-atlantic-yards-suit-
construction-workers-charge-article-1.977604. Members of the coalition of community groups include, inter alia, the New 
York State Association of Minority Contractors, Brooklyn United for Innovative Local Development, and Downtown 
Brooklyn Neighborhood Alliance. Atlantic Yards CBA, supra note 83, at 1. 

86 Enforcement of CBAs is beyond the scope of this Note, but will remain a challenge for beneficiaries, 
especially among the “weak CBAs.” Parks & Warren, supra note 3, at 92 (“[T]hree core features separate strong from 
weak CBAs. A broad scope, high level of transparency, and explicit and robust monitoring and enforcement 
mechanisms are the core characteristics of strong CBAs.”). 

87 Atlantic Yards CBA, supra note 83, at 13.  
88 Id. at 4, 12.  
89 However, census data from Kings County shows that the majority of Brooklyn residents are non-white. State 

& County QuickFacts, U.S. CENSUS BUREAU, http://quickfacts.census.gov/qfd/states/36/36047.html (last visited Mar. 
27, 2013). The areas immediately surrounding the development, such as Clinton Hill and Fort Greene, are sixty-four 
percent minority, though many minorities are leaving the area. Tamy Cozier & Annesofie Brochstedt, Census 2010: A 
Dramatic Decline in Black Residents, FORT GREENE/CLINTON HILL NEWS (May 23, 2011), http://fort-
greene.thelocal.nytimes.com/2011/05/23/census-2010-a-dramatic-decline-in-black-residents/#more-57249. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 231 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

The Milwaukee Park East Redevelopment Compact (PERC) in Milwaukee, Wisconsin was 
implemented in 2005 as part of the redevelopment of county land in downtown Milwaukee. The CBA 
required that during construction of the redevelopment, at least twenty-five percent of jobs would be in 
“Disadvantaged Business Enterprises/Minority Business Enterprises” (DBEs/MBEs)—defined as 
businesses that are at least fifty-one percent minority-owned and controlled—and at least twenty-five 
percent of employees would be minorities.90 However, the sixteen-acre project faced a significant 
obstacle when the city could not find a developer, so local community groups rallied around the Board 
of Supervisors and obtained the first CBA passed by legislation instead of negotiations with a private 
developer.91 This CBA therefore includes both state action and racial quotas, automatically requiring a 
strict scrutiny analysis.  

3. Christina Avenue Composting Facility CBA 

Finally, Peninsula Compost Company, a private developer, and members of a coalition of South 
Wilmington, Delaware community groups signed the Christina Avenue Composting Facility CBA in 
2007. The CBA maintained quotas by setting a goal that twenty percent of the subcontracted 
construction work on the Project was to be performed by DBEs. The CBA defines “Minority” as a 
person with origins in any of the Black African or Caribbean, Hispanic, Native American or Alaskan 
native, or Asian and Pacific Islander racial groups, and any other individual deemed to be 
disadvantaged.92 It also aims to have twenty percent of the labor employed during the construction phase 
to be performed by community residents.93 Paradoxically, the CBA acknowledges that the goal of twenty 
percent subcontracted work done by DBEs has never been achieved by any large urban development 
project in the community.94 Nevertheless, the CBA imposes quotas similar to those in Adarand by 
mandating that subcontracting work go to firms predominantly owned or controlled by minorities. 

                                                
90 Community Benefits Agreement Outline, Park East Redevelopment Area, INSTITUTE FOR WISCONSIN’S FUTURE, 

http://www.wisconsinsfuture.org/past_projects/econdev/gjln_agreement.htm (last visited May 3, 2013); Minority 
Business Certification Program, STATE OF WISCONSIN – DEPARTMENT OF ADMINISTRATION, 
http://www.doa.state.wi.us/section.asp?linkid=225&locid=171 (last visited May 3, 2013). 

91 The CBA was passed by the Milwaukee Board of Supervisors on Dec. 16, 2004. See Park East Redevelopment 
Compact (PERC), PARTNERSHIP FOR WORKING FAMILIES (Dec. 16, 2004), 
http://www.forworkingfamilies.org/sites/pwf/files/documents/PERC_0.pdf; Amy Lavine, Milwaukee Park East 
Redevelopment CBA, COMMUNITY BENEFITS AGREEMENTS (Jan. 30, 2008), 
http://communitybenefits.blogspot.com/2010/08/national-survey-on-cbas.html. The PERC states that the “Milwaukee 
County Board and the community asks and expects businesses and contractors to make a good faith effort to employ 
racial minorities consistent with their numbers in the County’s workforce[.] (The 2000 county census population . . . was 
68.7% White, 20.4% Black, 7.2% Hispanic and 3.7% other[.]).” Note, however, that the CBA on the Institute for 
Wisconsin’s Future website shows an employment requirement of twenty-five percent minority. Community Benefits 
Agreement Outline, Park East Redevelopment Area, INSTITUTE FOR WISCONSIN’S FUTURE, 
http://www.wisconsinsfuture.org/past_projects/econdev/gjln_agreement.htm (last visited May 3, 2013). 

92 Christina Avenue Composting Facility Community Benefits Agreement, NEW YORK CITY COMPTROLLER, 3–4, 
http://www.comptroller.nyc.gov/bureaus/opm/pba/pdf/CommunityBenefitsAgree-PCC-So-Wilmington-Coalition.pdf 
(last visited May 3, 2013) [hereinafter Christina Avenue CBA].  

93 Id. at 5. As of the 2010 census, the population of the relevant communities covered by the CBA was 67.4% 
non-white in Wilmington, DE. State & County QuickFacts, U.S. CENSUS BUREAU, 
http://quickfacts.census.gov/qfd/states/10/1077580.html (last visited Mar. 27, 2013). 

94 Christina Avenue CBA, supra note 92, at 4. 



232 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Naturally, these three generalized buckets of CBA structures still vary widely from place to place. 
The circumstances of Title VII or constitutional challenges to their provisions will also vary widely and 
ultimately depend on the facts unique to each case. However, given the general trends that have 
emerged, these groupings help to outline the possible scope of CBA challenges and their ultimate 
viability under Title VII and the Fourteenth Amendment. 

IV.  POTENTIAL LEGAL CHALLENGES TO CBAS AND HOW TO AVOID THESE 
CHALLENGES 

A. Title VII Challenges 

In the private sector, employers occasionally manipulate the gender or racial balance of their 
employees in favor of minority applicants when neutrally selecting from a qualified applicant pool would 
otherwise produce a different balance. These manipulations have a few judicially relevant manifestations. 
The first are affirmative action plans voluntarily negotiated between an employer and a union or group 
of employees. Under such a plan, the employer agrees with the union to hire or train a certain quota or 
percentage of minority applicants.95 In the second manifestation, an employer adopts a facially neutral 
hiring or promotion standard, then discovers that it may have a disparate impact on a minority 
population. For fear of disparate impact liability, the employer decides to invalidate or abandon the 
standard, to the detriment of those who met it.96 Whenever an employer makes an “adverse employment 
decision” on the basis of race, even if intended to avoid potential disparate impact liability, the employer 
violates Title VII.97 The Supreme Court has held that neither approach is per se invalid, but both are 
subject to narrow limitations that could affect the structure and implementation of CBAs, particularly 
those with racial quotas.98 

1. What are Affirmative Action Plans and do CBAs Meet Their 
Criteria? 

                                                
95 See United Steelworkers of Am. AFL-CIO-CLC v. Weber, 443 U.S. 193 (1979). The employer and the union 

bargained for a plan that reserved fifty percent of the openings in an in-plant craft training program to African 
Americans until the percentage of African Americans in the program was commensurate with the percentage in the local 
labor market. Id. at 198–99. 

96 See Ricci v. DeStefano, 557 U.S. 557 (2009). The city of New Haven tried to fill vacant lieutenant and captain 
jobs in its fire department by employing a promotional examination. Seven out of nine candidates eligible for the captain 
position based on exam performance were white; the other two were Hispanic. The department threw out the tests for 
fear of being found liable for adopting a practice with a disparate impact on minority firefighters. Id. at 557, 562. 

97 Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e–2(a), (j) (2006). These claims fall under a 
disparate treatment analysis because they make distinctions on the basis of race. Affirmative actions that are facially 
neutral but have the effect of creating more minority applicants or employees than exist in the relevant labor market 
would be subject to a disparate impact analysis. See also Ricci, 557 U.S. at 578–79 (city’s refusal to certify the test results 
violated disparate treatment provision of Title VII absent some strong basis in evidence that the city would be subject to 
disparate impact liability had they retained the test); McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273 (1976) (Title 
VII prohibits discrimination against whites as well as non-whites).  

98 In Ricci, the Supreme Court applied the rule that the city was required to demonstrate with a “strong basis in 
evidence” that it would be subject to disparate impact liability if it failed to take a race-conscious action. Under the 
disparate impact test, the Court held that the city officials lacked a strong basis in evidence both for the proposition that 
the tests were not job related and consistent with business necessity and for the belief that there existed an equally valid, 
less-discriminatory alternative to using examinations that adequately met the city’s needs. Ricci, 557 U.S. at 584–93.  



2013 COLUMBIA JOURNAL OF RACE AND LAW 233 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Most cases take for granted that the employer plan at issue is an affirmative action plan.  
However, the Second Circuit in United States v. Brennan recently sought to define the threshold issue of 
what actually constitutes an affirmative action plan.99 The court determined that in order to be an 
affirmative action plan, the employer action must benefit all members of a protected class and cannot be 
individualized.100 In other words, “when an employer, acting ex ante, although in the light of past 
discrimination, establishes hiring or promotion procedures designed to promote equal opportunity and 
eradicate future discrimination, that may constitute an affirmative action plan.”101  However, when an 
employer with established procedures changes those procedures ex post because of the racial composition 
of the results, the individualized remedy (also known as “make-whole relief” intended to remedy the 
effects of discrimination) does not warrant the affirmative action defense.102 Whether or not the plan is 
legally characterized as an affirmative action plan affects the relevant Supreme Court analysis. A 
voluntary affirmative action plan can be permissible under Weber and its progeny. However, an 
employer’s affirmative action remedy requires a different analysis under Ricci when the employer can 
justify that remedy only with a “strong basis in evidence” that the particular employment action caused a 
disparate impact in the first instance.103  

When considered with Brennan’s distinction in mind, all the CBAs discussed in this Note appear 
to create ex ante plans, in light of past discrimination, to promote equal opportunity and eliminate future 
discrimination. CBAs are designed to go into effect before the developer breaks ground on a new 
project, which means that the plans do not seek to remedy any current practice that may have had a 
discriminatory effect on individuals who previously applied for jobs with the developer. However, it is 
also debatable whether the CBAs could be affirmative action plans, given that some do not directly 
extend to “members of a racial or gender class.”104 Many CBAs, particularly those in San Francisco and 
Los Angeles, will have the effect of targeting minority candidates because the relevant geographic areas 
in the plans have majority minority populations.105 But because the criteria themselves are facially 
neutral, the CBAs might not be characterized as affirmative action plans. However, characterizing them 
as facially neutral is ultimately more beneficial to such agreements in the face of potential legal 
challenges. Therefore, assuming that potential plaintiffs will try to characterize them as affirmative action 
plans, the next two sections analyze CBAs in that light. 

2. Voluntarily Negotiated Affirmative Action Plans 

                                                
99 United States v. Brennan, 650 F.3d 65, 99–104 (2d Cir. 2011). 
100 Id. at 99. 
101 Id. at 102. 
102 Id. at 102–04. The court is directly referring to the situation in Ricci where the employer implemented a test 

as a prerequisite for promotions then threw out the test when it feared it would have a disparate impact on minority 
applicants. The action was individualized, the Second Circuit said, “for what it did, in essence, was to give promotion—
or at least another chance at promotion—to the individual black firefighters who had taken the test, at the expense of 
those . . . who would have [otherwise been eligible for promotion].” Id. at 102. 

103 Ricci, 557 U.S. at 563.  
104 Brennan, 650 F.3d at 104. 
105 See, e.g., William H. Frey, The New Metro Minority Map: Regional Shifts in Hispanics, Asians, and Blacks from Census 

2010, THE BROOKINGS INSTITUTE (Aug. 2011), 
http://www.brookings.edu/~/media/research/files/papers/2011/8/31%20census%20race%20frey/0831_census_race
_frey.pdf.  



234 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

The affirmative action plan in United Steelworkers v. Weber arose at the same time as the 
Philadelphia Plan, hometown approaches, and the wave of Title VII litigation in the late 1960s and early 
1970s.106 The steelworkers union and the employer negotiated the affirmative action plan to apply to the 
employer’s fifteen national plants. White workers at the employer’s Louisiana plant then sued for reverse 
discrimination. Before the employer negotiated and implemented the plan only 1.83% of the employees 
at the Louisiana plant were black, even though the local workforce was approximately thirty-nine percent 
black.107 Both the district court and a divided Fifth Circuit held for the plaintiffs and granted a 
permanent injunction prohibiting the plan on the grounds that it violated Title VII’s prohibition against 
race-based decisions in employment.108 The Supreme Court reversed and held that Title VII permits 
some voluntary race-conscious affirmative action plans in the private workplace. However, the Court 
also explicitly stated that its decision was based on the case’s narrow facts, including the fact that the 
union and private employer voluntarily agreed to the bona fide affirmative action plan.109 Although Weber 
held that some affirmative action plans are permissible, the Court did not give clear guidance as to which 
plans would be permissible and which would not.110 Nevertheless, the Court tried to be instructive to 
future plans by recognizing three factors important to the Weber plan. First, despite its narrow holding, 
the Court gave judicial notice to plans designed to remedy past discrimination or obvious racial 
imbalances in traditionally segregated job categories. Second, the plan did not “trammel the interests of 
the white employees” in that it did not fire white employees in order to hire black employees, nor did it 
serve as an “absolute bar” to white advancement because “half of those trained in the program would be 
white.”111 Third, the plan was a temporary measure intended to eradicate the racial imbalance rather than 
maintain a racial balance between blacks and whites.112 In Johnson v. Transportation Agency, Santa Clara 
County, the Court clarified Weber by stipulating that an employer does not need to have personally 
discriminated in the past in order to implement an affirmative action plan.113 Rather, the employer need 
only point to an obvious “imbalance in traditionally segregated job categories.”114 This ruling expanded 
Weber’s scope to include more employers in traditionally segregated industries, even if the segregation 
was unintentional. 

                                                
106 See Part II supra for this background. 
107 United Steelworkers of Am. AFL-CIO-CLC v. Weber, 443 U.S. 193, 198–99 (1979). 
108 Id. at 200. 
109 Id. at 200–01: 
The only question before us is the narrow statutory issue of whether Title VII forbids private 
employers and unions from voluntarily agreeing upon bona fide affirmative action plans that accord 
racial preferences in the manner and for the purpose provided in the Kaiser-USWA plan. That 
question was expressly left open . . . in a case not involving affirmative action, that Title VII protects 
whites as well as blacks from certain forms of racial discrimination. 

(internal citations omitted). Because the plan did not involve state action, it did not implicate the Equal Protection 
Clause of the Fourteenth Amendment. 

110 Id. at 208 (“We need not today define in detail the line of demarcation between permissible and 
impermissible affirmative action plans. It suffices to hold that the challenged Kaiser-USWA affirmative action plan falls 
on the permissible side of the line.”). 

111 Id. 
112 Id. 
113 Johnson v. Transp. Ag’y Santa Clara Cty., 480 U.S. 616, 617 (1987).  
114 Id. at 630 (quoting United Steelworkers of Am. AFL-CIO-CLC v. Weber, 443 U.S. 193, 209 (1979)). The 

requirements from Weber and Johnson are known as the “purpose, impact and duration standards.” Richard N. Appel, 
Alison L. Gray & Nilufer Loy, Affirmative Action in the Workplace: Forty Years Later, 22 HOFSTRA LAB. & EMP. L.J. 549, 564 
(2005).  



2013 COLUMBIA JOURNAL OF RACE AND LAW 235 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Significant gaps remain in the Weber and Johnson standards despite three decades of case law. 
First, it is unclear exactly what the Supreme Court would consider a “traditionally segregated” job 
category or those jobs in which past discrimination was sufficient to warrant current remedial action. 
Second, even if the Court did clearly establish what traditionally segregated job categories are, it is 
unclear what proof would be required to show whether a specific job falls into one of those categories. 
Third, it is unclear whether a non-remedial purpose could justify a voluntary affirmative action plan by a 
private employer.   

Even though the Supreme Court has never addressed whether a non-remedial purpose could 
justify a voluntary affirmative action plan, one circuit court has. In Schurr v. Resorts Intern. Hotel, Inc., the 
Third Circuit held that an affirmative action plan that was not based in any finding of historical or 
current discrimination in the casino industry or in the plaintiff’s particular job category violated Title VII 
under the Weber standard.115 In that case, a New Jersey casino director hired a qualified black male as a 
technician instead of a qualified white male because the director believed that such a decision was 
required by the casino’s affirmative action plan, which itself was required by the state of New Jersey for 
every casino licensee.116 The court observed that neither the New Jersey Casino Commission nor the 
defendant casino had designed the plans to correct a manifest imbalance in response to a job category 
that had ever been affected by segregation or in response to a finding that any relevant job category had 
been affected by segregation.117 Rather, the Casino Commission stated that the Casino Control Act was 
promulgated because the tourist area of Atlantic City had become “blighted” and when casino 
development resurged, the legislature wanted to benefit the large minority population through significant 
job creation.118 

CBAs with specific racial targets implicate both the issue of whether the affirmative action plan 
is in a “traditionally segregated industry” and whether a non-remedial purpose could ever warrant an 

                                                
115 Schurr v. Resorts Intern. Hotel, Inc., 196 F.3d 486 (3d Cir. 1999). The Third Circuit cited its decision in an 

earlier case for the rule that “unless the affirmative action plan has a remedial purpose, it cannot be said to mirror the 
purposes of the statute, and . . . cannot satisfy the first prong of [Weber].” Taxman v. Bd. of Educ., 91 F.3d 1547, 1557 
(3d Cir. 1996).  

116 Schurr, 480 U.S. at 488–90. The Casino Control Act required that every casino license holder undertake 
affirmative measures to ensure equal employment opportunities. A New Jersey regulation set specific minority and 
female percentage goals for particular job categories within the casinos and required casinos to file quarterly and annual 
reports on their affirmative actions efforts, including documentation of efforts to hire or promote a woman or minority 
to positions with a salary of $35,000 or higher. Furthermore, casino licensees would be subject to periodic hearings 
about the affirmative action plans to demonstrate compliance. Id. All of this indicates the extent of the state’s coercive 
power to enforce the affirmative action plans.  

117 Id. at 497–98. 
118 Id. at 498. The only other circuit court cases that analyzed non-remedial purposes involved public employees 

and an employer whose preferential treatment sought to promote racial diversity. See Cunico v. Pueblo Sch. Dist. No. 60, 
917 F.2d 431 (10th Cir. 1990) (court upheld a claim of reverse discrimination under Title VII because there was no prior 
discrimination in the school district, there was no statistical imbalance that would give rise to an inference of 
discrimination, and the plan was designed to maintain—not achieve—a racial balance, in contravention of the rule in 
Johnson); Taxman, 91 F.3d at 1547 (Board’s affirmative action plan preferring minority teachers over nonminority teachers 
in layoff decisions violated Title VII since it was adopted to promote racial diversity rather than to remedy past 
discrimination and it unnecessarily trammeled nonminority interests). However, these decisions predate the Supreme 
Court’s decisions in Gratz v. Bollinger, 539 U.S. 244 (2003) and Grutter v. Bollinger, 539 U.S. 306 (2003), which held 
that racial diversity in higher education—at least among the student population—can be a compelling state interest. 



236 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

affirmative action plan. Although the construction industry has a long and well-documented history of 
discrimination, most CBAs also affect jobs in the service sector, which do not necessarily have the same 
history.119 Cases like Schurr suggest, however, that a CBA could benefit from including both clear 
findings of historical segregation in the included job categories and an explicit intent to remedy 
discrimination in those categories.120 Some scholars theorize that a non-remedial purpose like racial 
diversity in the workplace could justify a voluntary affirmative action plan.121 However, if that were the 
case, the program would need to articulate how its plan is consistent with Title VII’s objectives of 
breaking down patterns of racial segregation and hierarchy and is not intended to maintain a specific 
racial balance.122 

The final remaining discrepancy after Weber is the duration of the affirmative action plan and 
how to measure whether a plan is truly “temporary” within the meaning of Weber. The idea of 
maintaining a racial balance usually comes into play after an employer has implemented an affirmative 
action plan and a minority employee has subsequently left a position. The employer will likely interview 

                                                
119 See, e.g., Gross, LeRoy & Janis-Aparicio, supra note 42, at 49 (“Collective bargaining agreements in retail, 

service, and manufacturing generally do not conflict with targeted hiring requirements”); id. at 63 (CBAs can implement 
protections similar to “worker retention” laws that “provide job security to long-term service workers when city 
contracts change hands. . . . This ensures that the service workers get to keep their jobs even when the specific hotel or 
theater operator changes.”). This is not to say the service industry has not had pervasive discrimination. See, e.g., Bonnie 
Kwon, Restaurants and Race: Discrimination and Disparity in the Food Service Sector, RESTAURANT OPPORTUNITIES CENTERS 

UNITED (Aug. 16, 2011), http://rocunited.org/restaurants-and-race-discrimination-and-disparity-in-the-food-service-
sector/; Sue Sturgis, Chick-fil-A’s history of workplace discrimination, THE INSTITUTE FOR SOUTHERN STUDIES (Aug. 2, 2011), 
http://www.southernstudies.org/2012/08/chick-fil-as-history-of-workplace-discrimination.html; Amanda Holpuch, 
Disney sued for discrimination by former employee over Muslim hijab, THE GUARDIAN (Aug. 14, 2012), 
http://www.guardian.co.uk/world/2012/aug/14/disney-sued-discrimination-muslim-employee. 

120 Although Johnson said that proof of the historical discrimination does not have to be from the actor 
imposing the affirmative action plan, Croson suggests that proof of historical discrimination should at least be localized 
and not based on national findings of discrimination in that industry. For a more in-depth discussion of Croson, see infra 
Part IV. 

121 Appel et al., supra note 114, at 570–73. Grutter and Gratz stated that there is a compelling state interest in 
promoting racial diversity in higher education. Some scholars have sought to extrapolate that argument into the 
employment context. For a more detailed analysis of these issues, see Eric A. Tilles, Lessons from Bakke: The Effect of 
Grutter on Affirmative Action in Employment, 6 U. PA. J. LAB. & EMP. L. 451 (2004) (concluding that even if Grutter may 
have laid the foundation for more expansive use of affirmative action in employment, the current paradigm used to 
examine affirmative action in private employment will likely prevent Grutter from having a large impact) and Rebecca 
Hanner White, Affirmative Action in the Workplace: The Significance of Grutter?, 92 KY. L.J. 263, 272 (2003) (stating that it 
“appears likely” that the Supreme Court would view a public employer’s need to take race into account for certain 
employment decisions as compelling). The irony of Grutter’s effect on Title VII jurisprudence is that the “Equal 
Protection Clause [imposes] fewer restraints on the government than Title VII imposes on private employers . . . 
because . . . private employer[s] would not be permitted to adopt an affirmative action plan to better community 
relations, or referring more directly back to Grutter, to sustain the ‘political and cultural heritage,’ ensure ‘[e]ffective 
participation by members of all racial and ethnic groups in . . . civic life,’ train ‘our Nation’s leaders,’ or ‘to cultivate a set 
of leaders with legitimacy.’”). Tilles at 463. White suggests differently—that Title VII will permit employers more 
flexibility than is constitutionally available. White, supra, at 275. 

122 United States v. Weber, 443 U.S. at 208. One of the stated reasons the Court upheld Weber plan was because 
the “plan [was] a temporary measure . . . not intended to maintain a racial balance, but simply to eliminate a manifest 
racial imbalance.” 



2013 COLUMBIA JOURNAL OF RACE AND LAW 237 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

both minority and nonminority candidates for the position. Failure to hire a minority candidate risks 
going against an affirmative action plan, while hiring a minority replacement could be seen as 
maintaining the new racial balance that now exists because of the affirmative action plan. 

In Sharkey v. Dixie Electricity Membership Corporation, the Fifth Circuit held that even though an 
employer had not explicitly stated when the affirmative action plan would end, it did not necessarily 
follow that the plan was intended to maintain a racial balance.123 Furthermore, the employer’s policies 
indicated that race was merely a “plus” to an otherwise qualified candidate’s application, not something 
that could transform an almost-qualified candidate into a qualified candidate.124 

For CBAs, this issue is only relevant for the permanent service-sector jobs generated by a given 
project. However, none of the aforementioned CBAs with racial quotas imposes those quotas for 
permanent jobs. Therefore, this inquiry is not entirely relevant. However, assuming arguendo that a CBA 
applies a racial quota to permanent jobs, Weber remains instructive. The Court stressed that the plan was 
aimed at eradicating a racial imbalance rather than maintaining a specific racial balance. Specifically, the 
plan’s preferential selection of black candidates would end as soon as the percentage of black workers in 
the plant approximated that of the local labor force.  

Some difficulties arise in applying this analysis to CBAs. First, in some areas surrounding a CBA 
project, the minority community is the majority. In such a situation, it is not entirely feasible to have the 
percentage of minority workers in the workforce equal that of the percentage of the minority in the 
surrounding community. If that outcome were the goal, the affirmative action program could continue in 
perpetuity, which could affect its viability under Weber. Instead, quotas like those in Milwaukee, Atlantic 
Yards, or Wilmington that applied to permanent jobs would be more realistic. However, to avoid 
challenges under the Weber rubric, those hypothetical CBAs should include a provision stating that once 
the employer reaches the quotas in the given job, it will not seek to maintain the racial balance but will 
instead evaluate further candidates on the merits, as the Fifth Circuit mentioned in Sharkey. 

3. Disparate Impact Liability 

If a plaintiff cannot prove disparate treatment on the basis of race, gender, or any other 
prohibited category, he may still make a claim for Title VII liability under the theory of disparate 
impact.125 The disparate impact standard provides that if an employer has a facially neutral hiring 
standard or policy—like a written test, for example—that has a disparate impact on a particular group, 
the Court can strike down the test as violating Title VII unless the standard is directly related to future 
job performance.126 The reasoning underlying this theory is that the disparate impact is an effective 
proxy for finding “artificial, arbitrary, and unnecessary barriers to employment when . . . [such barriers] 
operate invidiously to discriminate on the basis of [race].”127   

                                                
123 Sharkey v. Dixie Elec. Membership Corp., 262 Fed. App’x. 598, 604–08 (5th Cir. 2008).  
124 Id. at 606.  
125 This test was first articulated in Griggs v. Duke Power Co., 401 U.S. 424 (1971). 
126 Id. at 431. 
127 Id. But intent is ultimately irrelevant to the disparate impact analysis. “[A]bsence of discriminatory intent 

does not redeem employment procedures or testing mechanisms that operate as ‘built-in headwinds’ for minority groups 
and are unrelated to measuring job capability.” Id. at 432. 



238 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

In disparate impact cases, plaintiffs bear the initial burden of proving that a particular 
employment practice caused a disparate impact on the basis of race, color, religion, sex, or national 
origin.128 One way of showing an adverse impact from an employer’s hiring or promotion policies is by 
using the EEOC Guidelines’ “four-fifths” rule of thumb.129 If the selection rate for a protected group is 
less than four-fifths of the selection rate of the rest of the applicant pool, the process will be presumed 
to have an adverse impact.  

As a practical matter, cases brought under CBAs would presumably be disparate impact cases 
brought by white individuals, specifically white males, based on the disparate impact of facially neutral 
local hiring policies.130 Although whites bringing disparate impact claims may seem out of line with the 
initial policy behind the Civil Rights Act of 1964, the Supreme Court has held that disparate impact 
claims can be made by any race, not just minorities.131 A CBA that gives hiring preference to individuals 
in the area immediately surrounding a project as well as low-income individuals the larger area around 
the project could have a disparate impact on nonminority individuals who also live in the city but not in 
the targeted areas. For example, imagine a relatively segregated city of 100,000 with an eighty percent 
nonminority population and a twenty percent minority population. The smaller area immediately 
surrounding the CBA project has 20,000 people and is seventy-five percent minority, twenty-five percent 
nonminority.132 These numbers mean that the targeted applicants for the CBA project jobs would first 
be the 15,000 minority and 5,000 non-minority individuals around the project site, and then any low-
income individuals in the city, presumably also members of the larger minority population that does not 
live in the project area. In this hypothetical, the “facially-neutral” localized hiring plan would have an 
adverse impact on non-minority job seekers because of the racial demographics of the town. 

This situation is not uncommon. Many cities have areas dominated by one or a few racial groups 
that could stand to benefit greatly from these programs to the detriment of groups outside that 
immediate location. Furthermore, a localized hiring plan standing alone is not job-related or of business 
necessity, which is the legitimate defense to adverse impact claims. Therefore, CBAs that appear neutral 
need to be careful about unintentionally causing an adverse impact. CBAs like San Francisco’s Bayview 
and the Staples Center CBA are instructive because they suggest a quota of fifty percent of the targeted 
job applicants but allow employers to avoid liability for failing to meet that goal if they meet the other 

                                                
128 42 U.S.C. §2000e–2(k)(1)(A) (2006).  
129 Information on Impact, 29 C.F.R. §1607.4(D) (2013) (“A selection rate for any race, sex, or ethnic group 

which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be 
regarded by Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will 
generally not be regarded by Federal enforcement agencies as evidence of adverse impact”). However, the Supreme 
Court has noted that this standard has not provided much more than a “rule of thumb” for the courts. Watson v. Fort 
Worth Bank and Trust, 487 U.S. 977, 995 n.3 (1988).  

130 See supra notes 91 (for Brooklyn, NY), 93 (for Milwaukee, WI), and 95 (for Wilmington, DE) for evidence of 
the racial balance of certain areas where CBAs with local hiring policies are in effect.  

131 See McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273 (1976) (holding that Title VII prohibits 
discrimination in private employment to people of any race).  

132 Therefore, the city’s population is comprised of 80,000 non-minority and 20,000 minority individuals. 
However, 15,000 of these minority individuals live in the area around the development and only 5,000 non-minority 
individuals live in that area. Furthermore, of the remaining 5,000 minority individuals who do not live around the project 
area, perhaps about fifty percent of them are low-income. These low-income minority individuals will get job priority 
over the 75,000 non-minority individuals living outside the project area. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 239 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

provisions of the CBA. By providing an “out” for employers, some CBAs can ameliorate the potential 
adverse impacts of localized hiring policies. 

B. Constitutional Challenges to Race-Based CBAs Under the Equal 
Protection Clause 

An equal protection analysis logically follows the Title VII analysis because it is likely to be 
harder to withstand than Title VII challenges. Supreme Court jurisprudence is historically more 
permissive of race-based affirmative action programs under Title VII than under the Equal Protection 
Clause.133 Even though state action could create additional legal challenges, state sanction can also 
provide crucial enforcement and execution benefits. Therefore, the benefits of enforcement must be a 
well-planned trade off to the risks of subjecting these agreements to constitutional challenge. 

1. Government as Signer of or Party to the Agreement 

If the government—city, state, or federal—is a direct signatory or party to a CBA, any challenge 
objecting to racial and ethnic based affirmative action initiatives within the CBA will require a 
constitutional analysis under the Equal Protection Clause.134 Croson and Adarand require that hiring 
requirements for government contracts that are based on race will be subject to strict scrutiny; to 
withstand such scrutiny, they must be narrowly tailored to a compelling state interest.135 Part C of this 
Section will address whether current CBAs are likely to withstand this analysis. 

2. State Action Doctrine136 

                                                
133 See, e.g., Ronald W. Adelman, Voluntary Affirmative Action Plans by Public Employers: The Disparity in Standards 

Between Title VII and the Equal Protection Clause, 56 FORDHAM L. REV. 403, 421–22 (1987).  Adelman points out that for 
most courts, the equal protection standard for affirmative action plans is stricter than the Title VII standard:  

Earlier lower court cases that address both Title VII and the equal protection clause appear to fall into 
three camps. One view recognizes the disparity and, concluding that the stricter equal protection 
standard ultimately will govern, subsumes the Title VII standard into its equal protection analysis. A 
more common view sees the standards as distinct, the equal protection standard being stricter, but no 
court subscribing to this view has yet found a plan valid under Title VII and invalid under equal 
protection. Discussing this apparent disparity between statutory and constitutional standards, a court 
of appeals judge stated: “although I readily concede that interpreting Title VII to permit personnel 
practices that the Constitution prohibits seems anomalous, the Supreme Court has nevertheless 
concluded that Congress intended just that state of affairs.” 

Id. (internal citations omitted). The Supreme Court has also explicitly pointed out that the standard for a private 
employer under Title VII is not coextensive with the Constitution: “The fact that a public employer [under Title VII] 
must also satisfy the Constitution does not negate the fact that the statutory prohibition with which that employer must 
contend was not intended to extend as far as that of the Constitution.” Johnson v. Transp. Ag’y Santa Clara Cty., 480 
U.S. 616, 627 n.6 (1987) (emphasis in original). For a more in-depth analysis, see Chris Engels, Voluntary Affirmative Action 
in Employment for Women and Minorities Under Title VII of the Civil Rights Act: Extending Possibilities for Employers to Engage in 
Preferential Treatment to Achieve Equal Employment Opportunity, 24 J. MARSHALL L. REV. 731, 748 n.75 (1991). 

134 See Zehrt, supra note 34, at 5–6.  
135 See id at 6–8. 
136 For a comprehensive analysis of all components of the state action doctrine, see G. Sidley Buchanan, A 

Conceptual History of the State Action Doctrine: The Search for Governmental Responsibility, 34 HOUS. L. REV. 333 (1997) 

 



240 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

If the government is not an express signatory to a CBA, its involvement in funding and 
regulating parties to the agreement could nevertheless subject the CBA to a constitutional analysis under 
the state action doctrine.137 The state action doctrine first observes that the Constitution will not typically 
apply to private action that is not “fairly attributable” to the government.138 The doctrine nevertheless 
notes that the government is at least partially involved in many aspects of private activity, from land use 
to taxation, to subsidies and regulation, among others. Therefore, the doctrine contemplates a spectrum 
of government involvement with completely private action at one pole and completely governmental 
action at the other. A court will then use the state action doctrine to scan this spectrum of governmental 
involvement to determine when there is sufficient governmental involvement to render the private 
action now “fairly attributable” to the government and therefore subject to the relevant constitutional 
analysis. There are many analytical frameworks under which a private actor’s actions could sufficiently 
implicate governmental action to warrant a constitutional analysis.139 The structure and context of most 
CBAs most closely align with the state nexus issue: where one or more links exist between the 
government and the private actor such that the court must ask whether the amount or nature of those 
links is extensive enough to fairly attribute the private actor’s actions to the government.140   

The modern test for state action comes from the Supreme Court’s analysis in Lugar v. 
Edmondson.141 There, the Court established that the state action doctrine requires 1) a constitutional 
deprivation “caused by the exercise of some right or privilege created by the State or by a rule of conduct 
imposed by the State,” and 2) that the party charged with the deprivation is considered a state actor.142 

The Court later applied this Lugar test in Edmonson v. Leesville to address whether a defendant’s 
use of peremptory challenges could be considered state action for purposes of an equal protection 
violation.143 Applying the first Lugar prong, the Court found that the use of peremptory challenges was 
granted by Congress and had no purpose outside a court of law.144 Therefore, the ability to engage in the 
challenged act owed its origin and sanction to Congress itself.145   

Though the second prong of the Lugar state action test turns on a factual analysis, the Court in 
Edmonson observed principles of general application.146 To determine if an individual or organization’s 

                                                                                                                                                       
[hereinafter State Action Doctrine I] and G. Sidney Buchanan, A Conceptual History of the State Action Doctrine: The Search for 
Governmental Responsibility [Part II of II], 34 HOUS. L. REV. 665 (1997) [hereinafter State Action Doctrine II]. 

137 The state action doctrine owes its origin to Justice Bradley’s opinion in the Civil Rights Cases, 109 U.S. 3, 11 
(1883) (“It is state action of a particular character that is prohibited. Individual invasion of individual rights is not the 
subject-matter of the [Fourteenth] [A]mendment.”). This means that private acts of discrimination or infringement of 
due process rights under the Fourteenth Amendment would not warrant a constitutional analysis because they are not 
state action.) 

138 State Action Doctrine I, supra note 136, at 335. 
139 Id. at 333–34; State Action Doctrine II, supra note 136, at 665–67 (discussing each component of the state 

action doctrine, including the public function issue, the state nexus issue, the beyond-state-authority issue, the 
projection-of-state-authority issue, the state authorization issue, and the state inaction issue).  

140 State Action Doctrine I, supra note 136, at 346–47. 
141 Lugar v. Edmondson Oil Co., Inc., 457 U.S. 922 (1982). 
142 Id. at 937. 
143 Edmonson v. Leesville Concrete Co., Inc., 500 U.S. 614, 620 (1991). 
144 Id. 
145 Id. at 621. 
146 Id.  



2013 COLUMBIA JOURNAL OF RACE AND LAW 241 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

action is attributable to the state, several factors are relevant: first, “the extent to which the actor relies 
on governmental assistance and benefits;” second, “whether the actor is performing a traditional 
governmental function;” and third, “whether the injury caused is aggravated in a unique way by the 
incidents of governmental authority.”147   

Applying the first factor, the Court found that the private actor had sufficient governmental 
assistance and benefits because peremptory challenges require “extensive use of state procedures with 
‘the overt, significant assistance of state officials.’”148 The peremptory challenge system could not exist 
without significant participation by the government.149 Second, the Court found that selecting a jury is a 
traditional government function because a jury is a “quintessential governmental body, having no 
attributes of a private actor.”150 It asserted that “[i]f a government confers on a private body the power 
to choose the government’s employees or officials”—in this case, jurors—“the private body will be 
bound by the constitutional mandate of race neutrality.”151 Finally, the Court found that the alleged 
injury of discriminatory peremptory challenges was “made more severe” by its occurrence in a 
courtroom, which raised serious questions about the neutrality of the decisions made in that 
courtroom.152 Given that all these conditions were decisively met, the Court held that a defendant’s use 
of peremptory challenges could be attributable to state action.  

Despite Edmonson, however, an earlier Supreme Court decision applied a more constrained 
approach with a different analysis. In NCAA v. Tarkanian,153 the University of Nevada, Las Vegas fired 
its basketball coach, Mr. Tarkanian, at the behest of the NCAA.154  Tarkanian sued both the university 
and the NCAA alleging violations of his Fourteenth Amendment due process rights.155 Though the 
university was clearly a state actor, the Court considered whether the NCAA, too, could be considered a 
state actor based on its involvement in the case.156 The Court applied the first prong of the Lugar test, 
but analyzed different factors under the second prong to determine when the decisive action taken 

                                                
147 Id. at 621–22, 624. The Edmonson Court did not mention if these factors standing alone are dispositive of 

state action. Rather, the Court examined all aspects sequentially and found that they all apply to a defendant’s use of 
preemptory challenges on the basis of race during the voir dire process.  

148 Id. at 622. 
149 Edmonson, 500 U.S. at 622–24 (stating that Congress established the qualifications for jury service and the 

processes by which jurors are selected, that the procedures prescribed by the Administrative Office of the United States 
Courts also apply, that private parties can only exercise peremptory challenges with the assistance of the court itself, and 
that the judge, as a state actor, makes himself a party to the alleged discrimination). 

150 Id. at 624.  
151 Id. at 625.  
152 Id. at 628 (“Few places are a more real expression of the constitutional authority of the government than a 

courtroom, where the law itself unfolds.”).  
153 Nat’l Collegiate Athletic Ass’n v. Tarkanian, 488 U.S. 179 (1988). The Court held that the NCAA was not a 

state actor because its rules, ultimately executed by the University, did not originate in state law. Rather, the NCAA took 
its power from state law, but developed its rules and regulations on its own. Furthermore, the NCAA took no direct 
action against the coach; it could only threaten sanctions against the University actor. The University did not thereby 
delegate any power to the NCAA to make the decision. Finally, the Court said that even assuming the NCAA’s ability to 
sanction the university was so great that it effectively coerced the University to fire the coach, it did not follow that the 
NCAA was acting under color of state law.  

154 Id. at 180–81. 
155 Id. at 181. 
156 Id. at 181–82. 



242 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

against an individual may be deemed to be state action.157 The Court said such an attribution may be 
appropriate “if the State creates the legal framework governing the conduct, . . . if it delegates its 
authority to a private actor, . . . or . . . if it knowingly accepts the benefits derived from the 
unconstitutional behavior.”158 Although these factors differ from those later applied in Edmonson, the 
Edmonson Court did not seek to dismantle the Tarkanian analysis. Therefore, these factors still apply and 
overlap with those in Edmonson. 

An example of the first Tarkanian scenario occurs when the state compels or mandates the 
unconstitutional action. For example, in American Manufacturers Mutual Insurance Company v. Sullivan,159 the 
issue was whether a private insurer’s decision to refuse payment for medical treatment could be 
attributed to the state for purposes of the Fourteenth Amendment because the insurer was permitted to 
withhold that treatment under state law.160 The Court held that the insurer’s act was not state action 
because it is not enough that the state “authorized” or “encouraged” the act without more proof that it 
specifically wanted the insurers to withhold payment for medical treatment.161 Finding a “close nexus” 
between a state and the challenged action, the Court explained, requires the state to exert sufficient 
coercive power or to have provided significant overt or covert encouragement such that the action must 
have been that of the state.162 When the decision to withhold payment was made entirely by the 
judgment of the private party, the plaintiff would be hard-pressed to show a sufficient nexus for state 
action even though the state extensively regulated the insurer.163 By failing to show this nexus, the 
plaintiff in Sullivan failed to attribute the insurer’s conduct to the state.164 

The second Tarkanian scenario is when a state delegates power to a private individual. A 
common example is when the state delegates its Eighth Amendment obligation to provide medical 
treatment to incarcerated individuals to a private physician. In West v. Atkins, a doctor who contracted 
with the state to provide medical services to a prisoner was a state actor in light of his assumption of the 
state’s duty to provide constitutionally adequate medical care.165  

The third Tarkanian factor—whether the state knowingly accepts the benefits derived from 
unconstitutional behavior—comes from one of the earliest cases of the modern state action doctrine, 
Burton v. Wilmington Parking Authority.166 Burton more or less established the conceptual framework that 
guided most of the subsequent case law. In Burton, a restaurant located within a parking building owned 
by a Delaware state agency refused to serve an African American man because of his race.167 The 
plaintiff argued that because the restaurant leased its space from the state, its action was subject to the 
Equal Protection Clause of the Fourteenth Amendment.168 The Court ultimately agreed with the 

                                                
157 Id. at 192–94. 
158 Id. at 192.  
159 Am. Mfrs. Mut. Ins. Co. v. Sullivan, 526 U.S. 40 (1999). 
160 Id. at 51. 
161 Id. at 51–53. 
162 Id. at 52.  
163 Id. at 57–58. 
164 Id. at 58. 
165 West v. Atkins, 487 U.S. 42, 56–57 (1988).  
166 Burton v. Wilmington Parking Auth., 365 U.S. 715 (1961). See also State Action Doctrine I, supra note 136, at 

395. 
167 Burton, 365 U.S. at 716. 
168 Id.  



2013 COLUMBIA JOURNAL OF RACE AND LAW 243 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

plaintiff, conducting a fact-specific analysis that weighed several “contact factors” to determine that the 
restaurant had sufficient contact with the state to merit a constitutional analysis. First, the land and 
building were publicly owned and “dedicated to ‘public uses’ in performance of . . . ‘essential 
government functions.’”169 Second, the land and building were not “surplus state property” but rather 
integral—both financially and physically—to the state’s plan to operate a self-sustaining project.170 Third, 
the “peculiar relationship” of the restaurant and the parking facility “confer[red] on each an incidental 
variety of mutual benefits.”171 Finally, the Court found that it would be a “grave injustice” for the 
government to idly permit discrimination of a citizen in a government building, even if in good faith.172   

 Under this analysis, even if the state were not directly aware of the discriminatory conduct of 
the private actor, by virtue of the state’s physical and fiscal relationship to that actor, the conduct could 
become state action. The “knowingly accepting benefits” prong has been subject to judicial constraint. 
For example, in Rendell-Baker v. Kohn, the Supreme Court held that extensive government regulation and 
funding were generally not enough to qualify as state action.173 Even though the nonprofit school in 
question received almost all of its funding from and was heavily regulated by the state, the Court found 
that those factors did not override the school’s private judgment.174 The petitioners also claimed that 
because the school performs a “public function,” it is a state actor. The Court rejected that claim on the 
grounds that the function performed not only had to be “traditional,” but traditionally the “exclusive 
prerogative of the State.”175 Finally, in contrast to its decision in Burton, the Court found an insufficient 
“symbiotic relationship” between the school and the state in Rendell-Baker. Whereas in Burton the 
restaurant’s profits contributed to the support of the state-owned garage, the school in Rendell-Baker was 
unilaterally dependent on the government for support, but did not fiscally contribute to the government 
in return.176  

In Brentwood Academy v. Tennessee Secondary School Athletic Association, the most recent case on this 
issue, the Court said that “no one fact can function as a necessary condition across the board for finding 
state action; nor is any set of circumstances absolutely sufficient.”177 Here, the Court indicated a return 

                                                
169 Id. at 723. 
170 Id. at 723–24. 
171 Id. at 724. 
172 Burton v. Wilmington Parking Auth., 365 U.S. 715, 724–25 (1961). Buchanan refers to these factors as 1) 

the government and ownership factor, 2) the financial integration with government factor, 3) the symbiotic relationship 
factor, and 4) the governmental encouragement or endorsement factors. State Action Doctrine I, supra note 136, at 396.  

173 See Rendell-Baker v. Kohn, 457 U.S. 830 (1982) (nonprofit school receiving most of its funding from the 
state, subject to extensive state regulations, performing a “public function” and having a “symbiotic relationship” with 
the government did not rise to the level of state action). 

174 Id. at 841 (“Here the decisions to discharge the petitioners were not compelled or even influenced by any 
state regulation”). However, while regulation alone is not usually enough to prove state action, it may be sufficient where 
regulatory activity of a nominally private actor is sufficiently intertwined with the state. See Brentwood Acad. v. 
Tennessee Secondary Sch. Ath. Ass’n, 531 U.S. 288 (2001) (holding that regulatory activity of a statewide association 
made up primarily of public school officials and funded mostly by their dues, and that had traditionally regulated in lieu 
of the state Board of Education, was considered state action). However, mere receipt of dues from member universities 
who receive federal financial assistance will not subject the NCAA to constitutional requirements. See Nat’l Collegiate 
Athletic Ass’n v. Smith, 525 U.S. 459 (1999). 

175 Rendell-Baker, 457 U.S. at 842 (emphasis in original).  
176 Id. at 842–43. 
177 Brentwood Acad., 531 U.S. at 295. 



244 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

to Burton’s totality of the circumstances test for whether the combination of factors serves to push 
private action over the threshold into state action. In Brentwood, the persuasive thread running through 
precedent to its holding was the amount of management and control the State exerted over the private 
actor. The “pervasive entwinement” of the statewide association with state officials and institutions 
justified the Court’s holding that “there is no substantial reason to claim unfairness in applying 
constitutional standards to it.”178 However, the Court astutely noted that what is “fairly attributable” to 
the State “is a matter of normative judgment, and the criteria lack rigid simplicity.”179 Therefore, any 
honest analysis must be fact-specific. Still, the aforementioned examples provide some signposts as to 
relevant points of inquiry. 

3. Analysis of the State Action Doctrine as Applied to Atlantic Yards 

As a threshold matter, significant governmental funding for, regulation of, and authorization of a 
development and a CBA will not be enough to turn the developer into a state actor. Therefore, any 
direct hiring decision made by the developer or any private employer on the development site will not 
fall under the state action doctrine. This is true even when the project would be impossible but for 
governmental funding and regulation of the project.180 In the CBA context, this means that even the 
terms of a CBA such as Cherokee-Gates, according to which the community could not find a developer 
until the city agreed to significant subsidies, do not necessarily turn a developer action into state action. 
Furthermore, even where the building sites are subject to extensive regulation and approval processes, 
the Rendell-Baker and Sullivan analyses make it unlikely that any specific hiring decisions would be traced 
to state action when the state did not directly make or mandate those decisions. 

Nevertheless, legislation has been passed and is being contemplated that mandates CBAs 
between developers and the community for new projects.181 The legislation itself would only satisfy the 
first prong of the Lugar test; far more contact with the state than mere state legislation is required for the 
actual hiring action to be attributable to the state. However, if, as in the Milwaukee CBA example, the 
state legislation envisions racial or gender quotas for the CBA itself, those requirements would provide 
the requisite link between state action and the private developer’s hiring decisions. 

The trickier analysis is whether there are enough links between the private actor and the state to 
satisfy the Burton analysis. For the purposes of this analysis, I chose to look at the Atlantic Yards CBA 
because it involves a private developer, racial quotas, and significant New York City involvement. A 
project of this scope, although privately negotiated and executed, is ripe for a state action analysis to see 
if its racial hiring provisions require an equal protection analysis. 

The first Burton factor was that the city owned the land while the building and the restaurant 
leased its space. According to the 2009 Atlantic Yards lease agreement, Empire State Development 
Corporation (EDSC)—a New York State entity—is the landlord for the interim leases, ground leases, 
arena development, and non-arena development leases. For all except the Ground Leases, affiliates of 
Bruce Ratner’s private development company are the tenants. The Brooklyn Arena Local Development 

                                                
178 Id. at 298.  
179 Id. at 295.  
180 See Rendell-Baker v. Kohn, 457 U.S. 830, 842 (1982). 
181 See Milwaukee CBA, supra note 91. See also Proposed Michigan Senate Bill 379 (May 12, 2011), available at 

http://legiscan.com/MI/text/SB0379/2011 (requiring the developer of a new Detroit River bridge to sign a CBA with 
local governments and community representatives). 



2013 COLUMBIA JOURNAL OF RACE AND LAW 245 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Corporation (BALDC) leases the Ground Lease from the ESDC and then leases the Arena 
Development lease to a Ratner affiliate.182 The Arena lease is set to last for at least thirty-five years. This 
means that not only will the city own the land, but also will nominally own the arena and lease it to 
BALDC, which in turn will lease it to the developer.183 However, instead of paying money to the city or 
BALDC in rent or taxes, Ratner’s company will pay only a nominal dollar amount for “rent” and the rest 
will go toward the construction, operation, and maintenance of the arena in lieu of taxes.184 New York 
City is thus not getting the same fiscal return for the arena as the city received from the restaurant in 
Burton. However, to the extent that the Burton requirement envisions the apparent governmental sanction 
of private activity conducted on its property, this arrangement could satisfy that factor.  

The second Burton requirement was that the land not be “surplus” but rather financially integral 
to the city’s plan to operate a self-sustaining project. Without a doubt, the Barclays Center is an integral 
part of the city.  The arena hosts the New Jersey Nets, an NBA basketball team. The land is located in 
the heart of Brooklyn, and includes multiple housing units and retail space. However, in contrast to 
Burton, FCR and not the city owns those buildings. Nevertheless, the project is still funded with at least 
$100 million of ESDC funds and $100 million of city funds, and its plan envisions more city and state 
funding depending on further needs of the project.185 

The third Burton factor—the so-called “symbiotic relationship” factor—looks to the exchange of 
mutual benefits between the state and the private actor. In Burton, the Court looked to how the State’s 
parking garage gave guests of the parking facility an accessible place to park their cars and how that 
convenience for diners could, in return, increase demand for the city’s parking facilities.186 Similarly with 
the Barclays Center, the links between the city and this large-scale housing, stadium, and retail complex 
are multitudinous. In addition, although scholars and commentators remain divided about the net 
benefits of stadiums on local economic growth,187 the city and developer nevertheless project the 
benefits of the project with unified force.188 

                                                
182 Exhibit D: Project Leases Abstract, available at http://www.scribd.com/doc/78517384/2009-ESDC-Atlantic-

Yards-Lease-Abstract (last visited May 3, 2013) [hereinafter Atlantic Yards Abstract]. 
183 Norman Oder, Prokhorov filled arena financing gap not by buying bonds but offering a loan, ATLANTIC YARDS REPORT 

(Feb. 24, 2010), http://atlanticyardsreport.blogspot.com/2010/05/prokhorov-filled-arena-financing-gap.html; History of 
Empire State Development, EMPIRE STATE DEVELOPMENT, http://www.esd.ny.gov/AboutUs/History.html (last visited 
Mar. 31, 2013). 

184 Atlantic Yards Abstract, supra note 182. See also Atlantic Yards Land Use Improvement and Civil Project Modified 
General Project Plan 25–26, EMPIRE STATE DEVELOPMENT (June 23, 2009), 
http://www.esd.ny.gov/Subsidiaries_Projects/AYP/AtlanticYards/ModifiedGPP2009.pdf.  

185 Id. at 27–29. 
186 Burton v. Wilmington Parking Auth., 365 U.S. 715, 724 (1961). . 
187 Roger G. Noll & Andrew Zimbalist, Sports, Jobs & Taxes: Are New Stadiums Worth the Cost?, THE BROOKINGS 

INSTITUTE (Summer 1997), http://www.brookings.edu/research/articles/1997/06/summer-taxes-noll.aspx (discussing 
how local sports stadiums may not actually be as economically beneficial to New York City as typically claimed).  

188 Kareem Fahim, Ground Broken on Atlantic Yards Project, N.Y. TIMES (Mar. 11, 2010), 
http://cityroom.blogs.nytimes.com/2010/03/11/ground-broken-on-atlantic-yards-project (showing Mayor Bloomberg 
and Governor Paterson shoveling dirt for the groundbreaking at Atlantic Yards); Nicholas Confessore, To Build Arena in 
Brooklyn, Developer First Builds Bridges, N.Y. TIMES (Oct. 14, 2005), 
http://www.nytimes.com/2005/10/14/nyregion/14yards.html?ref=atlanticyardsbrooklyn (discussing the way in which 

 



246 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

The final Burton factor is whether, by virtue of the city’s implicit approval of the project and the 
CBA, it is putting its imprimatur on a “grave injustice” perpetrated by the private actor. Though 
maintaining racial quotas for a large-scale project is less grave than forbidding any African Americans 
from eating at a restaurant, the Supreme Court has upheld the notion that “any person, of whatever race, 
has the right to demand that any governmental actor subject to the Constitution justify any racial 
classification subjecting that person to unequal treatment under the strictest judicial scrutiny.”189 In 
Burton, the restaurant affirmatively alleged that its business would be injured if it served African 
Americans, and the Court found this indicative of the government’s direct benefit, through the lease, 
from discriminatory action.190  

The totality of the circumstances does not evince an obvious answer to the question of whether 
the Atlantic Yards CBA will be required to withstand strict scrutiny. Some commentators suggest that 
the state action analysis ultimately comes down to whether a reasonable person would think the private 
actor is actually a state actor.191 Cases like Sullivan, Brentwood Academy, and Tarkanian all also seem to 
adopt this pragmatic approach. When state officials are so involved in the decisions of a nominally 
private actor, the private actor’s judgments and decisions effectively become those of the state. When the 
appearance of state authority is so omnipresent and a private actor is literally encapsulated within a state 
building, its actions appear to be those of the state. However, the mere presence of state funding and 
regulation of an institution, without more, does not substitute state action for the independent 
judgments and decisions of private actors within that institution. Under the scope of this bigger picture, 
even a project like Atlantic Yards does not appear to have sufficient state entanglement to transform the 
private developer into a state actor for the purposes of the state action doctrine.  

4. Would a Race-Based CBA Withstand Strict Scrutiny? 

Even if the race-based programs in the Atlantic Yards, Milwaukee, and Peninsula CBAs likely do 
not meet the high bar of the state action doctrine, a thorough assessment requires a determination of 
whether if they would pass strict scrutiny. After Adarand, the Seventh, Eighth, and Ninth Circuits said 
that if a state is implementing a federal program within its jurisdiction, it does not have to provide its 
own compelling interest to satisfy strict scrutiny.192 But, because Croson applies to state and local 
programs akin to the localized provisions of CBAs, its analysis is more relevant. In 1983, Richmond, 
Virginia adopted a plan that required prime contractors with the city to subcontract at least thirty percent 
of the value of the contract to at least one Minority Business Enterprise (MBE).193 The Court struck 
down the plan on the grounds that it failed to show a compelling state interest in both imposing the plan 
and in arriving at the thirty percent quota.194  The Court observed that while a state or locality has the 

                                                                                                                                                       
Bruce Ratner’s ability to get support from prominent community figures helped move the Atlantic Yards development 
forward). 

189 Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 201 (1995).  
190 Burton, 365 U.S. at 724. 
191 See, e.g., John B. Owens, Westec Story: Gated Communities and the Fourth Amendment, 34 AM. CRIM. L. REV. 1127, 

1155 (1997) (“The reasonable person probably would believe that a coffee shop in a public parking structure was 
somehow owned by the state.”). 

192 Zehrt, supra note 34, at 14. The cases in these circuits all examined congressional findings regarding race 
discrimination in government highway contracting. The congressional findings were sufficient for the compelling state 
interest prong. 

193 Id. at 7. See also City of Richmond v. J.A. Croson Co., 488 U.S. 469, 477 (1988). 
194 Croson, 488 U.S. at 507–11. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 247 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

authority to remedy the effects of past discrimination within its jurisdiction, that authority must be 
strictly confined when making racial distinctions.195  In order to withstand strict scrutiny, the state or 
locality has to prove with enough evidence that it is remedying discrimination specifically within its 
jurisdiction.   

Richmond supplied evidence about the disparity between the number of prime contracts 
awarded to minority firms and the minority population of the city, but the Court found the comparison 
to be improper. It stated that when specific qualifications were necessary to fill certain jobs, the relevant 
comparison should instead be between qualified individuals and the current pool of minority contractors 
awarded prime contracts.196 The Court also did not accept the low MBE membership in local 
contractors’ associations because it found that low participation could be credited to a variety of 
explanations.197 Furthermore, the Court did not accept congressional findings of nationwide 
discrimination in the construction industry as sufficient justification for the Plan.198 On these grounds, 
the Croson Court held that Richmond had failed to supply sufficient evidence of any identified 
discrimination in the city’s construction industry sufficient to warrant a remedial plan.199 The Richmond 
Plan thus failed to identify a compelling state interest. 

However, the Court also observed many ways in which a city like Richmond could have shown a 
compelling state interest in creating the quota. First, instead of comparing contracts awarded to minority 
firms with the general minority population in the city, Richmond should have looked at qualified 
minority contractors.200 Second, instead of looking only at the fact that black membership in the trade 
organizations was low, the city should have linked low minority membership to the number of local 
MBEs eligible for membership. If the statistical disparity were great enough, an inference of 
discrimination could arise.201 Finally, instead of looking at nationwide discrimination in the construction 
industry, Richmond needed to determine if its own spending practices were exacerbating the pattern of 
prior discrimination, identify that discrimination with specificity, and then fashion relief based on its 
findings.202  Relying on Congress’s nationwide findings was not enough. 

Croson covers some of the field regarding how to prove a compelling state interest to remedy 
prior discrimination in a particular industry, but most of the battle has been waged in the Circuits. 
Professor Lynn Ridgeway Zehrt examined all the circuit court cases analyzing regional affirmative action 
plans and found that only one out of ten passed strict scrutiny. 203 The Ninth Circuit held that statewide 

                                                
195 Id. at 491. 
196 Id. at 501–02. The Court looks to Title VII cases for these examples of relevant comparisons. See, e.g., 

Hazelwood Sch. Dist. v. United States, 433 U.S. 299 (1977). In Croson, the city did not even know how many MBEs in 
the market were qualified to undertake prime or subcontracting work nor the percentage of total city construction 
dollars minority firms received at the time as subcontractors on prime contracts by the city. Croson, 488 U.S. at 502.  

197 Croson, 488 U.S. at 503 (“The mere fact that black membership in these trade organizations is low, standing 
alone, cannot establish a prima facie case of discrimination.”).  

198 Id. at 504 (“Congress has made national findings that there has been societal discrimination in a host of 
fields. If all a state or local government need do is find a congressional report on the subject to enact a set-aside 
program, the constraints of the Equal Protection Clause will, in effect, have been rendered a nullity.”).  

199 City of Richmond v. J.A. Croson Co., 488 U.S. 469, 505 (1988). 
200 Id. at 503. 
201 Id.  
202 Id. at 504. 
203 Zehrt, supra note 34, at 21. 



248 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

findings of discrimination were insufficient to justify an affirmative action plan within a state university. 
Instead, the university had to independently offer its own evidentiary support documenting how it had 
previously discriminated against the protected groups within the university. In the Sixth circuit, Ohio had 
dispatched a state task force that extensively studied the relationship between minorities and the state 
contracting industry and held numerous public hearings about the state’s historical exclusionary practices 
in construction.204 The task force determined that the legislature should adopt a statewide ten percent 
goal for construction contracts, and the Supreme Court of Ohio agreed.205 However, another group of 
plaintiffs filed a claim in the Sixth Circuit, which found the program not to be compelling because the 
statistical evidence was mostly outdated and either too limited in scope or irrelevant to the case.206   

Of the programs that had a compelling state interest, many were not narrowly tailored. The 
Supreme Court held that in order for a plan to be narrowly tailored, the relevant factors were “the 
necessity for the relief and efficacy of alternative remedies; the flexibility or duration of the relief[;] . . . 
the relationship of the numerical goals to the relevant labor market; and the impact of the relief on the 
rights of the parties.”207 The cases that have failed in the Circuits were unsuccessful either because the 
government had not considered race-neutral alternatives or because the term “minority” was 
overinclusive and not limited to minorities that had actually suffered discrimination in the jurisdiction.208 
The Croson Court even gave some suggestions of race-neutral ways to increase accessibility of contracting 
opportunities for minority contractors, including “simplification of bidding procedures, relaxation of 
bonding requirements, and training and financial aid for disadvantaged entrepreneurs of all races[.]”209 

For CBAs like the Milwaukee Park East Redevelopment Compact that directly involve the state 
and involve race-based quotas, the case law suggests that the city should conduct extensive studies to 
find proof of discrimination in the construction industry in the city or the surrounding area. The city of 
Denver proved a compelling state interest to the Tenth Circuit by presenting evidence of grievances filed 
by minority contractors, several public hearings and extensive testimony of racial discrimination within 
the city’s construction industry, several statistical disparity studies, telephone and mail surveys of local 
construction businesses, as well as a trial run of a voluntary program that ultimately failed.210 That this is 
the only program to prove a compelling interest indicates that the bar is set quite high. It suggests that a 
city needs to adopt a race-neutral program first before attempting to impose a racial quota and that the 
city must use a variety of survey methods to prove that discrimination actually exists. For those cities 
who have signed or legislatively enacted CBAs, this proof will be incredibly costly and burdensome. The 
better alternative is to let the private developer negotiate the CBA with the community groups directly to 
avoid a constitutional analysis. However, if having the city participate in the CBA is the only way to 
enforce the plan, then the quotas should not be race-based, but instead should follow the model of the 
Staples and San Francisco CBAs, which focuses on training and hiring local and low-income residents. 

Also, doing extensive research about the causes of local disparities is not enough.  The city’s 
remedy must also be narrowly tailored to the problem. First, the program cannot be overinclusive of 
other groups not subject to historical discrimination. In Croson, this requirement meant that “Spanish-

                                                
204 Id. at 22. 
205 Id.  
206 Id. at 23. 
207 Id. at 24. 
208 Id. 
209 City of Richmond v. J.A. Croson Co., 488 U.S. 469, 509–10 (1988). 
210 Zehrt, supra note 34, at 19–20. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 249 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

speaking, Oriental, Indian, Eskimo, or Aleut persons” could not benefit from the plan without this kind 
of proof even if there was proof of African-American discrimination.211 For a program like the Christina 
Avenue Compositing Facility CBA in Delaware, this point is directly relevant. Although the city is 
seventy-seven percent African American and the nature of historical discrimination seems confined to 
this group, the quota applies to minorities defined to include Hispanic persons, Native Americans or 
Alaskan natives, and Asian and Pacific Islander groups. Without any proof that these groups were 
similarly disadvantaged, this CBA could face a downfall similar to that of the plan in Croson. 

Finally, Croson also suggests that it is not enough to find that there was discrimination in the local 
area in the specific industry; there needs to be a reason for the targeted number.212 The Court did not say 
that percentage quotas were per se unconstitutional, but struck out strongly against them.213 At most, the 
decision suggests that where there is evidence of extreme discrimination, a “narrowly tailored racial 
preference might be necessary to break down patterns of deliberate exclusion,” but only after all race-
neutral non-numeric options have been exhausted.214 Unfortunately, this analysis strongly suggests that 
programs like the Composting Facility and the Milwaukee CBAs will not withstand a strict scrutiny 
analysis. Only the most thorough and iterative program has ever withstood this analysis, and these CBAs 
do not meet its high bar. 

V.  CONCLUSION  

Community benefit agreements have heretofore been the most promising approach to 
remedying discrimination against minority contractors and construction workers in urban communities. 
Nevertheless, these agreements require extensive safeguards to insulate them from potential legal 
challenges. Percentage goals will likely create the most cause for legal challenge, and CBAs probably 
should not include them. First source offices may well withstand legal challenge because they are not 
facially discriminatory and technically benefit all members of the community equally. Private CBAs are 
also more likely to withstand legal challenges than public CBAs in light of the Supreme Court’s decisions 
in Adarand and Croson.  

Furthermore, despite the benefit of these agreements, significant enforcement problems remain. 
In 2011, seven Brooklyn construction workers filed a lawsuit against developer Bruce Ratner and local 
group Brooklyn United for Innovative Local Development (BUILD) for failing to deliver 1,500 annual 
jobs and job training promised to mostly African American BUILD members who supported the 
project.215 In the CBA between Columbia University administrators and Harlem residents for the 

                                                
211 Croson, 488 U.S. at 506. 
212 Id. at 507 (“[T]he 30% quota cannot be said to be narrowly tailored to any goal, except perhaps outright 

racial balancing. It rests upon the ‘completely unrealistic’ assumption that minorities will choose a particular trade in 
lockstep proportion to their representation in the local population.”). 

213 Id. at 508 (“Since the city must already consider bids and waivers on a case-by-case basis, it is difficult to see 
the need for a rigid numerical quota.”).  

214 Id. at 509. 
215 Daniel Bush, Out of Work! Ratner ally BUILD got just 15 jobs for black Yard supporters, THE BROOKLYN PAPER 

(Dec. 1, 2011), http://www.brooklynpaper.com/stories/34/48/dtg_buildjobsrecord_2011_12_02_bk.html. One 
claimant stated, “I would never have joined this pre-apprenticeship program if it wasn’t agreed (guaranteed) to me that I 
would have a union card upon completion;” Amy Sara Clark, Construction Workers Sue Atlantic Yards Developer, Claiming 
They Were “Duped,” PROSPECT HEIGHTS PATCH (Nov. 15, 2011), 
http://prospectheights.patch.com/articles/construction-workers-sue-atlantic-yards-developer-claiming-they-were-

 



250 We Built This City Vol. 3:2 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Manhattanville project, many local minority architects claim they were offered fewer, smaller projects 
while European architects received the more lucrative ones.216 Lack of enforcement will likely remain the 
greatest problem facing CBAs—bigger than any of these legal challenges—particularly for New York-
based CBAs. 

This does not mean, however, that CBAs are not worthwhile. CBAs continue to break impressive 
ground all over the country.217 CBAs offer a promising approach to remedying past and precluding 
present discrimination against minority contractors and construction workers in areas of new 
development. They can promote normative values like democratic participation and cooperation within 
communities by encouraging community involvement in development projects.218 Contracts between 
private and public actors can promote accountability and flexibility in decision-making processes that 
affect local communities.219 CBAs also promote transparency, by letting private actors, public officials, 
and the local media keep track of the progress of a project and how it meets the promises set out in the 
agreement. 220 Finally, CBAs provide clarity about potential outcomes. This is crucial when local 
governments want to point to the successes of a particular project in terms of job creation or revenue 

                                                                                                                                                       
duped. See also Heather Haddon, Brooklyn Arena Criticized on Hiring, WALL STREET JOURNAL (Jan. 23, 2012), 
http://online.wsj.com/article/SB10001424052970203718504577177390852461060.html. As of November 2012, 
BUILD dissolved its operations citing funding difficulties, tax arrears and complaints about spending irregularities. Cate, 
Barclays Restaurant Unionizes, BUILD Dissolves, BROWNSTONER (Nov. 7, 2012), 
http://www.brownstoner.com/blog/2012/11/barclays-restaurant-unionizes-build-dissolves. 

216 Kimberly Shen & Chris Meyer, Black architects claim that they were passed over for Manhattanville, COLUM. DAILY 

SPECTATOR (Jan. 28, 2013), http://www.columbiaspectator.com/2013/01/28/black-architects-claim-they-were-passed-
over-manhattanville (“[The administrators] offer a project on Broadway that is worth $20,000, which is a drop in the 
bucket for professional architects like us. . . . Even after we have offered to work collectively and pair up with architects 
already hired for the project, the administrators still turned us down.”). 

217 For a new CBA on Long Island, see Aisha Al-Muslim, Hempstead Oks controversial agreement on downtown plan, 
NEWSDAY (Jan. 15, 2013), http://www.newsday.com/long-island/towns/hempstead-oks-controversial-agreement-on-
downtown-plan-1.4453020. For one in New Haven, see Thomas MacMillan, It’s a Deal—And a Sale, NEW HAVEN 

INDEPENDENT (Dec. 18, 2012), http://www.newhavenindependent.org/index.php/archives/entry/mlk-
amistad_community_benefits_deal_reached/id_53497 (CBA to sell an abandoned school to a not-for-profit charter 
school to rebuild a new high school). For one in Cleveland, see GCP Board votes to support establishment of Community Benefits 
Agreement, GREATER CLEVELAND PARTNERSHIP (Dec. 10, 2012), http://www.gcpartnership.com/News/2012/GCP-
Board-votes-to-support-establishment-of-Community-Benefits-Agreement-12-10-12.aspx. For one in Portland, Oregon, 
see Don McIntosh, City of Portland commits to build union and use minority workers and contractors, NWLABORPRESS.ORG (Sept. 
18, 2012), http://nwlaborpress.org/2012/09/cba-2/ (describing a “Model Community Benefits Agreement” where the 
City of Portland agrees that on future City construction projects, unions will represent workers and women and minority 
workers will have better opportunities). 

218 See Audrey G. McFarlane, When Inclusion Leads to Exclusion: The Unchartered Terrain of Community Participation in 
Economic Development, 66 BROOKLYN L. REV. 861, 910 (2000) (discussing the transformational benefits of democratic 
participation beyond the ballot box to the “process of participation [of] citizens themselves being involved in priority-
setting, as well as decision-making, dialogue and deliberation.”). See also Patricia E. Salkin & Amy Lavine, Community 
Benefits Agreements and Comprehensive Planning: Balancing Community Empowerment and the Police Power, 18 J.L. & POL’Y 157, 157 
(2009) (“Often low-income and minority communities are not fully engaged in the [land use regulatory] process, even 
when it may result in decisions negatively impacting their neighborhoods.”). 

219 See generally Jody Freeman, The Private Role in Public Governance, 75 N.Y.U. L. REV. 543 (2000).  
220 Gross, LeRoy & Janis-Aparicio, supra note 42, at 22. 



2013 COLUMBIA JOURNAL OF RACE AND LAW 251 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

brought into a local area.221 In light of these possible benefits, the most immediate upcoming challenge 
will be ensuring that CBAs have the teeth they need to be successful. This will take significant 
commitment from both the developer and the community groups involved. Until then, the legal issues 
mentioned in this Article may not even manifest. And, with enough endorsement and participation from 
the entire community, they may not have to. 

                                                
221 Id. 



252 We Built This City Vol. 3:2