MARK L. GROSS
LISA J. STARK
Department of Justice
P.O. Box 66078
Washington, D.C. 20035-6078
STATEMENT OF RELATED CASES
ISSUE PRESENTED 1
INTEREST OF THE UNITED STATES 2
SUMMARY OF ARGUMENT 7
I. THE DISTRICT COURT'S ANALYSIS IS
LEGALLY INCORRECT 9
A. The District Court's Decision Is Contrary
To Legal Precedent 9
B. On This Record, The District Court Erred
In Concluding That Denver's Statistics
Were Defective For Failing To Account
For Certain Variables 24
1. Size and Age of the Firm 24
2. Firm Specialization 33
3. Qualifications of a Firm 36
4. Prequalification 38
5. Number of Certified M/WBEs 40
C. The District Court's Description Of The
Necessary Bases For The City's Plan
Misinterprets Precedent 42
TABLE OF CONTENTS (CON'T) PAGE
STATEMENT REGARDING ORAL ARGUMENT
CERTIFICATE OF COMPLIANCE
CERTIFICATE OF SERVICE
TABLE OF AUTHORITIES PAGE
Adarand Constructors, Inc. v. Pena, 515 U.S. 200 (1995) 10
Allen v. Seidman, 881 F.2d 375 (7th Cir. 1989) 16
Appalachian Power Co. v. EPA, 135 F.3d 791
(D.C. Cir. 1998) 38
Bazemore v. Friday, 478 U.S. 385 (1986) 10, 11, 12
Berger v. Iron Workers Reinforced Rodmen Local 201,
843 F.2d 1395 (D.C. Cir. 1988) 15
Bullington v. United Airlines, Inc., 186 F.3d
1301 (10th Cir. 1999) 13, 15, 22
Carroll v. Sears, Roebuck & Co., 708 F.2d 183
(5th Cir. 1983) 17
Catlett v. Missouri Highway and Transport
Commission, 828 F.2d 1260 (8th Cir. 1987),
cert. denied, 485 U.S. 1021 (1988) passim
City of Richmond v. J.A. Croson, 488 U.S. 469
Coates v. Johnson & Johnson, 756 F.2d 524
(7th Cir. 1985) 16, 19, 27
Coble v. Hot Springs Sch. District No. 6, 682 F.2d
721 (8th Cir. 1982) 17
Concrete Works of Colo., Inc. v. Denver, 36 F.3d 1513
(10th Cir. 1994), cert. denied, 514 U.S. 1004
Concrete Works of Colo., Inc. v. Denver, 823 F. Supp.
821 (D. Colo., Feb. 26, 1993) 4
Concrete Works of Colo., Inc. v. Denver, 86 F. Supp.
2d 1042 D. Colo. Mar. 7, 2000) passim
Contractors Association v. Philadelphia, 6 F.3d 990
(3d Cir. 1993) 13, 22
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S.
579 (1993) 38
Dayton Bd. of Educ. v. Brinkman, 443 U.S. 526
Demedina v. Reinhardt, 686 F.2d 997
(D.C. Cir. 1982) 18
Dothard v. Rawlison, 433 U.S. 321 (1977) 12, 13, 48
EEOC v. General Tel. Co., 885 F.2d 575 (9th Cir.),
cert. denied, 498 U.S. 950 (1989) 13, 14, 23
EEOC v. Local 638, 81 F.3d 1162
(2d Cir. 1996) 16, 34, 38
EEOC v. Olsons Dairy Queens, Inc., 989 F.2d 165
(5th Cir. 1993) 16, 34
EEOC v. Sears, Roebuck & Co., 839 F.2d 302
(7th Cir. 1988) 14
Franks v. Bowman, 424 U.S. 747 (1976) 47
Fullilove v. Klutznick, 448 U.S. 448
(1980) 10, 44, 45
Hazelwood Sch. District v. United States, 433 U.S.
299 (1977) 11
James v. Stockham Valves & Fittings Co., 559 F.2d
310 (5th Cir. 1977), cert. denied, 434 U.S.
1034 (1978) 19, 27
Lau v. Nichols, 414 U.S. 563 (1974) 48
McClesky v. Kemp, 481 U.S. 279 (1987) 12
Ottavian v. State University, 875 F.2d 365
(2d Cir. 1989), cert. denied, 493 U.S. 1021
Palmer v. Shultz, 815 F.2d 84
(D.C. Cir. 1987) 14, 16, 23
Segar v. Smith, 738 F.2d 1249 (D.C. Cir.),
cert. denied, 471 U.S. 1115 (1985) 17, 18, 35
Smith v. Virginia Commonwealth Univ., 84 F.3d 672
Sobel v. Yeshiva University, 839 F.2d 18
(2d Cir. 1988), cert. denied, 490 U.S.
1105 (1989) 13, 15, 16, 23
Teamsters v. United States, 431 U.S. 324
Texas Department of Cmty. Affairs v. Burdine,
450 U.S. 248 (1981) 12
Trout v. Lehman, 702 F.2d 1094 (D.C. Cir.),
vacated and remanded on other grounds,
465 U.S. 1056 (1984) 18
Valentino v. United States Postal Service,
674 F.2d 56 (D.C. Cir. 1982) 19, 27
The Court has previously issued a decision in this case. Concrete Works of Colorado, Inc. v. Denver, 36 F.3d 1513 (10th Cir. 1994).
The United States' brief will address the following issue:
Whether the district court erred in presuming, without supporting empirical data, that numerous statistical studies, demonstrating a substantial underutilization of minority- and women- owned businesses (M/WBEs) as compared to their numbers in the local market, were unreliable merely because the studies failed to account for certain variables.
This case addresses the type of record a locality must present in order to adopt an affirmative action program in contracting. The federal government itself operates important affirmative action contracting programs, both in direct federal procurement and in projects funded in part by the Department of Transportation (DOT) or Environmental Protection Agency (EPA). In both instances, when these programs are challenged, the United States must demonstrate that it has a compelling interest in remedying the effects discrimination has had on the ability of minority-owned and, in the case of the DOT and EPA programs, women-owned firms, to participate in contracting on an equal basis. In many of these cases, the United States relies on statistical evidence similar to the disparity studies Denver offered to support its local program. Thus, the manner in which Denver's statistical evidence is analyzed, here, may impact how such evidence is evaluated with regard to a federal program.
In addition, the analysis in this case could impact the use of statistical evidence in discrimination cases in general. Should the district court's analysis rejecting Denver's statistical studies be sustained by this Court, then virtually every disparity study demonstrating discrimination, no matter how carefully done, will be in jeopardy.
In 1990, Denver, Colorado, adopted an affirmative action program for women and minorities with regard to certain public construction contracts. In January 1992, Concrete Works, a nonminority male-owned prime contractor filed suit challenging the constitutionality of the program. In February 1993, after extensive discovery, the district court (Chief Judge Finesilver) granted Denver's motion for summary judgment. Concrete Works of Colo., Inc. v. Denver, 823 F. Supp. 821 (D.Colo., Feb. 26, 1993). The court held that Denver's program satisfied the strict scrutiny standard set forth in City of Richmond v. J.A. Croson, 488 U.S. 469 (1989), because Denver's statistical and anecdotal evidence demonstrated a compelling interest and the program was narrowly tailored to further that interest.
On appeal, this Court reversed and remanded the case to the district court. Concrete Works of Colo., Inc. v. Denver, 36 F.3d 1513 (1994), cert. denied, 514 U.S. 1004 (1995). It concluded that although Denver had met its initial burden with regard to establishing a "strong basis in evidence" for its remedial program by offering several statistical studies showing an underutilization of minority- and women-owned businesses (M/WBEs) on the basis of their "absolute number[s] in the local marketplace," Concrete Works had raised several issues of fact that warranted a trial. 36 F.3d at 1522, 1528. This Court explained that while Denver's studies support its remedial program, Concrete Works had identified issues about the accuracy of Denver's data. Id. at 1528.
This Court refused to address the issue of whether Denver's program was narrowly tailored because it had not been preserved on appeal. It explained "[b]ecause Concrete Works does not challenge the district court's conclusion with respect to the second prong of Croson's strict scrutiny standard -- i.e. that the Ordinance is narrowly tailored to remedy past and present discrimination -- we need not address this issue[.]" Id. at 1531 n.24.
On remand, the district court, with a new judge assigned to the case, held a trial. Denver presented data from at least six statistical studies as well as expert opinion and anecdotal evidence demonstrating that M/WBES are underutilized when compared with their numbers in the local marketplace. In rebuttal, Concrete Works presented two experts who claimed, without supporting empirical evidence, that Denver's data were deficient because the studies failed to account for several variables including size and age of the firm, the firm's specialization, whether the firm satisfied prequalification standards, and the number of certified M/WBEs in the State.
On March 7, 2000, the district court held that the Ordinance was unconstitutional. Concrete Works of Colo., Inc. v. Denver, 86 F. Supp. 2d 1042 (D. Colo. Mar. 7, 2000). It concluded that because Denver's statistical studies do not account for "important variables * * * [they] do not generate a fair inference that there are discriminatory barriers to participation in the construction industry[.]" 86 F. Supp. 2d at 1070. The district court rejected the disparity studies and anecdotal evidence because they did not address actual qualifications and capacities of the MBEs and WBEs, did not take account of a firm's specialization, its experience, prequalification status, and the number of certified M/WBEs. Id. at 1065. The court also concluded that the program was not narrowly tailored.
The district court erred in rejecting Denver's disparity studies merely because the City allegedly failed to account for certain variables. The court's analysis is contrary to precedent, inconsistent with the evidence, and fundamentally flawed. Indeed, the district court erroneously presumed, without empirical support, that certain excluded variables caused the statistical disparities.
The district court's decision imposes a nearly insurmountable and unreasonable burden on a party seeking to support the use of affirmative action. It allows a party to defeat a prima facie showing of discrimination based on detailed statistical evidence by merely claiming, without supporting data, that unaccounted for variables caused the observed disparities, even when its own experts admit that statistics regarding many of those variables are currently unavailable and cannot be readily obtained because the variables are too subjective to measure. Consequently, it prevents a locality implementing a remedial preference program to rely, as does Denver, on data reflecting the number of M/WBEs in the marketplace, and requires the locality, regardless how impossible, to address any and all variables, regardless of their proven relevance.
I. THE DISTRICT COURT'S ANALYSIS IS LEGALLY INCORRECT
A. The District Court's Decision Is Contrary To Legal Precedent
In City of Richmond v. J.A. Croson, 488 U.S. 469 (1989), the Supreme Court struck down a city ordinance requiring every nonminority prime contractor to subcontract at least 30% of the value of city contracts to minority business enterprises (MBEs); the decision set forth the framework for evaluating the constitutionality of a race-conscious affirmative action program in public contracting. It explained that, under strict scrutiny, racial classifications are constitutional so long as they serve a compelling government interest and are "narrowly tailored" to further that interest. Id. at 493. It also was careful to point out that strict scrutiny does not automatically abolish all affirmative action programs. Id. at 519. See also Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 237 (1995), quoting Fullilove v. Klutznick, 448 U.S. 448, 507 (1980) ("we wish to dispel the notion that strict scrutiny is 'strict in theory, but fatal in fact.'")
In Croson, the Court also made clear that a city may rely exclusively on statistical evidence to establish the "strong basis in evidence" necessary to justify a race or gender conscious remedial contracting program. 488 U.S. at 500-501. The Court stated, "[w]here there is a significant statistical disparity between the number of qualified minority contractors willing and able to perform a particular service and the number of such contractors actually engaged by the locality or the locality's prime contractors, an inference of discriminatory exclusion could arise." Croson, 488 U.S. at 509. See also Bazemore v. Friday, 478 U.S. 385, 398 (1986); Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307-308 (1977); Teamsters v. United States, 431 U.S. 324, 337-339 (1977).
In Bazemore, the Supreme Court addressed the validity of statistical evidence that establishes a significant disparity supporting an inference of discrimination, but fails to consider all possible variables. The Supreme Court reversed the lower courts' refusals to accept plaintiffs' multiple regression analysis as proof of racial discrimination in employee salary levels. Noting that discrimination need "not [be] prove[d] * * * with scientific certainty," the Court flatly rejected the conclusion that "[a]n appropriate regression analysis * * * should * * * include all measurable variables thought to have an effect[.]" 478 U.S. at 399, 400 (emphasis in the original and internal quotation marks omitted). It also criticized defendants' trial strategy in which there was "no attempt * * * -- statistical or otherwise -- to demonstrate that when the[ ] [omitted] factors were properly organized and accounted for there was no significant disparity." Id. at 400, 403 n.14.
Accordingly, Bazemore establishes that a party's statistical evidence may prove discrimination so long as it accounts for the major measurable factors causing the racial disparity. See McClesky v. Kemp, 481 U.S. 279, 327-328 (1987) (Brennan, J., dissenting) (explaining that in Bazemore, the Court held that "a multiple regression analysis need not include every conceivable variable to establish a party's case, as long as it includes those variables that account for the major factors that are likely to influence decisions"); Texas Dep't of Cmty. Affairs v. Burdine, 450 U.S. 248, 254 (1981) (prima facie case "eliminates the most common nondiscriminatory reason for the employment decision); Dothard v. Rawlison, 433 U.S. 321, 331 (1977) (party establishing discrimination is "not required to exhaust every possible source of evidence [where] * * * [there] are fallacies or deficiencies in the data * * * [opposing party] is free to adduce countervailing evidence of [its] own.") See e.g., Bullington v. United Airlines, Inc., 186 F.3d 1301, 1314-1315 (10th Cir. 1999).
Nearly every circuit interpreting Bazemore has recognized that an inference of discrimination arising from statistical evidence cannot be rebutted merely by pointing out unaccounted for variables that might conceivably affect the analysis. Bullington, 186 F.3d at 1315; Contractors Ass'n v. Philadelphia, 6 F.3d 990, 1007 (3d Cir. 1993); EEOC v. General Tel. Co., 885 F.2d 575, 582 (9th Cir.), cert. denied, 498 U.S. 950 (1989); Sobel v. Yeshiva Univ., 839 F.2d 18, 34 (2d Cir. 1988), cert. denied, 490 U.S. 1105 (1989); Catlett v. Missouri Highway and Transp. Comm'n, 828 F.2d 1260 (8th Cir. 1987), cert. denied, 485 U.S. 1021 (1988); Palmer v. Shultz, 815 F.2d 84, 101 (D.C. Cir. 1987). But see EEOC v. Sears, Roebuck & Co., 839 F.2d 302, 350 (7th Cir. 1988). Rather, to dispel the inference of discrimination, a party must introduce "evidence to support the contention that the missing factor can explain the disparities." General Tel. Co., 885 F.2d at 581, quoting Palmer, 815 F.2d at 101. Indeed, in Concrete Works of Colo., Inc. v. Denver, 36 F.3d 1513, 1524-1525 (1994), cert. denied, 514 U.S. 1004 (1995), this Court stated that Concrete Works' attack on empirical data "quantify[ing] the actual disparity between the utilization of minority contractors and their representation in the local construction industry" "unconvincing" since contractor "present[ed] no evidence to challenge the validity of [the] * * * allegations * * * [and] "fail[ed] to introduce evidence refuting the substance of [City's] information." Accord Bullington, 186 F.3d at 1315 (reversing grant of summary judgment in favor of employer in a case alleging gender discrimination in hiring where plaintiff's statistical evidence did not account for experience level, but employer "did not correlate * * * experience level * * * [to] show [impact on] statistics"); Berger v. Iron Workers Reinforced Rodmen Local 201, 843 F.2d 1395, 1416 (D.C. Cir. 1988).
For example, in Sobel, the Second Circuit, relying on Bazemore, held that a university failed to dispel the inference of gender discrimination in faculty salaries arising from plaintiffs' statistical evidence when its "experts simply criticized plaintiffs' failure to include [certain variables] offering no reason, in evidence or analysis, for concluding that [the factors] correlated with sex and therefore were likely to affect" the analysis. 839 F.2d at 34. It explained that a party seeking to show that an opponent's statistical evidence fails to account for certain factors "must actually demonstrate" that those factors make a difference and "weaken the showing of a * * * disparity made by the analysis" Ibid. See also similar discussions in EEOC v. Local 638, 81 F.3d 1162, 1173 (2d Cir. 1996); EEOC v. Olson's Dairy Queens, Inc., 989 F.2d 165, 169 (5th Cir. 1993); Catlett, 828 F.2d at 1266; Palmer, 815 F.2d at 106 n.19; and Allen v. Seidman, 881 F.2d 375, 379 (7th Cir. 1989).
Even before Bazemore, several courts of appeals recognized that statistical evidence need "not control for every conceivable factor" and refused to assume, absent credible evidence, that an unaccounted for variable skewed or provided a non-racial explanation for the statistical disparity. Coates v. Johnson & Johnson, 756 F.2d 524, 544 n.20 (7th Cir. 1985). See Carroll v. Sears, Roebuck & Co., 708 F.2d 183, 193 (5th Cir. 1983); Coble v. Hot Springs Sch. Dist. No. 6, 682 F.2d 721, 732 (8th Cir. 1982). For example, in Segar v. Smith, 738 F.2d 1249, cert. denied, 471 U.S. 1115 (1985), the D.C. Circuit rejected an expert's opinion that plaintiffs' statistical methodology was flawed for failure to take account of prior specialized criminal investigative work in determining whether an employer had discriminated on the basis of race in making promotions. It explained because "nothing in the record so much as hints that black special agents are less likely than white special agents to possess this qualification" and "[t]here is simply no reason to assume * * * that such [a] skill [is] unevenly distributed[.] * * * [P]laintiffs' failure to account for this variable does not dilute the force of their statistical analysis; in the language of a statistician, absent any reason to conclude the omitted factor correlates with race, the omission of the variable will not affect the * * * analysis." 738 F.2d at 1276-1277 (emphasis added). See, e.g., Demedina v. Reinhardt, 686 F.2d 997, 1008 (D.C. Cir. 1982). The D.C. Circuit has succinctly explained:
[U]nquantified, speculative and theoretical objections to the proffered statistics are to be given little weight. * * * When a plaintiff submits accurate statistical data and a defendant alleges that relevant variables are excluded, defendant may not rely on hypotheses to lessen the probative value of plaintiff's statistical proof. Rather, defendant * * * must either rework plaintiff's statistics incorporating the omitted factors or present other proof undermining plaintiff's claims.Trout v. Lehman, 702 F.2d 1094, 1102 (D.C. Cir.), vacated and remanded on other grounds, 465 U.S. 1056 (1984).
In addition, a party challenging statistical evidence demonstrating a race or sex-based disparity cannot claim that a variable which itself is a product of or tainted by discrimination was not included in the analysis. For instance in employment discrimination cases, courts have refused to reject a statistical analysis demonstrating discrimination that fails to take account of a variable such as grade level, job assignments, or performance evaluations where there is evidence that the variable is itself tainted by discrimination. See, e.g., Coates, 756 F.2d at 544; Valentino v. United States Postal Serv., 674 F.2d 56, 71, n.26 & 73 n.30 (D.C. Cir. 1982); James v. Stockham Valves & Fittings Co., 559 F.2d 310, 322 (5th Cir. 1977), cert. denied, 434 U.S. 1034 (1978). Cf. Ottavian v. State Univ., 875 F.2d 365, 375 (2d Cir. 1989), cert. denied, 493 U.S. 1021 (1990).
Applying this precedent, the district court erred when it rejected the City's studies that demonstrated race and sex-based disparities in the participation of M/WBEs in City contracts because the "statistical studies do not account for "important variables * * * [also they] do not generate a fair inference that there are discriminatory barriers to participation in the construction industry." Concrete Works of Colo., Inc. v. Denver, 86 F. Supp. 2d 1042, 1070, 1074 (D. Col., Mar. 7, 2000). Because in no instance does the record contain evidence that consideration of any unaccounted for variable would change the results or eliminate the observed race- or sex-based disparities, Denver's statistical evidence demonstrates discrimination sufficient to justify its affirmative action program.
The record contains no evidence to reject or discredit the inference of discrimination Denver's various statistical studies established by showing an underutilization of M/WBEs on the basis of their "absolute number[s] * * * in the local market." Concrete Works, 36 F.3d at 1528. In the prior appeal, this Court cited various supportive statistical studies, including federal reports from the 1970s, data from the Department of Public Works (DPW) 1979 Bond Projects, statistics regarding contracting by Department of General Services (DGS) in 1989-1990, DPW's non-goal projects during the 1970s and 1980s, and 1989 data showing a decline in the utilization of M/WBEs after the relaxation of goals project. Id. at 1524-1528. At trial on remand, Concrete Works offered no empirical data, independent analysis, or statistical evidence to show that the allegedly unaccounted for variables correlate with race or gender or cause the proven race and sex-based statistical disparities. While plaintiff's two experts, Drs. John Lunn and George LaNoue, criticized Denver's methodology for failing to evaluate certain variables, they did not state that inclusion of the allegedly excluded variables would reduce or otherwise explain the statistical disparities Denver's data established in non race- or sex-based terms. In fact, Dr. Lunn acknowledged that he did not do any quantitative analysis to verify or determine the adequacy of Denver's statistics or the reason for the statistical disparities, that he relied "entirely [on] made-up data" and that the disparities reflected in Denver's statistics may reflect and be due to discrimination, and exist even after controlling for the excluded variables (Tr. 780, 797, 801, 856-857, 859-860, 863-864, 871-872). (1) Accordingly, the district court erred in presuming that Denver's data was deficient for failure to account for certain variables because the record contains no evidence that the variables would have affected the City's results. Bullington, 186 F.3d at 1315; Contractors Ass'n, 6 F.3d at 1007; General Tel. Co., 885 F.2d at 582; Sobel, 839 F.2d at 34; Catlett, 828 F.2d at 1260; Palmer, 815 F.2d at 101.
Moreover, to date, we are aware of no court that has interpreted Croson to require a locality implementing a remedial affirmative action program to present data regarding the numerous variables the district court mentioned. As this Court noted in its initial opinion, several circuits "have interpreted Croson * * * to permit a municipality to rely, as does Denver, on general data reflecting the number of MBEs and WBEs in the marketplace to defeat the challenger's summary judgment motion of request for a preliminary injunction." Concrete Works, 36 F.3d at 1528 (citing to various cases).
To conclude otherwise imposes an insurmountable burden on a party seeking to prove a compelling interest to support the use of affirmative action. If plaintiff here can defeat Denver's prima facie showing of discrimination by merely identifying variables and claiming, without empirical support, that the unaccounted variables explain the disparity, then contrary to established precedent, statistical evidence can virtually never be an adequate predicate to establish discrimination.
B. On This Record, The District Court Erred In
Concluding That Denver's Statistics Were
Defective For Failing To Account For Certain Variables
1. Size and Age of the Firm. The district court erred in concluding that Denver's statistics show nothing more than the fact that M/WBEs tend to be smaller and less experienced than nonminority male-owned firms and thus less capable of performing contracts. See Concrete Works, 36 F.3d at 1528; Concrete Works, 86 F. Supp. 2d at 1068. On this record, Denver should not have been required to consider the size and age of M/WBEs for at least two independent reasons.
First, it hardly can be disputed that discrimination has and continues to affect the size of M/WBEs. As this Court noted in its initial opinion, Concrete Works "present[ed] no evidence to challenge the validity" of statistical data reflecting discrimination in public construction contracting in the Denver metro area as reflected in federal agency reports. Concrete Works, 36 F.3d at 1524. As a result, the "empirical data, which quantifies the actual disparity between the utilization of minority contractors and their representation in the local construction industry * * * support[s] Denver's contention that race and gender discrimination existed prior to the enactment of the challenged Ordinance." Id. at 1524-1525.
Moreover, at trial, Dr. Lunn explained "it would take a long time before you see significant changes" when discrimination has effectively barred a group for a long time from entering the market altogether (Tr. 789). Such changes are even slower and less evident in a market like the construction industry, where Drs. Lunn and La Noue readily acknowledged some prime contractors even today still insist on working only with "a set of subcontractors" they know personally and have worked with "for years" (Tr. 789, 793, 836, 865-866, 906-907). Indeed, Dr. Lunn admitted the differentials in size and revenues between M/WBEs and nonminority male-owned firms could be the result of discrimination (Tr. 866-867). Further, the district court found that new construction firms are at a "significant disadvantage" because "reasonable contractors will tend to do repeat business with those with whom they have had previous success" and that "the anecdotal evidence shows that race, ethnicity, and gender affect the construction industry and those who work in it." Concrete Works, 86 F. Supp. 2d at 1065, 1074. Consequently, even if M/WBE's smaller size and younger age affects their utilization, those factors cannot be relied upon to excuse the observed disparities because there is ample evidence in the record to demonstrate an M/WBE's size and age are products of and tainted by discrimination. See Coates, 756 F.2d at 544; Valentino, 674 F.2d at 71, 73 n.26 & n.30; James, 559 F.2d at 322.
Further, the fact that minority- and women-owned construction businesses are smaller and younger is irrelevant to statistics reflecting their underutilization since the record in this case demonstrates little relationship between the size of a construction firm and its ability to do work and obtain contracts. One defense expert testified that firm size is an "especially bad measure of qualifications in the construction industry" since it is "the result [and] not the cause of * * * success" (Tr. 484). Several witnesses explained that the Denver construction market is "dominated by small firms," that 60% of all construction firms have four or fewer employees, and that only 50, or less than one percent of construction firms, have more than 100 employees (Tr. 30, 244, 288, 290-291, 476). Data and testimony reflect that while "minority firms are smaller, they're not so much smaller that * * * they could [not] expand" to accommodate more work (Tr. 47-48, 512-513). In fact, there was so much testimony on the flexibility and elasticity of the construction market and the ability of "firms [to] go back and forth in terms of the type of work they do," the district court commented at one point, "there really isn't any dispute about the ability of a small firm to do a big contract" and found that "most firms have few full-time permanent employees and [thus] must grow or shrink their performance capacity according to the volume of business they are doing." Concrete Works, 86 F. Supp. 2d at 1064; (see Tr. 44-45, 196, 293-295, 323- 325, 345, 481-482). Accordingly, contrary to the court's findings, size or age is not a nondiscriminatory explanation for the statistical disparities reflecting an underutilization of M/WBEs.
In any event, Denver did not ignore the variables of size and age when considering whether discrimination caused the observed disparities. Rather, the City presented statistical evidence from four separate studies that focused on size and/or age of M/WBE construction firms and together demonstrate that the observed disparities still exist.
For example, defendants presented data regarding approximately one-third of the City's construction contracts in 1989 or 1990, or nearly 5,000 non-goal construction projects offered by DGS (Tr. 106, 363; Exs. C-8 Pt. II at 11-12, K-23 at 5). Even though the contracts were extremely small (with an average price of only $10,000), the data nonetheless "reveal extremely low MBE and WBE utilization," and "disparity indices * * * approximat[ing] those in other cases where courts have applied the "strong basis in evidence" standard in Croson" to uphold the plan. Concrete Works, 36 F.3d at 1526. See Concrete Works, 86 F. Supp. 2d at 1043; Tr. 109-110, 357-363, 452; Exs. C-8, O-7. (2)
Denver also introduced the results of two very large telephone surveys of construction and design firms in the Denver metro area conducted by a private independent company in 1990, and again in 1994 and 1995. See Concrete Works, 86 F. Supp. 2d. at 1054, 1057 (describing the surveys). (3) Both studies considered the size and age of the firms and found significant statistical disparities demonstrating an underutilization of M/WBEs when compared to nonminority male-owned firms. Exs. E-15, C-8. "With the exception of WBEs in the professional design industry, the results reflect statistically significant" disparities showing an underutilization of M/WBEs as compared with nonminority male-owned firms regardless of age and size. Concrete Works, 86 F. Supp. 2d at 1057; see Tr. 381-383; Ex. E-15 Pt. II at 13, 15). The 1990 study reflects that revenues per employee for construction MBES and WBEs were only 63% and 58%, respectively, of revenues per employee for all firms, "suggest[ing] that even among firms of the same employment size, industry utilization of MBEs and WBEs was lower than that of non- minority male owned firms." Ex. K-23 at 3.
Denver also presented the results of a 1997 mail and telephone follow-up survey sent to all minority- and women-owned construction firms and a sample of nonminority male-owned firms in the Denver metro area that also considered the age and size of the construction firm. See Concrete Works, 86 F. Supp. 2d at 1060; Tr. 540-542; Ex. L-1. (4) After accounting for how long a firm had been in existence and its size, the survey compared similarly situated minority and female- owned firms with white male-owned firms and found a "statistically significant disparity" adverse to MBEs and that M/WBEs have "more difficulties" than firms owned by white males of similar size (Tr. 378, 546-548). Thus, contrary to the suggestion of plaintiffs' experts, Denver's empirical evidence not only considered the potential effects of size and age of the firms at issue, but demonstrated that those variables do not explain the statistical disparities.
2. Firm Specialization. The district court also erred in rejecting Denver's statistical evidence for failing to take account of firm specialization and the type of work being offered. See Concrete Works, 86 F. Supp. 2d at 1054, 1064, 1068. In fact, the record is devoid of evidence indicating that there are differences in training according to specialization, that those differences correlate to race or gender, or that differing skill levels account for the statistical disparities.
At the outset, Concrete Works offered no evidence that different specializations require different training, education, or skills. While different construction jobs clearly involve different types of work, there is no evidence that a significant difference in skill level exists among construction workers or that any area of work requires certain minimum qualifications.
Concrete Works also presented no evidence to show that individuals belonging to any particular gender, racial, or ethnic group are more likely to specialize in certain areas within the construction field. Likewise, it offered no statistical data or explanation as to why that would be the case or evidence to show that such differences account for the observed disparities. In the absence of such evidence, there is no reason to presume that there are fundamental differences in skill levels between specializations, that those differences correlate with race or gender, and that they explain the statistical disparities. See Smith, 84 F.3d at 687; EEOC v. Local 638, 81 F.3d at 1173; EEOC v. Olson's Dairy Queens, Inc., 989 F.2d at 169; Catlett, 828 F.2d at 1266; Segar, 738 F.2d at 1287.
In contrast, Denver presented statistics and expert testimony to explain why firm specialization does not cause the statistical disparities. Dr. Evans, the City's expert, testified that the data for the Denver area reflects that "MBEs are widely * * * distributed across the different construction specialties" (Tr.451).
Moreover, data regarding firm specialization were not readily available for analysis. Census data do not reflect firm specialization and Dr. La Noue explained that survey data accounting for firm specialization can create a sample size that is too small for analysis (Tr. 902). Accordingly, since there is no evidence in the record to suggest firm specialization affected the disparities, the data in the record demonstrate that specialization is irrelevant, and such data were not readily available, this issue was not a proper basis on which to reject Denver's statistics.
3. Qualifications of a Firm. To the extent that the district court discounted "Denver's disparity studies * * * [because they] say nothing about the actual qualifications and capacities of the MBEs and WBEs in * * * Denver * * * to perform the work[,]" Concrete Works, 86 F. Supp. 2d at 1065, its reasoning is unsound. Both plaintiff's and defendant's experts agreed that a firm's capacity and qualifications -- its readiness, willingness and ability to do particular work -- is a "vague term" based on a variety of unspecified "subjective" factors that cannot be quantified (Tr. 791, 856). Indeed, two witnesses for Denver "testified that capacity of business firms cannot be measured objectively and that data on this subject cannot be obtained from contractors". See Concrete Works, 86 F. Supp.2d at 1065. Concrete Works' own experts, Drs. George La Noue and John Lunn, explained that regression analysis "for the purposes of calculating a disparity is very rare" because there currently are "no data source[s]" to determine whether individual capacity and qualifications affect the observed disparities (Tr. 822, 859, 861, 863, 883, 902, 996, 1009, 1015).
Moreover, Concrete Works never offered evidence that any subcontracting opportunities required certain minimum objective qualifications beyond general licensing, that those qualifications correlate with race or gender, or explain the statistical disparities reflecting a substantial underutilization of M/WBEs. See discussion, (pp. 19-20, supra). Consequently, on this record, the district court erred when it rejected Denver's statistical evidence for failing to take account of each firm's qualifications and capacity to perform, even when Concrete Works' own experts testified that such data were not available and could not be accurately quantified and there was no evidence to show that such data would alter the analysis. See Appalachian Power Co. v. EPA, 135 F.3d 791, 805 (D.C. Cir. 1998) (statistical evidence cannot be challenged by merely identifying a variable that is too subjective to quantify or cannot be measured by existing data); Local 638, 81 F.3d at 1173 (same); Segar, 738 F.2d at 1287 (same). See also Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 592-593, (1993).
4. Prequalification. Nor does it matter whether Denver's statistics considered how many minority- and women-owned construction firms were prequalified or officially certified as M/WBEs. Concrete Works, 86 F. Supp. 2d at 1053-1054. First, Concrete Works offered no data to suggest that prequalification standards would have lessened the observed disparities. In fact, the prequalification standards, which specify certain requirements for bonding, licensing, and insurance, apply only to prime contracts over $750,000 that are offered by DPW (Tr. 1074). They do not apply to subcontracts, contracting opportunities offered by DGS, or prime contracts offered by DPW that are less than $750,000 (Tr. 106, 109, 452, 1040, 1074). Dr. La Noue acknowledged that M/WBEs are utilized more as subcontractors than prime contractors, and that prequalification has nothing to do with subcontracting and the vast majority of other contracting opportunities in the Denver metro area (Tr. 925, 1000, 1019).
In any event, the evidence establishes that prequalification relates to factors that have been found to be discriminatory. As this Court noted in its initial opinion, a 1978 GAO report identified the prequalification requirements as "adversly affect[ing] minority contractors' likelihood of participating in public works projects." Concrete Works, 36 F.3d 1524 & n.11. Accordingly, since prequalification standards are clearly not a nondiscriminatory factor that could affect the observed disparities, Denver should not be faulted for failing to control for that factor.
5. Number of Certified M/WBEs. To the extent that plaintiff's experts criticize Denver because its statistical studies analyze data regarding M/WBE availability and its remedial program credits only prime contractors that use certified M/WBEs, their criticism is unfounded. Concrete Works, 86 F. Supp. 2d at 1054.
First, the number of certified M/WBEs has nothing to do with whether Denver has established an adequate factual predicate for its remedial program. Because the failure of a firm to be certified "is not indicative of whether [it is] minority [or] women-owned" and does not control whether it can bid on projects, the number of certified M/WBEs suggests nothing with regard to whether M/WBEs are underutilized in comparison with their availability. In addition, the manner in which Denver counts whether a prime contractor has achieved a specified goal has nothing to do with whether there is an adequate factual predicate to justify implementation of the program. Accordingly, Denver's statistics need not consider the number of certified M/WBEs to establish discrimination.
In sum, the district court erred when it presumed that Concrete Works' identification of unaccounted for variables was sufficient to invalidate the inference of discrimination established by Denver's statistical studies, when the record contains no evidence to demonstrate that any of those variables were responsible for the differences in the success between white, male- owned firms and those owned by minorities and/or women. (5)
C. The District Court's Description Of The
Necessary Bases For The City's Plan
The Supreme Court's decision in Croson states that a municipality may take steps to ensure that it does not become a "passive participant" in a system in which minority or women-owned firms suffer the effects of private discrimination. 488 U.S. at 468. The district court identified six questions as the way in which the constitutionality of Denver's plan would be measured, (6) and found that Denver's statistical studies failed to provide sufficient answers to these questions. 86 F. Supp.2d at 1066, 1070. The questions demonstrate the court's fundamental misunderstanding of the ways in which a municipality may become such a "passive participant," and further that its review is legally erroneous.
In both Fullilove and Richmond, the Court identified the kinds of private actions that may affect the ability of M/WBEs to compete for public contracts and therefore may be considered when a municipality decides to enact Courieran affirmative action program. In Fullilove, Chief Justice Burger referred to evidence before Congress that minority businesses had encountered difficulties in gaining "working capital, in ability to meet bonding requirements, disabilities caused by an inadequate 'track record,' lack of awareness of bidding opportunities, unfamiliarity with biding procedures, preselection before the formal advertising process, and the exercise of discretion by government procurement officers to disfavor minority businesses." 488 U.S. at 467. Congress was also aware that the "business system  has traditionally excluded measurable minority participation" and is "perpetuat[ing] past inequities." Id. at 466 n. 48.
The program Fullilove upheld sought to ensure that grantees of federal funds "would not employ procurement practices that Congress had decided might result in perpetuation of the effects of prior discrimination which had impaired or foreclosed access by minority businesses to public contracting opportunities." Id. at 473. The opinion quite specifically stated that the discrimination that has such current effects need not be by the specific contractors now performing prime contracts, id. at 475, or by the governmental recipients of federal construction funds, id. at 477, or even that discrimination be current, as long as the effects of that prior discrimination still exist. Id. at 478.
While Croson may have applied a more exacting level of scrutiny to Richmond's program, there is no indication that it limited the kind of factors a jurisdiction could consider when determining whether an affirmative action program in municipal contracting was appropriate more narrowly than those Fullilove identified. See 488 U.S. at 491-492. Under Croson, these factors, if identified in the relevant local market, provide the record to sustain a municipality's decision to adjust its contracting processes so as to avoid extending the effects of private discrimination into current municipal contracting.
The district court's six questions evidence a much too cabined view of the predicate sufficient to permit a municipal affirmative action program. For example, question 1 looks for evidence of current discrimination, rather than the current effects of discrimination the Supreme Court has identified as the predicate for an affirmative action plan. Question 3 looks for evidence that policies and practices of private businesses are motivated by discriminatory intent, when the Supreme Court has stated that there need not be such proof. Question 4 suggests that an affirmative action program can be created only where the municipality is complicit in private discrimination, whereas the Court has denied that such proof is necessary. Croson, 488 U.S. at 491- 493. Question 6 appears to assume that a plan may require only those firms that actually have been proven to engage in prohibited intentional discrimination to play a part in a remedial program, when the Court has specifically denied that such proof is required. See Franks v. Bowman, 424 U.S. 747, 777 (1976).
These questions will not disclose whether a plan has an acceptable basis. What the Court said in Fullilove and Croson is that when there is statistical evidence that minority- or women-owned firms appear to be unable to participate equally in the competition for public contracts, and there is not a non-racial or non-sex based explanation for that inequality, we may presume that the effects of historical discrimination, coupled with bidding processes that fail to account for those historical disadvantages, are creating barriers that extend the effects of past discrimination. There need not be proof that discrimination continues to occur, as long as the effects of past practices continue to exist. Dayton Bd. of Educ. v. Brinkman, 443 U.S. 526, 538 (1979). There need not be proof that private firms adopt practices intentionally to exclude minority or women-owned firms, only that their practices may have the effect, intended or not, of extending that exclusion. See Dothard, 433 U.S. at 328-329, Lau v. Nichols, 414 U.S. 563 (1974). And there need not be proof that the municipality is intentionally complicit in discrimination, only that its current practices may unwittingly extend disadvantages bred in discrimination. Croson, 488 U.S. at 491-493. The district court here failed to recognize what constitutes an acceptable factual predicate for an affirmative action program, and its analysis is erroneously colored by that failure.
The district court's decision should be reversed and the case remanded.
MARK L. GROSS
LISA J. STARK
Department of Justice
P.O. Box 66078
The United States believes that oral argument in this case will be of assistance to the Court.
I hereby certify that this brief complies with the type volume limitation set out in Fed. R. App. P. 32(a)(7)(B). The brief was prepared using WordPerfect 7.0, and contains 6,890 words.
I hereby certify that on September 15, 2000, two copies of the Brief for the United States as Amicus Curiae were served by first-class mail, postage prepaid, on the following counsel of record:
William Perry Pendley
James Scott Detamore
Mountain States Legal Foundation
707 17th St.
Denver, CO 80202
Thomas C. Clark
4251 Kipling Street
Wheatridge, CO 80033
Todd Stubbs Welch
Colorado Interstate Gas Company
P.O. Box 1087
Colorado Springs, CO 80944
John R. Gross
J. Wallace Wortham, Jr.
Office of the City Attorney
1437 Banncock Street, Room 353
Denver, CO 80202
CERTIFICATE OF SERVICE (CON'T)
Andrew L. Frey
Thomas B. Colby
Mayer, Brown & Platt
1909 K Street, N.W.
Washington, DC 20006
City Attorney's Office
1445 Cleveland Place #303
Denver, CO 80202
David A. Strauss
University of Chicago Law School
1111 E. 60th Street
Chicago, IL 60637
Mara S. Georges
City of Chicago
Department of Law
City Hall, Room 600
Chicago, IL 60602
Bradford R. Jerbic
Las Vegas City Attorney
400 Stewart Avenue, 9th Floor
Las Vegas, NV 89101-2986
Jay M. Heffern
Minneapolis City Attorney
300 Metropolitan Centre
333 South 7th Street
Minneapolis, MN 55402
CERTIFICATE OF SERVICE (CON'T)
Office of the San Francisco
City Hall, Room 234
San Francisco, CA 94102
Henry W. Underhill, Jr.
Lani L. Williams
International Municipal Lawyers Assoc.
1110 Vermont Avenue, N.W.
Washington, DC 20005
Thomas J. Henderson
Lawyer's Committee for Civil Rights
1401 New York Avenue, N.W.
Washington, DC 20005
Paul C. Saunders
Cravath, Swaine & Moore
825 Eighth Ave.
New York, N.Y. 10019
LISA J. STARK
Department of Justice
P.O. Box 66078
1. References to "Tr.__" refer to page numbers of the Trial Transcript in this case. References to "Ex(s).__" refer to page numbers of Trial Exhibits used in this case.
2. The record reflects that a "consistent pool of firms" contracted for the construction work offered by DGS and DPW. In addition, several witnesses testified that in 1990, City personnel were steering contracts from DPW to DGS, because the latter, unlike the former, was not subject to the M/WBE goals (Tr. 87, 108-109, 226, 247, 350, 357-358).
3. The 1995 study contacted 2,920 construction firms in the Denver metro area. Concrete Works, 86 F. Supp. 2d at 1057; The 1990 study received "responses from 2,514 for a 59% 'completion rate.'" Concrete Works, 86 F. Supp. 2d at 1055.
4. The researchers mailed out 659 questionnaires and received 109 responses. Concrete Works, 86 F. Supp. 2d at 1061.
5. The United States has not addressed the issue of whether Denver's affirmative action program is narrowly tailored for at least two reasons. First, this Court held in its initial opinion that the issue was waived and not preserved on appeal. Concrete Works, 36 F.3d at 1531 n.24. Moreover, unlike the issue of whether Denver's statistics demonstrated a compelling interest --in which the district committed legal error and concluded contrary to precedent that Denver's studies were deficient merely because Concrete Works identified certain variables that had not been analyzed -- the question of whether Denver's program is narrowly tailored is a factual dispute, on which the United States has no interest.
6. The six questions are:
(1) Is there pervasive race, ethnic and gender discrimination throughout all aspects of the construction and professional design industry in the six county Denver MSA?
(2) Does such discrimination equally affect all of the racial and ethnic groups designated for preference by Denver and all women?
(3) Does such discrimination result from policies and practices intentionally used by business firms for the purpose of disadvantaging those firms because of race, ethnicity and gender?
(4) Would Denver's use of those discriminating firms without requiring them to give work to certified [M/WBEs] in the required percentages on each project make Denver guilty of prohibited discrimination?
(5) Is the compelled use of certified [M/WBEs] in the prescribed percentages on particular projects likely to change the discriminatory policies and programs that taint the industry?
(6) Is the burden of compliance with Denver's preferential program a reasonable one fairly placed on those who are justly accountable for the proven discrimination?
86 F. Supp. 2d at 1066-1067.