Kossman Contracting, Co. v. City of Hous.

Full title: KOSSMAN CONTRACTING, CO., Plaintiff, v. CITY OF HOUSTON, Defendant.

Court: UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION

Date published: Feb 16, 2016

CIVIL ACTION NO. H-14-1203 (S.D. Tex. Feb. 16, 2016)

Opinion

CIVIL ACTION NO. H-14-1203

02-16-2016

KOSSMAN CONTRACTING, CO., Plaintiff, v. CITY OF HOUSTON, Defendant.

MEMORANDUM AND RECOMMENDATION

Pending before the court¹ are Plaintiff's Motion for Summary Judgment (Doc. 22), Defendant's Motion to Exclude the Expert Testimony of John Sullivan ("Sullivan") (Doc. 30), and Defendant's Motion for Summary Judgment (Doc. 32). The court has considered the motions, the responses, all other relevant filings, and the applicable law. For the reasons set forth below, Defendant's Motion to Exclude the Expert Testimony of Sullivan is GRANTED, and the court RECOMMENDS that Plaintiff's Motion for Summary Judgment be GRANTED IN PART AND DENIED IN PART and Defendant's Motion for Summary Judgment be GRANTED IN PART AND DENIED IN PART.

1 The motions addressed in this Memorandum and Recommendation were referred to the undersigned magistrate judge pursuant to 28 U.S.C. § 636(b)(1)(A) and (B), the Cost and Delay Reduction Plan under the Civil Justice Reform Act, and Federal Rule of Civil Procedure 72. See Doc. 39.

I. Case Background

Plaintiff filed this action to challenge the constitutionality of the Defendant's 2013 Small/Minority Business Enterprise Program for Construction Contracts ("2013 Program"), which set percentages of its construction contracts as goals to be awarded on the basis of race or gender. A. Factual Background

Defendant first implemented a remedial, preferential program in the area of awarding construction contracts in 1984.² Years later, Defendant commissioned NERA Economic Consulting ("NERA") to examine the past and present status of minority-owned business enterprises ("MBE") and women-owned business enterprises ("WBE") (collectively "MWBE" or "M/WBE") in the geographic and product markets of Defendant's construction contracts.³ On April 18, 2012, NERA issued its study, entitled "The State of Minority- and Women-Owned Business Enterprise in Construction: Evidence from Houston" ("NERA Study").⁴ Based on its findings, NERA also provided recommendations for revising and augmenting Defendant's remedial program ("NERA Recommendations").⁵

2 Doc. 32-3, Ex. 1.1A to Def. City's Mot. for Summ. J., "The State of Minority- & Women-Owned Bus. Enter. in Constr.: Evid. from Hous." ("The NERA Study") p. 1.

3 See id.; Doc. 32-13, Ex. 5 to Def. City's Mot. for Summ. J., City of Hous., Tex., Ordinance No. 2013-428 pp. 1-2.

4 See Doc. 32-3, Ex. 1.1A to Def. City's Mot. for Summ. J., The NERA Study cover page.

5 See Doc. 32-4, Ex. 1.1.A-1 to Def. City's Mot. for Summ. J., NERA Recommendations.

NERA employed regression analysis, a method of analyzing statistics that examines the correlation between two or more variables "in a mathematical model by determining the line of best fit through a series of data points."⁶ The method is commonly used in econometric research, which "concerns itself with the application of statistical inference to the empirical measurement of relationships postulated by economic theory."⁷

6 Doc. 32-3, Ex. 1.1A to Def. City's Mot. for Summ. J., The NERA Study p. xvii.

7 Id. p. xv.

NERA began its study by defining the relevant market area along the lines of geography and industry for Defendant's construction contracts.⁸ NERA extracted data from Defendant's constructon contract records.⁹ By geography, NERA defined the relevant market area by relying on zip codes to determine where the majority of Defendant's contractors and subcontractors were located (accounting for at least seventy-five percent of Defendant's construction spending) as reflected in Defendant's construction contracts from fiscal years 2005-2009 and one-half of 2010.¹⁰ According to NERA's calculations, the relevant geographic market area consisted of the counties of Austin, Brazoria, Chambers, Fort Bend, Galveston, Harris, Liberty, Montgomery, San Jacinto, and Waller.¹¹

8 See id. pp. 2, 59.

9 See id. pp. 2, 60.

10 See id. pp. 59, 63. The database created from Defendant's construction contract records contained information from 756 prime contracts and 7,440 associated subcontracts in a total award amount of $2.82 billion. Id. pp. 2, 60. The NERA Study provides results based on both award amounts and paid amounts. To simplify matters, the court refers only to the results based on award amounts.

11 See id.

By industry, NERA defined the relevant market area by "estimating which North American Industrial Classification System (NAICS) codes best described each identifiable contractor, subcontractor, subconsultant, or supplier" (accounting for more than ninety-nine percent of Defendant's construction spending) in Defendant's construction contract records.¹² NERA weighted the averages by the dollars spent within each industry.¹³ "The weights . . . allow[ed] the overall availability measure to be influenced more heavily by availability in those industries where more contracting dollars [were] spent, and less heavily by availability in those industries where relatively few contracting dollars [were] spent."¹⁴ NERA identified 134 NAICS code industries that made up Defendant's market area, forty of which accounted for ninety-nine percent of Defendant's construction spending.¹⁵

12 Id. pp. 2, 59.

13 See id. p. 59.

14 Id.

15 See id. p. 2

To calculate MWBE availability in the Houston market area, NERA employed the "custom census" method, a multi-step process that included verification of the ownership status of all firms, MWBE and nonMWBE, in the relevant market area.¹⁶ "For each industry category, M/WBE availability [was] defined as the number M/WBE establishments divided by the total number of business establishments in [Defendant's] construction contracting market area, weighted by the dollars attributable to each detailed industry category."¹⁷

16 See id. pp. 69-70.

17 Id. p. 2.

NERA relied on data acquired from Dun & Bradstreet's Hoovers subsidiary on the total number of businesses in the defined market area.¹⁸ The Dun & Bradstreet data contained more than fifteen million records nationally that were updated continuously, and the data record on each business included, among other information, the business name, address, telephone number, NAICS code, business type, and descriptive information.¹⁹ Because the Dun & Bradstreet data did not adequately identify all MWBEs, NERA collected information on MWBEs in Texas and surrounding states through lists from public and private entities, as well as prior NERA studies, and culled records for MWBEs within Defendant's defined market area.²⁰ In determining the availability of MWBEs, the study adjusted for the following firm characteristics: industry affiliation, geographic location, owner labor market experience, and educational attainment.²¹ The study was "careful not [to] adjust for capacity factors that are themselves likely to be influenced by discrimination."²²

18 See id. p. 71.

19 Id.

20 See id. pp. 75-76.

21 See id. p. 84.

22 Id. The NERA Study explains that focusing on "capacity" of firms as defined statistically "in terms of employment, annual individual firm revenues, bonding limits, number of trucks, and so forth is simply wrong as a matter of economics because it can obscure the existence of discrimination." Id. p. 85. "A truly 'effective discriminatory system would lead to a finding of no 'capacity,' and[,] under the 'capacity' approach, a finding of no discrimination." Id.

After verifying ownership status, NERA opined that overall current availability of MWBEs for construction contracting with Defendant was 34.73 percent as weighted for award amounts.²³ This estimation of availability consisted of the following percentages weighted for award amounts by group: African American—4.95; Hispanic—13.12; Asian—4.29; Native American—1.04; Non-minority Female—11.34.²⁴

23 See id. pp. 3, 87, 90.

24 See id. pp. 3, 90.

The next step of the study was to "examine disparities in business formation and earnings principally in the private sector, where contracting activities are generally not subject to M/WBE or other affirmative action requirements."²⁵ The results were consistent across categories considered (wage and salary, business- owner earnings, and business formation).²⁶ Within both of the earnings categories, NERA observed that minorities and women earned "substantially and significantly less" than non-minority men.²⁷ Within the business-formation category, NERA observed "large, adverse, and statistically significant disparities consistent with the presence of discrimination."²⁸ The study also examined whether discrimination existed in the small-business credit market and found evidence of discrimination there as well, particularly against African-American businesses.²⁹

25 Id. p. 91.

26 See id. pp. 3-5, 91-123.

27 See id. pp. 4, 91-107.

28 Id. p. 5; see also id. pp. 107-17.

29 See id. pp. 5-7, 124-79.

The NERA Study then combined the statistical evidence of MWBE availability in the defined market area with Defendant's construction contract records from the prior five and one-half years to determine whether there was "statistical evidence of disparities in the public sector construction contracting activities supported by [Defendant]."³⁰ The utilization percentages weighted for award amounts were: African American—2.86; Hispanic—13.66; Asian—2.12; Native American—1.42; Non-minority Female—9.14.³¹ Overall, NERA found statistically significant adverse disparity for African Americans, Asians, MBEs as a group, and MWBEs as a group.³² NERA observed adverse disparities "among all minority and gender groups and in a wide variety of industry categories," many of which were statistically significant.³³

30 Id. p. 180.

31 See id. pp. 9, 191.

32 See id. pp. 9, 190, 191.

33 Id. p. 192.

NERA did not find statistically significant disparities for Hispanics, Native Americans, and non-minority women, but offered interpretive opinions. NERA pointed out that Defendant's long history of remedial efforts would be reflected in its contracting records such that the records would not reveal underutilization in Defendant's contracting even if underutilization exists in the private sector.³⁴

34 See id. p. 180.

[I]t will usually be counterproductive to suspend or significantly curtail M/WBE programs at the first sign of the elimination of public sector disparities. Given the presence of proactive efforts to remedy discrimination, we would expect public sector disparities to lessen or even disappear. . . . But as long as private sector disparities remain, and private sector efforts to increase utilization of M/WBEs remain limited or non-existent, public sector disparities are likely to reemerge if M/WBE programs are weakened or suspended.³⁵

35 Id.

That rationale led to NERA's opinion that, based on the significant statistical disparity in utilization of Hispanic-owned businesses in the unremediated, private sector, utilization of Hispanic-owned businesses would decline precipitously if that group was dropped from the program.³⁶ With regard to Native Americans, NERA opined that the statistical evidence should be interpreted with caution because the result was largely due to the work of two Native-American businesses.³⁷ If those two businesses were not considered, then the utilization of Native-American businesses would yield statistically significant disparity ratios.³⁸

36 See Doc. 32-4, Ex. 1.1A-1 to Def.'s Mot. for Summ. J., NERA Recommendations p. 6.

37 See Doc. 32-3, Ex. 1.1A to Def. City's Mot. for Summ. J., The NERA Study p. 190.

38 See id.

Non-minority women were dropped from Defendant's remedial program in March 2009, during the period considered by NERA.³⁹ NERA observed that the utilization rate of non-minority women was 10.14 percent by dollars awarded while they were included in the program but dropped to 5.01 percent by dollars awarded after they were eliminated.⁴⁰

39 See id. p. 182.

40 See id.

NERA also took into consideration anecdotal evidence as a check on the statistical findings.⁴¹ NERA started with a large-scale mail survey of business establishments in the market area, both MWBEs and nonMWBEs, asking about contemporary business-related acts of discrimination.⁴² To determine whether the responses were representative, NERA surveyed a random sample of MWBE and nonMWBE firms and compared the responses to the first set of responses.⁴³ Finally, NERA held business-experience group interviews.⁴⁴ NERA found that the anecdotal evidence supported the statistical conclusions.⁴⁵

41 See id. pp. 11-12, 212-63.

42 See id. pp. 11, 212.

43 See id. p. 212.

44 See id.

45 See id. pp. 12, 212.

The NERA Study concluded that, based on both statistical and anecdotal evidence, business discrimination against MWBEs existed in the geographic and industry markets for Defendant's awarding of construction contracts:

[W]e conclude that there is strong evidence of large, adverse, and frequently statistically significant disparities between minority and female participation in business enterprise activity in [Defendant's] relevant market area and the actual current availability of those businesses. We further conclude that these disparities cannot be explained solely, or even primarily, by difference between M/WBE and non-M/WBE business populations in factors untainted by discrimination, and that these differences therefore give rise to a strong inference of the continued presence of discrimination in [Defendant's] market area. There is also strong anecdotal evidence of continuing barriers to the full and fair participation of M/WBEs on [Defendant] contracts and subcontracts, despite the implementation of the M/W/SBE Program, and in the wider Houston construction economy. Remedial efforts remain necessary to ensure that Houston does not function as a passive participant in discrimination.⁴⁶

46 Id. p. 13; see also id. p. 1.

Based on its findings, NERA recommended that Defendant continue and augment the then-current preferential program with neutral and race- and gender-based measures.⁴⁷ The suggestions included increasing communication and outreach with the contracting community and making greater efforts to unbundle the size and complexity of construction contracts.⁴⁸ NERA also recommended that Defendant review and revise as necessary surety bonding, insurance, and experience requirements and implement a program to aid MWSBEs with bonding and working capital.⁴⁹ Among other suggestions were to ensure prompt payments, to resume WBE contract goals and revise the SBE element of the program, and to improve data collection and retention.⁵⁰

47 Doc. 32-4, Ex. 1.1A-1 to Def.'s Mot. for Summ. J., NERA Recommendations pp. 1-14.

48 See id. p. 1.

49 See id. p. 2.

50 See id. pp. 3-5.

Specifically regarding the race- and gender-based remedial measures, NERA suggested adopting overall MWBE goals, setting contract-specific goals, counting MWBE prime contractor participation toward the contract goals, enhancing the program for good-faith reviews and approvals, and ensuring the monitoring of contract performance.⁵¹ According to NERA, Defendant should develop measures for success and review the program every five years.⁵²

51 See id. pp. 7-13.

52 See id. p. 14.

Less than two weeks after NERA issued the study, Defendant held its first stakeholders meeting; approximately eighty individuals attended.⁵³ Months later, approximately forty individuals attended a focus-group meeting.⁵⁴

53 See Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 2; Doc. 32-8, Ex. 3.1 to Def. Mot. for Summ. J., Constr. Disparity Study Stakeholders Meeting Sign-in Sheet.

54 See Doc. 32-7, Ex. 3 to Def. Mot. for Summ. J., Aff. of Marsha Murray ¶ 2; Doc. 32-8, Ex. 3.1 to Def. Mot. for Summ. J., Disparity Study Focus Grp. Sign-in Sheet.

The mayor of Defendant formed a Disparity Study Working Group to analyze the NERA Study and the NERA Recommendations.⁵⁵ The Disparity Study Working Group, through the Office of Business Opportunity, developed and recommended that the City Council adopt a revised preferential program consistent with NERA's findings and recommendations.⁵⁶ At a City Council Public Session on May 7, 2013, several organizations, including the Houston Contractors Association, offered support for the proposed program.⁵⁷

55 See Doc. 32-13, Ex. 5 to Def. City's Mot. for Summ. J., City of Hous., Tex., Ordinance No. 2013-428 p. 2.

56 See id.

57 See Doc. 32-12, Ex. 4 to Def. City's Mot. for Summ. J., Excerpts of Tr. of Hous. City Council Public Session on May 7, 2013.

On May 8, 2013, Defendant unanimously passed Ordinance No. 2013-428 to be effective July 1, 2013.⁵⁸ The ordinance established an MWBE annual goal of thirty-four percent in Defendant's construction contracts.⁵⁹ Chapter 15, Article V of the Code of Ordinances, Houston Texas, ("Code") explains the details of the 2013 Program. The Code states:

58 See Doc. 32-13, Ex. 5 to Def. City's Mot. for Summ. J., City of Hous., Tex., Ordinance No. 2013-428. p. 5.

59 See id. pp. 3-4.

(b) This article is intended to be remedial in nature and to continue only until its purposes and objectives are achieved. At least every five years [Defendant] shall make its best efforts to initiate a review of its minority and women business enterprise program, the results of which shall be provided to city council, who shall determine, upon its receipt of recommendations and the consideration of other relevant information from the [Office of Business Opportunities] director, whether there is strong statistical and anecdotal evidence of discrimination against minority and women business enterprises in [Defendant's] contracting warranting the continuation of a race and gender conscious minority and women business enterprise program.⁶⁰

60 Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code § 15-81.

The Code's definition of minority persons lists Black American, Hispanic American, Asian-Pacific American, Native American, and subcontinent Asian American.⁶¹ The Code provides examples of the ancestry, nationality, lineage, or country for each identified minority group.⁶² An MBE or WBE is defined as a firm that is owned fifty-one percent or more by a minority or woman respectively.⁶³ An SBE is a firm with three-year average gross revenues or number of employees that does not exceed the size standards of the Federal Small Business Act.⁶⁴ To have their bids considered, MWSBEs must be certified as such through Defendant's Office of Business Opportunity.⁶⁵

61 See id. Code § 15-82.

62 See id.

63 See id.

64 See id.

65 See id. Code § 15-84; Doc. 32-10, Ex. 3.3 to Def.'s Mot. for Summ. J.,Requirements for the City of Hous. Prog. for Minority, Women, & Small Bus. Enters. & Persons with Disabilities Enters. ("2013 Program Requirements") p. 5.

The 2013 Program places initial responsibility on each department to determine those construction contracts that are goal-oriented contracts and those that are regulated contracts.⁶⁶ A goal-oriented contract is defined as one in excess of $1,000,000 for which competitive bids are required and with significant subcontracting potential in fields in which there are sufficient known MWSBEs.⁶⁷ A regulated contract is defined as one for which competitive bids are not required, that is not covered by national MBE/WBE programs, and that has significant subcontracting potential in fields in which there are sufficient known MWSBEs or is a type for which sufficient known MWSBEs have represented the ability to perform the prime contract service so as to be able to bid competitively.⁶⁸

66 See Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code § 15-83(c).

67 See id. Code § 15-82.

68 See id.

After classifying the construction contract, the department then must determine whether the MWSBE provisions should be applied to the contract at issue.⁶⁹ The ordinance lists four exceptions to the application of the 2013 Program.⁷⁰ If the department determines that the contract does not fall within an exception, the department must assign an appropriate MWSBE participation level.⁷¹ Goals are set on a contract-by-contract basis based on availability of MWBEs for each NAICS code and on divisibility of contract elements.⁷² Thus, not every construction contract necessarily has an MWBE goal as high as thirty-four percent.⁷³

69 See id. Code § 15-83(c).

70 See id. Code § 15-83(c)(1).

71 See id. Code § 15-83(c)(2).

72 See id.; Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 13.

73 See Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 13.

The 2013 Program also allows administrative flexibility in the application of the MWSBE provisions through waivers.⁷⁴ A contractor may show that, notwithstanding good-faith efforts to meet or exceed the goal, it was unable to do so.⁷⁵ "While it is not a requirement that a contractor meet its goal, it is required that the contractor objectively demonstrate to the office of business opportunity that it has made good faith efforts to meet the goal."⁷⁶ A contractor may demonstrate good-faith efforts through documentation of attempts to obtain bids from MWBEs or documentation that no MWBEs were available.⁷⁷ The Office of Business Opportunity is responsible for reviewing documentation concerning good-faith efforts.⁷⁸

74 See id. ¶ 11; Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code §§ 15-83, 15-84.1.

75 See Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code §§ 15-83(d), 15-85.

76 See id. Code § 15-85.

77 See Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 11; Doc. 32-10, Ex. 3.3 to Def.'s Mot. for Summ. J., 2013 Program Requirements, Attach. A, Good Faith Efforts Policy; Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code §§ 15-82, 15-85.

78 Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code §§ 15-84(b)(5), 15-85.

Other options are built into the 2013 Program. For example, a non-minority, male-owned SBE may be substituted for an MBE or WBE for up to four percent of the value of a contract that is subject to an MWSBE goal.⁷⁹ Also, a contractor may enter a joint venture with one or more MWSBEs and receive a participation percentage appropriate to the details of the joint venture.⁸⁰

79 See Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 8.

80 See Doc. 32-10, Ex. 3.3 to Def.'s Mot. for Summ. J., 2013 Program Requirements p. 5.

During the course of the contract, the contractor is required to complete monthly MWSBE utilization reports.⁸¹ The director of the Office of Business Opportunity is authorized to suspend contractors who fail to make good-faith efforts to meet contract goals or MWSBEs that fail to make good-faith efforts to meet all participation requirements.⁸² A decision to implement a suspension must follow notice and the opportunity for a hearing before an impartial individual.⁸³ The hearing officer's decision may be appealed to the director of the Office of Business Opportunity.⁸⁴

81 See id. pp. 1, 4.

82 See Doc. id. pp. 7-9; Doc. 32-13, Ex. 5 to Def.'s Mot. for Summ. J., Attach. A, Code § 15-86.

83 See Doc. 32-10, Ex. 3.3 to Def.'s Mot. for Summ. J., 2013 Program Requirements pp. 8-9.

84 See id. p. 9.

As far as efforts beyond preferential participation percentages, Defendant continued or adopted numerous neutral measures, including Build Up Houston, a seven-month bonding education program to help small businesses obtain bonding, which replaced a bonding education program of another name in November 2014.⁸⁵ Defendant's Public Works and Engineering Department began the Small Contractor Rotation Program to aid SBEs in acquiring small contracts of less than $50,000.⁸⁶ In partnership with Turner Construction Company, Defendant offered the Turner School of Construction Management, an eight-session program "designed to enhance the technical and managerial skills of small contractors and entrepreneurs in the construction industry."⁸⁷

85 Doc. 32-7, Ex. 3 to Def.'s Mot. for Summ. J., Aff. of Marsha Murray ¶ 6.

86 See id. ¶¶ 9, 10.

87 Id. ¶ 6.

In partnership with Metropolitan Transit Authority of Harris County, Houston Independent School District, and the Port of Houston, Defendant created the Interagency Mentor-Protégé Program designed to enhance business skills of certified firms through workshops and mentorships.⁸⁸ Defendant also offered Kauffman Fasttrac NewVenture Program, a ten-week program designed "for entrepreneurs in the early stages of business development."⁸⁹ In partnership with Vinson & Elkins Law Firm, Defendant created the Small Business Legal Academy, a one-day event featuring workshops on the most common legal issues facing small businesses and one-on-one sessions with lawyers.⁹⁰

88 See id. ¶ 7.

89 Id.

90 See id.

Plaintiff is a small business that is more than fifty-one percent owned by Robert Kossman, a white male, and that offers various types of vegetative erosion control services.⁹¹ Prior to the implementation of the 2013 Program, Plaintiff had worked as a subcontractor on Defendant's construction projects.⁹² Ninety percent of Plaintiff's revenue derived from public contracts, but changes in the MWBE goals resulted in a decrease of the award of subcontracts on Defendant construction projects and a loss of profits.⁹³ B. Procedural Background

91 See Doc. 1, Compl. pp. 2, 8.

92 See id. p. 4.

93 See id. pp. 6-7.

In September 1996, Plaintiff filed a lawsuit challenging Defendant's then-in-effect MWBE program.⁹⁴ The parties settled the case in May 2006, and the court entered a final judgment in June 2006.⁹⁵ The settlement required that Defendant decrease the MWBE utilization goal in its MWBE program for construction contracting from seventeen percent to fifteen percent pending the completion of a new disparity study and that Defendant add a subcontracting goal of five percent utilization of small business enterprises ("SBEs").⁹⁶ Pursuant to the settlement agreement, a special master decided the amount of attorneys' fees and expenses to be awarded to Plaintiff and settled on approximately $800,000.⁹⁷

94 See Kossman Contracting v. The City of Hous., Civ. Action No. H-96-3100, Doc. 1, Compl.

95 See id., Doc. 138, Notice of Settlement; Doc. 141, Final J.

96 See id., Doc. 138-1, Ex. A to Notice of Settlement, Settlement Memo.

97 See id., Doc. 140, Order of the Special Master.

In March 2007, Defendant enacted new utilization goals for participation by MWBEs and SBEs: MBEs at fourteen percent; WBEs at five percent; and SBEs at three percent.⁹⁸ Plaintiff filed a motion to reopen the case and a motion to enforce settlement, asserting that the commissioned disparity study did not support the percentages enacted by Defendant and that the settlement agreement required a five-percent subcontracting goal for SBEs.⁹⁹ The court reopened the case and held a hearing on enforcement of the settlement agreement.¹⁰⁰ In August 2007, the court ordered the parties to meet face-to-face and discuss settlement.¹⁰¹

98 See id., Doc. 143-5, Ex. D to Def. City's Report of Performance of Settlement Agreement, City of Hous., Tex., Ordinance No. 2007-293 p. 2.

99 See id., Doc. 145, Pl.'s Mot. to Enforce Settlement pp. 2-3.

100 See id., Doc. 152, Order Dated Apr. 30, 2007; Doc. 153, Min. Entry Dated May 11, 2007.

101 See id., Doc. 161, Order Dated Aug. 17, 2007.

Settlement negotiation and litigation continued until December 16, 2008, when the parties reached an agreement to modify the settlement agreement.¹⁰² They agreed to utilization goals of fourteen percent for MBEs and eight percent for SBEs, eliminating any utilization goal for WBEs.¹⁰³ Defendant agreed to pay Plaintiff $150, 00 in damages and $125,000 in attorneys' fees.¹⁰⁴ The court entered a final order of dismissal but reopened the case in February 2009 on motion of the parties to allow more time for compliance with the modified settlement agreement.¹⁰⁵

102 See id., Doc. 193, Agreement to Modify Settlement.

103 See id., Doc. 193, Agreement to Modify Settlement p. 1.

104 See id., Doc. 193, Agreement to Modify Settlement p. 1.

105 See id., Doc. 194, Final Order of Dismissal Dated Dec. 16, 2008; Doc. 196, Jt. Mot. for Extension of Time for Compliance with Settlement Agreement; Doc. 198, Order Reopening Case Dated Feb. 2, 2009.

In April 2009, Defendant filed a status report reflecting that it had passed an ordinance setting the percentage goals at fourteen percent for MBEs and eight percent for SBEs with no preference for WBEs.¹⁰⁶ The court entered an order of dismissal.¹⁰⁷ On January 13, 2014, after the enactment of the 2013 Program, Plaintiff filed another motion to reopen the case; the court denied that motion.¹⁰⁸

106 See id., Doc. 205, Status Report; Doc. 205-1, City of Hous., Tex., Ordinance No. 2009-280 p. 4.

107 See id., Doc. 206, Order of Dismissal Dated Apr. 8, 2009.

108 See id., Doc. 210, Mot. to Reopen File & Enforce Settlement; Doc. 212, Order Dated Feb. 4, 2014.

On May 1, 2014, Plaintiff filed the current lawsuit, challenging the goals in the 2013 Program related to MWBEs.¹⁰⁹ Plaintiff asserted that the MWBE goals of the 2013 Program deny it the opportunity to compete on equal footing in the bidding process and have resulted in the denial of Plaintiff's quotes and bids in favor of MWBEs even when the MWBEs bids were not the lowest.¹¹⁰ Plaintiff requested a declaratory judgment that voids the portions of the 2013 Program that unconstitutionally deny Plaintiff equal protection as well as a permanent injunction prohibiting Defendant from enforcing the unconstitutional portions of the 2013 Program.¹¹¹ Plaintiff also requested the award of attorneys' fees, expenses, and costs.¹¹²

109 See Doc. 1, Compl. Plaintiff does not challenge the SBE aspect of the 2013 Program.

110 See id. p. 7.

111 See id. p. 9.

112 See id.

In lieu of an answer, Defendant filed a motion to dismiss in June 2014.¹¹³ After the completion of briefing, the court denied the motion.¹¹⁴ Defendant answered Plaintiff's Complaint.¹¹⁵ On July 24, 2015, Plaintiff filed its pending motion for summary judgment, attaching no evidence.¹¹⁶ Plaintiff concurrently filed Sullivan's expert report.¹¹⁷ The court granted Defendant an extension on the time to file a response to Plaintiff's motion for summary judgment, and Defendant responded on October 1, 2015.¹¹⁸

113 See Doc. 8, Def.'s Mot. to Dismiss.

114 See Doc. 13, Pl.'s Resp. to Def.'s Mot. to Dismiss; Doc. 14, Def.'s Reply to Pl.'s Resp. to Def.'s Mot. to Dismiss; Doc. 15, Pl.'s Surreply in Support of its Mot. to Dismiss; Doc. 17, Mem. Op. & Order Dated Mar. 3, 2016.

115 See Doc. 18, Answer.

116 See Doc. 22, Pl.'s Mot. for Summ. J.

117 See Doc. 21, Sullivan's Expert R.

118 See Doc. 23, Agreed Mot. for Extension of Time to File Resp. to Pl.'s Mot. for Summ. J.; Doc. 25, Order Dated Aug. 21, 2015; Doc. 26, Def.'s Resp. to Pl.'s Mot. for Summ. J.

On November 10, 2015, Defendant filed a motion to exclude Sullivan's expert testimony.¹¹⁹ Two days later, Defendant filed a motion for summary judgment with evidence attached.¹²⁰ Plaintiff responded to both and filed a motion for leave to file an amicus brief.¹²¹ Defendant filed a reply in support of its motion for summary judgment on December 10, 2015.¹²²

119 See Doc. 30, Def.'s Mot. to Exclude the Expert Test. of Sullivan.

120 See Doc. 32, Def.'s Mot. for Summ. J.

121 See Doc. 35, Pl.'s Resp. to Def.'s Mot. to Exclude the Expert Test. of Sullivan; Doc. 36, Pl.'s Resp. to Def.'s Mot. for Summ. J.; Doc. 38, Pl.'s Mot. for Leave to File Amicus Brief.

122 See Doc. 37, Def.'s Reply to Pl.'s Resp. to Def.'s Mot. for Summ. J.

Shortly thereafter, the parties' motions for summary judgment, Defendant's motion to exclude Sullivan's testimony, and the motion for leave to file an amicus brief were referred to the undersigned.¹²³ After Defendant responded to Plaintiff's amicus motion, the court denied leave to file the amicus brief.¹²⁴ On January 5, 2016, Defendant filed a reply in support of its motion to exclude Sullivan's testimony.¹²⁵

123 See Doc. 39, Order Dated Dec. 17, 2015.

124 See Doc. 40, Def.'s Resp. to Pl.'s Mot. for Leave to File Amicus Brief; Doc. 42, Order Dated Jan. 14, 2016.

125 See Doc. 41, Def.'s Reply in Support of its Mot. to Exclude the Expert Test. of Sullivan.

The court now addresses the motion to exclude expert testimony and the two motions for summary judgment.

II. Motion to Exclude Expert Testimony

Defendant moves to exclude the testimony of Sullivan, whom Plaintiff retained as an expert in the field of disparity studies with the purpose of having Sullivan evaluate the NERA Study.¹²⁶ A. Legal Standard for Expert Testimony

126 See Doc. 30, Ex. 1 to Def.'s Mot. to Exclude the Expert Test. of Sullivan, Sullivan's Expert Report p. 5; Doc. 35, Pl.'s Resp. to Def.'s Mot. to Exclude the Expert Test. of Sullivan pp. 3, 6.

Under the Federal Rules of Evidence and related case law, an expert may be qualified by "knowledge, skill, experience, training, or education." Fed. R. Evid. 702. "[T]o qualify as an expert, the witness must have such knowledge or experience in his field or calling to make it appear that his opinion or inference will probably aid the trier in [the] search for truth." United States v. Hicks, 389 F.3d 514, 524 (5^th Cir. 2004)(quoting United States v. Bourgeois, 950 F.2d 980, 987 (5^th Cir. 1992)). If an opinion is based solely or primarily on experience, it "must be grounded in an accepted body of learning or experience in the expert's field." Fed. R. Evid. 702, advisory committee's note, 2000 Amends. The witness must connect the experience to the conclusion offered, must explain why the experience is a sufficient basis for the opinion, and must demonstrate the appropriateness of the application of the experience to the facts. Id.

Although an expert need not be highly qualified to testify on a given topic, his testimony on subjects in which he is not minimally qualified must be excluded. See Huss v. Gayden, 571 F.3d 442, 452 (5^th Cir. 2009). An additional limitation on expert witnesses is that they may not offer conclusions of law. C.P. Interests, Inc. v. Cal. Pools, Inc., 238 F.3d 690, 697 (5^th Cir. 2001)(citing Owen v. Kerr McGee Corp., 698 F.2d 236, 240 (5^th Cir. 1983)).

The expert's testimony must be both relevant and reliable. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999); Smith v. Goodyear Tire & Rubber Co., 495 F.3d 224, 227 (5^th Cir. 2007); see also Fed. R. Evid. 702 & advisory committee's note, 2000 Amends. To be relevant, the testimony must assist "the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702; see also Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 591 (1993). The Federal Rules of Evidence define relevant evidence as that which "has any tendency to make a fact more or less probable than it would be without the evidence" and concerns facts that are of consequence in resolving the case. Fed. R. Evid. 401.

Reliability hinges on the sufficiency of the facts or data upon which the opinion is based, the dependability of the principles and methods employed, and the proper application of the principles and methods to the facts of the case. See Fed. R. Evid. 702. Daubert outlined several reliability factors for scientific testimony: (1) whether the theory or technique has been tested; (2) whether it has been subjected to peer review and publication; (3) whether it has a known or potential rate of error; and (4) whether it is generally accepted or has gained only minimal support within the relevant community. Daubert, 509 U.S. at 593-94. When the testimony is not scientific, these factors may apply even though the focus is on experience and personal knowledge. See Kumho Tire Co., 526 U.S. at 150-51. The trial judge must make certain that the expert applied "the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." Id. at 152.

The burden of establishing that the predicate for the expert's testimony falls on the party producing the expert. See Mathis v. Exxon Corp., 302 F.3d 448, 459-60 (5^th Cir. 2002). The trial court has the responsibility of determining whether that party has met its burden. Fed. R. Evid. 104(a); Mathis, 302 F.3d at 459-60. see also Daubert, 509 U.S. at 589. The trial judge has "wide latitude in determining the admissibility of expert testimony" even though "the rejection of expert testimony is the exception rather than the rule." Fed. R. Evid. 702, advisory committee's note, 2000 Amends. B. Sullivan's Experience

Sullivan received a Bachelor of Arts in 1976 from Loyola College and a Juris Doctor in 1983 from the University of Maryland Law School.¹²⁷ From 1983 to 1990, Sullivan worked for the Department of Legislative Research, the Bureau of National Affairs, and Housing and Urban Development.¹²⁸ In 1990, he founded Croson Legal Services and, since that time, has worked as a "[l]egal consultant in the areas of affirmative action and public contracting."¹²⁹

127 See Doc. 21, Sullivan's Preliminary Expert Report p. 68.

128 See id.

129 Id. p. 66; see also id. p. 68.

Sullivan listed five cases, other than this one, in which he was designated as an expert witness, one in 1998 and four during the years 2008 to 2013.¹³⁰ In a non-expert capacity, Sullivan has "[a]ssisted on more than thirty cases" in the federal court system "involving the constitutionality of racial, gender, and ethnic preferences in public contracting, the sufficiency of disparity studies, and the appropriateness of including certain groups in preferential contracting programs."¹³¹ His experience in those cases included consulting, sitting second chair, representing clients at depositions, strategizing, developing trial tactics, preparing witnesses, drafting certiorari petitions, and conducting discovery.¹³² Sullivan also reported extensive experience representing construction contractors and subcontractors, specifically concerning prequalification and bonding.¹³³

130 See id. p. 68.

131 Id. p. 66.

132 See id.

133 See id. p. 67.

In addition to his more concrete legal roles, Sullivan has studied and analyzed statutes and regulations that establish guidelines for contracting with disadvantaged, minority-owned, and women-owned businesses, disparity studies, procurement law and regulation, and construction bids.¹³⁴ He has worked with five local governments during the years 1993 to 2000 to produce constitutionally consistent disparity studies and to design race- neutral alternatives to racial preferences in public contracting.¹³⁵

134 See id.

135 See id.

Sullivan's research and writing experience includes writing three and cowriting six scholarly articles addressing aspects of race preferences, particularly in public contracting.¹³⁶ His publications also include nineteen print-media editorial articles during the years 1990 to 2014, newspaper book reviews in 1990 and 1991, and four blog entries in 2013.¹³⁷

136 See id. p. 69.

137 See id. pp. 69-70, 72.

In the academic setting, Sullivan serves as the associate director of the Project on Civil Rights and Public Contracts at the University of Maryland Baltimore County and has taught constitutional law at the university on four occasions as an adjunct professor.¹³⁸ He gave one public lecture in 2003 and presented at one conference in 1993.¹³⁹ In 1998, Sullivan testified before the House Subcommittee on the Judiciary on the subject of "the federal Benchmark disparity study."¹⁴⁰ C. Sullivan's Opinion Testimony

138 See id. pp. 67, 71.

139 See id. p. 70.

140 Id. p. 71.

Sullivan began his report with a history of Houston's minority- and women-owned business enterprise program since its inception in 1984 and the recurrent litigation challenging the program's constitutionality.¹⁴¹ From there, Sullivan presented what he termed the "fundamental flaws of the study."¹⁴²

141 See id. pp. 6-8.

142 See id. pp. 8-41.

Sullivan opined that the heart of City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989) [hereinafter Croson], is the "qualified, willing and able" test for availability.¹⁴³ The NERA Study fails this test, according to Sullivan, because: (1) the data on which it is was based was over-inclusive in that it provided "no measure of the qualifications or willingness of the firms" to perform the type of construction work the City sought; (2) the study provided only a headcount of firms without regard for availability; (3) the study combined prime contractors and subcontractors in its availability and utilization data, prohibiting a determination of where, if any, discrimination had occurred and what would be an appropriate remedy; (4) the study did not reveal statistically significant underutilization for all MWBE groups in all construction sub-industries.¹⁴⁴

143 Id. p. 8.

144 Id. pp. 9-10.

Additionally, Sullivan took issue with: (1) the study's use of regression analysis because it was based on data that was more than five years old and because regression analysis could not identify who, if anyone, was discriminating; (2) the low return rate of surveys and the failure to verify or investigate anecdotal evidence of discrimination; (3) the failure, in the study and by the City, to demonstrate that race-neutral alternatives were implemented, evaluated, and found to be insufficient; and (4) the 2013 Program's placing of an undue burden on Plaintiff.¹⁴⁵ D. Discussion

145 See id. pp. 10-11.

Defendant argues that Sullivan is not qualified by knowledge, skill, training, education, or experience, that his report consists of legal opinions, and that his methodology is not commonly relied upon by experts in the field. Plaintiff asserts that Sullivan's experience in researching, analyzing, writing, consulting, litigating, and opining on disparity studies more than qualifies him as an expert on the validity and reliability of the NERA Study. Plaintiff describes Sullivan's method as "the application of extensive experience."¹⁴⁶

146 Doc. 35, Pl.'s Resp. to Def. City's Mot. to Exclude the Expert Test. of Sullivan pp. 9, 16.

Sullivan's qualifications are those of a legal scholar; he has spent many years researching, analyzing, and writing about race-and gender-based remedial public programs that seek to remedy past discrimination. His opinions originate in his interpretation of constitutional principles as applied to race- and gender-based remedial programs, particularly his interpretation of Croson. The court has no doubt that Sullivan has devoted thousands of hours of contemplation to the issue; however, his experience does not include designing or conducting statistical studies.

Sullivan has no education or training in statistics or economics. He is not qualified by any of his experience to collect, organize, or interpret numerical data. He has absolutely no experience extrapolating general conclusions about a subset of the population by sampling it. He has demonstrated no knowledge of sampling methods or understanding of the mathematical concepts used in the interpretation of raw data. Thus, Sullivan is not qualified to challenge the methods and calculations of NERA's economically sophisticated study. The methods employed in selecting and refining statistical data, in calculating market area, availability, and disparity, and in drawing meaningful and practical conclusions from the mathematical results are not tools that a lay person can master through common experience.

Sullivan fails to connect his legal experience or his research, writing, and public speaking on disparity studies to his conclusions. Sullivan's report is a theoretical attack on the NERA Study with no basis in objective evidence, such as data about or testimony of construction firms in the relevant market area that support his assumptions about available MWBEs or comparative studies that controlled for the factors about which he complained. Sullivan challenged the anecdotal evidence because the accounts were not verified or investigated, but he did not rely on evidence that the anecdotal evidence was not trustworthy or on other witness testimony that countered the evidence.

Sullivan criticized the NERA Study for disregarding capacity. NERA explained that capacity is not a proper economic consideration. Sullivan is not an economist and is not qualified to challenge NERA's explanation. He also championed the use of data reflecting the MWBEs that actually submitted bids or quotes to better determine firm availability. Not only did he fail to provide econometric support for the use of bidder data, he cited no personal experience with the use of bidder data and provided no proof that it would more accurately reflect availability of MWBEs absent discriminatory influence. Moreover, he acknowledged that, in this case, no bidder data had been collected for the years covered by the NERA Study.¹⁴⁷

147 See id. p. 21.

The court does not find Sullivan's testimony to be helpful. It is incumbent upon the court to decide, as a matter of law, whether the NERA Study forms a strong evidentiary basis of past and/or present discrimination to support Defendant's implementation of a race- and gender-based remedial program. Sullivan's attempts to poke holes in the NERA Study by pointing out what he considers to be flaws based on his legal studies are inadmissible. Although Sullivan is an attorney and an expert in the constitutional evaluation of preferential contracting programs, he is not permitted to offer conclusions of law. C.P. Interests, Inc., 238 F.3d at 697. The central theme of his report is that the NERA Study does not meet all of the requirements articulated in Croson and, as a result, is not a strong basis in evidence of the need for remedial action. The court does not need Sullivan's help in making a legal determination as to the evidentiary strength of the NERA Study or, for that matter, the constitutionality of the 2013 Program.

Sullivan articulates no method at all, and Plaintiff is of no help to Sullivan by describing Sullican's method as "the application of extensive experience."¹⁴⁸ Experience is not a method. It qualifies an expert, and it may prove the reliability of a method; but it is not, on its own, a method. Sullivan certainly expresses opinions about how disparity studies are flawed or, in some instances, could be better performed and about how narrowly defined the data must be to produce constitutionally permissible results. But these ideas of Sullivan are nothing more than untested hypotheses. Sullivan's criticisms are not founded in cited professional social science or econometric standards, and his suggestions are not backed by his personal prior application or by cites to professional application by experts in the social sciences. Sullivan uses terms, such a qualified, willing, and able, that may have attained legal definitions in this context, but he fails to define them in any statistical sense so that they may be measurable through refined data.

148 Doc. 35, Pl.'s Resp. to Def. City's Mot. to Exclude the Expert Test. of Sullivan pp. 9, 16.

For these reasons, the court finds that Sullivan is not qualified to offer the opinions contained in his report and that his report is not relevant, not reliable, and, therefore, not admissible.

III. Motions for Summary Judgment

Relying solely on Sullivan's report as evidence, Plaintiff argues that the 2013 Program is incompatible with the Equal Protection Clause of the Fourteenth Amendment. Defendant defends the constitutionality of the 2013 Program. A. Applicable Law

The current legal framework for analyzing equal protection concerns in the context of MWBE preferential programs has evolved over the past three decades. Initially, however, the court devotes special consideration to the Croson case and its continuing significance.

1. Croson's Continuing Significance

Croson ¹⁴⁹ was decided twenty-seven years ago. The case was the first in which the United States Supreme Court addressed race-based remedial programs in state and local public contracting. As the first case to analyze the issue in detail, to explain the court's rationale, and to formulate an approach for the evaluation of race-based remedial programs in state and local public contracting, it is a seminal case on which courts continue to rely. Croson synthesized prior Supreme Court decisions addressing "the tension between the Fourteenth Amendment's guarantee of equal treatment to all citizens[] and the use of race-based measures to ameliorate the effects of past discrimination on the opportunities enjoyed by members of minority groups in our society" and applied principles from those cases to the facts of the case before it. Croson, 488 U.S. at 476-77.

149 Unfortunately, Croson was a fractured opinion. A majority of six carried the judgment, but only three joined the entire opinion.

The facts underlying the Croson decision involved a self-described remedial plan adopted by the Richmond City Council that required prime contractors on city construction contracts to subcontract a minimum of thirty percent of the dollar amount of each public contract to one or more MBEs. Id. at 477, 478. The plan exempted minority-owned prime contractors from the minority set-aside requirements. Id. at 477-78. The plan benefitted not only local MBEs but MBEs from anywhere in the United States. Id. at 478. The Richmond City Council adopted the plan in April 1983, and the plan expired at the end of June 1988. Id. at 477, 478.

A waiver of the set-aside percentage was available "in exceptional circumstances" upon a showing that "every feasible attempt ha[d] been made to comply" and that insufficient MBEs were available and willing to participate in the contract. Id. at 478-79. The director of the Department of General Services made the final decision whether a contract bid complied with the plan's set-aside provisions and whether a waiver was appropriate. Id. at 479. A bidding construction firm that was denied a contract for failure to comply was not entitled to an appeal but had a general right of protest under the city's procurement policies. Id.

In September 1983, Richmond issued an invitation to bid on a plumbing project at the city jail. Id. at 481. Interested in submitting a bid, the regional manager of J.A. Croson Company contacted five or six MBEs to supply the fixtures for the project but found none that were interested in the project or that tendered a quote. See id. at 481-82. On the day the bid was due, J.A. Croson Company's regional manager made another effort to solicit a fixture quote from an MBE. See id. at 482. A local MBE indicated interest but was unable to obtain a price quote from a fixture manufacturer. See id.

J.A. Croson Company submitted the bid with a direct fixture quote from one of the manufacturers. See id. J.A. Croson Company was the only bidder and, a week after the bids were due, sought waiver of the minority set-aside requirement. See id. Before the city ruled on the waiver request, the local MBE that J.A. Croson Company's regional manager had previously contacted notified the city that the MBE could supply the fixtures. See id. at 482-83. The city denied J.A. Croson Company's waiver and allowed ten days for the firm to submit a form indicating utilization of an MBE. See id. at 483. J.A. Croson Company sent the city two letters explaining that the MBE was not an authorized supplier of the fixtures and was subject to credit approval and that the bid was twenty-one days late and substantially higher than the direct quote given to J.A. Croson Company. See id. The city refused to grant J.A. Croson Company a waiver or to raise the overall contract price and informed the firm that it had decided to rebid the project. See id.

The Supreme Court began its analysis by discussing Fullilove v. Klutznick, 448 U.S. 448 (1980). See Croson, 488 U.S. at 486. In Fullilove, the Supreme Court reviewed a ten-percent minority set-aside for federally funded construction grants under the Public Works Employment Act. See Croson, 488 U.S. at 486. The Supreme Court noted that the decision in Fullilove did not employ any standard of equal-protection review but, rather, deferred to Congress on the basis that the objectives of the legislation were within Congress's power and that Congress was permitted to use racial criteria in carrying out its objectives. See Croson, 488 U.S. at 487. The Croson opinion noted that state and local governments do not have the same power as Congress to "identify and redress the effects of society-wide discrimination", but that they do have "the authority to eradicate the effects of private discrimination within [their] own legislative jurisdiction[s]." Id. at 490, 491-92.

The Court then turned to Wygant v. Jackson Bd. of Educ., 476 U.S. 267 (1986), which addressed whether a local school district could rely on the use of racial quotas set by agreement with the teachers' union. See Croson, 488 U.S. at 492. The plurality opinion in Wygant determined that a showing of prior discrimination by the school district was required to justify a race-based layoff program. Croson, 488 U.S. at 492. Drawing from that case, Croson held that, if a local government could show passive participation in discrimination, it could take steps to eradicate the discrimination. Id.

The Croson opinion also reviewed Regents of the University of California v. Bakke, 438 U.S. 265 (1978), a case that involved a racial quota in medical school admissions. Croson, 488 U.S. at 496. The Court drew from Bakke the notion that neither a desire to have more minority medical students or doctors nor the history of discrimination in society generally justified a racial quota. Croson, 488 U.S. at 496. Rather, the governmental interest must be based on findings of prior constitutional or statutory violations. See id. at 497.

The Croson court held that the proper standard of review for racial classifications, regardless of the race benefitted or burdened, was strict scrutiny. See id. at 493-94. Applying Wygant and Bakke, the Croson Court viewed the generalized assertion of past discrimination in the construction industry to lack "the precise scope of the injury" to be remedied and determined a history of discrimination in the country could not justify Richmond's rigid quota system. Id. at 498-99. The Court then analyzed the facts on which the district court had relied in finding the quota system justified. Id. at 499. The facts were: (1) the ordinance stated that it was remedial; (2) proponents of the system opined that there had been discrimination in the industry; (3) minority firms received less than one percent of Richmond's prime contracts while the minority population of Richmond was fifty percent; (4) very few minority contractors were members in local and state contractors' associations; and (5) six years prior to the adoption of the quota system, "Congress made a determination that the effects of past discrimination had stifled minority participation in the construction industry nationally." Id. The Court found that those five factual findings, individually or collectively, did not amount to a strong basis in evidence to support remedial action and held that "the city ha[d] failed to demonstrate a compelling interest in apportioning public contracting opportunities on the basis of race."¹⁵⁰ Id. at 500, 505.

150 The Court limited its analysis to the preferential treatment of African Americans, noting that "absolutely no evidence of past discrimination against Spanish-speaking, Oriental, Indian, Eskimo, or Aleut persons" was presented. Croson, 488 U.S. at 506.

The Court briefly addressed whether the remedial plan was narrowly tailored and determined that it was obviously not, making two observations. Id. at 507, 508. First, nothing suggested that Richmond had given any consideration to the use of race-neutral means to increase the awarding of construction contracts to minority firms. Id. Second, the thirty-percent quota rested solely upon "the assumption that minorities [would] choose a particular trade in lockstep proportion to their representation in the local population" and was obviously "not narrowly tailored to remedy the effects of prior discrimination." Id. at 508.

Although the Croson opinion did not articulate a precise framework for evaluating preferential programs, it highlighted several important factors to consider in evaluating race-based preferential programs. First among the important take-aways from Croson was the clarification of the parameters of a strong evidentiary basis of past discrimination that was sufficiently particular to both justify and fashion a remedial program. Croson recognized that evidence of passive participation in discrimination could give rise to a compelling interest, but emphasized that a history of societal discrimination was not enough. The evidence needed to be particular to the governmental entity enacting a remedy. That is, the evidence must reveal past discrimination: (1) within the geographical bounds of the governmental entity; (2) against minorities qualified¹⁵¹ to undertake the particular task, and (3) against separately identified minority groups. The facts before the Croson court met none of the three factors: (1) the data was nationwide; (2) the disparity comparison was made with Richmond's population as a whole; and (3) the evidence related only to African Americans but the program benefitted numerous other minority groups.

151 Plaintiff relies heavily on what it refers to as the Croson "qualified, willing, and able test" in criticizing the NERA Study. Doc. 22, Pl.'s Mot. for Summ. J. p. 14. The Croson opinion mentions those three words together only once, in its conclusion: "Where 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. According to Plaintiff, that sentence launched a test that "is the key to any effectively done disparity study." Doc. 22, Pl.'s Mot. for Summ. J. p. 14. The court disagrees. In context, the Court, consisting, at that point, of only four justices, was emphasizing that evidence of past discrimination must relate to minorities within the relevant field as opposed to minorities in the locality or nation generally. The Court was not establishing a mechanical test for the evaluation of disparity studies. Cf., e.g., Concrete Works of Colo., Inc. v. City & Cty. of Denver, 321 F.3d 950, 983 (10^th Cir. 2003)("[W]e do not read Croson to require disparity studies that measure whether construction firms are able to perform a particular contract.").

The Court's two observations on narrowly tailoring are also important focal points: (1) race-neutral programs may eradicate barriers to minority participation and should be tried before enacting a preferential program; and (2) a rigid, arbitrary quota system is not narrowly tailored. In Croson, no evidence suggested that Richmond had even considered race-neutral alternatives before enacting its so-called remedial plan; and that plan was a rigid, race-based quota percentage based on assumptions and bureaucratic simplicity.

In its conclusion, the Court provided guidance for the evaluation of future cases. It explained that a governmental entity could "tak[e] action to rectify the effects of identified discrimination within its jurisdiction" if it had evidence "that non-minority contractors were systematically excluding minority businesses from subcontracting opportunities." Croson, 488 U.S. at 509. Evidence of a significant statistical disparity between qualified minority contractors in the relevant areas and the number of contractors actually engaged through the entity's process of awarding contracts would give rise to an inference of discriminatory exclusion. Id.

In the intervening twenty-seven years, federal courts across the nation have refined and updated Croson's rationale and have applied it on a case-by-case basis to a wide variety of factual scenarios. Here, the court presents current constitutional standards, relying primarily on Croson, more recent Supreme Court opinions, and Fifth Circuit analysis.

2. Equal Protection Review of MWBE Remedial Programs

The Equal Protection Clause of the Fourteenth Amendment commands, "No State shall . . . deny to any person within its jurisdiction the equal protection of the laws." U.S. Const. amend. XIV, § 1. This clause protects the personal right "to be treated with equal dignity and respect," which is "implicated by a rigid rule erecting race [(or gender)] as the sole criterion in an aspect of public decisionmaking." Croson, 488 U.S. at 493. Courts are responsible for determining whether race-based programs are justified as remedial and necessary or are, in fact, "motivated by illegitimate notions of racial inferiority or simple racial politics." Id.

Preferential programs based on race must be reviewed under the tenets of strict scrutiny.¹⁵² See Dean v. City of Shreveport, 438 F.3d 448, 454 (5^th Cir. 2006)(reviewing a city's race-based hiring practices for firefighters).

152 Gender-based programs must also be justified but are subject to intermediate scrutiny, a less strenuous review standard. See United States v. Virginia, 518 U.S. 515, 532-33 (1996); H.B. Rowe Co. v. Tippett, 615 F.3d 233, 242 (4^th Cir. 2010). Because the court finds that the remedial plan at issue here survives strict scrutiny, the court does not discuss the tenets of intermediate scrutiny or evaluate the 2013 Program under that standard.

[T]he purpose of strict scrutiny is to "smoke out" illegitimate uses of race by assuring that the legislative body is pursuing a goal important enough to warrant use of a highly suspect tool. The test also ensures that the means chosen "fit" this compelling goal so closely that there is little or no possibility that the motive for the classification was illegitimate racial prejudice or stereotype.

Croson, 488 U.S. at 493. Accordingly, to survive review under strict scrutiny, "a race-conscious measure [must] be (1) justified by a compelling government interest and (2) narrowly tailored to further that interest." Dean, 438 F.3d at 454 (citing Police Ass'n of New Orleans ex rel. Cannatella v. City of New Orleans, 100 F.3d 1159, 1167 (5^th Cir. 1996)); Hous. Contractors Ass'n v. Metro. Transit Auth. of Harris Cty., 189 F.3d 467, *2 (5^th Cir. 1999)(unpublished)(citing Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227 (1995))(addressing a constitutional challenge to the transit authority's disadvantaged-business-enterprise ("DBE") program).

"[C]ombating racial discrimination is a compelling government interest." W.H. Scott Constr. Co. v. City of Jackson, Miss., 199 F.3d 206, 217 (5^th Cir. 1999)[hereinafter Scott](stating that Croson made this point clear). More specifically, a governmental entity "has a compelling interest in assuring that public dollars, drawn from the tax contributions of all citizens, do not serve to finance the evil of private prejudice." Croson, 488 U.S. at 492; see also Scott, 199 F.3d at 217 (citing Croson, 488 U.S. at 492). The challenged governmental entity must identify the targeted discrimination with sufficient particularity to create a strong basis in evidence justifying remedial action. Scott, 199 F.3d at 217 (quoting Croson, 488 U.S. at 500); see also Dean, 438 F.3d at 455. Unfortunately, the "quantum of evidence" required to establish a "strong basis in evidence" cannot be determined by a "precise mathematical formula." Scott, 199 F.3d at 218 n.11. Rather, the evidence in each case must be evaluated for its sufficiency. Id.

The targeted discrimination need not be the work of the governmental entity itself; evidence of passive participation in the discriminatory awarding of public contracts may establish a compelling interest in remedial action. See id. at 217 (citing Croson, 488 U.S. at 492). To justify remedial action based on passive participation, "a governmental entity must establish a factual predicate, tying its set-aside percentage to identified injuries in the particular local industry." Id. (citing Croson, 488 U.S. at 499). "[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" would support a finding that non-minority contractors were systematically excluding minority businesses from subcontracting opportunities and would justify remedial action. Id. at 218 (emphasis omitted)(quoting Croson, 488 U.S. at 509).

Given Croson's emphasis on statistical evidence, other courts considering equal protection challenges to minority-participation programs have looked to disparity indices, or to computations of disparity percentages, in determining whether Croson's evidentiary burden is satisfied. Disparity studies are probative evidence of discrimination because they ensure that the "relevant statistical pool" of qualified minority contractors is being considered.

Id. (citations omitted)(quoting Croson, 488 U.S. at 501). If the past discrimination is satisfactorily demonstrated, the resulting remedial program must be narrowly tailored to achieve its end. See Croson, 488 U.S. at 507-08. The Fifth Circuit relies on the factors enunciated in United States v. Paradise, 480 U.S. 149, 171 (1987), to ensure that a race-based remedy is narrowly tailored. See Dean, 438 F.3d at 458. Those factors are: (1) "the necessity of the particular relief and the efficacy of alternative remedies;" (2) "the flexibility and duration of the relief, including the availability of waiver provisions;" (3) "the relationship between the numerical goal of the relief and the relevant labor market;" and (4) "the impact of the relief on the rights of third parties." Id. (citing Paradise, 480 U.S. at 171).

The first factor requires evidence of statistical disparity during the relevant period supporting the race-based remedial program, as well as information about the implementation and efficacy of race-neutral remedies. See id. at 458-59. The Supreme Court has repeated that "narrow tailoring does not require exhaustion of every conceivable race-neutral alternative" but does require the court to examine the governmental entity's efforts without deferring to its "serious, good faith consideration" of race-neutral alternatives. Fisher v. Univ. of Tex. at Austin, ___ U.S. ___, 133 S.Ct. 2411, 2420 (2013)(quoting Grutter v. Bollinger, 539 U.S. 306, 339 (2003)).

The flexibility aspect of the second factor focuses on whether the program allows for a waiver of the requirements or an exception for lack of qualified candidates. See Dean, 438 F.3d at 459 (citing Paradise, 480 U.S. at 177-78). The duration aspect of the second factor equates a short life-span of the remedy with narrow tailoring. Id. at 460.

The third factor requires that the "long-term numerical goal of a race-conscious remedy . . . be closely tied to the relevant [available] labor market." Id. at 462. The final factor protects innocent third parties from the imposition of an unacceptable burden, as in when the race-based remedial program allows for contracting with unqualified MWBEs or when the remedy does more than merely postpone benefits to third parties. See id. (citing Paradise, 480 U.S. at 182).

A plaintiff bears the burden of putting the adoption of the particular preferential program in issue. Fisher, 133 S.Ct. at 2420. A defendant, however, bears the burden of proving that race-neutral alternatives are not sufficient. See id.; Croson, 488 U.S. at 500, 510. The Seventh Circuit explained that the burden shifts back to the plaintiff to prove unconstitutionality upon the defendant's production of acceptable proof of a compelling interest and narrow tailoring. Majeske v. City of Chicago, 218 F.3d 816, 820 (7^th Cir. 2000); see also Aiken v. City of Memphis, 37 F.3d 1155, 1162 (6^th Cir. 1994); Concrete Works of Colo., Inc. v. City & Cty. of Denver, 36 F.3d 1513, 1522 (10^th Cir. 1994).

One court characterized this standard as requiring the governmental entity to make a prima facie showing of the necessary interest and the plaintiff to overcome that showing by the greater weight of evidence. Builders Ass'n of Greater Chi. v. Cty. of Cook, 123 F. Supp. 2d 1087, 1094 (N.D. Ill. 2000); see also H.B. Rowe Co. v. Tippett, 615 F.3d 233, 241 (4^th Cir. 2010)(explaining the governmental entity is not required to conclusively prove existence of discrimination to establish a strong evidentiary basis); Rothe Dev. Corp. v. U.S. Dep't of Def., 262 F.3d 1306, 1317 (Fed. 2001)(explaining that the governmental entity's burden is one of production); O'Donnell Constr. Co. v. D.C., 963 F.2d 420, 424 (D.C. Cir. 1992)(explaining that a governmental entity's "legislation must rest on evidence at least approaching a prima facie case of racial discrimination in the relevant industry").

The challenger must respond with more than speculation that the governmental entity's evidence is insufficient or flawed. H.B. Rowe Co., 615 F.3d at 241-42. What is necessary is "credible, particularized evidence," such as contrasting statistical data, testimony or documentation of neutral justifications for the statistical results, or proof that the disparity study results are flawed or insignificant. Id. (collecting cases).

The court bears the responsibility for employing strict scrutiny in determining whether a remedial measure passes muster. See Dean, 438 F.3d at 455. "[A] trial court must make a factual determination that there was a strong basis in evidence for the conclusion that remedial action was necessary." Id.; see also Croson, 488 U.S. at 493 (emphasizing the importance of a "searching judicial inquiry into the justification" for preferential programs); Wygant, 476 U.S. at 277 (stating that it's the trial court's responsibility to "make a factual determination that the [governmental entity] had a strong basis in evidence for its conclusion that remedial action was necessary"). The determination of whether a strong basis in evidence of past or present discrimination establishes a compelling interest to enact a remedial program is a question of law but rests on factual determinations about the accuracy and validity of the governmental entity's evidence. Concrete Works of Colo., Inc., 36 F.3d at 1522. B. Analysis

Plaintiff presents no evidence to raise any dispute of fact regarding the commissioning of the NERA Study or the enactment or implementation of the 2013 Program. Plaintiff's focus on what it calls "flaws" in the NERA Study is a legal argument challenging whether Defendant had a compelling interest in implementing a preferential program for the awarding of construction contracts. Likewise, Plaintiff's issues with the 2013 Program itself mount the legal challenge that the program is not narrowly tailored to remedy the discrimination identified in the NERA Study.

Frankly, Plaintiff's briefs leave the impression that Plaintiff finds no need for any remedial program and that none would be satisfactory. Plaintiff's owner confirmed as much at his deposition:

Q. Is there any sort of minority- or gender-based, gender-conscious affirmative action program that Kossman would be satisfied with?

A. Race and gender neutral, all small business.

Q. Okay. Period?

A. Period.

Q. No race and gender conscious —

A. No. No —

Q. — that you believe —

A. No. No. Nowhere in our real world do we make decisions based on race and gender. This is 2015. You never make those decisions. Every time you do, you're looking for trouble.

Q. Okay. So in other words, Kossman — there's no program the City could implement that would involve race- or gender-conscious preferences that Kossman

A. That was our motion for summary judgment, yes.

Q. — would ever be satisfied with?

A. No. It's antiquated. . . . ¹⁵³

153 Doc. 32-14, Ex. 6 to Def.'s Mot. for Summ. J., Dep. of Robert Kossman pp. 127-28.

Defendant disagrees with Plaintiff's position, offering point-by-point responses to Plaintiff's arguments. Defendant defends the NERA Study as providing a strong basis in evidence of ongoing discrimination in its construction contracting process and justifying by compelling interest its remedial program to combat that discrimination. Defendant also defends the 2013 Program as narrowly tailored to further that interest.

The court bears the responsibility of reviewing the NERA Study and the 2013 Program to determine whether Defendant has complied with the Equal Protection Clause. In the process, the court addresses each of Plaintiff's challenges to the NERA Study and the 2013 Program.

First, the court notes that the facts on which it must base the constitutional analysis are a far cry from those on which the Supreme Court based the Croson analysis. In the disparity analysis here, the market area was geographically confined to area codes in which the majority of the public contracting construction firms were located whereas Richmond relied on nationwide data. Also, the relevant market area here was weighted by industry whereas Richmond relied on the African-American population in Richmond. In actuality, Richmond did not make any effort to circumscribe its disparity analysis according to the relevant market area whereas NERA limited the relevant market area by geography and industry based on Defendant's records from prior construction contracts.

Unlike in the Croson case, Defendant relied on an updated, comprehensive database, which it supplemented and verified. Richmond offered no basis for its chosen goal, but Defendant relied on sophisticated calculations of weighted availability percentages in setting an overall goal. Richmond did not attempt to differentiate discrimination from other causes whereas the NERA Study considered other causes and determined they were not devoid of discrimination. The evidence on which Richmond built its race-based remedial program is now, nearly thirty years later, obviously inadequate.

The court looks to Fifth Circuit cases but finds that they too are distinguishable. In Scott, the City of Jackson, Mississippi, did not adopt the conclusions of a disparity study that it commissioned and chose to retain its prior MBE utilization goal of fifteen percent without any basis in fact. See Scott, 199 F.3d at 210. In Houston Contractors Association and Dean, the Fifth Circuit found conflicts in evidence that needed to be resolved before the court could apply strict scrutiny. See Hous. Contractors Ass'n, 189 F.3d at *2; Dean, 438 F.3d at 458-62. In the former, the conflict concerned whether the program was implemented as written, and, in the latter, one of the evidentiary issues concerned the implementation and efficacy of alternative remedies. See Hous. Contractors Ass'n, 189 F.3d at *2; Dean, 438 F.3d at 458-62. Here, Defendant did adopt the NERA Study, and there are no fact issues to prevent the court from evaluating the 2013 Program pursuant to the applicable constitutional standards of review.

Plaintiff does not dispute that Defendant has a compelling interest in making sure that its public dollars do not finance racial discrimination. Courts have confirmed as much. See Croson, 488 U.S. at 492; Scott, 199 F.3d at 217. The compelling-interest dispute between the parties here is whether Defendant has demonstrated a strong basis in evidence that remedial action is necessary. Plaintiff argues that the NERA Study "is not persuasive evidence of discrimination by [Defendant] or within the Houston construction marketplace."¹⁵⁴

154 Doc. 22, Pl.'s Mot. for Summ. J. p. 12.

Since Croson was decided, disparity studies that compare the availability of MWBEs in the relevant market with their utilization in local public contracting have been widely recognized as strong evidence. See Scott, 199 F.3d at 218. The NERA Study defined the market area by reviewing past contract information. By looking at information from Defendant's own contracts, NERA properly defined the relevant market according to two critical factors, geography and industry. Those parameters, weighted by dollars attributable to each industry, were used to identify for comparison MWBEs that were available and MWBEs that had been utilized in Defendant's construction contracting over the last five and one-half years. NERA did not merely count firms to determine availability; NERA adjusted for owner labor market experience and educational attainment in addition to geographic location and industry affiliation. Plaintiff has produced no evidence that the availability estimate was inadequate.

Plaintiff complains that the NERA Study ignored capacity, but NERA did so for the stated rationale that capacity factors are not economically reliable in determining availability in the absence of discrimination. Plaintiff's counter-position is not supported by any contrary evidence or expert opinion. NERA sufficiently limited the statistical data to available MWBEs.

Along those same lines, Plaintiff suggests that analysis of bidder data is a better way to identify MWBEs that are truly available in the sense of being qualified, willing, and able. Plaintiff presents no comparative evidence based on bidder data, and the court finds that bidder data may produce availability statistics that are skewed by active and passive discrimination in the market. For example, the documentation regarding bids and quotes of subcontractors is in the possession of prime contractors, who may or may not be motivated to be forthcoming with accurate information. Also, the solicitation of bids itself may have been infused with active or passive discrimination.

In addition to being underinclusive due to discrimination, it may be overinclusive due to inaccurate self-evaluation by firms offering bids despite the inability to fulfill the contract. See Concrete Works of Colo., Inc. v. City & Cty. of Denver, 321 F.3d 950, 983 (10^th Cir. 2003)("If availability is calculated on an individual basis using [this] approach, it is possible that unqualified firms would be included in the availability figure simply because they bid on a particular project."). As an illustration of this point, the court need look no further than J.A. Croson Company's argument to the city in Croson that the local MBE which sought to submit a bid on fixtures was not qualified to do so because it was not an authorized supplier and was subject to credit approval. See Croson, 488 U.S. at 483. Whether the MBE was qualified to make the bid is not point, only that, even though it sought to make a bid, it was arguable whether it was in fact qualified.¹⁵⁵ The law does not require an individualized approach that measures whether MWBEs are qualified on a contract-by-contract basis. See Concrete Works of Colo., Inc., 321 F.3d at 983.

155 The same is true of bid-submitting nonMWBEs.

Defendant commissioned NERA to determine whether its process of awarding construction contracts was tainted by discrimination. NERA designed a study and defined the relevant market by geography and industry so as identify the targeted discrimination with sufficient particularity. The results of the NERA Study, indicating significant statistical adverse disparities as to businesses owned by African Americans and Asians, provide more than a prima facie case of a strong basis in evidence. That evidence justifies the 2013 Program's utilization goals for businesses owned by Black American, Asian-Pacific American, and subcontinent Asian American.

The disparity analysis did not reflect significant statistical disparities as to businesses owned by Hispanic American, Native American, or, over the course of the entire period, non-minority women. However, the evidence of significant statistical adverse disparity in the utilization of Hispanic-owned businesses in the unremediated, private sector meets Defendant's prima facie burden of producing a strong evidentiary basis for the continued inclusion of businesses owned by Hispanic Americans. The difference between the private sector and Defendant's construction contracting is especially notable because the utilization of Hispanic-owned businesses by Defendant has benefitted from Defendant's remedial program for many years. Without a remedial program, the evidence suggests, and no evidence contradicts, a finding that utilization would fall back to private sector levels.

With regard to businesses owned by Native Americans, the NERA study indicated that they were utilized to a higher percentage than their availability in the relevant market area. In other words, the disparity in utilization favored Native-American-owned businesses in contracting with Defendant. NERA suggested that a significant statistical disparity would exist if two of the contracting Native-American-owned businesses were disregarded. That opinion is not evidence of the need for remedial action. The court finds no equal-protection significance to the fact that the majority of contracts let to Native-American-owned businesses were to only two firms. The utilization goal for businesses owned by Native Americans is not supported by a strong evidentiary basis.

The utilization of WBEs declined by fifty percent when they no longer benefitted from remedial goals. Because WBEs were eliminated during the period studied, the significance of statistical disparity is not reflected in the numbers for the period as a whole. During the time that WBEs were not part of the program, the statistical disparity between availability and utilization was significant. The precipitous decline in the utilization of WBEs after WBEs were eliminated and the significant statistical disparity when WBEs did not benefit from preferential treatment provide a strong basis in evidence for the necessity of remedial action. Plaintiff offers no evidence of a gender-neutral reason for the decline.

Plaintiff's argument that prime contractor and subcontractor data should not have been combined is nothing more than a vacuous criticism. Plaintiff cites Contractors Association of Eastern Pennsylvania v. City of Philadelphia, 91 F.3d 586, 599 (1996), in support of its argument that prime contractor and subcontractor data must be evaluated separately in order to determine where the discrimination is occurring. Plaintiff misapplies the cited case. In that case, the city had proposed three theories of discrimination. See id. The portion cited by Plaintiff dealt with the city's evidence in support of the theory that prime contractors were discriminating in the selection of subcontractors. The court found that the record did not provide a "firm basis for inferring discrimination by prime contractors in the subcontracting market during that period." Id. The point was not that prime contractor and subcontractor data must be evaluated separately but that the city in that case failed to produce reliable subcontractor data to indicate discrimination by prime contractors. That is not a problem in this case.

Here, NERA was able to identify MWBEs that had contracted with Defendant by industry and those available in the relevant market by industry. Unlike the evidence in Contractors Association of Eastern Pennsylvania, the data here was specific and complete. Separately considering prime contractors and subcontractors is not only unnecessary but may be misleading. The anecdotal evidence indicates that construction firms had served, on different contracts, in both roles.¹⁵⁶ The law requires that the targeted discrimination be identified with particularity, not that every instance of explicit or implicit discrimination be exposed. NERA defined the relevant market at a sufficient level of particularity to produce evidence of passive discrimination in Defendant's awarding of construction contracts and to reach constitutionally sound results. Plaintiff's argument on this point is not evidence of the absence of discrimination in the bidding process.

156 Doc. 32-2, Ex. 1.1 to Def.'s Mot. for Summ. J., Expert Report of Jon Wainwright pp. 28-30.

Plaintiff criticizes the anecdotal evidence with which NERA supplemented its statistical analysis as not having been verified and investigated. Anecdotes are not the sole or even primary evidence of discrimination in this case. As with the above arguments, Plaintiff certainly could have presented its own evidence but did not. Plaintiff presents no contrary body of anecdotal evidence and points to nothing that calls into question the specific results of the NERA's market surveys and focus groups. One reason anecdotal evidence is valuable supplemental evidence is that it reaches what statistics cannot: "a witness' narrative of an incident told from the witness' perspective and including the witness' perceptions." H.B. Rowe Co., 615 F.3d at 249 (quoting Concrete Works of Colo., Inc., 321 F.3d at 989).

Plaintiff's final compelling-interest argument concerns the regression analyses of business formation, earnings, and capital markets. Plaintiff criticizes regression analysis for failing to precisely point to where the identified discrimination was occurring. Plaintiff also contends that the data used in the regression analyses was stale.

Plaintiff's focus on identifying where discrimination is occurring misses the point. As suggested by Defendant's expert, discrimination is not likely to stand up and be counted.¹⁵⁷ Regression analysis is not intended to point to specific sources of discrimination but to eliminate factors other than discrimination that might explain disparities. Plaintiff fails to convince the court that the precise "where" of discrimination plays any significant role in the analysis or is necessary to formulate a remedy. Discrimination is not revealed through evidence of explicit discrimination but is revealed through unexplainable disparity.¹⁵⁸

157 See id. pp. 36-39.

158 See id.

The data that NERA used in the regression analyses were the most current available data at the time. For the most part, the data dated from within a couple of years or less of the start of the study period. Where NERA used earlier data, it looked at longer periods to demonstrate that disparities remained relatively constant.¹⁵⁹ Plaintiff has produced no evidence that the data on which the regression analyses were based were invalid.

159 See id. p. 33.

Having found a strong basis in evidence to support a remedial program, the court turns to consideration of the narrow tailoring factors. The lack of parity between MWBEs and nonMWBEs in Defendant's letting of contracts requires continued remedial action, as discussed above. Defendant has augmented the 2013 Program with a variety of race-neutral remedies, including many educational opportunities for young businesses. Some of the programs have been continued, and some are new. For those that are continuing, evidence of their efficacy or lack thereof is found in the disparity analysis. While they may have a positive effect, they have not eliminated the discrimination. For the new programs, evaluation is premature. Additionally, the ability to substitute SBEs for a percentage of the contract is a race-neutral alternative that Plaintiff agrees is effective, although Plaintiff views the current percentage as too low. The court finds Defendant's race-neutral programming sufficient to satisfy the requirements of narrow tailoring.

Regarding flexibility and duration of the 2013 Program, the court also finds that these aspects satisfy narrow tailoring. The 2013 Program employs goals as opposed to quotas, sets goals on a contract-by-contract basis, allows substitution of SBEs for MWBEs for up to four percent of the contract, includes a process for allowing good-faith waivers, and builds in due process for suspensions of contractors who fail to make good-faith efforts to meet contract goals or MWSBEs that fail to make good-faith efforts to meet all participation requirements. Defendant committed to use its best efforts to review the 2013 Program at least every five years, a reasonably brief duration.

The thirty-four percent annual goal is proportional to the availability of MWBEs historically suffering discrimination. Although the program is overinclusive with regard to Native-American-owned businesses, their elimination will not affect the proportionality of the thirty-four-percent annual goal. The availability estimate for MWBEs was 34.73 percent, and the availability of Native-American-owned businesses was 1.04 percent. The removal of Native-American-owned businesses from the overall MWBE availability percentage lowers it to 33.69. The court finds the proportionality factor satisfied.

Finally, the court finds that effect of the 2013 Program on third parties is not so great as to impose an unconstitutional burden. The burden on non-minority SBEs, such as Plaintiff, is lessened by the four-percent substitution provision. As one astute district court noted, "the mere possibility that innocent parties will share the burden of a remedial program is itself insufficient to warrant the conclusion that the program is not narrowly tailored." Kornhass Constr., Inc. v. State of Okla., Dep't of Cent. Servs., 140 F. Supp.2d 1232, 1247 (W.D. Okla. 2001). In fact, the proportional sharing of opportunities is, at the core, the point of a remedial program.

Ultimately, after a searching examination of the NERA Study and the 2013 Program, the court is satisfied that Defendant established a prima facie case of compelling interest and narrow tailoring for all aspects of the 2013 Program except goals for Native-American-owned businesses. Plaintiff failed to produce any evidence, much less the greater weight of evidence, calling into question the constitutionality of the 2013 Program.

IV. Conclusion

Based on the foregoing, the court RECOMMENDS that Plaintiff's Motion for Summary Judgment be GRANTED as to the utilization goal for Native-American-owned businesses and DENIED in all other respects and Defendant's Motion for Summary Judgment be DENIED as to the Native-American-owned businesses and GRANTED in all other respects. Defendant's Motion to Exclude the Expert Testimony of John Sullivan is GRANTED.

The Clerk shall send copies of this Memorandum and Recommendation to the respective parties who have fourteen days from the receipt thereof to file written objections thereto pursuant to Federal Rule of Civil Procedure 72(b) and General Order 2002-13. Failure to file written objections within the time mentioned shall bar an aggrieved party from attacking the factual findings and legal conclusions on appeal.

The original of any written objections shall be filed with the United States District Clerk electronically. Copies of such objections shall be mailed to opposing parties and to the chambers of the undersigned, 515 Rusk, Suite 7019, Houston, Texas 77002.

SIGNED in Houston, Texas, this 16^th day of February, 2016.

/s/_________

U.S. MAGISTRATE JUDGE