FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

ASSOCIATED GENERAL No. 11-16228

CONTRACTORS OF AMERICA , SAN

DIEGO CHAPTER, INC., a non profit D.C. No. California corporation, 2:09-cv-01622- Plaintiff-Appellant, JAM-GGH

v.

OPINION

CALIFORNIA DEPARTMENT OF

TRANSPORTATION ; WILL KEMPTON ,

individually and in his official capacity as Director of the California Department of Transportation;

OLIVIA FONSECA ,

Defendants-Appellees,

COALITION FOR ECONOMIC EQUITY ;

NATIONAL ASSOCIATION FOR THE

ADVANCEMENT OF COLORED

PEOPLE , San Diego Chapter, Intervenor-Defendants-Appellees.

Appeal from the United States District Court for the Eastern District of California John A. Mendez, District Judge, Presiding

Argued and Submitted February 11, 2013—San Francisco, California 2 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

Filed April 16, 2013

Before: Jerome Farris, Sidney R. Thomas, and N. Randy Smith, Circuit Judges.

Opinion by Judge Farris

SUMMARY*

Civil Rights

The panel dismissed an appeal from the district court’s summary judgment in a 42 U.S.C. § 1983 action in which plaintiffs, Associated General Contractors of America, sought declaratory and injunctive relief against the California Department of Transportation (Caltrans) and its officers, on the grounds that Caltrans’ 2009 Disadvantaged Business Enterprise program unconstitutionally provided race- and sex- based preferences to African American-, Native American-, Asian-Pacific American-, and women-owned firms on certain transportation contracts.

The panel held that AGC did not identify any of its members who have suffered or will suffer harm as a result of Caltrans’ affirmative action program, and therefore AGC had not established that it has associational standing to bring suit. The panel further held that even if AGC could establish standing, its appeal would fail. The panel held that Caltrans’ program survived strict scrutiny by (1) having a strong basis

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 3

in evidence of discrimination within the California transportation contracting industry; and (2) being narrowly tailored to benefit only those groups that have actually suffered discrimination.

COUNSEL

Ralph W. Kasarda, Pacific Legal Foundation, Sacramento, California, for Plaintiff-Appellant.

G. Scott Emblidge, Moscone Emblidge & Sater LLP, San Francisco, California, for Defendant-Appellee; Oren M. Sellstrom, San Francisco, California, for Intervenors- Defendants-Appellees.

Angela C. Thompson, Sacramento, California, for Amicus Curiae United States Justice Foundation.

Sharon M. McGowan, United States Department of Justice, Civil Rights Division, Washington, D.C., for Amicus Curiae United States of America.

OPINION

FARRIS, Senior Circuit Judge:

Associated General Contractors of America, San Diego Chapter, appeals from the district court’s adverse summary judgment rulings. AGC sought declaratory and injunctive relief against the California Department of Transportation and its officers, on the grounds that Caltrans’ 2009 Disadvantaged Business Enterprise program unconstitutionally provided 4 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

race- and sex-based preferences to African American-, Native American-, Asian-Pacific American-, and women-owned firms on certain transportation contracts. The Coalition for Economic Equity and the National Association for the Advancement of Colored People, San Diego Chapter, intervened to defend the program.

On summary judgment, the district court upheld the constitutionality of Caltrans’ program and entered judgment for the defendants. Following Western States Paving Co. v. Washington Sate Department of Transportation, 407 F.3d 983 (9th Cir. 2005), the district court held that Caltrans’ program would satisfy strict scrutiny if it had a strong basis in evidence of discrimination in the California transportation contracting industry, and the program was narrowly tailored to those groups that actually suffered discrimination. The court held that Caltrans’ substantial statistical and anecdotal evidence provided a strong basis in evidence of discrimination against the four named groups, and that the program was narrowly tailored to benefit only those groups. AGC appealed. We DISMISS the appeal because AGC did not identify any of its members who have suffered or will suffer harm as a result of Caltrans’ program, and therefore AGC has not established that it has associational standing to bring suit.1

1 Caltrans’ request for judicial notice is GRANTED to the extent that it is compatible with F ED . R. E VID . 201 and “do[es] not require the acceptance of facts subject to reasonable dispute.” California ex rel. RoNo, LLC v. Altus Fin. S.A., 344 F.3d 920, 931 n.8 (9th Cir. 2003) (internal quotation marks omitted). The motion of the United States Justice Foundation for leave to file a brief as amicus curiae is

GRANTED.

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 5

I. BACKGROUND AND STATEMENT OF FACTS

A. Statutory and Regulatory Background

The Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users, Pub. L. No. 109-59, § 1101(b), 119 Stat. 1144 (2005), authorizes the U.S. Department of Transportation to distribute funds to states for transportation-related projects. The Act is the most recent federal statute providing for race- and gender-based preferences in the transportation contracting industry in response to pervasive and ongoing discrimination. See Western States, 407 F.3d at 988 & n.3. The Act directs the Secretary of Transportation to ensure that 10% of funds distributed to states and municipalities are expended on “disadvantaged business enterprises.” § 1101(b)(2), 119 Stat. at 1156.

The Act does not establish a uniform national affirmative action program. Each state that receives federal funds must implement a preference program that complies with federal regulations. See 49 C.F.R. § 26.1 et. seq. The regulations define “disadvantaged business enterprises” as small businesses owned or controlled by “socially and economically disadvantaged” individuals. Id. § 26.5. There is a reubuttable presumption that African Americans, Hispanic Americans, Native Americans, Asian-Pacific Americans, Subcontinent Asian Americans, and women are socially and economically disadvantaged. Id. § 26.67.

States that receive federal funding must establish overall goals for disadvantaged business participation in federally assisted contracts. Id. § 26.45(a). In the process of setting a goal, the state must first determine the availability of 6 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

disadvantaged businesses in its jurisdiction. Id. § 26.45(c). Then, the state may make an upward or downward adjustment to account for factors affecting the availability of disadvantaged businesses. Id. § 26.45(d). After comparing availability data with the actual utilization of disadvantaged businesses, the state sets an overall goal to address significant disparities. Id. § 26.45(e).

States must use race- and gender-neutral means to meet their goals to the maximum extent possible, but may use race- and gender-conscious means if necessary. Id. § 26.51(a)–(d). Generally, race- and gender-conscious means may not be targeted at specific groups. Id. § 26.51(e)(4). However, a state may use race-conscious means directed at specific minority groups, if it obtains a waiver. See id. § 26.15. States must seek approval of their affirmative action programs by the U.S. Department of Transportation every three years. Id. § 26.45(f)(1)(i).

B. Ninth Circuit Decision in Western States

In 2005, the Ninth Circuit decided Western States Paving Co. v. Washington State Department of Transportation, 407 F.3d 983 (9th Cir. 2005), which involved a facial challenge to the constitutional validity of a predecessor law to the Act, as well as an as-applied challenge to the Washington program implementing the federal mandate. Applying strict scrutiny, we upheld the constitutionality of the federal statute and regulations. Id. at 990–95. However, we struck down Washington’s program because it was not narrowly tailored. Id. at 999–1002. In so doing, Western States announced a two-prong test for narrow tailoring: (1) the state must establish the presence of discrimination within its transportation contracting industry, and (2) the remedial

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 7

program must be “limited to those minority groups that have actually suffered discrimination.” Id. at 997–98.

C. Caltrans’ Implementation of the Act

Caltrans receives up to $3 billion annually from the federal government for transportation projects. Prior to 2006, Caltrans administered a race- and gender-conscious affirmative action program on federally assisted contracts. However, on May 1, 2006, Caltrans ceased to use race- and gender-conscious measures while it gathered evidence in an effort to comply with Western States.

1. Evidence Gathering and the 2007 Disparity Study

Caltrans commissioned a disparity study by BBC Research and Consulting to determine whether there was evidence of discrimination in California’s contracting industry. Disparity analysis involves making a comparison between the availability of minority- and women-owned businesses and their actual utilization, producing a number called a “disparity index.” An index of 100 represents statistical parity between availability and utilization, and a number below 100 indicates underutilization. An index below 80 is considered a substantial disparity that supports an inference of discrimination. See H.B. Rowe Co. v. Tippett, 615 F.3d 233, 243–44 (4th Cir. 2010).

The research firm gathered extensive data to calculate disadvantaged business availability in the California transportation contracting industry. Based on review of public records, interviews, assessments as to whether a firm could be considered available for Caltrans contracts, as well as 8 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

numerous other adjustments, the firm concluded that minority- and women-owned businesses should be expected to receive 13.5% of contract dollars from Caltrans administered federally assisted contracts.

The research firm then examined over 10,000 transportation-related contracts administered by Caltrans between 2002 and 2006 to determine actual DBE utilization. The firm assessed disparities across a variety of contracts, separately assessing contracts based on funding source (state or federal), type of contract (prime or subcontract), and type of project (engineering or construction). A key difference between federally funded and state funded contracts is that race-conscious goals were in place for the federally funded contracts during the 2002–2006 period, but not for the state funded contracts. Thus, state funded contracts functioned as a control group to help determine whether previous affirmative action programs skewed the data. Moreover, the research firm measured disparities in all twelve of Caltrans’ administrative districts, and computed aggregate disparities based on statewide data.

The firm evaluated statistical disparities by race and gender. To control for gender, the firm grouped women who were members of racial minorities with male members of the same minority. As such, its report describes its gender control group as “white women-owned firms” and measures disparities for this group, as well as firms owned by African Americans, Native Americans, Asian-Pacific Americans, Subcontinent Asian Americans, and Hispanic Americans.

The research firm published its results in June 2007. Within and across many categories of contracts, it found substantial statistical disparities for African American, Asian-

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 9

Pacific, and Native American firms. For example, in aggregated state funded contracts, African Americans received only 15% of the contract dollars that would be expected, given their availability, and Asian-Pacific and Native Americans earned less than one-third and two-thirds, respectively. However, there were not substantial disparities for these minorities in every subcategory of contract. For example, the disparity indices for Native and Asian-Pacific Americans were above 80 in federally funded construction subcontracts.

The disparity study also found substantial disparities in utilization of women-owned firms for some categories of contracts. For example, the disparity index for white women on aggregated state funded contracts was 48. After publication of the disparity report, BBC calculated disparity indices for all women-owned firms, including female minorities. The results showed substantial disparities in the utilization of all women-owned firms similar to those measured for white women. For some subcategories of contracts, there was no substantial disparity for white women- owned firms.

Caltrans and the research firm also gathered extensive anecdotal evidence by (1) conducting twelve public hearings to receive comments on the firm’s findings; (2) receiving letters from business owners and trade associations; and (3) interviewing representatives from twelve trade associations and 79 owners/managers of transportation firms. Some of the anecdotal evidence indicated discrimination based on race or gender. 10 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

2. Design of Caltrans’ Affirmative Action Program

Caltrans concluded that the evidence supported an inference of discrimination in the California transportation contracting industry. Specifically, Caltrans concluded that it had sufficient evidence to make race- and gender-conscious goals for African American-, Asian-Pacific American-, Native American-, and women-owned firms. Caltrans adopted the recommendations of the disparity report and set an overall goal of 13.5% for disadvantaged business participation. Caltrans expected to meet one-half of the 13.5% goal using race-neutral measures.

On November 14, 2007, Caltrans submitted its proposed program to the U.S. Department of Transportation for approval. The proposal included a request for a waiver to implement the program only for the four identified groups. The program included 66 race-neutral measures that Caltrans already operated or planned to implement. Subsequent proposals increased the number of race-neutral measures to 150.

On August 7, 2008, the Department of Transportation granted the waiver, but did not approve Caltrans’ program. On April 2, 2009, the Department of Transportation approved Caltrans’ program designed for fiscal year 2009.

D. District Court Proceedings

On June 11, 2009, AGC filed a complaint alleging that Caltrans’ implementation of the Act violated inter alia the Fourteenth Amendment of the U.S. Constitution, Title VI of the Civil Rights Act, and Article I, section 31 of the

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 11

California Constitution. On December 17, 2010, AGC voluntarily dropped its state constitutional claim, and only argued an as-applied challenge to Caltrans’ affirmative action program. On December 23, 2009, the district court granted a motion filed by the Coalition for Economic Equity and the NAACP, San Diego Chapter to intervene as defendants.

Following discovery, all parties filed summary judgment motions. On March 23, 2011, the district court granted Caltrans’ and Intervenors’ motions and denied AGC’s motion. The court held that Caltrans’ program was “clearly constitutional,” as it was supported by a strong basis in evidence of discrimination in the California contracting industry and was narrowly tailored to those groups which had actually suffered discrimination.

E. Subsequent Caltrans Program

While this appeal was pending, Caltrans commissioned a new disparity study from the research firm to update its preference program as required by the federal regulations. See 49 C.F.R. § 26.45(f)(1)(i). On August 31, 2012, the research firm published its second disparity report. Caltrans concluded that the updated study provided evidence of continuing discrimination in the California transportation contracting industry against African Americans, Native Americans, Asian-Pacific Americans, Hispanic Americans, and women. Caltrans submitted a modified disadvantaged business enterprise program that is nearly identical to the program approved in 2009, except that it now includes Hispanic Americans and sets an overall goal of 12.5%, of which 9.5% will be achieved through race- and gender-conscious measures. On November 29, 2012, the U.S. Department of Transportation approved Caltrans’ updated program. 12 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

II. JURISDICTION

Before reaching the merits, we must determine whether we have jurisdiction over AGC’s appeal. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 94–95 (1998). Caltrans argues that jurisdiction is precluded by the doctrines of mootness and standing.

A. Mootness

In a letter filed shortly before oral argument, Caltrans contends that many issues raised in AGC’s appeal are moot because Caltrans has enacted a new affirmative action program since AGC filed its appeal.

The Supreme Court rejected a similar argument in Northeastern Florida Chapter of Associated General Contractors of America v. City of Jacksonville, 508 U.S. 656 (1993). In Northeastern Florida, the plaintiff challenged a Jacksonville ordinance establishing a disadvantaged business enterprise program granting race-based preferences in public contracting. Id. at 658–59. Shortly after the Supreme Court granted certiorari, Jacksonville repealed the ordinance and replaced it with a substantially similar one. The Court held that the original case was not moot because “a defendant’s voluntary cessation of a challenged practice does not deprive a federal court of its power to determine the legality of the practice.” Id. at 661–62 (quoting City of Mesquite v. Aladdin’s Castle, Inc., 455 U.S. 283, 289 (1982)). In language equally applicable to the instant case, the Court reasoned that there was not merely a risk that the government would repeat the challenged conduct, it had already done so by instituting a substantially similar preference program. Id. at 662. Similarly, the appeal in the instant case is not moot.

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 13

Caltrans’ new preference program is substantially similar to the prior program and is alleged to disadvantage AGC’s members “in the same fundamental way” as the previous program. See id.

B. Standing

To establish associational standing, AGC must show:

(a) its members would otherwise have standing to sue in their own right;

(b) the interests it seeks to protect are germane to the organization’s purpose; and

(c) neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.

Associated Gen. Contractors of Am. v. Metro. Water Dist. of S. Cal., 159 F.3d 1178, 1181 (9th Cir. 1998) (citation omitted); see also Summers v. Earth Island Inst., 555 U.S. 488, 494 (2009). The second and third prongs are not at issue here. See Metro. Water, 159 F.3d at 1181 (concluding that AGC’s lawsuit was germane to its purpose and that the injunctive and declaratory relief sought did not require the individual participation of members).

To meet the first prong, AGC must show that a member suffers an injury-in-fact that is traceable to the defendant and likely to be redressed by a favorable decision. See Braunstein v. Arizona Dep’t of Transp., 683 F.3d 1177, 1184 (9th Cir. 2012); see also Summers, 555 U.S. at 494; Lujan v. Defenders of Wildlife, 504 U.S. 555, 573 (1992). AGC’s effort to prove 14 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

the requisite injury to a member requires, first, “specific allegations establishing that at least one identified member had suffered or would suffer harm.” Summers, 555 U.S. at 498 (emphasis added). The “requirement of naming the affected members has never been dispensed with in light of statistical probabilities.” Id. at 498–99. Moreover, on summary judgment, AGC was required to submit competent evidence, not mere allegations, to demonstrate that at least one of its members had standing. See Lujan, 504 U.S. at 561; see also Metro. Water, 159 F.3d at 1181 (holding that AGC established associational standing where AGC’s standing argument was supported by affidavits from member companies showing harm).

AGC fails to meet this standard. AGC does not identify any affected members by name nor has it submitted declarations by any of its members attesting to harm they have suffered or will suffer under Caltrans’ program. The only evidence AGC relies on to satisfy standing is a declaration from James Ryan, AGC’s Executive Vice President, and many relevant portions of the declaration were struck from the record by the district court in an evidentiary ruling that AGC does not challenge. In any event, the Ryan declaration does not name any specific members of AGC who would be harmed by Caltrans’ program.

At oral argument, AGC contended that the general allegations in its complaint asserting that its members would suffer harm is sufficient to establish standing under Northeastern Florida. AGC’s reliance on Northeastern Florida is misplaced. In Northeastern Florida, standing was upheld based on uncontested allegations in a verified complaint that the plaintiff’s members suffered the requisite harm. See 508 U.S. at 668–69. Because the allegations were

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 15

not challenged, the Court reasoned that it had to accept them as true. Id. In contrast, Caltrans disputes AGC’s allegations and undermined any evidentiary support that AGC offered to substantiate those allegations. Additionally, AGC concedes that unlike Northeastern Florida, its complaint was not verified. An unverified complaint cannot form the basis of evidence considered at summary judgment. Moran v. Selig, 447 F.3d 748, 759–60 (9th Cir. 2006).

Because AGC has failed to establish standing, we must DISMISS the appeal for lack of jurisdiction. See Summers, 555 U.S. at 492–93.

III.

Further, even if AGC could establish standing, its appeal would fail. Caltrans’ affirmative action program is constitutional, so long as it survives the applicable level of scrutiny required by Equal Protection jurisprudence.

Race-conscious remedial programs must satisfy strict scrutiny. Western States, 407 F.3d at 990 (citing Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 227 (1995) (Adarand III)). Racial classifications survive strict scrutiny if they are “narrowly tailored measures that further compelling governmental interests.” Id. “The burden of justifying different treatment by ethnicity . . . is always on the government.” Id. Although stringent, strict scrutiny is not “fatal in fact.” Adarand III, 515 U.S. at 237. “The unhappy persistence of both the practice and the lingering effects of racial discrimination against minority groups in this country is an unfortunate reality, and government is not disqualified from acting in response to it.” Id. 16 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

Gender-conscious programs must satisfy intermediate scrutiny. Western States, 407 F.3d at 990 n.6. Intermediate scrutiny requires that gender-conscious programs be “supported by an ‘exceedingly persuasive justification’ and substantially related to the achievement of that underlying objective.” Id. (quoting United States v. Virginia, 518 U.S. 515, 524 (1996)).

Caltrans’ affirmative action program contains both race- and gender-conscious measures. The entire program passes strict scrutiny. It is therefore unnecessary to undertake a separate analysis under intermediate scrutiny. See id.

A. Application of Strict Scrutiny Standard Articulated in Western States

The framework for AGC’s as-applied challenge to Caltrans’ affirmative action program is governed by Western States, 407 F.3d at 995–1002. Western States held that the state “need not demonstrate an independent compelling interest for its [affirmative action] program” because the state’s program rested upon the compelling nationwide interest identified by Congress in passing the federal statute. Id. at 997. The Court then devised a two-prong test for narrow tailoring: (1) the state must establish the presence of discrimination within its transportation contracting industry, and (2) the remedial program must be “limited to those minority groups that have actually suffered discrimination.” Id. at 997–99.

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 17

1. Evidence of Discrimination in California Contracting Industry

In Equal Protection cases, courts consider statistical and anecdotal evidence to identify the existence of discrimination. E.g., Western States, 407 F.3d at 991; Adarand Constructors, Inc. v. Slater, 228 F.3d 1147, 1166 (10th Cir. 2000). The Supreme Court has suggested that a “significant statistical disparity” could be sufficient to justify race-conscious remedial programs. City of Richmond v. J.A. Croson Co., 488 U.S. 469, 509 (1989). Although generally not sufficient, anecdotal evidence complements statistical evidence because of its ability to bring “the cold numbers convincingly to life.” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 339 (1977).

Western States concluded that Washington’s affirmative action program was not supported by sufficient evidence. Id. at 999–1002. Washington had performed no statistical studies and offered no anecdotal evidence. Id. at 1000–01. Instead, Washington merely compared the availability of disadvantaged businesses to the percentage of contract dollars awarded to such businesses. Id. at 1000. The Court criticized Washington’s oversimplified methodology, and gave little weight to the purported disparity because Washington’s data did not account for the relative capacity of disadvantaged businesses to perform work, nor did it control for the fact that existing affirmative action programs skewed the prior utilization of minority businesses in the state. Id. The Court struck down Washington’s program after determining that the record was “devoid of any evidence suggesting that minorities currently suffer—or have ever suffered— discrimination in the Washington transportation contracting industry.” Id. at 1002. 18 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

In contrast, Caltrans’ affirmative action program is supported by substantial statistical and anecdotal evidence of discrimination in the California transportation contracting industry. The 2007 disparity study documented disparities in the utilization of African American-, Native American-, Asian-Pacific American-, and women-owned firms in many categories of transportation contracts. The study accounted for the factors mentioned in Western States as well as others, adjusting availability data based on capacity to perform work and controlling for previously administered affirmative action programs. See Western States, 407 F.3d at 1000. Moreover, the statistical evidence from the disparity study is bolstered by anecdotal evidence supporting an inference of discrimination. The substantial statistical disparities alone would give rise to an inference of discrimination, see Croson, 488 U.S. at 509, and certainly Caltrans’ statistical evidence combined with anecdotal evidence passes constitutional muster.

AGC urges that strict scrutiny requires Caltrans to provide evidence of “specific acts” of “deliberate” discrimination by Caltrans employees or prime contractors, which Caltrans has failed to do. AGC derives this purported rule from Croson’s requirement that race-conscious measures be limited to address discrimination that the state has identified “with some specificity.” 488 U.S. at 504.

AGC reads Croson too broadly. Croson explicitly states that “[t]he degree of specificity required in the findings of discrimination . . . may vary.” Id. at 489 (quotation marks and citation omitted). Moreover, a rule requiring the state to show specific acts of deliberate discrimination by identified individuals would run contrary to the statement in Croson that statistical disparities alone could be sufficient to support

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 19

race-conscious remedial programs. Id. at 509. This Court has previously rejected a similar interpretation of Croson. See Associated Gen. Contractors of Cal., Inc. v. Coal. for Econ. Equity, 950 F.2d 1401, 1416 n.11 (9th Cir. 1991) (AGCC II) (rejecting AGC’s attempt to “seek[ ] to have us engraft the framework for title VII . . . onto Croson’s equal protection framework.”). We reject AGC’s argument that Caltrans’ program does not survive strict scrutiny because the disparity study does not identify individual acts of deliberate discrimination.

Second, AGC argues that the 2007 disparity study shows inconsistent results for utilization of minority businesses depending on the type and nature of the contract, and thus cannot support an inference of discrimination in the entire transportation contracting industry. Specifically, AGC asserts that the statistical results vary depending on whether the contracts at issue are prime or subcontracts, and within each of those categories, results differ as between construction and engineering contracts. AGC appears to contend that each of these subcategories of contracts must be viewed in isolation when considering whether an inference of discrimination arises.

AGC’s argument overlooks the rationale underpinning the constitutional justification for remedial race-conscious programs: they are designed to root out “patterns of discrimination.” Croson, 488 U.S. at 504. The issue is not whether Caltrans can show underutilization of disadvantaged businesses in every measured category of contract. Rather, Caltrans can meet the evidentiary standard required by Western States if, looking at the evidence in its entirety, the data show substantial disparities in utilization of minority firms suggesting that public dollars are being poured into “a 20 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

system of racial exclusion practiced by elements of the local construction industry.” Id. at 492.

The 2007 disparity study and anecdotal evidence document a pattern of disparities for African American-, Native American-, Asian-Pacific American-, and women- owned firms. The study found substantial underutilization of these groups in numerous categories of California transportation contracts, which the anecdotal evidence confirms. This is sufficient to enable Caltrans to infer that these groups are systematically discriminated against in publicly-funded contracts.

Third, AGC contends that the anecdotal evidence has little or no probative value in identifying discrimination because it is not verified. AGC cites to no controlling authority for a verification requirement. Both the Fourth and Tenth Circuits have rejected the need to verify anecdotal evidence. Rowe, 615 F.3d at 249; Concrete Works of Colo., Inc. v. City & Cnty. of Denver, 321 F.3d 950, 989 (10th Cir. 2003). AGC makes no persuasive argument that we should hold otherwise.

AGC also discounts the anecdotal evidence because some accounts ascribe minority underutilization to factors other than overt discrimination, such as difficulties with obtaining bonding and breaking into the “good ole boy” network of contractors. However, federal courts and regulations have identified precisely these factors as barriers that disadvantage minority firms because of the lingering effects of discrimination. See, e.g., Western States, 407 F.3d at 992; AGCC II, 950 F.2d at 1414; 49 C.F.R. § 26.45(d)(2)(i). Morever, AGC ignores the many incidents of racial and gender discrimination presented in the anecdotal evidence.

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 21

Caltrans does not claim, and the anecdotal evidence does not need to prove, that every minority-owned business is discriminated against. It is enough that the anecdotal evidence supports Caltrans’ statistical data showing a pervasive pattern of discrimination. See AGCC II, 950 F.2d at 1414. The individual accounts of discrimination offered by Caltrans and Intervenors meet this burden.

Fourth, AGC contends that Caltrans’ evidence does not support an inference of discrimination against all women because gender-based disparities in the 2007 study are limited to white women. AGC misunderstands the statistical techniques used in the disparity study. The 2007 report correctly isolates the effect of gender by limiting its data pool to white women, ensuring that statistical results for gender- based discrimination are not skewed by discrimination against minority women on account of their race. The original disparity report discusses this standard social science technique and explains that “[e]vidence of discrimination against white women-owned firms should be considered evidence of discrimination against women of any race.” Moreover, after AGC’s early objections to the methodology, the research firm conducted a follow-up analysis of all women-owned firms, which produced a disparity index of 59. This index is evidence of a substantial disparity that raises an inference of discrimination and is sufficient to support Caltrans’ decision to include all women in its affirmative action program. See Rowe, 615 F.3d at 243–44.

2. Program Tailored to Groups Who Actually Suffered Discrimination

The second prong of the test articulated in Western States requires that an affirmative action program be limited to those 22 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

groups that actually suffered discrimination in the state’s transportation contracting industry. 407 F.3d at 998–99. When explaining the justification for the second prong, this Court seemed primarily concerned with the “random inclusion of racial groups” in affirmative action programs. Id. at 998 (quoting Croson, 488 U.S. at 506). For example, the Court cited the quota system in Richmond, Virginia, at issue in Croson, which encompassed U.S. citizens who were “Blacks, Spanish-speaking, Orientals, Indians, Eskimos, and Aleuts.” Id. (quoting Croson, 488 U.S. at 478). In Croson, the Supreme Court reasoned that lumping together random racial groups—particularly Aleuts and Eskimos, for which there was no evidence of discrimination in the Richmond construction industry—suggested that the city’s program was not intended to remedy past discrimination. Croson, 488 U.S. at 506. After discussing Croson, the Court in Western States cited to numerous other cases striking down preference programs that included racial groups for which there was no evidence that they had actually been discriminated against in the relevant industry. Western States, 407 F.3d at 998–99 (citing cases).

The issue is whether the Caltrans’ affirmative action program “is limited to those minority groups that have actually suffered discrimination.” Id. at 998. It is. The 2007 disparity study showed systematic and substantial underutilization of African American-, Native American-, Asian-Pacific American-, and women-owned firms across a range of contract categories. These disparities support an inference of discrimination against those groups. See Rowe, 615 F.3d at 243–44. Caltrans concluded that the statistical evidence did not support an inference of a pattern of discrimination against Hispanic or Subcontinent Asian Americans. Although the federal regulations generally do not

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 23

allow states to create affirmative action programs that benefit some groups but not others, California applied for and received a waiver in order to limit its 2009 preference program to African American, Native American, Asian- Pacific American, and women-owned firms. Caltrans’ program adheres precisely to the narrow tailoring requirements of Western States.

AGC contends that the program is not narrowly tailored because it creates race-based preferences for all transportation-related contracts, rather than distinguishing between construction and engineering contracts. However, AGC cites to no case that requires a state preference program to provide separate goals for disadvantaged business participation on construction and engineering contracts. To the contrary, the federal guidelines for implementing the Act instruct states not to separate different types of contracts. There are sound policy reasons to not require such parsing, including the fact that there is substantial overlap in firms competing for construction and engineering contracts, as prime and subcontractors. See N. Contracting, Inc. v. Illinois, 473 F.3d 715, 723 (7th Cir. 2007) (explaining that “[i]t would make little sense to separate prime contractor[s] and subcontractor[s]” because the same firms compete for both types of contract).

B. Consideration of Race-Neutral Alternatives

Additionally, AGC asserts that Caltrans’ program is not narrowly tailored because it failed to evaluate race-neutral measures before implementing the system of racial preferences. The law, however, imposes no such requirement. First, Western States does not require states to independently meet this aspect of narrow tailoring, and instead focuses on 24 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

whether the federal statute sufficiently considered race- neutral alternatives. 407 F.3d at 995, 997–98. Second, even if this requirement does apply to Caltrans’ program, narrow tailoring only requires “serious, good faith consideration of workable race-neutral alternatives[.]” Grutter v. Bollinger, 539 U.S. 306, 339 (2003). The Caltrans program has considered an increasing number of race-neutral alternatives, starting at 45 in 2008 and reaching 150 in 2010. We reject AGC’s claim that Caltrans’ program does not sufficiently consider race-neutral alternatives.

C. Certification Affidavits for Disadvantaged Business Enterprises

AGC argues that Caltrans’ program is not narrowly tailored because affidavits that applicants must submit to obtain “disadvantaged business enterprise” certification do not require applicants to assert that they have suffered discrimination in California. AGC relies on language in Western States criticizing similar affidavits for not providing “any evidence of discrimination within Washington’s transportation contracting industry.” 407 F.3d at 1002. AGC takes the statements from Western States out of context. Western States did not criticize the affidavits for their role in the certification process; it chastised Washington for using the affidavits to serve a completely different purpose: to demonstrate the existence of discrimination within Washington transportation industry. Id. at 1001–02. Caltrans completed its own comprehensive disparity study and does not rely on the certification affidavits for this purpose.

The certification process employed by Caltrans follows the process detailed in the federal regulations. See 49 C.F.R. §§ 26.67(a)(1), 26.83(c)(7)(ii). To the extent that AGC

A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP . 25

contends that Caltrans’ program is over-inclusive because the certification form does not require that minority firms attest to the fact that they have been discriminated against in California, this is an impermissible collateral attack on the facial validity of the federal Act and regulations. See N. Contracting, 473 F.3d at 722 (plaintiff “cannot collaterally attack the federal regulations through a challenge to [a state’s affirmative action] program”).

D. Application of Program to Mixed State and Federally Funded Contracts

AGC challenges Caltrans’ application of its affirmative action program to transportation contracts funded by both federal and state money. This is another impermissible collateral attack on the federal program, which explicitly requires goals to be set for mix-funded contracts. See 49 C.F.R. § 26.45 (recipients “must set an overall goal for DBE participation in your DOT-assisted contracts”); id. § 26.5 (defining DOT-assisted contracts as any contract “funded in whole or in part with DOT financial assistance”); see also N. Contracting, 473 F.3d at 722 (no collateral attacks on federal regulations in challenge to state program).

IV. CONCLUSION

AGC did not identify any of its members that would be harmed by Caltrans’ affirmative action program. AGC has failed to establish standing. Further, Caltrans’ program survives strict scrutiny by 1) having a strong basis in evidence of discrimination within the California transportation 26 A SSOC . G EN . C ONTRACTORS V . C AL . D EP ’ T OF T RANSP .

contracting industry and 2) being narrowly tailored to benefit only those groups that have actually suffered discrimination.

DISMISSED.