IN THE UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF WISCONSIN
STATE OF WISCONSIN DEPARTMENT OF WORKFORCE DEVELOPMENT, DIVISION OF VOCATIONAL REHABILITATION,
Petitioner, v. OPINION and ORDER
UNITED STATES DEPARTMENT OF EDUCATION, 18-cv-220-jdp BETSY DeVOS, in her official capacity as Secretary of the United States Department of Education, and THERESA TAYLOR,
Respondents.
This case involves a dispute over the award of a vending machine services contract for two Wisconsin correctional facilities in Racine County, Wisconsin. The contract was awarded under the Randolph-Sheppard Act, a federal law that gives priority to blind persons in awarding contracts to operate vending machines on government property. Theresa Taylor, a blind vending machine operator, sought the Racine contract from the Wisconsin Department of Workforce Development (DWD), through its Division of Vocational Rehabilitation, which administers the Randolph-Sheppard program in Wisconsin. But DWD awarded the contract to another blind vendor, Joelyn Belsha. As provided under the Randolph-Sheppard Act, Taylor asked the United States Department of Education to convene a panel to arbitrate the dispute between Taylor and DWD over the vending contract. Taylor prevailed in the arbitration, and the arbitration panel ordered that Taylor be given the contract and awarded money damages. DWD has petitioned this court to review the decision of the arbitration panel. Now before the court are two mirror-image motions: DWD’s motion to vacate the arbitration panel’s decision, Dkt. 26, and Taylor’s motion to confirm and enforce it. Dkt. 24. The Department of Education filed the administrative record, Dkt. 21, but it has not otherwise participated in the case. Taylor asks for oral argument on her motion. The procedural history
of this case is protracted, but the court reviews the arbitration decision under the Administrative Procedure Act, which limits review to the administrative record. The parties have adequately briefed the issues and oral argument is unnecessary. The arbitration panel found that DWD was persistently biased against Taylor, and that she was decisively disadvantaged by DWD’s decision to use 2013 profitability data to evaluate the competing applicants. But the key factual findings are not supported by substantial evidence, and the arbitration panel’s ultimate conclusions are arbitrary and capricious. The court concludes that there were no material deficiencies in the 2013 selection process and that
Taylor unreasonably chose not to participate in that process. The court will deny Taylor’s motion to enforce the arbitration award, grant DWD’s motion to vacate and reverse the arbitration decision, and remand the case to the Department of Education with instructions to affirm the award of the contract to Belsha.
BACKGROUND A. The Randolph-Sheppard Act The Randolph-Sheppard Act, enacted in 1936 and codified at 20 U.S.C. § 107, establishes a voluntary federal-state program to provide employment opportunities to the blind
by giving them priority for contracts to operate vending machines on government property. Responsibility for administering this program is divided between state and federal agencies. At the federal level, the Secretary of Education interprets and enforces the Act’s provisions and designates state licensing agencies to implement the program in the states. 20 U.S.C. § 107a. The designated state licensing agency for Wisconsin is DWD, which carries out its Randolph- Sheppard Act responsibilities through its “business enterprise program.” DWD’s business
enterprise program is governed by Wis. Stat. § 47.03(4)–(8) and by implementing regulations in the Wisconsin Administrative Code, Chapter DWD 60. As a condition of participating in the Randolph-Sheppard program, DWD must afford certain procedural protections to blind vendors. A blind vendor licensed under the program who is dissatisfied with DWD action relating to the program can file a grievance, and if dissatisfied with the resolution of the grievance, the vendor can ask for further review, including ultimately a full evidentiary hearing. Wis. Admin. Code DWD 60.05; 20 U.S.C. § 107d-1(a). If the vendor is dissatisfied with the results of the full evidentiary hearing, the vendor can
submit the matter to arbitration before a panel to be convened for that dispute by the Department of Education. Wis. Admin. Code DWD 60.05(4); 20 U.S.C. § 107d-1(a). Decisions by Department of Education arbitration panels are final agency actions reviewable in federal court under the Administrative Procedure Act. 20 U.S.C. § 107d-2(a). B. Procedural history 1. The Racine Correctional Institution and Sturtevant Transitional Facility contract Taylor has worked as a licensed blind vendor since 2006, when she was first appointed to run vending machines at the Southern Wisconsin Center—which serves veterans, the cognitively disabled, and female prisoners in a minimum-security facility. AR-143.1 In October
1 Record citations are to the administrative record, filed in multiple parts at Dkt. 21. A useful of 2007, Taylor was asked to temporarily take over vending operations at three Wisconsin correctional facilities: Racine Correctional Institution, Sturtevant Transitional Facility, and the Racine Youthful Offender Correctional Facility. She served as the interim operator of those facilities for four years.
In the meantime, DWD was preparing to select permanent operators for those three sites. DWD decided that Racine Correctional Institution and Sturtevant Transitional Facility together would be bid out for selection of a permanent operator as a “stand-alone” facility, because they were considered large enough to provide full-time employment for a single operator. AR-236. That meant that the operator who won that bid would have to give up her existing sites. By contrast, DWD decided to bid out the Racine Youth Offender Correctional Facility vending machines as an “add-on” site, meaning that it would be added to an operator’s existing sites. AR-237.
2. 2011 interviews In July 2011, DWD sent out the bid announcement for Racine Youth Offender Correctional Facility. AR-23. Taylor and Joelyn Belsha, another licensed blind operator, interviewed for the position. Taylor scored 110 points based on scores from three interviewers on the panel; Belsha scored 104 points. AR-412. So DWD offered Taylor the site, and she accepted. AR-345. DWD sent out the bid announcement for Racine Correctional Institution and Sturtevant Transitional Facility (the “RCI/STF” site) in August 2011. Once again, Taylor and
Belsha interviewed for the position, along with two other blind vendors. Taylor scored a total
index of the record can be found at Dkt. 21-4. of 96 points from the interview panel, whereas Belsha scored 101 points. AR-410. Belsha was awarded the RCI/STF site. 3. Taylor’s grievance and the full evidentiary hearing Taylor sent a grievance letter to DWD contesting the selection of Belsha as the
permanent operator of the RCI/STF site. AR-414–15. Specifically, Taylor contended that DWD had failed to consider a letter of recommendation that she received from the RCI/STF facilities manager, which she had presented at the interview. AR-414. By ignoring her recommendation letter and selecting Belsha, Taylor said, DWD had violated Wisconsin Administrative Code § DWD 60.08, which requires that “the licensee deemed to be best suited for an available business enterprise” be selected. DWD held a hearing on Taylor’s grievance on November 8, 2011, after which business enterprise program director Kim Pomeroy denied Taylor’s grievance, explaining that
“[r]esumes and references are not requested of the interviewees because [DWD] is responsible for personnel records of the [business enterprise program] operators and is familiar with the training and performance of each operator.” AR-418. At that point, Taylor asked for a full evidentiary hearing in accordance with § 107d-1(a) of the Randolph-Sheppard Act and the corresponding state statute, Wisconsin Statute § 47.03. The full evidentiary hearing took place on May 8, 2012. (The hearing was audio recorded, but the recording was too poor to transcribe, so there is no official transcript of the proceedings.) According to records from that hearing, Taylor testified that Greg Feypel (a DWD
employee who administered the business enterprise program) had contacted her on August 26, 2011 to discuss the RCI/STF bid. According to Taylor, Feypel “expressed the thought of giving Belsha RCI/STF due to her UW-Parkside location not being profitable and to ‘even out’ things.” AR-423. Feypel denied that he had said this, and he testified that that “his conversation with Taylor was just trying to explain the ‘stand-alone’ versus ‘add-on’ site” arrangement. AR-425. The hearing panel determined that the interviews for the RCI/STF contract did not
comply with DWD’s own policies, and that the selection of Belsha over Taylor had not been “adequately justified” under § DWD 60.08’s requirement that the “best suited” licensee be selected. Id. The hearing panel identified the following problems with the process: (1) there had been no site facilities manager or designee on the interview panel, as required by DWD policy, see AR-347; (2) the selection had been based solely on the interview score, whereas DWD policy said that scores would be considered as part of the selection process; (3) the panel had not submitted a written recommendation with justification of their selection to the business enterprise program director; and (4) the panel hadn’t accepted Taylor’s letter of
recommendation. AR-425. The panel acknowledged that DWD’s policy allowed for discretion in determining whether to accept recommendation letters and other such materials, but it noted that “resumes and references are almost universally recognized as appropriate and helpful in job interviews,” and that it seemed “unreasonable to ignore a candidate’s successful operation of the precise facility at issue for such a long time.” AR-426. The panel also acknowledged Taylor’s concern that DWD had improperly factored Belsha’s perceived financial need into its selection decision, noting that the evidence and their findings “clearly suggest[ed] that [DWD] might well have considered factors outside the ambit of selecting the
candidate best suited for the business being bid.” Id. In light of these findings, the hearing panel recommended that DWD: (1) set aside the selection of Belsha as the permanent operator for the RCI/STF site; (2) give the candidates an opportunity to interview again with a different panel comprised as the selection policy directs; (3) prepare new interview questions and benchmarks; (4) accept resumes and references if offered by a candidate and consider relevant factors other than just the interview scores; (5) require the interview panel to submit a written recommendation to DWD justifying its
selection; and (6) select the candidate best suited for the business enterprise, regardless of the candidate’s need or the desire to even out sites among operators. Id. The hearing panel did not say who should serve as interim operator of the RCI/STF site during the re-interview process. 4. DWD’s final decision on Taylor’s grievance The evidentiary hearing panel’s recommendations were passed along to Michael Greco, the acting administrator of DWD, in accordance with the procedures outlined in Wisconsin Administrative Code § DWD 60.05(3). On June 12, 2012, Greco issued a final decision on Taylor’s grievance in which he mostly accepted the evidentiary hearing panel’s
recommendations. He ordered that: (1) DWD would conduct a new interview process to decide on the permanent operator for the RCI/STF site; (2) the same candidates would be invited to re-interview in front of a new panel, the composition of which would reflect DWD policy; (3) the interviews would use new questions and benchmarks; and (4) the interview panel members would submit a written recommendation to DWD justifying its selection; but (5) resumes and references would not be accepted. AR-429. Greco also decided that Belsha would continue to operate the RCI/STF vending machines in the meantime 5. Taylor’s appeal to U.S. Department of Education
Taylor disagreed with Greco’s handling of her grievance, so on July 23, 2012, she filed a complaint against DWD with the U.S. Department of Education. In that complaint, Taylor alleged that DWD had discriminated and retaliated against her, and she requested as a remedy that she “be placed as permanent operator of the vending sites” and receive “financial compensation for the loss of revenue and hardship resulting from [DWD’s] actions.” AR-3. More than nine months passed after Taylor’s filing with the Department of Education. On May 6, 2013, the Department of Education acknowledged receipt of Taylor’s complaint.
AR-6. That same day, the Department of Education sent Greco a letter notifying him that Taylor had lodged a complaint against DWD seeking a Randolph-Sheppard Act arbitration panel and that the complaint was “being reviewed.” AR-4. 6. 2013 interviews While Taylor’s complaint was pending before the Department of Education, DWD formulated new interview questions and benchmarks for selecting best-suited operators for stand-alone sites. See AR-578–83. This was an involved process, during which DWD sought review and comment from the Elected Committee of Blind Vendors (a committee of operators
that coordinates with DWD in administering the business enterprise program, see Wis. Admin. Code § DWD 60.03), as well as from the larger pool of blind vendors. AR-281–82. On March 12, 2013, DWD and the Elected Committee jointly approved the new benchmarks. AR-438– 39. On June 18, 2013, DWD sent a re-interview notice to the candidates who had originally interviewed for the RCI/STF site in 2011. Taylor’s interview was scheduled for June 27, 2013. AR-572. On June 24, Taylor emailed Feypel, the DWD employee who administered the business enterprise program, and Lorie Lange, the director of DWD’s business enterprise
program, to ask that the re-interview be delayed until after her appeal to the Department of Education was completed. AR-438. Lange responded that, in accordance with Greco’s decision on Taylor’s grievance, DWD would proceed with the re-interview process as planned. AR-437. Taylor replied to ask that Lange explain how the interview would be administered, how points would be awarded, and whether the candidates would be evaluated based on the status of their businesses as of the original date of the RCI/STF site interview in 2011. Id. Lange sent Taylor a copy of the recently approved benchmarks. Id. And in a subsequent email, Ralph Mikell
(another DWD employee) confirmed to Taylor that candidates would be evaluated based on their current business data rather than their data from 2011, when the original interviews had been conducted. AR-436. When she learned that she would be evaluated based on her current business data, Taylor withdrew from the re-interview process. In a June 26, 2013 email to Lange and Mikell, Taylor wrote: Due to my pending appeal regarding this matter, I am informing [DWD] that I will not be attending the re-interview on Thursday, June 27, 2013. Also, I strongly disagree with the re-interview, it is unethical and does not f[o]llow the decision rendered by the panel members of my evidentiary hearing. As well as, the fact that it is unacceptable to conduct a re-interview for a site bid out 2 years ago, based on an operator[’]s current business status. All candidates[’] statuses have drastically changed during the last two years, including many aspects that directly [a]ffects the outcome of Thursday[’]s re-interview. The arbitration process is moving forward and both parties can express their facts and reasoning at that time. I look forward to resolving this issue at that time. Id. The re-interviews went forward without Taylor, and Belsha was once again selected as the permanent operator for the RCI/STF site. AR-282. Greco informed the Department of Education about these developments in a letter dated July 11, 2013. AR-576–77. 7. The Department of Education arbitration On October 8, 2013, Taylor filed an amended complaint filed with the Department of Education in which she expanded the scope of her original complaint to incorporate objections to the re-interview process—specifically, DWD’s determination that it would evaluate re-interview candidates based on their current business data instead of their business data from 2011. AR-16–18. Almost two years later, on July 23, 2015, the Department of Education notified Taylor and DWD that it was convening a three-member arbitration panel to hear and
render a decision on the issues raised in Taylor’s complaint. AR-457–58. The arbitration hearing took place on September 26, 2017, and the panel issued its decision in early 2018. AR-856. C. The arbitration panel’s decision In a written decision, a two-member majority of the panel concluded that DWD had “arbitrarily, capriciously and in a biased manner failed to award Taylor the RCI/STF site during the two selection processes.” AR-835. One panel member dissented from the conclusion that DWD had acted improperly and from the remedies awarded to Taylor. AR-858–65. The court
will refer to the majority decision as the decision of the panel. The arbitration panel concluded that the 2011 interviews were not conducted in accordance with DWD policies and procedures, and that “consequently the [initial] selection of Belsha was not adequately justified” under Wisconsin Administrative Code § DWD 60.08. AR-843. The arbitration panel fully agreed with the results of the full evidentiary hearing, concluding that the recommendations “should have been followed in their entirety.” AR-846. The arbitration panel found four flaws with DWD process after the full evidentiary hearing. First, the arbitration panel concluded that Greco was arbitrary and capricious in not
accepting resumes and references. AR-848, 849. Second, the arbitration panel concluded Greco had been arbitrary and capricious to allow Belsha to serve as the interim operator of the RCI/STF site while the second interview process was underway. AR-847. Third, the arbitration panel concluded that it was unfair to Taylor to use profitability
data from 2013, rather than 2011. AR-850. Fourth, the arbitration panel faulted the 2013 interviews because of the “the year plus delay between [the] decision of the acting DVR administrator to order a new interview process and the date the new interview took place.” AR-850. The panel concluded that “the year plus delay in holding the new interview was not fair or equitable” and was “arbitrary and capricious.” AR-851. Ultimately, the panel concluded that DWD had violated the requirement that it select the best-suited operator for the RCI/STF site. AR-851. The panel determined for itself that
Taylor was the best-suited operator and that she should have been selected. The panel decided that DWD had demonstrated “bad faith throughout this matter” and that “Taylor must be made whole by making her the permanent operator at the RCI/STF site.” AR-855. The panel also awarded Taylor compensatory damages at the rate of $105.38 per day, along with her legal fees and costs. Id. DWD has not yet implemented the award.
ANALYSIS The Randolph-Sheppard Act, 20 U.S.C. § 107d-2(a), provides for district court review
of the arbitration panel decision under the Administrative Procedure Act. APA review is limited and deferential: the court does not reweigh the evidence or substitute its judgment for the agency’s. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971). But the reviewing court will set aside an agency action if it is: (A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; (B) contrary to constitutional right, power, privilege, or immunity; (C) in excess of statutory jurisdiction, authority, or limitations, or short of statutory right; (D) without observance of procedure required by law; (E) unsupported by substantial evidence in a case subject to sections 556 and 557 of this title or otherwise reviewed on the record of an agency hearing provided by statute; or (F) unwarranted by the facts to the extent that the facts are subject to trial de novo by the reviewing court. 5 U.S.C. § 706(2). In this case, several key factual findings by the panel are not supported by substantial evidence. The panel has ordered money damages, a remedy that is contrary to the state’s sovereign immunity. Before discussing specific conclusions of law and findings of fact, the court will begin with two more foundational problems. A. Foundational problems The Randolph-Sheppard Act specifies that arbitration panel decisions are reviewed under the APA, but it provides little guidance on the conduct of the arbitration itself. The Department of Education adopted a set of policies and procedures in 1978 to govern Randolph-Sheppard Act arbitrations. AR-786–92. The arbitration panel apparently applied these policies and procedures in this case. But the policies and procedures establish only a few procedural rules; they provide no guidance to arbitration panels concerning such important issues as the scope of the arbitration panel’s review or its authority to order relief. The arbitration panel here took it upon itself to define the scope of the arbitration because the parties could not agree on the issues. AR-828. Drawing some guidance from the letter from the Department of Education convening the arbitration, AR-113–14, the panel stated the issues as follows:
1. Did the [DWD] act in an arbitrary, capricious and biased manner by failing to follow its rules, regulations, policies and procedures when it selected a permanent operator for the RCI/STF business site and/or by failing to place Theresa Taylor as permanent operator for RCI/STF thereby violating the Randolph- Sheppard Act, implementing regulations, state rules, regulations, policies and procedures? 2. If so, what is the appropriate remedy? AR-828. It’s not clear from this statement whether the panel thought its task was a review of DWD’s actions under an arbitrary and capricious standard, or whether it believed that it was empowered to re-do the bid award process. The panel decision appears at times to apply the arbitrary and capricious standard to DWD decisions. But reading the panel decision as a whole, particularly its decision about the remedy, it is clear that the panel felt empowered to redo the bid process and install Taylor as the permanent operator at the RCI/STF site. The broad authority that the arbitration panel arrogated to itself leads to a first foundational problem: the ordered remedy is unfair because none of the other blind vendor candidates participated in the arbitration. The panel’s remedy would deprive Belsha of the permanent contract for the RCI/STF site, yet Belsha had no opportunity to participate in the arbitration. The panel decided that the rules for the contract award process should have been different in that letters of recommendation should have been considered and that DWD should have considered profitability data from 2011, not 2013. But Belsha may have been able to supplement or amend her application and prevail under the revised rules. The panel’s failure to consider the rights of the competing blind vendors, particularly Belsha’s, renders its remedy arbitrary and capricious. See Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (a decision is arbitrary and capricious if it “entirely failed to consider an important aspect of the problem.”).
A second foundational problem is that the panel applied the wrong burden of proof. The panel correctly recognized that Taylor had the burden of proof because she was the proponent of a ruling that DWD had violated the law. 5 U.S.C. § 556(d) (“Except as otherwise provided by statute, the proponent of a rule or order has the burden of proof.”) But the panel wrongly concluded that Taylor was required to prove her case only by “substantial evidence” rather than a “preponderance of the evidence.” AR-838–39. “Substantial evidence” is not a burden of proof; it is a standard of review generally applicable to judicial review of agency fact-finding under the APA. This standard is deferential,
requiring less than a preponderance of the evidence. Young v. Sec'y of Health & Human Servs., 957 F.2d 386, 389 (7th Cir. 1992). Substantial evidence means more than a mere scintilla of proof, but it requires no more than “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Kepple v. Massanari, 268 F.3d 513, 516 (7th Cir. 2001) (quoting language that originated in Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)). The arbitration panel decided to apply the deferential substantial evidence standard over DWD’s objections that Taylor needed to prove her case by a preponderance of the
evidence. The panel’s reasons for doing so were misguided. The panel said that this court had adopted the substantial evidence standard in Wisconsin Department of Workforce Development, Division of Vocational Rehabilitation v. United States Department of Education, 667 F. Supp. 2d 1007, 1017–18 (W.D. Wis. 2009). AR-839. But that is incorrect. In that case, Judge Crabb applied the substantial evidence standard in her own review of the arbitration panel decision under § 706(2) of the APA; she did not say that substantial evidence was the burden of proof assigned to the aggrieved party in the arbitration itself. The arbitration panel in this case also
cited 5 U.S.C. § 556(d), which says that the panel’s decision must be “supported by and in accordance with the reliable, probative, and substantial evidence.” It is true that the panel decision must be supported by substantial evidence, lest it be reversed on APA review. But that doesn’t mean that the aggrieved vendor need only prove her case by substantial evidence. By using a deferential standard of review as Taylor’s burden of proof in this case, the arbitration panel effectively flipped the burden to DWD.2 For example, the arbitration panel cited DWD’s failure to adduce evidence to support the explanation of its decisions, even though it was Taylor’s burden to show that these decisions were arbitrary and capricious. See
e.g., AR-847 (no evidence to support Greco’s decision that it would be disruptive to remove Belsha as interim operator); AR-848 (no evidence to support DWD’s claim that it was familiar with both Taylor and Belsha as existing blind operators). The arbitration panel hedged a bit by saying that “even assuming, arguendo, that the preponderance of evidence test should be used, the panel finds that Taylor in any event has met that heavier burden.” AR-839. This statement adds nothing of substance. As the discussion of the panel’s substantive decisions will show, it’s apparent that the panel gave Taylor the
2 The mistake is understandable given the lack of more specific guidance in the APA, the Randolph-Sheppard Act, or even in the Department of Education’s arbitration policies and procedures. At least one other arbitration panel has made the same error. See Opportunities for Ohioans with Disabilities v. United States Department of the Air Force, Wright-Patterson Air Force Base, No. R-S/16-08 (Feb. 22, 2018), available at https://www2.ed.gov/programs/rsarsp/ arbitration-decisions/r-s-16-08.pdf. benefit of the doubt at many points, confirming that the panel applied a standard more deferential the preponderance of the evidence. Nowhere does the arbitration panel explain how Taylor’s evidence would have met the heavier burden of proof to a preponderance of the evidence.
B. Problems with the panel’s substantive rulings The arbitration panel decision includes a long discussion of DWD’s actions before the full evidentiary hearing. The panel focused particularly on statements attributed to Feypel that Belsha’s earnings had diminished significantly, and that DWD wanted to give her a site that would allow her to make enough to survive. The arbitration panel said that the record was “replete” with evidence that DWD was intent on placing obstacles in front of Taylor and that it was biased against her. AR-845. In support, it provided one string cite of many pages from the hearing transcript as “uncontroverted evidence” showing that DWD broke its own rules
and was biased against Taylor. Id. But the finding that DWD was relentlessly biased against Taylor is unsound for several reasons. First, the court has reviewed the entire transcript of the hearing, including specifically the pages cited by the panel. The panel did not explain what in the cited pages demonstrated consistent bias against Taylor; that conclusion seems to be based entirely on Feypel’s statements. DWD did not call Feypel or Greco to testify, but it is not a fair to say that there was no testimony to controvert Taylor’s allegation of bias. Lorie Lang, the director DWD business enterprise program, testified extensively about DWD’s policies on temporary
assignments and promotions and transfers. No witness directly refuted Taylor’s testimony that Feypel made the statements about Belsha, but Lang’s testimony rebutted the idea that DWD considered “social need” in awarding vending contracts. AR-153–54. The arbitration panel offered no criticism of Lang’s testimony or her credibility. Second, loss of income to an incumbent blind vendor is a permissible consideration under Wisconsin’s Randolph-Sheppard regulations. The arbitration panel concluded that
Feypel’s consideration of Belsha’s declining income was an unfair consideration of “social need.” But the regulations governing the Randolph-Sheppard program in Wisconsin allow an operator faced with significant declining sales due to reasons beyond her control to request a transfer to another location. Wis. Admin. Code § DWD 60.07(2). And such a transferring vendor is entitled to preference in selection for future vacancies. Wis. Admin. Code § DWD 60.08(3). To be clear, DWD did not argue that Belsha sought a transfer under these transfer- preference provisions. But these provisions undercut the panel’s conclusion that Feypel’s consideration of Belsha’s declining income can be explained only as unfair bias against Taylor.
Third, the record as a whole does not show consistent bias against Taylor. Feypel himself did not demonstrate bias against Taylor in the 2011 interviews: he scored Taylor and Belsha exactly the same. AR-410. Taylor won the bid for the RYOC facility. AR-412. And after the 2011 interviews, DWD awarded Taylor two additional vending sites outside the normal bidding process. AR-193. Fourth, and most important, the influence of Feypel’s bias (if there were any) was eliminated by re-doing the interviews without his participation. After the full evidentiary hearing, DWD substantially sustained Taylor’s grievance, and DWD decided to re-do the
selection process using new criteria. The arbitration panel determined that the decision of the full evidentiary hearing panel was correct in every respect and should have been fully implemented. Because the full evidentiary hearing addressed Taylor’s grievance, what happened before the full evidentiary hearing should have been immaterial. What really matters is what happened after the full evidentiary hearing, when DWD implemented the decision of the full evidentiary hearing, but only partially. The arbitration
panel found four specific deficiencies in the implementation of the recommendations of the full evidentiary hearing panel. None is supported by substantial evidence. 1. Use of 2013 business data in the 2013 interviews The cornerstone of the arbitration panel’s analysis was its conclusion that DWD should have evaluated Taylor based on her 2011 profitability data rather than her 2013 data in the 2013 re-interviews. The arbitration panel concluded that it was “clear that Taylor was disadvantaged in comparison to Belsha” in answering the interview questions about profit margins, because Taylor’s profit margins in 2013 were lower than they would have been in
2011, when she was the interim operator of the RCI/STF site. AR-850. The arbitration panel relied exclusively on Taylor’s arbitration hearing testimony that prison sites, such as RCI/STF, are more profitable than the non-prison sites. But the panel’s finding that Taylor would have been disadvantaged under the 2013 data is not supported by substantial evidence. There were only two questions in the 2013 selection criteria that assessed the applicant’s profitability data—the fifth and sixth questions in the “performance review” section. See AR-433.3 Differences in profit margin would have had a
3 Those questions were: 5. Operator cost of goods sold percentage. a. 0%–40% - 3 points b. 40.01%–50% - 2 points c. Greater than 50.01% - 1 points minimal impact on the overall scores of the candidates. And, critically, the record of the arbitration hearing does not contain evidence of what Taylor’s and Belsha’s scores would have been using the 2013 data. Taylor testified at the arbitration hearing that she did not know what she would have scored on either question had she gone through with the 2013 interview.
AR-191–92. There is no evidence in the record of any profitability data for Belsha, so even if the arbitration panel had Taylor’s 2013 profitability, the panel would have had nothing to compare it to. Taylor’s testimony that her profitability was significantly diminished by the loss of the RCI/STF site is undercut by the evidence of record. Taylor testified that in 2011 she had 24 percent net profit. AR-184. But, as DWD points out, Taylor’s net profit in 2011 at the Southern Wisconsin Center—a primarily non-prison site—was 23 percent. AR-350. Either way, she would have scored four points on question 6 (for having net profit between 20 and 24.99
percent). So the arbitration panel’s finding that Taylor was significantly disadvantaged by the use of 2013 profitability data is not supported by substantial evidence. Indeed, the arbitration panel itself seemed to recognize this uncertainty: “The scores on these two questions might have made the difference between Belsha and Taylor as to who scored the highest on the interview.” AR-850 (emphasis added). See Rapanos v. United States, 547 U.S. 715, 786 (2006) (“conditional language” can “suggest an undue degree of speculation” that renders an agency decision unsupported by substantial evidence). If Taylor’s loss of profitability were critical to her case,
6. Operator net profit percentage prior to commissions and/or rents to state agencies. Vending only a. Greater than 25% - 6 points b. 20%–24.99% - 4 points c. 0%–19.99% - 2 points one would have expected Taylor, as the party with the burden, to produce documentary evidence to support it. She did not. 2. Greco’s decision not to accept letters of recommendation The arbitration panel concluded that DWD’s failure to accept letters of
recommendation was arbitrary and capricious because Greco “gave no explanation for his decision not to comply with the [final evidentiary hearing] panel’s recommendation” that they be accepted. AR-848. The arbitration panel thought recommendation letter would have been “a powerful asset” that “would have helped negate the internal bias [DWD] demonstrated during the selection process in favor of selecting Belsha as permanent operator.” AR-848, 849. The panel’s finding that there was persistent bias against Taylor is unfounded, for reasons explained above. But the panel’s conclusions about Taylor’s letter of recommendation is unsound for additional reasons. DWD had explained why it would not accept
recommendation letters early in the process. In preparation for the 2011 interviews, Pomeroy said that DWD already had “personnel records of the [business enterprise program] operators and is familiar with the training and performance of each operator.” AR-418. Both Belsha and Taylor were long-serving incumbent vendors well known in DWD’s business enterprise program, so letters of recommendation were not necessary to inform DWD about these two candidates. Greco did not expressly reiterate that explanation in his 2013 decision, but that alone would not make his decision arbitrary and capricious. The arbitration panel thought that Taylor’s letter of recommendation would counteract
bias against her, but the panel did not explain how it would do that. Taylor’s actual letter of recommendation is favorable, but it is very brief and it provides little detail about Taylor’s performance. AR-353. It simply does not contain enough information to be fairly characterized as a “powerful asset.” And, if letters were considered in the 2013 interview process, Belsha could have provided letters in support of her application. There is no evidence that Belsha’s performance as a vendor was ever deficient. The arbitration panel’s finding that Taylor’s letter of recommendation would have had a significant effect on the contract award process is mere
speculation. 3. Greco’s decision not to make Taylor the interim operator of the RCI/STF site The arbitration panel also concluded that Greco’s decision to retain Belsha as the interim operator of the RCI/STF site during the re-interview process was arbitrary and capricious. AR-847. Greco explained that it would have been “unnecessarily disruptive” to appoint a different interim operator until the permanent operator was selected. According to the arbitration panel, DWD had “offered no evidence to support this contention,” and “Taylor’s past performance [as interim operator] proves that someone could have taken over
the RCI/STF site on an interim basis without disrupting the vending business.” Id. The panel’s reasoning is unsound for several reasons. The panel criticized Greco for not following the recommendation of the full evidentiary hearing on this point: “The [DWD] did not offer any explanation or reason for not expressly complying with this recommendation.” Id. But the panel also acknowledged that the full evidentiary hearing panel hadn’t made such a recommendation: “The FEH panel did not say who should serve as interim operator while a second interview process was held.” Id. The arbitration panel appears to have been confused about the recommendation of the full evidentiary hearing panel and about Belsha’s status as
the operator. The arbitration panel is right that a vendor could, if necessary, step up on short notice and run the RCI/STF site on an interim basis, as Taylor once did. But a change in vendor would be disruptive to the vendors themselves, and it would inevitably disrupt the client institutions to some degree. Greco’s decision to maintain Belsha as the operator of the RCI/STF facilities while the re-interviews went forward meant that there might be no operator change-over (if Belsha were re-selected), or at most one change-over (if Taylor were selected). But replacing
Belsha with Taylor as the interim operator would have guaranteed at least one change of vendor, and created the possibility of two (if Belsha were selected in the 2013 re-interviews). Taylor was the interim operator with no permanent claim to the RCI/STF site, and Greco’s decision to leave Belsha in place until the re-interviews limited potential disruption. The arbitration panel’s conclusion that Greco’s decision was arbitrary and capricious is unsound, and its finding that this decision would have disadvantaged Taylor in the re- interviews is not supported by substantial evidence, as discussed in the next section. 4. Delay between Greco’s decision and 2013 re-interviews
The arbitration panel criticized DWD for the delay between Greco’s decision to order new interviews and the date of the re-interviews. The panel concluded that DWD had failed to “give a good reason or explanation for the lengthy delay.” AR-850. According to the panel, DWD “simply dropped the ball” by failing to “do on a timely basis what it said it was going to do to comply with the [full evidentiary hearing] panel’s recommendation.” AR-850, 851. The panel thought this was unfair to Taylor and that she was severely harmed as a result. DWD contends that the delay between the first and second selection processes was attributable to the requirement that DWD seek active participation from the Elected
Committee of Blind Vendors in establishing the new selection benchmarks called for by the full evidentiary hearing panel. The evidence amply supports DWD’s account. The arbitration panel based its conclusion that DWD had “dropped the ball” solely on arbitration-hearing testimony from Lange, who testified that she became manager of the business enterprise program in December 2012 and had not become aware that the re-interviews were “an outstanding item” until May 2013. AR-273. But documents in the record show that Lange was simply uninformed: DWD had been working on developing the new benchmarks in
collaboration with the Elected Committee of Blind Vendors over the course of the year. See AR-438 (indicating that the Elected Committee of Blind Vendors approved new benchmarks on March 12, 2013). There is no reliable evidence that DWD actually “dropped the ball,” and certainly nothing to suggest that DWD intentionally delayed the proceedings. And even if there were a delay in conducting the re-interviews, the arbitration panel does not explain how the delay “made it difficult for Taylor to compete for the permanent operator position at RCI/STF.” AR-850. The arbitration panel seems to have assumed that Belsha’s interim operation of the RCI/STF site would allow Belsha to build up her profitability,
thus gaining a decisive advantage over Taylor. But, because there is no evidence in the record of either Belsha’s or Taylor’s profitability in 2013, this yet more speculation. The arbitration panel’s findings regarding the reason for delay in conducting the re-interviews, and the consequence of that delay, are not supported by substantial evidence. 5. The arbitration panel’s remedies Ultimately, the panel concluded that Taylor was the best-suited operator for the RCI/STF site, it ordered that she be installed as the permanent operator of the site, and it ordered money damages against DWD for its failure to award Taylor the contract earlier.
The substantive decision that Taylor was the best-suited operator for the RCI/STF site is arbitrary and capricious because information about the competing vendors, particularly Belsha, was not in the record, and thus the panel could not reliably gauge the relative merits of their applications. Blame for this lies with Taylor herself, because she chose not to participate in the 2013 re-interview process. Taylor’s choice was unreasonable. She though the use of 2013 data was unfair to her, but she did not actually know whether this would pose any meaningful disadvantage, and she forfeited the opportunity to explain her profitability data in the
interviews. And as it turns out, even at the arbitration she did not adduce evidence that her 2013 profitability data was significantly different from her 2011 data. Taylor’s choice prevented DWD from considering her application, and it foreclosed the possibility that the arbitration panel would have a record that would show how Taylor’s application compared to those of the competing applicants. The arbitration panel’s decision that Taylor was the best-suited operator for the RCI/STF site is thus unsupported. And the panel’s decision that Taylor should replace Belsha as the permanent operator of the RCI/STF site cannot stand. Because the court will set aside
the arbitration panel’s substantive decision that Taylor is the best-suited operator, the panel’s award of damages must also be set aside. The panel’s award of damages would have to be set aside for an additional reason: it would be barred by sovereign immunity. The court need not address the matter in great depth, given the decision on the merits of the arbitration panel decision. The court will adopt Judge Crabb’s analysis in Wisconsin Department of Workforce Development, Division of Vocational Rehabilitation v. United States Department of Education, 667 F. Supp. 2d 1007 (W.D. Wis. 2009). Although courts of appeals in other circuits have held otherwise, see id. at 1012–13 (collecting
cases), Supreme Court and Seventh Circuit case law make clear that “if Congress intends to condition the states’ participation in a federal program or receipt of federal funds on their waiver of sovereign immunity to damage suits, it must do so unambiguously so that the states are fully aware of the conditions to which they are agreeing.” Id. at 1014 (citing Lane v. Pena, 518 U.S. 187, 192 (1996); United States v. Nordic Village Inc., 503 U.S. 30, 34 (1992); Nelson v. Miller, 570 F.3d 868, 883 (7th Cir. 2009)). The Randolph-Sheppard Act arbitration provisions, and Wisconsin’s acquiescence to them, abrogated state sovereign immunity for
injunctive relief. But the Randolph-Sheppard Act lacks the “‘unequivocal textual expression’ required to waive sovereign immunity for monetary relief.” Id. (quoting Nelson, 570 F.3d at 884). Nothing in Taylor’s briefs persuades the court to deviate from Judge Crabb’s analysis of the sovereign immunity issue. Taylor does not cite more recent authority that undermines that analysis. And since that 2009 decision, the Supreme Court has reaffirmed that a state’s consent to suit “must be ‘unequivocally expressed’ in the text of the relevant statute. . . . Waiver may not be implied.” Sossaman v. Texas, 563 U.S. 277, 284 (2011) (quoting Pennhurst
State Sch. & Hosp v. Halderman, 465 U.S. 89, 99 (1984)). The arbitration panel’s award of compensatory damages, attorney fees, and costs would have to be aside as contrary to law. C. Instructions for remand The court will vacate and reverse the arbitration panel’s decision and remand the case to the Department of Education. Because the court concludes that there was no deficiency in the 2013 re-interviews and that Taylor unreasonably chose not to participate in those re- interviews, the Department of Education is instructed to affirm DWD’s award of the permanent contract for the RCI/STF site to Joelyn Belsha. ORDER IT IS ORDERED that: 1. Respondent Theresa Taylor’s motion to confirm and enforce the arbitration award, Dkt. 24, is DENIED. 2. Petitioner’s petition for judicial review of the final federal agency decision, Dkt. 26, is GRANTED. The arbitration panel’s decision is VACATED and REVERSED, and the case is REMANDED to the United States Department of Education to affirm DWD’s award of the permanent contract for the RCI/STF site. 3. The clerk of court is directed to enter judgment accordingly and close the case. Entered December 3, 2019. BY THE COURT:
/s/ ________________________________________ JAMES D. PETERSON District Judge