545 F.3d 1023 | Fed. Cir. | 2008
Before MICHEL, Chief Judge, MAYER, Circuit Judge, and STEARNS, [*] District Judge. MICHEL, Chief Judge.
This case concerns the constitutionality of 10 U.S.C. § 2323 (“Section 1207”), which, in relevant part, (1) sets a “goal” that five percent of federal defense contracting dollars for each fiscal year be awarded to certain entities including small business concerns owned and controlled by “socially and economically disadvantaged individuals”; (2) incorporates the Small Business Act’s presumption that Black Americans, Asian Americans, Hispanic Americans, and Native Americans are socially disadvantaged individuals; and (3) provides that the Department of Defense shall give specified forms of assistance to the listed entities and may, when practicable and necessary to achieve the five percent goal, make advance payments to those entities and award contracts to them at prices up to ten percent above fair market cost.
Plaintiff-Appellant Rothe Development Corporation (“Rothe”) filed its first complaint in this case in November, 1998, bringing an equal protection challenge to Section 1207 both on its face and as applied by Defendants-Appellants the Department of Defense and Department of the Air Force (together, “DOD”) earlier that year, when DOD awarded a contract to an Asian-American-owned business despite the fact that Rothe—owned by a Caucasian woman—was the lowest bidder. Since this case began in 1998, Congress has reenacted Section 1207 a number of times, the district court has rendered judgment in this case three times, and we have remanded the case twice without reaching the ultimate question of constitutional muster. Most recently, the district court granted summary judgment to DOD on Rothe’s facial challenge, Rothe Dev. Corp. v. U.S. Dep’t of Def., 499 F. Supp. 2d 775 (W.D. Tex. 2007) (“Rothe VI”), and, after Rothe’s claim for monetary relief became moot, entered a final judgment in favor of DOD on September 25, 2007. Rothe appealed, and we heard oral argument on September 2, 2008.
Now, in our third opinion in this case, we must decide whether Section 1207, on its face, as reenacted in 2006, violates the right to equal protection (as incorporated against the federal government by the Due Process Clause of the Fifth Amendment). Because we will hold that Congress did not have a “strong basis in evidence” before it in 2006, upon which to conclude that DOD was a passive participant in racial discrimination in relevant markets across the country and that therefore race-conscious remedial measures were necessary, we will reverse the district court’s judgment in part, and will hold that Section 1207 (i.e., 10 U.S.C. § 2323) is unconstitutional on its face.
BACKGROUND
A. Relevant Statutes and Regulations Section 1207, relevant sections of the Small Business Act, pertinent regulations, and their history are set out in detail in the district court’s thorough opinion. See Rothe VI, 499 F. Supp. 2d at 784-94. A brief review follows here.
1. Section 1207 as Originally Enacted Congress first enacted Section 1207 in 1986, for a three-year period. See
National Defense Authorization Act of 1987, Pub. L. No. 99-661, § 1207, 100 Stat. 3816, 3973 (Nov. 14, 1986). The statute, titled “Contract Goal for Minorities,” provided that, except where compelling national security considerations required otherwise:
a goal of 5 percent of the amount of [DOD procurement, R&D, military construction, and maintenance contracts] shall be the goal of the Department of Defense in each of fiscal years 1987, 1988, and 1989 for the total combined amount obligated for contracts and subcontracts entered into with [inter alia,] small business concerns, including mass media, owned and controlled by socially and economically disadvantaged individuals (as defined by section 8(d) of the Small Business Act (15 U.S.C. 637(d)) and regulations issued under such section), the majority of the earnings of which directly accrue to such individuals . . . .
Id. § 1207(a) (emphasis added). Section 8(d) of the Small Business Act and relevant regulations, in turn, provided at that time that Black Americans, Hispanic Americans, Native Americans, Asian Pacific Americans, and other minorities were presumed to be “socially and economically disadvantaged individuals.” See 15 U.S.C. § 637(d)(3)(C)(ii) (1990).
Subsection (e) of Section 1207, titled “Competitive Procedures and Advanced Payments,” directed DOD to take certain measures to attain the goal of directing five percent of contract dollars to businesses owned by socially and economically disadvantaged individuals (such businesses, “SDBs”). Most prominently, the statute provided that
[t]o the extent practicable and when necessary to facilitate achievement of the 5 percent goal described in subsection (a), the Secretary of Defense may enter into contracts using less than full and open competitive procedures (including awards under section 8(a) of the Small Business Act), but shall pay a price not exceeding fair market cost by more than 10 percent in payment per contract to contractors or subcontractors described in subsection (a).
Pub. L. No. 99-661, § 1207(e)(3) (emphasis added). DOD implemented this directive by applying a price evaluation adjustment (“PEA”) to bids submitted by non-SDB bidders, increasing those bids by ten percent before comparing them to the bids submitted by SDBs.
2. History of Section 1207 In 1989, before Section 1207 was set to expire, Congress reenacted the statute
for another three years. See National Defense Authorization Act for Fiscal Years 1990 and 1991, Pub. L. No. 101-189, § 831(b), 103 Stat. 1352, 1507 (Nov. 29, 1989). In 1992, Congress reenacted Section 1207 again, this time for seven years. See National Defense Authorization Act for Fiscal Year 1993, Pub. L. No. 102-484, § 801(a)(1)(B), 106 Stat. 2315, 2442 (Oct. 23, 1992).
In 1998, Congress amended the statute without yet reenacting it. This amendment required DOD to suspend the PEA mechanism for an entire year after any fiscal year in which the five percent goal had been met. See Strom Thurmond National Defense Authorization Act for Fiscal Year 1999, Pub. L. No. 105-261, § 801, 112 Stat. 1920, 2080-81 (Oct. 5, 1999). Because the five percent goal was met for fiscal year 1998, the PEA was suspended in 1999. See Suspension of the Price Evaluation Adjustment for Small Disadvantaged Businesses, 64 Fed. Reg. 4847 (Feb. 1, 1999).
Congress reenacted Section 1207 in 1999, again in 2002, and most recently in 2006. See National Defense Authorization Act for Fiscal Year 2000, Pub. L. No. 106- 65, § 808, 113 Stat. 512, 705 (Oct. 5, 1999); Bob Stump National Defense Authorization Act for Fiscal Year 2003, Pub. L. No. 107-314, § 816, 116 Stat. 2610 (Dec. 2, 2002); National Defense Authorization Act for Fiscal Year 2006, Pub. L. No. 109-163, § 842, 119 Stat. 3136 (Jan. 6, 2006). The statute is now set to expire at the end of fiscal year 2009 if it is not reenacted before then. See 10 U.S.C. § 2323(k)(1) (“This section applies in the Department of Defense to each of fiscal years 1987 through 2009.”).
As the district court noted, DOD met the five percent goal in every fiscal year from 1998 to 2006, and thus the PEA was serially suspended through March 9, 2008. See Rothe VI, 499 F. Supp. 2d at 792-93. In February of 2008, while this appeal was pending, DOD published notice that it met the five percent goal in fiscal year 2007 as well, and that DOD would continue to suspend the PEA through March 9, 2009. Suspension of the Price Evaluation Adjustment for Small Disadvantaged Businesses, 73 Fed. Reg. 9304 (Feb. 20, 2008).
3. Section 1207 Today The present Section 1207, i.e., as reenacted in 2006, and relevant regulations
differ from the original enactment and regulations to some degree, as the district court discussed. See Rothe VI, 499 F. Supp. 2d at 790-94. First, in addition to the PEA suspension clause added by the 1998 amendments, the present statute provides that whenever the PEA is not suspended, the size of the adjustment made to non-SDB bids need not be ten percent, and must be smaller if non-SDBs are being denied a reasonable opportunity to compete. [2] See 10 U.S.C. § 2323(e)(3)(A) (“The head of an agency shall adjust the percentage specified in the preceding sentence for any industry category if available information clearly indicates that nondisadvantaged small business concerns in such industry category are generally being denied a reasonable opportunity to compete for contracts because of the use of that percentage in the application of this paragraph.”); see also id. § 2323(g)(1)(A) (“To the maximum extent practicable, the head of the agency shall ensure that no particular industry category bears a disproportionate share of the contracts awarded to attain the goal established by subsection (a).”); id. § 2323(g)(2) (“Upon making a determination that a particular industry category is bearing a disproportionate share, the head of the agency shall take appropriate actions to limit the contracting activity’s use of set asides in awarding contracts in that particular industry category.”).
Second, through changes to pertinent regulations, minorities are no longer presumed to be economically disadvantaged regardless of their wealth, and instead must represent that they meet the same economic requirement met by any SDB applicant. See 13 C.F.R. § 124.1002(c) (2008) (“In assessing the personal financial condition of an individual claiming economic disadvantage, his or her net worth must be less than $750,000 after taking into account [certain] exclusions . . . .”). [3]
Members of certain minority groups are still presumed to be socially disadvantaged, however. See id. § 124.103(b) (“There is a rebuttable presumption that the following individuals are socially disadvantaged: Black Americans; Hispanic Americans; Native Americans (American Indians, Eskimos, Aleuts, or Native Hawaiians); Asian Pacific Americans (persons with origins from Burma, Thailand, Malaysia, Indonesia, Singapore, Brunei, Japan, China (including Hong Kong), Taiwan, Laos, Cambodia (Kampuchea), Vietnam, Korea, The Philippines, U.S. Trust Territory of the Pacific Islands (Republic of Palau), Republic of the Marshall Islands, Federated States of Micronesia, the Commonwealth of the Northern Mariana Islands, Guam, Samoa, Macao, Fiji, Tonga, Kiribati, Tuvalu, or Nauru); Subcontinent Asian Americans (persons with origins from India, Pakistan, Bangladesh, Sri Lanka, Bhutan, the Maldives Islands or Nepal) . . . .”); id. § 124.1002(a) (“In determining whether a firm qualifies as an SDB, the criteria of social and economic disadvantage and other eligibility requirements established in subpart A of this part [including § 124.03] apply, including the requirements of ownership and control and disadvantaged status, unless otherwise provided in this subpart.”).
Third, a person who is economically disadvantaged but is not a member of one of the minority groups listed in the Small Business Administration (“SBA”) regulations may nevertheless pursue SDB status if the person can “establish individual social disadvantage by a preponderance of the evidence.” Id. § 124.103(c)(1) (emphasis added). Although non-minorities have long had the option to demonstrate social disadvantage, the standard of proof was once higher. See 13 C.F.R. § 124.105(c)(1) (1990) (“An individual who is not a member of one of the above-named groups must establish his/her individual social disadvantage on the basis of clear and convincing evidence.” (emphasis added)).
Fourth, a business that loses a bid to a SDB because of a PEA may protest the disadvantaged status of the successful bidder. See 37 C.F.R. § 124.1017(a) (2007); see also Small Disadvantaged Business Program, 73 Fed. Reg. 57,490, 57,495 (Oct. 3, 2008) (to be codified at 37 C.F.R. pt. 123) (redesignating 13 C.F.R. § 124.1017(a) as § 124.1007(a)). The SBA can revoke an entity’s SDB status if the protest contains credible evidence that the entity no longer meets SDB requirements or that the application contains false and misleading information. See 37 C.F.R. § 124.1021(c) (2007); see also Small Disadvantaged Business Program, 73 Fed. Reg. at 57,495 (redesignating 37 C.F.R. § 124.1021(c) as § 124.1011(c)). B. Facts and Procedural History
The procedural history of this case, like the history of the statute itself, is substantial. It is set out in significant detail in the district court’s most recent opinion, see Rothe VI, 499 F. Supp. 2d at 794-815, as well as in our two prior opinions. Here, we briefly review the case history with a focus on those arguments and developments that are relevant to the question before us today.
1. Genesis of This Dispute Starting in the late 1980’s, the Department of the Air Force contracted with Rothe
to maintain, operate, and repair computer systems at Columbus Air Force Base in Mississippi. In the late 1990’s, the Air Force decided to consolidate Rothe’s contract with a contract for communications services, to issue a solicitation for competitive bids, and to let the contract pursuant to the Section 1207 program. Rothe, owned by a Caucasian female, bid $5.57 million. International Computer and Telecommunications, Inc. (“ICT”), a competitor to Rothe owned by a Korean-American couple and certified as a SDB, bid $5.75 million. Although Rothe’s bid was lower than ICT’s bid and was in fact the lowest bid, the Air Force considered Rothe’s bid to be $6.1 million, higher than ICT’s bid, because of the PEA. Therefore the Air Force awarded the contract to ICT. See Rothe VI, 499 F. Supp. 2d at 780-83.
2. Rothe’s Suit and the First Appeal—Rothe I, Rothe II, and Rothe III Rothe filed suit against DOD in November of 1998, and moved for a preliminary
injunction against DOD’s award of the contract to ICT. The district court denied Rothe’s motion, and on February 2, 1999, Rothe filed its First Amended Complaint (“FAC”). The FAC, which is still the operative complaint in this case, included claims for (1) a declaratory judgment that Section 1207 is unconstitutional on its face, (2) an injunction prohibiting DOD from proceeding with the contract Rothe had lost to ICT, and (3) recovery of Rothe’s bid preparation costs, “in an amount not to exceed $10,000,” under the Little Tucker Act.
On April 27, 1999, the district court granted summary judgment to DOD, upholding the constitutionality of Section 1207 and denying Rothe any relief. Rothe Dev. Corp. v. U.S. Dep’t of Def., 49 F. Supp. 2d 937, 954 (W.D. Tex. 1999) (Prado, J.) (“Rothe I”). Rothe appealed to the Fifth Circuit, and DOD moved to dismiss the appeal or to transfer it to the Federal Circuit because of Rothe’s Tucker Act claim. On October 27, 1999, the Fifth Circuit transferred the appeal to us, explaining that the Federal Circuit had exclusive appellate jurisdiction because the district court’s jurisdiction in Rothe I was based in part on the Little Tucker Act. Rothe Dev. Corp. v. U.S. Dep’t of Def., 194 F.3d 622, 625 (5th Cir. 1999) (“Rothe II”).
We heard oral argument on November 8, 2000, and issued our opinion on August 20, 2001. Rothe Dev. Corp. v. U.S. Dep’t of Def., 262 F.3d 1306 (Fed. Cir. 2001) (“Rothe III”). We agreed with the Fifth Circuit that appellate jurisdiction lay exclusively in the Federal Circuit, id. at 1316, but we did not affirm or reverse the district court on the merits. Rather, we vacated and remanded for further proceedings, because “the district court improperly applied a deferential legal standard rather than ‘strict scrutiny,’ and also impermissibly relied on post-reauthorization evidence to support [Section 1207’s] constitutionality as reauthorized.” Id. at 1312. We directed the district court to apply strict scrutiny on remand, “particularly in accordance with the principles set forth in [Richmond v. J. A. Croson Co., 488 U.S. 469 (1989) and Adarand Constructors v. Peña, 515 U.S. 200, 226 (1995) (‘Adarand III’)].” Rothe III, 262 F.3d at 1329.
3. The Second Judgment and Appeal—Rothe IV and Rothe V On remand, DOD moved to dismiss Rothe’s claims as moot in light of factual
developments since the decision in Rothe I, namely: (1) the contract at issue had been superseded by a replacement contract, which was put up for bid outside of the Section 1207 program, with Rothe rebidding and losing to a lower bidder; (2) DOD had offered to tender $10,000 to Rothe, the full amount Rothe could recover on its Little Tucker Act claim; and (3) the PEA had been serially suspended since 1999. On July 5, 2002, the district court granted DOD’s motion in part, agreeing that Rothe’s Little Tucker Act claim and its requests for an injunction against further work on the contract and for an equitable award of the contract were moot, but holding that Rothe could still maintain its claims for declaratory judgments that Section 1207 is unconstitutional on its face and was unconstitutional as applied to Rothe in 1998. Rothe Dev. Corp. v. U.S. Dep’t of Def., No. 5:98-cv-01011-XR, Docket No. 98 (W.D. Texas).
In 2003, the district court issued a series of discovery orders that unintentionally but improperly limited discovery regarding Rothe’s facial challenge. These orders repeatedly referenced the 1992 reenactment of Section 1207 (which was in effect in 1998, the relevant time for Rothe’s as-applied challenge), and were construed by the parties to prohibit some discovery relevant to the facial constitutionality of Section 1207, i.e., discovery related to the evidence before Congress during the statute’s 2002 reenactment. See Rothe VI, 499 F. Supp. 2d at 812 & n.39 (acknowledging that “taken out of context, some loose language in those three discovery orders could be construed to limit the issues on remand,” but maintaining that scope of discovery “could have been resolved in a one-page motion for clarification” had the parties chosen to file one).
On July 2, 2004, the district court issued an opinion on Rothe’s remaining claims. Rothe Dev. Corp. v. U.S. Dep’t of Def., 324 F. Supp. 2d 840 (W.D. Tex. 2004) (Rodriguez, J.) (“Rothe IV”). The district court held that Section 1207 was indeed unconstitutional as applied in 1998, because DOD failed to demonstrate that there was sufficient statistical evidence of discrimination before Congress when the statute was reenacted in 1992. Id. at 850. But the district court held that the statute was constitutional on its face, because there was sufficient statistical evidence of discrimination before Congress when the statute was then most recently reenacted in 2002. Id. at 860. In a separate opinion, in August of 2004, the district court also declined to award Rothe attorney’s fees on its as-applied challenge. Rothe Dev. Corp. v. U.S. Dep’t of Def., No. SA-98-CA-1011-XR, 2004 U.S. Dist. LEXIS 17305, at *7 (W.D. Tex. Aug. 31, 2004).
Rothe appealed. We heard argument on March 7, 2005, and issued an opinion on June 28, 2005. Rothe Dev. Corp. v. Dep’t of Def., 413 F.3d 1327 (Fed. Cir. 2005) (“Rothe V”). We vacated the district court’s holding that Rothe’s Little Tucker Act claim was moot, explaining that DOD’s offer to tender $10,000 did not moot the claim because an actual tender is required for mootness, and the DOD had not yet tendered the money. Id. at 1331-32. We affirmed the district court’s holding that Rothe’s claim for an equitable award of the contract was moot, but for slightly different reasons than given by the district court. Id. at 1332. In our view, the claim was moot because the disputed contract expired in 2001, not merely because it had been rebid outside of the Section 1207 program. Id. And we affirmed the district court’s holding that Rothe’s facial challenge was not moot, explaining that though the PEA has been serially suspended since 1999, it will only continue to be suspended as long as DOD meets the five percent goal, which is not a certainty. Id. at 1333.
But, as in Rothe III, in Rothe V we did not rule on the merits of Rothe’s facial challenge. Rather, we explained that the district court’s ruling on Rothe’s facial challenge in Rothe IV was an unwarranted sua sponte grant of summary judgment. DOD had moved to dismiss Rothe’s facial challenge as moot, but DOD had not asked the district court to reach the merits of Rothe’s facial challenge, and thus the district court’s ruling on the merits, in combination with the court’s errant discovery orders, had unfairly deprived DOD of “an opportunity to introduce its evidence and arguments” on the facial challenge. Rothe V, 413 F.3d at 1336; id. at 1335 (noting that DOD agreed with Rothe “that the district court’s findings and legal analysis do not support a holding that the present reauthorization of section 1207 is facially constitutional”). Because “the district court’s discovery orders [had] clearly limited the issues to exclude the present reauthorization of section 1207,” we held that “the government was not on notice that it was required to come forward with all of its evidence,” and we therefore vacated the grant of summary judgment. Id.
Though we remanded the case “for development of the record,” id. at 1339, we preemptively addressed three specific questions pertinent to the facial challenge. First, we held that United States v. Salerno, 481 U.S. 739 (1987), in which the Court stated that, to win a facial challenge, a challenger “must establish that no set of circumstances exists under which the Act would be valid,” was a pronouncement “of limited relevance here, at most describing a conclusion that could result from the application of the strict scrutiny test.” Rothe V, 413 F.3d at 1337-38. Second, we confirmed that the district court could not rely on any evidence that was not before Congress in 2002. Id. at 1338 (“[T]o be relevant in the strict scrutiny analysis, the evidence must be proven to have been before Congress prior to enactment of the racial classification.”). And third, we clarified that Rothe’s argument to exclude certain statistical evidence as stale “relates to whether the data itself is ‘outdated,’ . . . not [to] whether Congress was aware of the data,” and thus “the district court should consider on remand whether the data presented is so outdated that it does not provide a strong basis in evidence for the most recent reauthorization of section 1207.” Id.
Finally, we held that Rothe failed to preserve for appeal the issue of attorney fees for its successful as-applied challenge, because Rothe had improperly attempted to incorporate arguments into its opening brief by reference to the Joint Appendix, rather than articulating substantive arguments in the brief. Id. at 1339.
4. The Third Judgment and Appeal—Rothe VI On July 28, 2005, one month after our decision in Rothe V, Rothe moved in the
district court for a preliminary injunction against DOD’s continued use of any race-based procurement programs. The district court denied that motion and Rothe’s motion for reconsideration, and noted that Rothe was now attempting to challenge not only the PEA mechanism, but also the other provisions of Section 1207. See Rothe Dev. Corp. v. U.S. Dep’t of Def., No. SA-98-CA-1011-XR, 2006 U.S. Dist. LEXIS 53738, at *3 (W.D. Tex. July 24, 2006). The court wrote that
Rothe also objects to the following elements of the § 1207 program: (1) subcontracting incentive programs; (2) awards using less than full and open competition/set asides to the designated groups of 10 U.S.C. § 2323(a) with no competition, to include awards under section 8(a) of the Small Business Act; (3) advance payments; (4) assistance as provided in 10 U.S.C. § 2323(c); and (5) the provision of SDB status to historically black colleges and universities.
Id. (internal footnotes and quotations omitted). Though Rothe had only made its additional objections clear to the district court “for the first time” in 2005, the court held that it would consider the new arguments to be within the parameters of the FAC. Id.; see also Rothe VI, 499 F. Supp. 2d at 814 (“Prior to the filing of the motion for reconsideration, this Court had always construed Plaintiff’s First Amended Complaint as objecting only to the PEA. . . . Nevertheless, the Court stated that it would consider Rothe’s challenge to the entire 1207 Program as coming within the parameters of the current complaint.” (footnote omitted).
Because Section 1207 was reenacted in January of 2006, see National Defense Authorization Act for Fiscal Year 2006, Pub. L. No. 109-163, § 842, 119 Stat. 3136 (Jan. 6, 2006), the district court directed the parties “to present what, if any, [non-stale] evidence was before Congress at the time of this 2006 reauthorization,” and allowed Rothe to conduct some additional discovery for that purpose. 2006 U.S. Dist. LEXIS 53738, at *14-15. After that discovery, Rothe moved for summary judgment on September 8, 2006, and DOD cross-moved for summary judgment on November 6, 2006. See Rothe VI, 499 F. Supp. 2d at 815.
On August 10, 2007, the district court granted summary judgment to DOD on Rothe’s facial challenge. Id. at 883. In light of the 2006 reenactment, the district court held that the facial constitutionality of the 1999 and 2002 reenactments were moot and outside the scope of our remand in Rothe V, and that the proper question was the facial constitutionality of Section 1207 as reenacted in 2006. Rothe VI, 499 F. Supp. 2d at 818-25. On the merits, the court applied strict scrutiny and upheld the statute against Rothe’s facial challenge, concluding that Congress sought to further a compelling interest supported by a “strong basis in evidence,” and that the statute was narrowly tailored to that interest. Id. at 835-83. We will review this holding in detail below.
In Rothe VI, the district court also addressed Rothe’s claim under the Little Tucker Act, which DOD had previously attempted to satisfy by offering to tender $10,000 to Rothe. Recounting our holding in Rothe V that this claim would not be moot until DOD “produced the subject matter of the tender, e.g., by providing Rothe with a $10,000 check or depositing such a check with the court,” Rothe V, 413 F.3d at 1331-32, the district court ordered DOD to deposit $10,000 with the court registry as a formal act of tender, and held that “[i]t is unnecessary for the Court to render a judgment on the merits of Rothe’s Little Tucker Act [c]laim because the claim will become moot after the Government deposits $10,000 into the registry of the Court.” Rothe VI, 499 F. Supp. 2d at 817.
Thereafter, on September 20, 2007, DOD deposited $10,000 into the court registry, and Rothe made a motion to disburse the money a few days later. The court granted Rothe’s motion on September 25, 2007, and entered final judgment for DOD that same day. Rothe filed a timely notice of appeal on September 27, 2007.
DISCUSSION
A. Jurisdiction and Rothe’s Tucker Act Claim Before turning to the merits of Rothe’s appeal, we briefly explain why appellate jurisdiction lies exclusively in this Court rather than in the Fifth Circuit. With certain exceptions not relevant here, we possess appellate jurisdiction “of an appeal from a final decision of a district court of the United States . . . if the jurisdiction of that court was based, in whole or in part, on [28 U.S.C. § 1346, known as the Tucker Act.]” 28 U.S.C. § 1295(a)(2). In Rothe III, we possessed jurisdiction because Rothe’s FAC contained a claim for recovery of Rothe’s bid preparation costs under the Tucker Act. See Rothe III, 262 F.3d at 1316 (“As a suit to recover its bid preparation costs, Rothe’s complaint invoked the Tucker Act. Thus, this court has exclusive jurisdiction over all issues Rothe raises in this appeal.”).
Rothe’s claim under the Tucker Act is no longer at issue; it has been satisfied by DOD’s tender, and Rothe’s acceptance, of $10,000. We retain appellate jurisdiction, however, because under 28 U.S.C. § 1295(a)(2), as under § 1295(a)(1) (regarding patent law), “[t]he path of appeal is determined by the basis of jurisdiction in the district court, and is not controlled by . . . the substance of the issues that are appealed.” Abbott Labs. v. Brennan, 952 F.2d 1346, 1349-50 (Fed. Cir. 1991). Of course, the basis of a district court’s jurisdiction—and thus the path of appeal—may change over time in a case, for example, if certain claims are dismissed without prejudice to later refiling. See, e.g., Nilssen v. Motorola, Inc., 203 F.3d 782, 785 (Fed. Cir. 2000); Gronholz v. Sears, Roebuck & Co., 836 F.2d 515, 518 (Fed. Cir. 1987). But that is not what happened here. Rather, Rothe’s claim under the Tucker Act was satisfied, and thus may not be refiled later. Cf. Chamberlain Group, Inc. v. Skylink Techs., Inc., 381 F.3d 1178, 1190 (Fed. Cir. 2004) (“Dismissals divest this court of jurisdiction only if ‘[t]he parties were left in the same legal position with respect to [all] patent claims as if they had never been filed.’” (quoting Nilssen., 203 F.3d at 285)).
Further, although the district court explicitly declined to render a judgment on the merits of Rothe’s Tucker Act claim, Rothe VI, 499 F. Supp. 2d at 817, the district court ordered DOD to deposit $10,000 with the court in satisfaction of that claim, and thus the court changed the positions of the parties with respect to that claim. Cf. Chamberlain Group, 381 F.3d at 1190 (“[W]henever the complaint included a patent claim and the trial court’s rulings altered the legal status of the parties with respect to that patent claim, we retain appellate jurisdiction over all pendent claims in the complaint.”). Therefore, jurisdiction is proper in this Court. B. Rothe’s Facial Challenge
We now turn to the merits of Rothe’s appeal. Rothe argues that Section 1207, on its face as reenacted in 2006, violates the right to equal protection, that the district court was wrong to conclude otherwise, and that the district court was wrong to grant summary judgment to DOD instead of to Rothe. We agree with Rothe, and will reverse the judgment of the district court in this respect. [4]
1. Standard of Review and Strict Scrutiny Framework Though we possess jurisdiction over this appeal, Rothe’s facial challenge does
not implicate matters within our exclusive jurisdiction, and we will apply the law of the Fifth Circuit. [5] Hutchins v. Zoll Med. Corp., 492 F.3d 1377, 1383 (Fed. Cir. 2007). We will review the district court’s grant of summary judgment de novo, “applying the same criteria used by the district court in the first instance.” W.H. Scott Constr. Co. v. City of Jackson, 199 F.3d 206, 211 (5th Cir. 1999). Summary judgment is appropriate “if the pleadings, the discovery and disclosure materials on file, and any affidavits show that there is no genuine issue as to any material fact and that the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(c).
Because Section 1207 incorporates an explicit racial classification—the presumption that members of certain minority groups are “socially disadvantaged” for purposes of obtaining SDB status and the benefits that flow from that status under Section 1207 itself—the statute is subject to strict scrutiny. See Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 127 S. Ct. 2738, 2751 (2007) (“It is well established that when the government distributes burdens or benefits on the basis of individual racial classifications, that action is reviewed under strict scrutiny.”). Thus, to survive Rothe’s facial challenge, Section 1207 “must serve a compelling governmental interest, and must be narrowly tailored to further that interest.” Adarand III, 515 U.S. at 235; Rothe III, 262 F.3d at 1322 (“Strict scrutiny is a single standard and must be followed here.”).
The Supreme Court has held that government may have a compelling interest in “remedying the effects of past or present racial discrimination.” Shaw v. Hunt, 517 U.S. 899, 909 (1996); see also Croson, 488 U.S. at 492 (“It is beyond dispute that any public entity, state or federal, 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.”). However, “an effort to alleviate the effects of societal discrimination is not a compelling interest.” Shaw, 517 U.S. at 909-10. Therefore, before resorting to race- conscious measures, the government must “identify [the] discrimination [to be remedied], public or private, with some specificity,” and must have a “strong basis in evidence” upon which “to conclude that remedial action [is] necessary.” Croson, 488 U.S. at 500, 504; see also Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 274-77 (1986).
Although the party challenging a statute bears the ultimate burden of persuading the court that it is unconstitutional, the government first bears a burden to produce strong evidence supporting the legislature’s decision to employ race-conscious action. See Rothe III, 262 F.3d at 1317. “[T]he court must review the government’s evidentiary support to determine whether the legislative body had a ‘strong basis in evidence’ to believe that remedial action based on race was necessary.” Id.; see also Wygant, 476 U.S. at 278.
Even where there is a compelling interest supported by a strong basis in evidence, the statute must be narrowly tailored to further that interest. A narrow tailoring analysis commonly involves six factors: “(1) the necessity of relief; (2) the efficacy of alternative, race-neutral remedies; (3) the flexibility of relief, including the availability of waiver provisions; (4) the relationship of the stated numerical goals to the relevant labor market; (5) the impact of relief on the rights of third parties; and (6) the overinclusiveness or underinclusiveness of the racial classification.” Rothe III, 262 F.3d at 1331; see also Adarand III, 515 U.S. at 238-39; Croson, 488 U.S. at 506; United States v. Paradise, 480 U.S. 149, 171 (1987).
2. Compelling Interest—Strong Basis in Evidence Here, the district court concluded that “the Government has satisfied its burden of
producing a strong basis in the evidence for remedial action.” Rothe VI, 499 F. Supp. 2d at 877. In particular, the district court held that the non-stale statistical and anecdotal evidence before Congress “constitute[d] prima facie proof of a nationwide pattern or practice of discrimination in both public and private contracting,” and that the “[s]tatistical evidence supports the conclusion that African Americans, Hispanic Americans, Asian Americans, and Native Americans are currently and have been subject to discrimination in state and local contracting throughout the United States,” including “in Rothe’s relevant industry-professional services.” Id. at 877-78.
This statistical and anecdotal evidence, discussed by the district court in some detail, included the following: six disparity studies of state or local contracting, conducted by private research and consulting firms between 2002 and 2005 at the behest of state or local government in the cities of Dallas, Cincinnati, and New York, in Cuyahoga County, Ohio and Alameda County, California, and in the Commonwealth of Virginia; see id. at 835-64; a September 2005 study by the United States Commission on Civil Rights (“USCCR”) titled “Federal Procurement After Adarand,” see id. at 864-65; letters from individual business owners describing incidents of perceived discrimination in state, local, and private contracting, see id. at 865-68; various anecdotes regarding discrimination recounted by members of Congress in floor statements or remarks, see id. at 868-69; testimony by small business owners before the House Small Business Committee in 2001 and 2004, see id. at 869-71; and three reports from the Small Business Administration—two from 2005 and one from 2000— regarding the ownership and success rates of small businesses, see id. at 871-72.
The district court also discussed three sources of statistical analysis that were available to Congress in 2006, but were based on data gathered many years earlier, between the late 1980’s and mid 1990’s, namely: the appendix to a 1996 Department of Justice Study, see Proposed Reforms to Affirmative Action in Federal Procurement, Appendix—The Compelling Interest for Affirmative Action in Federal Procurement: A Preliminary Survey, 61 Fed. Reg. 26,042, 26,050 (May 23, 1996) (the “Appendix“); a 1997 report by the Urban Institute titled Do Minority-Owned Businesses Get a Fair Share of Government Contracts? (the “Urban Institute Report”); [6] and the 1998 Department of Commerce Benchmark Study, see supra note 2. However, the district court found that “the data contained in the Appendix, the Urban Institute Report, and the Benchmark Study is stale for purposes of strict scrutiny review of the 2006 Reauthorization,” and therefore the court concluded that it “[would] not rely on those three reports as evidence of a compelling interest for the 2006 Reauthorization of the 1207 Program.” Rothe VI, 499 F. Supp. 2d at 875. DOD does not challenge this finding on appeal, DOD Br. at 49 n.13, so we will not consider the Appendix, the Urban Institute Report, or the Department of Commerce Benchmark Study, and will instead determine whether the evidence relied on by the district court is indeed sufficient to demonstrate a compelling interest.
a. Six State and Local Disparity Studies The primary focus of the district court’s compelling interest analysis, and of the parties’ arguments on appeal, is the evidentiary strength of six particular disparity studies conducted at the state or local level. A disparity study, in this context, is a study attempting to measure the difference—or disparity—between the number of contracts or contract dollars actually awarded to minority-owned businesses in a particular contract market, on the one hand, and the number of contracts or contract dollars that one would expect to be awarded to minority-owned business given their presence in that particular contract market, on the other hand. Disparity studies can be relevant to the compelling interest analysis because, as Justice O’Connor has explained, “[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 [a] locality or the locality’s prime contractors, an inference of discriminatory exclusion could arise.” Croson, 488 U.S. at 509; see also W.H. Scott Constr., 199 F.3d at 218 (“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.”).
Here, the district court reviewed the parameters and findings of one state-wide and five local disparity studies, and concluded that these studies “analyze evidence of discrimination from a diverse cross-section of jurisdictions across the United States, and they constitute prima facie evidence of a nation-wide pattern or practice of discrimination in public and private contracting.” Rothe VI, 499 F. Supp. 2d at 838-39. On appeal, Rothe argues that the district court was wrong to rely on these studies because (1) the data analyzed by the studies was stale by the time of the 2006 reenactment, (2) the studies were not truly “before Congress,” (3) the studies are methodologically flawed and therefore unreliable, and (4) the studies do not establish that DOD itself has played any role in the discriminatory exclusion of minority-owned contractors. We will address each argument in turn.
i. Staleness Rothe argues that “[t]he extremely great weight of authority” holds that data more than five years old is necessarily stale for purposes of measuring contracting disparities, and that therefore “most of the data in most of the six studies and all of the data in some of the studies was stale” by the time of the 2006 reenactment. Rothe Br. at 29-30. [7] Rothe points to the recommendation of the USCCR that “Federal officials must discard disparity studies conducted using data that is more than five years old,” [8] and to the statements of experts relied on by the USCCR in reaching this recommendation (including Professor George R. La Noue, Rothe’s expert in previous stages of this case). Beyond the USCCR report, however, Rothe points to no judicial authority finding that data more than five years old is stale per se, and we decline to adopt such a per se rule here.
Indeed, as the district court noted, other circuit courts have relied on studies containing data more than five years old when conducting compelling interest analyses. See Rothe VI, 499 F. Supp. 2d at 839 n.86; W. States Paving Co. v. Wash. State Dep’t of Transp., 407 F.3d 983, 992 (9th Cir. 2005) (relying on the Appendix, published in 1996); Sherbrooke Turf, Inc. v. Minn. Dep’t of Transp., 345 F.3d 964, 970 (8th Cir. 2003) (same). Here, the district court considered the USCCR’s recommendation but declined to follow it, explaining that the court “must realistically focus on the availability of current data,” and finding that the data used in the six studies is not stale because it “was the most current data available at the time that these studies were performed [between 2002 and 2005].” Rothe VI, 499 F. Supp. 2d at 840. While we certainly agree with the USSCR that researchers should use current data when possible, we agree with the district court that Congress “should be able to rely on the most recently available data so long as that data is reasonably up-to-date.” Id.; see also Rothe III, 262 F.3d at 1331 (whether evidence is stale “is a factual question for the district court to resolve”). Because these disparity studies analyzed data pertaining to contracts awarded as recently as 2000 or even 2003, and because Rothe does not point to more recent, available data, [9] we affirm the district court’s conclusion that the data analyzed in these six disparity studies was not stale at the relevant time.
ii. “Before Congress” Rothe next argues that the district court was wrong to rely on the six disparity studies because “[t]here is no proof [the studies] were ever ‘before’ Congress,” or that they were ever “subject to any kind of analysis, hearing, or findings.” Rothe Br. at 27. In Rothe V, we explained that for evidence to be relevant in the strict scrutiny analysis, it “must be proven to have been before Congress prior to enactment of the racial classification.” 413 F.3d at 1338. It would be error for the district court to rely on studies without “a finding that [they] were put before Congress prior to the date of the present reauthorization in relation to section 1207 and to ground its enactment.” Id. Taking note of our holding, the district court found in Rothe VI that “[a]lthough the full text of these six disparity studies was not printed in the Congressional Record, . . . the repeated reference to these studies in Congressional hearings and floor debates indicates that these studies were before Congress.” 499 F. Supp. 2d at 839 n.83. The district court cited floor speeches by Senator Ted Kennedy and Representative Cynthia McKinney, in which these members of Congress referred to the six disparity studies by title, author, and date. See id. at 835-38. [10] Beyond these floor speeches, however, the district court did not identify any further references to the studies in Congressional proceedings—in particular, the court did not identify a single hearing at which the studies were named or discussed. On appeal, DOD does not identify any hearings to which the district court might have been referring, and Rothe contends—apparently uncontroverted by DOD—that these six studies were not in fact discussed at any Congressional hearings. Audio Recording of Oral Arg., available at http://oralarguments.cafc.uscourts.gov/mp3/2008-1017.mp3, at 10:30-11:00.
Although we are mindful that Congress has broad discretion to regulate its internal proceedings, see Am. Fed’n of Gov’t Employees v. United States, 330 F.3d 513, 522 (D.C. Cir. 2003), we are hesitant to conclude that the mere mention of a statistical study in a speech on the floor of the House of Representatives or the Senate is sufficient to put the study “before Congress” for purposes of Congress’ obligation to amass a “strong basis in evidence” for race-conscious action. We recognize that there is no dispute that these six studies were completed prior to the 2006 reenactment of Section 1207, and in that sense they were indeed “before” the acting legislature. But beyond their mere mention, there is no indication that these studies were debated or reviewed by members of Congress or by any witnesses. Cf. Sherbrooke Turf, 345 F.3d at 969-70 (relying on the Appendix, itself “a Department of Justice summary of more than fifty documents and thirty congressional hearings on minority-owned businesses prepared in response to the Adarand decision” (emphasis added)). And because Congress made no findings concerning these studies, we cannot even broach the question of whether to defer to Congress in any respect regarding them. [11] Cf. Croson, 488 U.S. at 500 (“The factfinding process of legislative bodies is generally entitled to a presumption of regularity and deferential review by the judiciary. . . . But when a legislative body chooses to employ a suspect classification, it cannot rest upon a generalized assertion as to the classification’s relevance to its goals.”).
Ultimately, however, we need not decide whether these six studies were put before Congress, because we will hold in any event that the studies do not provide a substantially probative and broad-based statistical foundation necessary for the “strong basis in evidence” that must be the predicate for nationwide, race-conscious action.
iii. Methodology Rothe contends that the six disparity studies contain methodological defects, relating primarily to the studies’ availability analyses, which render their conclusions about the existence of certain disparities unreliable. The district court acknowledged Rothe’s contentions, but rejected them as unsupported by “expert report[s] or other competent summary judgment evidence.” Rothe VI, 499 F. Supp. 2d. at 847; see also id. at 847 n.96 (“None of [Rothe’s] expert reports address the six disparity studies cited by McKinney and Kennedy in support of the 2006 Reauthorization.”); id. at 848 (“Rothe failed to rebut the statistical evidence contained in any of the six disparity studies with ‘credible, particularized’ evidence from its own expert reports.”); id. at 851 (“Rothe’s generalized objections regarding the alleged deficiencies of the Ohio Disparity Study are conclusory and are not competent summary judgment evidence.”); id. at 859 (“[T]he argument of counsel regarding the alleged deficiencies of the [Cincinnati] study are not competent summary judgment evidence.”).
This was error. In Rothe III, we instructed the district court to “undertake the same type of detailed, skeptical, non-deferential analysis undertaken by the Croson Court,” because “Congress is entitled to no deference in determining whether Congress had a compelling interest in enacting the racial classification” (as opposed to deference in the conduct of its factfinding proceedings). 262 F.3d at 1321. Many of Rothe’s objections to the six disparity studies at issue here are of the same general character as the objections articulated by Justice O’Connor to the statistical evidence offered by the government in Croson—i.e., objections to the parameters used to select the relevant pool of contractors. The potential pitfalls of race-conscious legislation are far too great for a court to dismiss such objections as incompetently offered, rather than to address them on their merits. See Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 291 (1978) (opinion of Powell, J.) (“Racial and ethnic distinctions of any sort are inherently suspect and thus call for the most exacting judicial examination.”), quoted in Parents Involved, 127 S. Ct. at 2764-65 (2007); Fullilove v. Klutznick, 448 U.S. 448, 491 (1990) (“Any preference based on racial or ethnic criteria must necessarily receive a most searching examination.”), quoted in Adarand III, 515 U.S. at 223, and Wygant, 476 U.S. at 273 (plurality opinion of Powell, J.).
Rather than remand this case a third time, however, we will consider here whether these studies are sufficiently probative for purposes of Congress’ burden to amass a “strong basis in evidence.” We note that although there are six studies, four of them were conducted by the same research consultant—Mason Tillman Associates— and employ very similar methodologies. The two remaining studies—the Cincinnati study, by Griffin & Strong, and the Virginia study, by MGT of America—are similar to the others in some basic respects.
As the district court explained, all of the studies sought to calculate a ratio “between the expected contract amount of a given race/gender group and the actual contract amount received by that group.“ Rothe VI, 499 F. Supp. 2d at 842 (citing the New York City study). In general, “[a] disparity ratio less than 0.80”—i.e., a finding that a given minority group received less than eighty percent of the expected amount— “indicates a relevant degree of disparity,” and might support an inference of discrimination. Id.; see, e.g., Eng’g Contractors Ass’n of S. Fla., Inc. v. Metro. Dade County, 122 F.3d 895, 914 (11th Cir. 1997) (“In general . . . disparity indices of 80% [i.e., 0.80] or greater, which are close to full participation, are not considered indications of discrimination.”). The district court reviewed the various disparity ratios found by each of the six studies for each of several minority groups in each of several industry categories, see Rothe VI, 499 F. Supp. 2d at 835-64, and we will not repeat them all. For example, however, the New York City study determined that Black Americans represented 16.72% of available construction firms but received only 1.7% of prime construction contract dollars, a statistically significant underutilization which can be expressed as a disparity ratio of 0.10. Other disparity ratios found by the studies included, for further example, 0.30 for Asian American firms in Alameda County construction subcontracts, 0.03 for Hispanic firms in Virginia professional services contracts, and 0.14 for Native American firms in Cincinnati supplies and services contracts during fiscal year 2000.
Rothe’s primary objection to the six disparity studies regards their availability analysis, or benchmark analysis—i.e., the steps taken to ensure that only those minority-owned contractors who are qualified, willing, and able to perform the prime contracts at issue are considered when forming the denominator of a disparity ratio. See Croson, 488 U.S. at 501-02 (“[W]here special qualifications are necessary, the relevant statistical pool for purposes of demonstrating discriminatory exclusion must be the number of minorities qualified to undertake the particular task.”). As Professor Ian Ayres, DOD’s expert in previous stages of this case, explained in a written statement before the USCCR, “the crucial question in disparity studies is to develop a credible methodology to estimate this benchmark share of contracts minorities would receive in the absence of discrimination,” and “[t]he touchstone for measuring the benchmark is to determine whether the firm is ‘ready, willing, and able’ to do business with the government.” J.A. at A777 (USCCR Disparity Studies, supra note 7, at 66).
Rothe contends that these six studies misapplied this “touchstone” of Croson, and erroneously included any minority-owned firm that was deemed willing or potentially willing and able, without regard to whether that firm was qualified. In particular, Rothe objects to the studies’ use of lists compiled by local business associations, and of community outreach, to identify minority-owned businesses. After reviewing the availability analyses contained in the six studies, we conclude that this defect does not substantially undercut the results of the four studies conducted by Mason Tillman Associates, because the bulk of the businesses considered in these studies were identified in ways that would tend to establish their qualifications, such as by their presence on city contract records and bidder lists. See, e.g., J.A. at A2999 (New York City study: of available prime contractors owned by minorities or women (“M/WBEs”), 23.48 percent were identified via “Prime Contractor Utilization,” 76.22 percent identified via “Certification Lists,” and 0.30 percent identified via “Willingness Survey of Chamber Membership and Trade Association Membership Lists”); J.A. at A1578 (Dallas study: “77.04 percent of the prime contractors available in the four industries combined were obtained from public agency and certification lists,” including “[m]ore than 75 percent of the M/WBEs,” while “[c]ompanies identified through outreach only were 26.35 percent of the M/WBEs”); J.A. at 1943 (Cuyahoga County study: “90.71 percent of the prime contractors available in the four industries combined were obtained from either utilized prime contractors, bidder lists, or certification lists,” including “73.42 percent of the [M/WBEs],” while “4.52 percent [of the M/WBEs] were identified solely through community meetings”); J.A. at A2674 (Alameda County study: of available prime contractor M/WBEs, 55.17 percent identified via “Alameda County and Other Government Records,” 37.43 percent identified via “Agency Certification Lists,” 0.57 percent identified via “Business Outreach Events,” 1.54 percent via “Trade Association Membership Lists,” and 5.29 percent identified via “Chamber Membership Lists”).
We are less confident in this aspect of the Virginia and Cincinnati studies, because the availability methodology employed in those studies appears less likely to have weeded out unqualified businesses. See J.A. at A1780 (Cincinnati study: requiring only that a firm “does business within an industry group from which the City of Cincinnati makes certain purchases”; that “[t]he firm’s owner has demonstrated that he or she believes the firm is qualified and able to perform the work” (emphasis added); and that “[b]y the owner’s actions, he or she has demonstrated an interest in obtaining work from the entity”); J.A. at A2128 (Virginia study: “For our analysis we used vendor data as the basis of the availability component . . . . Using this approach, we assume that all firms in the relevant market area are ready, willing, and able to do work for the Commonwealth at the prime or sub level.” (emphasis added)).
We are even more troubled, however, by the failure of five of the studies to account sufficiently for potential differences in size, or relative capacity, of the businesses included in those studies. As Professor Ayres explained to the USCCR, “‘qualified’ firms may have substantially different capacities,” and thus might be expected to bring in substantially different amounts of business even in the absence of discrimination:
Firms A and B may both be qualified to do some business with the government, but one firm may be a multinational with many plants, while the other firm may be a sole proprietorship with only a single plant. The ‘qualified-firm counting’ approach ignores differences in capacity and deems the single-plant firm to be equally ‘available’ to serve the government as the multiplant firm. It might assume, for example, that the manufacturers of a small micro-brewery brand and Budweiser are equally available to sell beer.
J.A. at A778 (USCCR Disparity Studies at 67) (emphasis in original). The Eleventh Circuit has explained similarly that “[b]ecause they are bigger, bigger firms have a bigger chance to win bigger contracts. It follows that, all other factors being equal and in a perfectly nondiscriminatory market, one would expect the bigger (on average) non-MWBE firms to get a disproportionately higher percentage of total construction dollars awarded than the smaller MWBE firms.” Eng’g Contractors Ass’n, 122 F.3d at 917. And we ourselves criticized a statistic, offered by DOD in an earlier stage of this case, in part because it “[did] not take into account the fact that the sheer number of businesses owned by minorities may not be significantly correlated with the volume of business conducted by minority-owned businesses.” Rothe III, 262 F.3d at 1324; see also W. States Paving Co., 407 F.3d at 1000 (“[T]he fact that [disadvantaged business enterprises] constitute 11.17% of the Washington market does not establish that they are able to perform 11.17% of the work.”).
Here, each of the six disparity studies accounted for the relative sizes of contracts awarded to minority-owned businesses, by measuring the utilization of minority-owned contractors—i.e., the numerator in a disparity ratio—in terms of contract-dollars directed to minority-owned businesses rather than in the raw number of contracts awarded. But none of the studies took complementary account of the relative sizes of the businesses themselves. Rather, each of the studies measured the availability of minority-owned businesses—i.e., the denominator in a disparity ratio—by the percentage of firms in the market owned by minorities, instead of by the percentage of total marketplace capacity those firms could provide. [12] See J.A. at A1603 (Dallas study: introducing concept of disparity ratio as comparison between “the proportion of contract dollars awarded to [MBEs and WBEs]” and “the proportion of available MBEs and WBEs in the relevant market area” (emphases added)); J.A. at A1962 (Cuyahoga County study: same); J.A. at A2702 (Alameda County study: same); J.A. at A3031 (New York City study: “Under a fair and equitable system of awarding contracts, the proportion of contract dollars awarded to M/WBEs would be approximate to the proportion of available M/WBEs in the relevant market area.” (emphases added)); J.A. at A1768 (Cincinnati study: defining “disparity index” as “the percentage of M/WFBE participation in government contracts,” i.e., “the percentage of contracting dollars paid to M/WFBE” firms, divided by “the percentage of M/WFBEs in the relevant population of local firms” (emphases added)); J.A. at A2211 (Virginia study: describing “underlying assumption” of study’s disparity analysis that “absent discrimination, the proportion of dollars received by a particular MBE group should approximate that group’s proportion of the relevant population of vendors” (emphases added)).
We do not mean to suggest that the studies completely ignored the question of firm size. In the New York City study, for example, the disparity analysis was “restricted to an examination of the prime contract awards of $1,000,000 and under to limit the capacity required to perform the contracts subjected to the statistical analysis.” J.A. at A3005. Similarly, the Dallas study attempted to control for capacity by ensuring that firms had a “demonstrated ability to win large competitively bid contracts,” J.A. at A1582; the study concluded that “the majority of the City’s contracts are small,” and “[t]herefore, to perform on most City contracts, even the competitively bid construction projects, the available firms only required minimal capacity.” J.A. at A1602; see also J.A. at A1928 (Cuyahoga County study: similar approach).
But while these parameters may have ensured that each minority-owned business in the studies met a capacity threshold—i.e., had the capacity to bid for and to complete any one contract—these parameters simply fail to account for the relative capacities of businesses to bid for more than one contract at a time, or, borrowing from Professor Ayres’s example, the difference between the volume of a particular micro- brewed beer in the marketplace at a given time and the volume of Budweiser. This failure renders the disparity ratios calculated by the studies substantially less probative, on their own, of the likelihood of discrimination. [13]
Of course, the studies could have accounted for firm size even without changing their disparity-ratio methodologies; they could have employed regression analysis to determine whether there was a statistically significant correlation between the size of a firm and the share of contract-dollars awarded to it. See Eng’g Contractors Ass’n, 122 F.3d at 917 (explaining that “regression analysis is a statistical procedure for determining the relationship between a dependent and independent variable, e.g., the dollar value of a contract award and firm size,” and that “[t]he point of a regression analysis is to determine whether the relationship between the two variables is statistically meaningful”). But of the six studies, only the Virginia study conducted this type of regression analysis:
The regression analysis which included the independent variables of a firm—age of company, owner education level, number of employees, percent of revenue from private sector, and owner experience for industry groupings—had an R square of .18, indicating that the independent variables explained only 18 percent of the variations in firm revenue categories.
J.A. at A2234. And although the Virginia study did find “a consistent and negative relationship between MBE status and revenue” even after accounting for such “independent” variables, J.A. at A2235, the relatively lax “qualified, willing, and able” requirements of the Virginia study render this conclusion less probative.
To be clear, we do not hold that the defects in the availability and capacity analyses in these six disparity studies render the studies wholly unreliable for any purpose. Where the calculated disparity ratios are low enough, we do not foreclose the possibility that an inference of discrimination might still be permissible for some of the minority groups in some of the studied industries in some of the jurisdictions. And we recognize that a minority-owned firm’s capacity and qualifications may themselves be affected by discrimination. [14] But we hold that the defects we have noted detract dramatically from the probative value of these six studies, and, in conjunction with their limited geographic coverage, render the studies insufficient to form the statistical core of the “strong basis in evidence” required to uphold the statute.
iv. Geographic Coverage In Rothe III, we explained that, although Section 1207 is subject to scrutiny that is no less strict than the scrutiny applied to the race-based policies of municipalities, Congress nevertheless has, in a sense, a “‘broader brush’ than municipalities for remedying discrimination,” because Congress has the power to legislate for the entire nation. 262 F.3d at 1329. In particular, we wrote that “[w]hereas municipalities must necessarily identify discrimination in the immediate locality to justify a race-based program, we do not think that Congress needs to have had evidence before it of discrimination in all fifty states in order to justify the 1207 program.” Id.; see also Adarand Constructors, Inc. v. Slater, 228 F.3d 1147, 1165 (10th Cir. 2000) (“Adarand VII”) (“The fact that Congress’s enactments must serve a compelling interest does not necessitate the conclusion that the scope of that interest must be as geographically limited as that of a local government.”). However, we were clear that “evidence of a few isolated instances of discrimination would be insufficient to uphold the nationwide program,” and we left to the district court the question of “[w]here to draw the line . . . in the first instance.” Rothe III, 262 F.3d at 1330.
The district court has now drawn a line, holding that “[t]hese six state and local disparity studies analyze evidence of discrimination from a diverse cross-section of jurisdictions across the United States, and they constitute prima facie evidence of a nation-wide pattern or practice of discrimination in public and private contracting.” Rothe VI, 499 F. Supp. 2d at 838-39 (footnote omitted). It is now up to us to review this holding, and we cannot affirm it. We take judicial notice that the United States comprises over three thousand counties and county-equivalent regions, [15] and, as of July 1, 2007, there are at least two hundred cities or metropolitan areas with populations above 200,000 people. [16] It may be reasonable to assume that there are some demographic and industrial similarities between many of the larger cities and counties across the country. And we still think that Congress need not amass evidence of discrimination in all fifty states to meet its burden. But we would be hesitant to conclude even from methodologically unimpeachable disparity studies of one state, two counties, and three cities that there is a “nation-wide pattern or practice of discrimination in public and private contracting,” where the discrimination is on the same order as the local discrimination that might be inferred from the six studies. Here, given the weaknesses in the six studies’ benchmark analyses as described above, we simply cannot agree with the district court’s conclusion.
We stress that in holding these six studies insufficient in this case, we do not necessarily disapprove of decisions by other circuit courts that have relied, directly or indirectly, on municipal disparity studies to establish a federal compelling interest. Different studies, in the context of different legislative history, may support different conclusions. In particular, the Appendix, relied on by the Ninth and Tenth Circuits in the context of certain race-conscious measures pertaining to federal highway construction, references the Urban Institute Report, which itself analyzed over fifty disparity studies and relied for its conclusions on over thirty of those studies, a far broader basis than the six studies here provide. See Adarand VII, 228 F.3d at 1172-73; W. States Paving Co., 407 F.3d at 992-93; Urban Institute Report, supra note 5, at 9. [17]
b. Other Evidence Before Congress We now turn to the remaining evidence relied on by the district court to hold that Congress had a strong basis in evidence for reenacting Section 1207 in 2006. Rather than take each item in turn, however, we will organize our discussion by statistical and anecdotal evidence. Ultimately, we will conclude that the remaining statistical evidence is not substantially more probative than that contained in the six disparity studies. Without strong statistical evidence, the statute cannot be upheld. See Rothe III, 262 F.3d at 1323-24 (“Statistical evidence is particularly important to justify race-based legislation. . . . Indeed, nearly every court of appeals upholding the constitutionality of a race-based classification has relied in whole or in part on statistical evidence.”); Adarand VII, 228 F.3d at 1166 (explaining that “[b]oth statistical and anecdotal evidence are appropriate in the strict scrutiny calculus,” but “anecdotal evidence by itself is not”).
i. Other Statistical Evidence The district court considered statistics presented by USCCR Commissioner Michael Yaki in his dissent to a September, 2005 USCCR study titled “Federal Procurement After Adarand.” For example, the district court noted that although the revenue of minority-owned businesses grew by sixty percent between 1992 and 1997, more than the rate of growth for all United States firms, “the revenue of African American-owned firms grew only half as much as minority-owned businesses generally, and less than all U.S. firms.” Rothe VI, 499 F. Supp. 2d at 865. The court also noted that “[m]inority-owned firms with paid employees were much less likely to survive from 1997 to 2001 than from 1992 to 1996,” and that “African American-owned enterprises were less likely to survive than other groups in either period.” Id. Although this data is national in scope and is therefore in a sense more probative of nationwide discrimination than are municipal disparity studies, the data from Commissioner Yaki’s dissent dates to the early 1990’s. Thus, as the district court recognized, this data is not by itself highly probative of the state of affairs at the time of the 2006 reenactment of Section 1207. See id. at 865 n.107 (“The Court finds that this statistical evidence is not categorically stale, but it is less probative of present-day discrimination than the six state and local disparity studies previously discussed.”).
The district court also referred to studies beyond the six disparity studies discussed above, but only in quoting speeches made by members of Congress. For example, in a July 2005 speech made to support the Department of Transportation’s disadvantaged business program, Senator Kennedy stated that “studies completed since 1998 show that minority and women-owned companies are underutilized in government contracting,” and listed studies identified by the Department of Transportation that he alleged
show[] significant disparities between the availability and utilization of minority and women-owned firms in government contracting . . . in Nebraska; in Maryland; in Colorado; in Georgia; in Kentucky; in Ohio; in Wilmington, DE; in Dekalb County, GA; in Broward County, FL; in Dallas, TX; in Cincinnati, OH; in Tallahassee, FL; and in Baltimore, MD.
Rothe VI, 499 F. Supp. 2d at 869 n.114 (quoting 151 Cong. Rec. S9442-02, at *S9443 (Jul. 29, 2005)). These studies themselves, however, are not in the record here, and are not even sufficiently described in this floor excerpt for us to locate them, let alone subject them to “detailed, skeptical, non-deferential analysis.” See Rothe III, 262 F.3d at 1321. Likewise, the district court quoted floor speeches given by Representatives Watt and Menendez in December of 2005, in which they cited, respectively, to “a 2004 disparity study for North Carolina that was performed by MGT America,” and “a New Jersey disparity study by MGT America,” and in which these Representatives concluded that these disparity studies demonstrated underutilization of minority contractors. Rothe VI, 499 F. Supp. 2d at 869. These studies are not in the record, and we cannot defer to Representatives Watt and Menendez about the studies’ probative value, particularly in light of our concerns about the Virginia study by MGT of America discussed above. See Adarand VII, 228 F.3d at 1167 (“We cannot merely recite statements made by members of Congress alleging a finding of discriminatory effects and the need to address those effects . . . .”).
We conclude that the remaining statistics cited by members of Congress in the floor speeches quoted by the district court cannot serve as the foundation of a “strong basis in evidence,” because they are not sufficiently probative of nationwide discrimination against the range of minority groups afforded a presumption under Section 1207. Nor are the statistics quoted by the district court from the three SBA reports sufficient, because they do not account for firm size or qualifications. See, e.g., Rothe VI, 499 F. Supp. 2d at 871 (noting that the 2000 State of Small Business Report “found that minority-owned businesses ‘make [up] 9 percent of the business population of the United States, but small minority-owned businesses won just 6 percent of the award dollars in FY 1998 and 1999’”).
ii. Anecdotal Evidence Given our holding regarding statistical evidence, we will not review the anecdotal evidence before Congress in detail. Beyond the anecdotal evidence contained in the six municipal disparity studies in the record, the district court discussed floor speeches by members of Congress, in which they related anecdotes submitted by minority business owners concerning their experiences with private sector and state- and city- level “good old boy” networks and discriminatory government agencies. See Rothe VI, 499 F. Supp. at 866-67. The House Small Business Committee also heard in hearings about the difficulties faced by minority business owners in federal procurement. See id. at 869-71. Rothe contests the accuracy of some of these anecdotes, but we need not make credibility determinations, because anecdotal evidence is insufficient by itself to support Section 1207. Adarand VII, 228 F.3d at 1166.
We do pause, however, to consider that while the district court cited complaints about “slow payment by government agencies,” and “the effects of contract bundling in continuing the disparities of contracts awarded to minority business contractors,” Rothe VI, 499 F. Supp. 2d at 870, and cited the Small Business Committee’s conclusion that “much work remains to be done” in “increasing access to capital and federal procurement markets for minority entrepreneurs,” id., neither the district court in its opinion, nor DOD on appeal, cited to a single instance of alleged discrimination by DOD in the course of awarding a prime contract, nor to a single instance of alleged discrimination by a private contractor identified as the recipient of a prime defense contract.
This lacuna is quite significant. Justice O’Connor wrote in Croson that if a government has become “a ‘passive participant’ in a system of racial exclusion practiced by elements of the local construction industry,” then that government may take “affirmative steps to dismantle” the exclusionary system. 488 U.S. at 492. Here, the district court concluded that DOD “is a ‘passive participant’ in a system of racial exclusion practiced by elements of various state and local contracting sectors because the Government has compiled evidence of marketplace discrimination and linked its spending practices to that private or public discrimination.” Rothe VI, 499 F. Supp. 2d at 827 (citing Concrete Works, 321 F.3d at 976). But without a single identified incident of discrimination by DOD or a recipient of DOD funds, the only evidence recited by the district court to establish a link between DOD’s spending practices and discrimination is the fact that DOD “is one of the largest buyers of goods and services in the world,” having procured “approximately $268 billion of goods and services in fiscal year 2005, with much of that money being distributed to state and local contractors.” Rothe VI, 499 F. Supp. 2d at 827.
Undoubtedly, some state and local contractors have engaged in discrimination. And given the amount of money spent by DOD, it is likely that some money injected by DOD into the nationwide contract market has made its way into the hands of contractors, subcontractors, or sub-subcontractors who have engaged in discrimination. But we are skeptical that this bare likelihood would be sufficient to establish DOD’s “passive participation” in discrimination within the meaning of Croson, even if—unlike the present case—Congress were to possess a strong evidentiary basis upon which to conclude that there was a nationwide pattern or practice of pervasive discrimination by state, local, and private contractors in the relevant contract markets at the relevant time. Even in Concrete Works, for example, the City of Denver offered more than mere dollar amounts to link its spending to private discrimination. As the Tenth Circuit explained, Denver provided testimony from minority business owners that “general contractors who use them in City construction projects refuse to use them on private projects,” with the result that Denver had “pa[id] tax dollars to support firms that discriminate against other firms because of their race, ethnicity, and gender.” Concrete Works, 321 F.3d at 976-77. We do not know whether, in the present case, DOD could have obtained similar testimony from minority-owned subcontractors at the relevant time, only that DOD did not do so.
To conclude our compelling interest analysis, we hold that Congress did not have before it, at the time of the 2006 reenactment of Section 1207, a “strong basis in evidence” for the proposition that DOD was a passive participant in racial discrimination in relevant markets across the country and that therefore race-conscious remedial measures were necessary. We stress that our holding is grounded in the particular items of evidence offered by DOD and relied on by the district court in this case, and should not be construed as stating blanket rules, for example about the reliability of disparity studies. As the Fifth Circuit has explained, there is no “precise mathematical formula to assess the quantum of evidence that rises to the Croson ‘strong basis in evidence’ benchmark.” W.H. Scott Constr. Co., 199 F.3d at 218 n.11. Rather, “[t]he sufficiency of a [government’s] findings of discrimination in a local industry,” or for that matter in a state-wide or nationwide industry, “must be evaluated on a case-by-case basis.” Id. Thus, if Congress reenacts Section 1207 again before it is set to expire in 2009—as Congress is free to do—we cannot now predict, nor do we intend to prejudge, whether any such new enactment will be supported by a “strong basis in evidence.”
3. Narrow Tailoring Because we hold that Congress lacked the evidentiary predicate for a compelling
interest, we cannot determine whether Section 1207, as reenacted in 2006, is narrowly tailored to a compelling interest. We make two observations about narrow tailoring, however.
First, we held in Rothe III that the district court in Rothe I had “thoroughly analyzed and correctly concluded that the 1207 program was flexible in application, limited in duration, and that it did not unduly impact on the rights of third parties.” 262 F.3d at 1331. We have no occasion to revisit that holding today, but note that Section 1207 and pertinent regulations have been amended over time, and that the amendments have tended to limit, rather than broaden, the application of preferences based on racial classifications.
Second, even if we were to reach the other narrow tailoring factors, the absence of strongly probative statistical evidence makes it impossible to evaluate at least one of those factors. Without solid benchmarks for the minority groups covered by the statute, we simply cannot determine whether Section 1207’s five percent goal is reasonably related to the capacity of firms owned by members of those minority groups—i.e., whether that goal is comparable to the “share of contracts minorities would receive in the absence of discrimination.” J.A. at A777 (USCCR Disparity Studies, supra note 7, at 66). Such benchmarks might have been available here, had the Department of Commerce made good on its 1999 intention to “develop new benchmarks and utilization estimates every three years.” See supra note 2. But Commerce did not do so. See J.A. at A779 (USCCR Disparity Studies, supra note 7, at 68). C. Rothe’s Request for Attorney Fees
There is one more matter we must address. Rothe argues that DOD’s defense of Section 1207 in this litigation is not substantially justified, and seeks attorney fees and costs under the Equal Access to Justice Act (“EAJA”). Section 2412 (d)(1)(A) of Title 28 sets forth the basis for attorney’s fees under the EAJA:
Except as otherwise specifically provided by statute, a court shall award to a prevailing party other than the United States fees and other expenses . . . incurred by that party in any civil action (other than cases sounding in tort) . . . brought by or against the United States in any court having jurisdiction of that action, unless the court finds that the position of the United States was substantially justified or that special circumstances make an award unjust. We deny Rothe’s request as premature. EAJA provides that a party seeking an
award of fees and other expenses shall submit an application therefore “within thirty days of final judgment in the action.” Id. § 2412(d)(1)(B). There is not yet a final judgment in this case. We will remand the case for the district court to enter judgment; Rothe may submit an application for attorney fees at the appropriate time, though we take no position on the merits of such an application.
CONCLUSION
For the foregoing reasons, we hold that Section 1207, on its face, as reenacted in 2006, violates the equal protection component of the Fifth Amendment right to due process. Because the statute authorizes DOD to afford preferential treatment on the basis of race, we must apply strict scrutiny. And because Congress did not have a “strong basis in evidence” upon which to conclude that DOD was a passive participant in pervasive, nationwide racial discrimination—at least not on the evidence produced by DOD and relied on by the district court in this case—the statute fails strict scrutiny. We reverse the judgment of the district court in part, and remand with instructions to enter a judgment (1) denying Rothe any relief regarding the facial constitutionality of Section 1207 as enacted in 1999 or 2002, (2) declaring that Section 1207 as enacted in 2006 (i.e., the current 10 U.S.C. § 2323) is facially unconstitutional, and (3) enjoining application of the current 10 U.S.C. § 2323.
AFFIRMED-IN-PART, REVERSED-IN-PART and REMANDED
NOTES
[*] Honorable Richard G. Stearns, District Judge, United States District Court for the District of Massachusetts, sitting by designation.
[2] A present regulation directs that the sizes of price evaluation adjustments are to be determined by the Department of Commerce “on an annual basis, by North American Industry Classification System (NAICS) Industry Subsector, and region, if any.” 48 C.F.R. § 19.201(b) (2007). Although Commerce did issue determinations of price evaluation adjustments in 1998 and 1999, and proposed to “develop new benchmarks and utilization estimates every three years” thereafter, no new benchmarks or price evaluation adjustment determinations have been issued, presumably because the PEA has been serially suspended since 1999. See Small Disadvantaged Business Procurement; Reform of Affirmative Action in Federal Procurement, 63 Fed. Reg. 35,714 (June 30, 1998) (“Department of Commerce Benchmark Study”); Office of Federal Procurement Policy Small Disadvantaged Business Procurement: Reform of Affirmative Action in Federal Procurement, 64 Fed. Reg. 52,806 (Sept. 30, 1999).
[3] In a letter dated October 23, 2008, DOD counsel brought to the Court’s attention the recent publication by the Small Business Administration (“SBA”) of an interim final rule. See Small Disadvantaged Business Program, 73 Fed. Reg. 57,490 (Oct. 3, 2008) (to be codified at 37 C.F.R. pt. 123). The interim final rule relieves SBA of the task of certifying SDBs and places that responsibility on the procuring agencies, such as DOD. As DOD counsel notes, and we agree, the substantive standards governing SDB certification have not changed. Thus, although the new rules alter the process of qualifying as an SDB, the new rules have no effect on our constitutional analysis.
[4] Rothe claims, without support, that the district court should have also declared the statute unconstitutional as reenacted in 1999 and 2002. We agree with the district court that the facial constitutionality of these earlier enactments is a moot question; they have been superceded by the 2006 reenactment and are no longer the law. Thus, “a declaratory judgment on the validity of these intervening reauthorizations is ‘a textbook example of advising what the law would be upon a hypothetical state of facts.’” Rothe VI, 499 F. Supp. 2d at 783 (quoting Concrete Works of Colo., Inc. v. City & County of Denver, 321 F.3d 950, 954 n.1 (10th Cir. 2003)). Rothe’s facial challenge to the 2006 reenactment, however, is not moot. While the PEA has been suspended since 1999 and will continue to be suspended until March 9, 2009, the 2006 reenactment of Section 1207 will remain in effect until September 30, 2009 (i.e., the end of the fiscal year), and it is not absolutely certain that the PEA will be suspended for the period between March 9 and September 30, 2009. See Rothe V, 413 F.3d at 1334 (“[B]ecause the government has not proven that the suspension of the price-evaluation adjustment will remain in place, it has failed to prove mootness.”).
[5] We note that while we stand in the shoes of the Fifth Circuit, the bulk of relevant, controlling authority comes directly from the Supreme Court, and we have interpreted much of that authority in our prior opinions in this case.
[6] Available at http://www.urban.org/UploadedPDF/DMOBGFSGC.pdf (last visited Sept. 27, 2008).
[7] In the aggregate, the studies covered data pertaining to contracts awarded as early as 1995 and as late as 2003. See Rothe VI, 499 F. Supp. 2d at 839 (New York City data from 1997-2002; Alameda County data from 2000-2003; Cuyahoga County data from 1998-2000; Dallas data from 1997-2000; Cincinnati data from 1995-2001; Virginia data from 1997-2002).
[8] USCCR, Disparity Studies as Evidence of Discrimination in Federal Contracting 79 (May 2006) (“USCCR Disparity Studies”), available at http://www.usccr.gov/pubs/DisparityStudies5-2006.pdf (last visited Sept. 27, 2008).
[9] Rothe notes that DOD and the SBA collect data on the race of certain subcontractors, and implies that this data is more recent than the data analyzed in the disparity studies. Rothe Br. at 5. But Rothe concedes that DOD does not yet “centrally access” this data, and thus Rothe fails to establish that DOD could have compiled this data for Congress at the time of the 2006 reenactment. Id.
[10] For example, Senator Kennedy stated in a floor speech on November 10, 2005, that “[y]ears of Congressional hearings have shown that minorities historically have been excluded from both public and private construction contracts in general,” and later stated that “[such] problems are detailed in many recent disparity studies, including [the six studies at issue here].” 151 Cong. Rec. S12668-01, 2005 WL 3018127 (Nov. 10, 2005).
[11] Findings regarding the disparity studies, however, are to be distinguished from formal findings of discrimination by DOD, which Congress was emphatically not required to make. See Dean v. City of Shreveport, 438 F.3d 448, 455 (5th Cir. 2006) (“[T]he government need not incriminate itself with a formal finding of discrimination prior to using a race-conscious remedy.”).
[12] Constance F. Citro, one of the experts appearing before the USCCR panel on disparity studies in December of 2005, wrote that comparing utilization measured in terms of contract-dollars against availability measured in terms of number of firms is akin to comparing apples to oranges. See J.A. at A767 (USCCR Disparity Studies, supra note 7, at 56). While we do not adopt that statement as a general prohibition, we certainly agree with Dr. Citro that for a disparity ratio to have significant probative value, “the same time period and metric (dollars or numbers) should be used in measuring the utilization and availability shares.” J.A. at A772 (USCCR Disparity Studies, supra note 7, at 61).
[13] According to Professor Ayres, the Department of Commerce Benchmark Study employed a true “capacity” methodology, which “calculates in dollar terms the capacity of qualified firms to do business with the government” and uses as a benchmark for each industry the minority-owned businesses’ “share of the industry’s total capacity.” J.A. at A778 (USCCR Disparity Studies, supra note 7, at 67).
[14] But see Concrete Works of Colo., Inc. v. City & County of Denver, 540 U.S. 1027, 1032 (2003) (Scalia, J., dissenting from denial of cert.) (“The Tenth Circuit accepted the city’s contention that . . . MBEs . . . ‘are generally smaller and less experienced because of industry discrimination.’ The argument fails because it rests on nothing but speculation.” (internal citation omitted)).
[15] See, e.g., United States Geological Survey, Frequently Asked Questions, “How many counties are there in the United States?” available at http://www.usgs.gov/faq/list_faq_by_category/get_answer.asp?id=785 (last visited September 28, 2008).
[16] See United States Census Bureau, Annual Estimates of the Population for Incorporated Places Over 100,000, Ranked by July 1, 2007 Population: April 1, 2000 to July 1, 2007, available at http://www.census.gov/popest/cities/tables/SUB-EST2007- 01.csv (last visited September 28, 2008).
[17] The disparity studies analyzed in the Urban Institute Report may themselves be subject to methodological criticism, but we have no need to further consider those studies or the Urban Institute Report itself in this case, and decline to do so.
