American Association of Cosmetology Schools v. DevosAmerican Association of Cosmetology Schools v. Devos
Ms. Samuel has not merely strewn together unrelated acts of alleged discrimination and retaliation to create a hostile-work-environment claim. She alleges that from when she first joined MPD, Ms. Haines-Walton “maintained an ongoing pattern of harassing [her] because of [her] national origin” to the point where “[t]he environment in HR[] became unbearable” due to Ms. Haines-Walton‘s “creat[ion] [of] such a hostile work environment that [she] was fearful that [her] job and career were in jeopardy.” Samuel Decl. ¶¶ 22, 49, 56, 61. Whatever criticisms MPD may have about these allegations, they are not mere cobbled-together discrete acts to which the D.C. Circuit is averse. See Dudley, 924 F. Supp. 2d at 164. And, even though Plaintiff alleges disparate treatment based on some discrete acts that she also includes under the hostile-work-environment umbrella, that fact alone is not a legally sufficient reason to reject her claim. Baird, 662 F.3d at 1252. It is not as if plaintiffs must choose between asserting a disparate-treatment claim or a hostile-work-environment claim. See id. Thus, Ms. Samuel‘s hostile-work-environment claim survives MPD‘s motion for summary judgment - at least for now.
Because Ms. Samuel‘s hostile-work-environment claim is raised so ambiguously in the complaint and pleadings, the Court will allow MPD to file a renewed motion for summary judgment on Plaintiff‘s claim that she suffered a hostile work environment. The Court allows such briefing so that Plaintiff can clarify the contours of her claim and MPD can squarely raise its defenses to such a claim.
V. CONCLUSION
For the foregoing reasons, the Court grants in part and denies in part Defendant‘s motion for summary judgment. An order consistent with this Memorandum Opinion is separately and contemporaneously issued.
Kathryn L. Wyer, U.S. Department of Justice, Washington, DC, for Defendant.
MEMORANDUM OPINION
GRANTING PLAINTIFF‘S MOTION FOR SUMMARY JUDGMENT; DENYING DEFENDANT‘S MOTION FOR SUMMARY JUDGMENT; DENYING PLAINTIFF‘S MOTION FOR A PRELIMINARY INJUNCTION
RUDOLPH CONTRERAS, United States District Judge
I. INTRODUCTION
In this case, the Court considers whether the Department of Education (“DOE“)
In setting standards that determine which proprietary schools’ graduates are entitled to federally backed student loans, the DOE looks to the rates at which the schools’ graduates are “gainfully employed.” To determine whether graduates are gainfully employed, the DOE has adopted a test that compares the graduates’ income levels to their levels of debt. To determine the graduates’ income, the DOE presumptively uses the Social Security Administration‘s (“SSA“) income data. This data does not account for income that is not reported to the Internal Revenue Service. Schools may appeal the DOE‘s use of SSA data through “alternate earnings appeals,” which, if successful, allow them to use alternate measures of income before the debt-to-earnings rates become final. To submit such an appeal, a school is required to use either state-sponsored data pertaining to over half of its graduates during the relevant timeframe or gather income data on almost all of its graduates through a survey. Schools that fail the debt-to-earnings test for a long enough time lose eligibility for federal loans. Schools at immediate risk of losing federal-loan eligibility are required to warn their students and prospective students that they may be ineligible for student loans in the near future.
During the notice-and-comment period, several commenters argued that use of SSA data would be unfair to, among others, cosmetology programs, because their graduates disproportionately underreport their income due to high levels of cash-based and self-employment-based earnings, including tips. The commenters suggested that the Bureau of Labor Statistics‘s survey-based data would better account for unreported income.
The DOE rejected the commenters’ objection to the data and their proposed solution. In rejecting the objection, the DOE reasoned that graduates who do not report their income are subject to civil and criminal penalties, and noted that the SSA data is only one means of determining income; programs can submit alternate earnings appeals to show the DOE that its graduates’ average incomes are actually higher than the SSA data indicates. In rejecting the proposed solution, the DOE noted that the Bureau of Labor Statistics data cannot be tied to particular programs, and thus would undermine the purpose of the regulations - to identify individual programs whose graduates are not gainfully employed.
The American Association of Cosmetology Schools (“AACS“) sued under the Administrative Procedure Act, arguing that these responses were unsatisfactory, and thus the DOE arbitrarily and capriciously failed to adequately consider the unreported-income issue when it promulgated the regulations. AACS is a nonprofit association of cosmetology schools, many of which are at risk of failing the DOE‘s debt-to-earnings test. In fact, at least three schools have already posted warnings to their students, because they could not feasibly appeal their failing grades. Defendant moves for summary judgment on two procedural grounds and on the basis that the regulations were not arbitrary or capricious.
Both of the DOE‘s procedural arguments concern subject-matter jurisdiction. It first argues that AACS does not challenge final agency action, because none of
The DOE argues that it did not act arbitrarily and capriciously because it provided a reasoned explanation for its rejection of commenters’ concerns about unreported income and their proposed solution. The Court holds that, although the DOE was rational in rejecting the commenters’ proposed alternative, it did not adequately address the issue of unreported income. Neither side disputes that SSA data is highly accurate, even though it may be insufficient in certain circumstances. But, as the commenters pointed out - and the government did not dispute - there is a significant problem with this data. Despite the illegality of failing to report earnings to the IRS, many programs’ graduates fail to report substantial portions of their income. The DOE proffered two responses to the commenters’ problem. First, it argued that civil and criminal penalties deter underreporting. Second, it noted that institutions may submit an alternate earnings appeal to have their graduates’ earnings more accurately calculated.
The existence of penalties - which existed when the commenters’ data was collected and will continue to exist into the future, absent some added deterrent - is irrelevant to the issue of undercounting income. In comparison, an alternative means of measuring earnings data is responsive to the problem. The DOE justifiably used SSA data as the default measure of earnings, but supplemented that default methodology with an alternate means of determining income. However, by inexplicably requiring high response rates to submit state-sponsored or survey-based alternate earnings calculations, the DOE narrowly circumscribed the alternate-earnings appeal process, making it unfeasible for certain programs to appeal their designations. To remedy the DOE‘s arbitrary and capriciously narrowing of appellate recourse - and to avoid upending the entire administrative scheme - the Court removes barriers to appeal, making it more widely available for programs subject to the regulations.
II. REGULATORY BACKGROUND
Congress passed Title IV of the Higher Education Act of 1965 (“HEA“) to make postsecondary education more widely available to the general public. See
From 2009 to 2011, the DOE promulgated regulations defining the term “gainful employment” (“GE“) in terms of a debt-to-income test and a debt-repayment test. See Program Integrity: Gainful Employment - Debt Measures, 76 Fed. Reg. 34,386 (June 13, 2011); see also Ass‘n of Private Colls. & Univs. v. Duncan, 870 F. Supp. 2d 133, 153-54 (D.D.C. 2012). This Court determined that parts of that rule were arbitrary and capricious under the Administrative Procedure Act (“APA“). See Ass‘n of Private Colleges & Universities, 870 F. Supp. 2d at 154, 158.
The DOE again set out to define the term “gainful employment,” this time focusing only on program graduates’ overall debt compared to their earnings, which it refers to as “D/E” rates. See Program Integrity: Gainful Employment, 79 Fed. Reg. 64,890, 64,891 (Oct. 31, 2014) (to be codified at
The GE regulations use two tests to determine whether programs “pass,” each of which is designed to determine whether programs’ graduates are earning enough income to justify the government‘s guarantee of their loans. The first test divides the program‘s graduates’ average loan payment amounts by their average discretionary income amounts.1 The resulting rate is called the program‘s “discretionary income rate.”2
The second test divides the program‘s graduates’ average annual loan payments by their average annual income. The resulting rate is called the program‘s “annual earnings rate.”
A program is classified as “failing” the GE regulations if its discretionary income rate is greater than 30 percent and its annual earnings rate is more than 12 percent.
To measure program graduates’ income rates in calculating a program‘s discretionary income rate or annual earnings rate, the DOE presumptively uses income data from the Social Security Administration.
If the DOE‘s presumptive D/E rates, as calculated using SSA data, would leave a program either failing or in the zone, it may file an “alternate earnings appeal” to request an alternative formulation to determine the program‘s final D/E rates.
Notably, if a program successfully appeals the DOE‘s use of SSA data for the determination of its D/E rates, it can avoid any negative repercussions altogether. Under
However, if a school does not timely appeal its designation or is unsuccessful in seeking an alternative earnings calculation, it begins to suffer serious consequences. Under the Gainful Employment rules, schools that could lose HEA loan eligibility the following year must provide written warnings to current and prospective students informing them that their “program has not passed standards established by the U.S. Department of Education,” and that if the institution does not obtain a passing grade, the students may not be able to receive federal student grants or loans the following year.
Federal courts have jurisdiction to review final agency action.
III. FACTUAL BACKGROUND
The facts in this case are largely undisputed. Plaintiff, the American Association of Cosmetology Schools brings this action under the APA, seeking to enjoin the DOE from enforcing its GE rules. Compl. at 1. AACS is a nonprofit association of accredited cosmetology programs. Compl. ¶ 1. At core, AACS takes issue with the DOE‘s use of the SSA annual earnings data. See Compl. ¶¶ 15, 35-38. Plaintiff notes that the SSA data is incomplete because it tends to undercount the income of self-employed individuals and individuals who receive a significant amount of their income in the form of tips and cash. J.A. 797, ECF No. 26-1; see also Decl. of Anthony Civitano (“Civitano Decl.“) ¶¶ 25, 28, ECF No. 8-1. These two groups, the argument goes, can underreport their income more easily than persons earning paychecks from employer payrolls. J.A. 606; Civitano Decl. ¶ 28. Thus, cosmetology school graduates - who Plaintiff claims are disproportionately self-employed and paid in cash - on average have higher incomes than the SSA‘s numbers reflect, in many cases as much as double. J.A. 595, 787; Civitano Decl. ¶ 28.
According to Plaintiff, the DOE was made aware of “this well-known reported income gap phenomenon” by several commenters during the notice-and-comment period of the GE rule promulgation. Compl. ¶¶ 47-48. Indeed, the DOE explicitly acknowledges that these comments included claims that “about half of earnings in service occupations such as cosmetology” are made up of tips, and that many people in the industry are self-employed. See 79 Fed. Reg. at 64,955. A report by Stanford professor Dr. Eric Bettinger, which was submitted to the agency during the notice-and-comment period, found that both tip income and self-employment in-
Even aside from formal public comments, AACS cites to several sources suggesting that the underreporting of cash-based earnings - including in the cosmetology industry - is widely known to be a serious problem. See Brodie Decl. ¶ 3. For example, AACS provides the House Budget Committee testimony of Russell George, Treasury Inspector General of the Tax Administration, wherein Mr. George states that “self-employed individuals who formally operate ... businesses ... are estimated to report only about 68 percent of their income for tax purposes” and “self-employed individuals operating businesses on a cash basis report just 19 percent of their income to the IRS.” Brodie Decl. at 1258. Although these materials were apparently not submitted in connection with notice-and-comment and have little direct relevance, they lend credence to the general idea that the income gap phenomenon is a significant problem.
The DOE acknowledged that the use of SSA data was imperfect, but generally viewed civil and criminal penalties as sufficient deterrents to underreporting. See 79 Fed. Reg. at 64,955-56. The pertinent portions of the DOE‘s responses to the public comments are as follows:
We do not agree that our reliance on reported earnings is flawed because of its treatment of self-employment earnings and tips .... We acknowledge that some self-employed individuals may fail to report, or underreport, their earnings. However, [the Internal Revenue Code] requires self-employed individuals to file a return if the individual earns $400 or more for the taxable year. ... Underreporting subjects the individual to penalty or criminal prosecution. ...
With respect to the earnings of workers who regularly receive tips for their services, [the Internal Revenue Code] requires individuals to report to IRS their tip earnings for any month in which those tips exceeded $20, and ... individuals who fail to do so are subject to penalties. ...
For these reasons, we do not agree with the commenters’ assertion that aggregate earnings data provided by SSA ... are unreliable with respect to workers in occupations that involve significant tip income or a high percentage of income from self-employment. ... Moreover, the regulations allow an institution to submit an alternate earnings appeal using State databases or a survey.
79 Fed. Reg. at 64,955-56.
At least one commenter also proposed an alternate means of calculating income. See
[i]mputing some percentage of added earnings to account for underreported
tips and other compensation could only be done by generalizations drawn from some source of data on earnings, but none has been suggested that would permit doing so in a way that would distinguish between programs. To assess the bias that the commenter asserted ... the commenter relies on earnings data from ... surveys of households. The survey samples data on a selection of all households, and relies on earnings data as provided by the individuals included in the survey. As the commenter noted, there are no data ... that allow one to associate a particular respondent with a particular GE program.
Unlike the approach taken in these regulations, which captures all earnings of the cohort of students completing a program and credits those earnings to the program completed by the wage earners, the analysis proposed by the commenter does the reverse: It extrapolates from earnings reported by those survey recipients who identify their occupation as one that appears related to GE programs of that general type, and then projects an increase in aggregate earnings for all GE programs in the category of programs that appears to include that occupation. In fact, even if the respondents were all currently employed in occupations for which a category of GE programs trains students, the respondents’ earnings will almost certainly have no connection with a particular GE programs [sic] we are assessing. Because any inference drawn from ... respondents’ earnings could only benefit a whole category of programs - improving the D/E rates for every program in that category - using such inferences would mask poorer performing programs and thwart a major purpose of the GE assessment.
In addition, by the time the survey is conducted, the respondent can be expected to identify his or her current or most recent job, which may be different than the occupation for which training was received years before in a GE program. Thus, to draw a usable inference about D/E earnings from data gathered in the CPS one must connect a particular GE program now being offered and evaluated with earnings and occupations disclosed by the CPS respondents years, even decades, into their careers, during which they may have worked in different kinds of occupations.
Id. at 64890, 64,955-56. When the DOE rejected the entirety of the underreporting problem identified by commenters, it stated that the commenters’ most “important[]” shortcoming was the lack of an alternative.
AACS suggests - and the DOE does not directly dispute - that it is very difficult for cosmetology schools to gather their own data for use in administrative appeals. Many cosmetology schools operate in states that do not maintain state-sponsored data and the schools anticipate that they will be unable to obtain the student responses required to use institutional data. See Decl. of Mez Varol, President of International Academy (“Varol Decl.“) ¶¶ 8-9, ECF No. 8-3; Decl. of Robert
Plaintiff argues that, despite no school having been deemed ineligible for federal student loans under the HEA, the schools are already suffering harm due to the GE rules. Plaintiff‘s affiants state that the mere publication of the 2015 D/E rates has resulted in a drop in enrollment, in one case by at least 30 percent. Varol Decl. ¶ 13; Rosenberg Decl. ¶¶ 22-23. For the handful of schools that have failed the D/E ratio and must post a warning to their students, Plaintiff states that such a warning would result in “insurmountable” drops in enrollment that would force those schools to end certain programs. See Rosenberg Decl. ¶¶ 24-25 (“Based on my experience to date, the compelled ... warning ... will result in an insurmountable drop in enrollments ... [and] ultimately force the school to end its barbering course within weeks,” which is a necessary course to cover its operating expenses, and thus lead to closure of the school.); Varol Decl. ¶¶ 17-18 (reasoning that because the warning would lead to a 50% drop in enrollment and “the school needs a constant flow of new enrollees ... to remain in business,” publication of the warning would “likely lead to the closure of the school almost immediately“). At least three AACS member schools have already had to post the warnings. See Decl. of Edward M. Cramp (“Cramp Decl.“) ¶ 3, ECF No. 24-1.4 AACS specifically identified three schools - located in California, Illinois, and Texas - after the parties briefed summary judgment. Id.
IV. LEGAL STANDARD
In a typical case, the Court must grant summary judgment to a movant who
V. ANALYSIS
The DOE argues that the Court should enter summary judgment in its favor for two reasons. First, it argues that the Court does not have subject-matter jurisdiction. Def.‘s Cross-Mot. Summ. J. (“Def.‘s Mot. Summ. J.“) at 16-18, ECF No. 18-1. It contends that AACS member schools have not been subject to final agency action - a prerequisite to suit under the APA - because none of them have lost their federal student loan eligibility. Def.‘s Mot. Summ. J. at 16-18. And because AACS did not identify any school that has incurred an injury in its complaint, the DOE also argues that AACS lacks standing to sue on behalf of its members. Def.‘s Reply Supp. Def.‘s Cross-Mot. Summ. J. (“Def.‘s Reply“) at 8, ECF No. 23. Second, Defendant contends that it did not act arbitrarily or capriciously, because the DOE reasonably concluded that underreporting of income is adequately deterred by legal penalties and the SSA data is supplemented by an alternate-earning appeal process, and because the alternatives proposed during notice and comment would not allow the DOE to differentiate among programs to identify which are not producing gainfully employed graduates. AACS argues instead that it is entitled to summary judgment in its favor. See generally Pl.‘s Mot. Summ. J., ECF No. 8.5
The Court holds that AACS is entitled to summary judgment. Because the GE designations need not be appealed to be final and requiring failing schools to issue written warnings constitutes final agency action, at least some AACS member schools have been subject to final agency action. AACS has associational standing to bring claims on those schools’ behalf because it identified the particular programs that have been injured.
Moreover, the DOE acted arbitrarily and capriciously with respect to the underreporting issue identified by commenters. The DOE openly acknowledged that underreporting is an issue, even identifying cosmetology schools by name. Commenters pointed to data that seem to show the issue is significant. In response to public comments, the DOE addressed the underreporting issue by noting that underreporters are subject to civil and criminal penalties and that programs may seek to have separate earnings data used
But the Court goes no further. The DOE adequately explained its decision to reject the alternatives presented during the notice-and-comment period: it stated that the alternative data was too disconnected from particular programs to allow the DOE to identify which programs are underperforming. The explanation that the DOE provided sufficed to show that it was not acting arbitrarily or capriciously. Given that no better data or methodology was presented during notice-and-comment and AACS has failed to show that it is DOE‘s responsibility to come up with a better formula for each distinct type of program, it is incumbent on each school to provide better data or methodology - if it exists - during the appeal process.
A. Subject-Matter Jurisdiction
The DOE first argues that there has been no “final” agency action over which the Court has jurisdiction for an as-applied challenge. See Def. Mot. Summ. J. at 16-18. It contends that the earliest a program could be deemed ineligible for federal loan guarantees would be 2018, and that schools could still submit an alternate earnings appeal. Def.‘s Mot. at 17-18. Without a “final agency action” having been taken against any AACS member school, the DOE argues that AACS does not have associational standing to pursue this case. See Def.‘s Mot. at 17 (“Plaintiff here points to no [final] action taken against it or any of its member cosmetology schools.“); Def.‘s Reply at 8. However, the DOE concedes that the schools that have already had to post warnings have been subject to final agency action. See Tr. at 40-42, 50. The Court will first analyze whether the DOE has taken final agency action, and then turn to whether AACS has standing to maintain this action. Because requiring schools to post warnings constitutes final agency action and some schools have already posted those warnings, the Court concludes that final agency action has been taken against, at least, three member-schools, and the AACS has associational standing to pursue claims on these schools’ behalf.
1. AACS Challenges Final Agency Action
The DOE argues that AACS has not identified a final agency action that DOE has applied to AACS or a member-school. Def.‘s Mot. Summ. J. at 16-17. In support of this argument, DOE notes that “no program can become ineligible for Title IV funding because no such decision has been issued,” and that member-schools can submit alternate earnings appeals to avert losing federal funding. Def.‘s Mot. Summ. J. at 17-18. As noted above, the DOE concedes that the schools that have already had to post warnings have been subject to final agency action. See Tr. at 40-42, 50.
Plaintiff seeks judicial review of the DOE‘s GE regulations pursuant to the APA. Compl. ¶ 53. The APA provides for judicial review of “final agency action.”
In Dole v. United Steelworkers of America, the Supreme Court analyzed OSHA disclosure rules, which the Court characterized as rules that “protect by providing access to information about what dangers exist and how these dangers can be avoided.” Id. at 28. In deciding what constitutes final agency action, the Court said that
[t]he promulgation of a disclosure rule is a final agency action that represents a substantive regulatory choice. An agency charged with protecting employees from hazardous chemicals has a variety of regulatory weapons from which to choose: It can ban the chemical altogether; it can mandate specified safety measures, such as gloves or goggles; or it can require labels or other warnings alerting users to dangers and recommended precautions. An agency chooses to impose a warning requirement because it believes that such a requirement is the least intrusive measure that will sufficiently protect the public, not because the measure is a means of acquiring information useful in performing some other agency function.
Consistent with the reasoning under Dole, the warning DOE regulations require for schools failing the GE rules constitutes final agency action once they are triggered. The requirement for schools to post warnings was not an interlocutory measure aimed at acquiring information for a later agency action, but rather a policy implemented with the aim of protecting students. See 79 Fed. Reg. at 64,901 (“[T]he [warning] requirements ... balance the need to provide prospective students with critical information at a time when they can most benefit from it with ensuring that the administrative burden for institutions is not unnecessarily increased.“). It was the consummation of the DOE‘s decisionmaking process with respect to schools that failed under the GE rules and did not file appeals, thus triggering the regulations’ warning obligations. Thus, at least some of the AACS schools have been subject to this final agency action for two reasons. First - and most obviously - at least three of AACS‘s member-schools have already posted the required warnings. See Cramp Decl. ¶ 3. Indeed, the government admitted as much at oral argument, and conceded that those schools
2. AACS Has Associational Standing
Because the Court has a sua sponte obligation to ascertain that it has subject-matter jurisdiction and Defendant mentioned the issue in its reply, the Court next analyzes whether AACS has standing to bring claims on behalf of any of its member schools that has suffered an injury as a result of the GE regulations. The DOE asserts that AACS lacks standing because it was required to identify, by name, at least one member-school who actually had to post the warnings at the time of plead-
“An organization can have standing on its own behalf ... or on behalf of its members.” Abigail All. for Better Access to Developmental Drugs v. Eschenbach, 469 F.3d 129, 132 (D.C. Cir. 2006). Standing based on an organization‘s own injury - “organizational standing” - requires an organization, “like an individual plaintiff, to show actual or threatened injury in fact that is fairly traceable to the alleged illegal action and likely to be redressed by a favorable court decision.” Equal Rights Ctr. v. Post Properties, Inc., 633 F.3d 1136, 1138 (D.C. Cir. 2011). “Associational standing,” in comparison, allows an organization to sue on behalf of its members to protect their interests. Common Purpose USA, Inc. v. Obama, 227 F. Supp. 3d 21, 26-27 (D.D.C. 2016).
For an organization to sue on behalf of its members through associational standing, it must show that (1) “its members would otherwise have standing to sue in their own right,” (2) “the interests it seeks to protect are germane to the organization‘s purpose,” and (3) “neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.” United Food & Commercial Workers Union Local 751 v. Brown Grp., Inc., 517 U.S. 544, 553 (1996) (quoting Hunt v. Wash. State Apple Advert. Comm‘n, 432 U.S. 333, 343 (1977)). The third requirement “ha[s] been understood to preclude associational standing when an organization seeks damages on behalf of its members.” Id. at 554. Thus, when an association seeks prospective relief on behalf of its members, actual participation by individual members is generally not required. Warth v. Seldin, 422 U.S. 490, 515 (1975) (“If in a proper case the association seeks a declaration, injunction, or some other form of prospective relief, it can reasonably be supposed that the remedy, if granted, will inure to the benefit of those members of the association actually injured.“). But see Ctr. for Auto Safety v. Nat‘l Highway Traffic Safety Admin., 793 F.2d 1322, 1329 n.44 (D.C. Cir. 1986) (“Courts have required individual participation in circumstances where there are conflicts of interest within the organization or when a specific factual setting is needed to illuminate the issues.“).
Courts are somewhat split on whether an organization must identify affected members by name to meet the first requirement - that members would otherwise have standing to sue in their own
Cases in this district suggest that a plaintiff-association must identify injured parties at the pleading stage, not necessarily that they be named in the complaint. In Western Wood Preservers Institute v. McHugh, the court, ruling on a motion to reconsider, merely identified that courts had divergent views on whether an individual member needed to be identified by name at the pleading stage. 292 F.R.D. 145, 148 (D.D.C. 2013). Rather than state that any court was correct on the issue, the court simply concluded that the existence of diverging viewpoints did not make the court‘s underlying decision “clearly erroneous.” Id. In the two cases that the Western Wood Preservers Institute court identified, the court did not squarely address the issue of whether an individual member must be named, so long as it is identified. In Californians for Renewable Energy v. Department of Energy, the court analyzed the injury of the only named associational member because “it is ‘not enough to aver that unidentified members have been injured.‘” 860 F. Supp. 2d 44, 48 (D.D.C. 2012) (quoting Chamber of Commerce v. EPA, 642 F.3d 192, 199 (D.C. Cir. 2011)). To conclude that an individual member must be named to ensure that it is identified is unwarranted. In the second case cited by the court in Western Wood Preservers Institute, the court merely held that because the plaintiffs did not respond to the defendants’ argument that an organization must “specifically identif[y] ... members who suffered the requisite harm,” the plaintiffs conceded they did not have associational standing. Common Cause v. Biden, 909 F. Supp. 2d 9, 21 n.6 (D.D.C. 2012), aff‘d on other grounds, 748 F.3d 1280 (D.C. Cir. 2014). These cases do not require a plaintiff to specifically name an injured member in the complaint so long as it is identified. The DOE did not take a contrary position at oral argument once it became clear that AACS would name specific members in a supplemental filing. Tr. at 41-43.
AACS meets all three requirements for associational standing. AACS has shown that at least some of its members would have standing to sue in their own right. AACS identified certain member-schools that did not appeal the final GE rates, and thus were required by the regulations to issue the warnings. Civitano Decl. ¶ 32 (stating that “AACS member schools who cannot appeal ... and who did not file a notice of appeal ... [were] required to deliver the warning ... no later than February 8, 2017“). And even assuming that AACS was required to identify specific schools by name at the summary judgment stage, it did so, naming three schools who have already had to post the warnings. Cramp Decl. ¶ 3.
Second, this lawsuit is germane to the purpose of AACS. AACS is a nonprofit association representing cosmetology programs, most of whom are subject to the GE rules promulgated by the DOE. Compl. ¶ 1; Civitano Decl. ¶ 6. Ensuring that member schools maintain Title IV federal loan assistance eligibility is in line with the purpose of the group, which is to represent the interests of its member cosmetology schools.
Third, this case does not require the participation of individual member-schools. AACS seeks injunctive and declaratory relief on behalf of the member-schools, not damages. See Compl. at 10. Thus, as a general rule, individual participation is not necessary. See Warth, 422 U.S. at 515. The DOE has not identified
B. Arbitrary and Capricious
Finally, having run the gauntlet of the DOE‘s threshold arguments, the Court now reaches the merits of AACS‘s challenge to the GE regulations. AACS argues that the DOE acted arbitrarily and capriciously in disregarding the widespread underreporting of income in the cosmetology industry. See Pl.‘s Mot. Summ. J. at 12. AACS specifically contends that the DOE (1) knowingly used income data that was faulty and that would lead to seriously flawed results, despite commenters’ concerns, and (2) did not consider credible alternatives. See Pl.‘s Mot. Summ. J. at 12. In support of its first argument, AACS asserts that not only is the SSA data incorrect, but that the DOE actually acknowledged that it failed to take underreporting into account in response to commenters. See Pl.‘s Mot. Summ. J. at 17-19. Plaintiff further asserts that the DOE failed to adequately explain why it did not account for underreporting, because civil and criminal penalties are irrelevant to the issues of gainful employment. See Pl.‘s Mot. Summ. J. at 19-20. To support its argument that viable alternatives existed, Plaintiff cites to proposals by commenters during notice and comment, and argues that the DOE could have conducted a survey to estimate unreported income levels in the industry and adjusted each school‘s graduates’ earnings accordingly. See Pl.‘s Mot. Summ. J. at 8-9. The Court will analyze each of these arguments in turn, but first addresses the DOE‘s argument that the availability of appeal gives its regulations more leeway.
1. The Court Analyzes Whether the Entire Scheme is Arbitrary or Capricious
The DOE argues that the Court should analyze whether the adoption of the entire D/E-calculation scheme was arbitrary and capricious as applied to AACS member schools, as opposed to only focusing on the DOE‘s presumptive use of SSA data. Def.‘s Mot. Summ. J. at 8-9, 17-18; Def.‘s Reply at 4-5. AACS argues that the Court must look at each piece separately, meaning that the alternate-earnings appeal process has no bearing on whether the DOE‘s decision to presumptively use SSA data was arbitrary and capricious. The Court thus addresses, as a threshold matter, whether the existence of the alternate-earnings appeal process alters the Court‘s standard of review with respect to the DOE‘s use of SSA data. See Connors v. Cedar Coal Co., No. 90-cv-0260, 1991 WL 102640, at *2 (D.D.C. June 4, 1991) (“As a threshold matter the Court must determine the appropriate standard of review to apply ....“).
In general, the court must look to an agency‘s final rule as a whole to determine whether the agency gave substantial consideration to issues raised during the notice-and-comment period. See Chamber of Commerce of United States of Am. v. NLRB, 118 F. Supp. 3d 171, 217-18 (D.D.C. 2015). One cannot assert that a particular portion of a regulation is arbitrary or capricious because it does not address a shortcoming when another part of the regulation squarely addresses it. See id.
The sole authority that AACS cites for the proposition that the Court should consider the DOE‘s use of SSA data in a
Later, in American Iron and Steel Institute v. EPA, the D.C. Circuit analyzed EPA limits on water pollutants promulgated under the Clean Water Act. 115 F.3d 979, 985 (D.C. Cir. 1997). To discharge pollutants into the Great Lakes, the Clean Water Act requires entities to have permits, each of which comes with specific discharge limits based on, among other things, the quality of water and the levels of pollution-control technology. Id. at 990. To formulate specific discharge limits for entities, the EPA used two different methodologies, known as “Tier I” and “Tier II.” Id. at 991. The difference between Tier I and Tier II is the amount of toxicological data they required to calculate discharge limits. Id. Tier I data required significantly more toxicological data than Tier II. Tier II was used as a fallback methodology when insufficient toxicology data was available. Id. at 991 n.5. Tier II also resulted in more stringent toxicology limits, making compliance more difficult for certain regulated entities. Id. at 993. The plaintiff, the American Iron and Steel Institute, whose limits were determined by Tier II, complained that the regulations unfairly subjected it to more stringent requirements than it did to entities in situations where more toxicological data was available. See id. The D.C. Circuit upheld the tiered approach because the methodologies were based on the best-available data in the face of imperfect data, and differing compliance burdens on industry did not automatically make something arbitrary or capricious. Id. Distinguishing Leather Industries of America, Inc., the circuit held that “[t]he Tier II methodology, by scaling the uncertainty factors to reflect existing data, properly correlates risk with knowledge.” Id.
These cases are no doubt similar. The major issue that the D.C. Circuit found with the “optional” regulatory caps in Leather Industries was that meeting them excused certain regulated entities from significant legal obligations. 40 F.3d at 399. Thus, if the caps were arbitrary, entities would be subjected to burdensome regulations randomly based on the types of pollutants they produce, while others would avoid the scheme altogether. Id. In comparison, in American Iron and Steel Institute, the differing standards were created based on the relative availability of data. 115 F.3d at 993. Although entities able to use the Tier I methodology were ultimately subjected to laxer standards, those standards were the result of having more accurate data available. See id. It was not as if those subjected to Tier II standards were forced to adhere to more stringent standards separate from the overall regulatory scheme; the EPA decided to “err on the side of overprotection” in the face of un-
There is no reason for the Court to analyze the DOE‘s use of SSA data in a vacuum. As a general matter, the Court looks to the entire regulatory scheme to determine whether an agency acted arbitrarily and capriciously. Chamber of Commerce of United States of Am., 118 F. Supp. 3d at 217-18. Leather Industries, the sole case relied upon by AACS, does not affect the Court‘s analysis. The DOE‘s GE regulations do not exempt programs from the regulatory scheme altogether if they happen to have a certain level of reported income. See
2. The DOE Acted Arbitrarily and Capriciously with Respect to the Problem of Underreporting
AACS argues that the DOE uses “income data that is manifestly bad in the context of the application of the GE Rule” to its member-schools. Pl.‘s Mot. at 17; see also Compl. ¶ 1. Plaintiff notes that the DOE even acknowledged that the data was flawed when it stated that it was aware that “some self-employed individuals may fail to report, or underreport, their earnings,” and that experts submitted evidence that tip income and self-employment income are underreported by as much as 60% in, among others, the cosmetology industry. Pl.‘s Mot. at 18-19. AACS further argues that the DOE‘s justification for accepting the flaws in the data - that there are civil and criminal penalties for failing to report taxable income - is not reasoned, and has no rational connection to the issues raised by commenters. Pl.‘s Mot. at 19-20. DOE responds that its justification regarding civil and criminal penalties was rational and that programs can seek a different metric through an alternate-earnings appeal. Def.‘s Mot. Summ. J. at 21-23.
Under
An agency action is arbitrary and capricious if the agency fails to consider all of the relevant factors in reaching its decision. See N.Y. Cross Harbor R.R. v. Surface Transp. Bd., 374 F.3d 1177, 1181 (D.C. Cir. 2004). An agency rule is also arbitrary and capricious if the agency “entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before [it], or [the explanation] is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” State Farm, 463 U.S. at 43. Thus, an “agency must examine the relevant data and articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” Int‘l Ladies’ Garment Workers’ Union v. Donovan, 722 F.2d 795, 814 (D.C. Cir. 1983) (quotation marks and citation omitted). An agency also must give a reasoned response explaining how it resolved any significant problems raised during notice and comment. Id. at 818. When an agency‘s reasoning involves a non-obvious, essential factual assumption, the agency must justify that assumption “notwithstanding a party‘s failure to challenge [it], as part of its affirmative duty to engage in rational decisionmaking.” Am. Mar. Ass‘n v. United States, 766 F.2d 545, 566 n.30 (D.C. Cir. 1985).
With that said, “[a]n agency has discretion to design rules that can be broadly applied, sacrificing some measure of ‘fit’ for administrability,” but “must justify its failure to take account of circumstances that appear to warrant different treatment for different parties.” Leather Industries of America, Inc., 40 F.3d at 403 (quoting Petroleum Commc‘ns v. FCC, 22 F.3d 1164, 1172 (D.C. Cir. 1994)); see also Am. Pub. Gas Ass‘n v. Fed. Power Comm‘n, 567 F.2d 1016, 1046 (D.C. Cir. 1977) (“Courts cannot fairly demand the perfect at the expense of the achievable.“). Shortcomings and analytical weak points, if justified or explained in light of practical constraints, do not render a regulation arbitrary or capricious. See City of Brookings Mun. Tel. Co. v. FCC, 822 F.2d 1153, 1168 (D.C. Cir. 1987). Indeed, even the “best available data standard leaves room for error, so long as more reliable data did not exist at the time of the agency decision.” Baystate Med. Ctr. v. Leavitt, 545 F. Supp. 2d 20, 49 (D.D.C. 2008), amended in part, 587 F. Supp. 2d 37 (D.D.C. 2008), judgment entered, 587 F. Supp. 2d 44 (D.D.C. 2008).
But wooden use of this SSA data is problematic. As commenters noted during the notice-and-comment period, graduates of certain programs - including cosmetology programs - underreport their income with far greater frequency than graduates of other programs. J.A. 797; see also Civitano Decl. ¶ 28. One commenter, citing a Stanford study, noted that both tip and self-employment income are underreported by around 60%. J.A. 595, 787; see also Civitano Decl. ¶ 28; Brodie Decl. at 493. The DOE explicitly acknowledged this problem, noting that tips make up “about half of earnings in service occupations such as cosmetology,” and that many people in the industry are self-employed. See 79 Fed. Reg. at 64,955. So, although SSA data is highly accurate in some respects and is thus a useful starting point for determining average earnings, it has significant shortcomings - at least for certain occupations.
Pursuant to its obligations to respond to significant problems raised during the notice-and-comment period, see Int‘l Ladies’ Garment Workers’ Union, 722 F.2d at 818, the DOE dealt with the underreporting issue in two ways. First, it relied on the civil and criminal penalties associated with failing to report income. 79 Fed. Reg. at 64,955-56. Second, it noted that SSA data is only one of three possible sources of earnings data; programs could also appeal for the use of state-sponsored data or survey-based data.
The first justification does not grapple with the issue in any meaningful way. The DOE obviously did not demonstrate that the legal disincentives for failing to report meant that the commenters’ data was incorrect; the DOE itself acknowledged that the problem existed. See 79 Fed. Reg. at 64,955. There is also no basis for the inference that the issue of underreporting would somehow subside under the new GE regulations. Civil and criminal penalties existed before and after the regulations took effect, and the DOE regulations did not add any additional deterrent for underreporting. To the extent the DOE believed that the penalties somehow would correct for underreporting, it was required to explain this assumption explicitly. See
The DOE‘s second response more directly addresses the commenters’ concerns; but nonetheless comes up short. Although the commenters identified the issue of underreporting, they did not provide a better data set that the DOE could use to measure overall income, nor did they suggest a specific adjustment factor that the DOE could use for the various programs that would inevitably claim underreporting as a problem with their earnings data. APSCU II, 110 F. Supp. 3d at 195; 79 Fed. Reg. at 64,955-56. Nor did the commenters propose an alternative calculus to balance fit and administrability.
The DOE‘s attempt to address these many issues came in the form of an alternate earnings appeal process, which the DOE cited as its other justification for its use of SSA data in the face of the underreporting problem. 79 Fed. Reg. at 64,956. On the surface, this justification seems reasonable. Assuming that every program is capable of mounting an appeal, the data would yield perfectly tailored D/E rates for individual programs regardless of their levels of underreporting and the unique characteristics of the occupations in which their particular graduates are employed. But that assumption is the fly in the DOE‘s reasoned-decisionmaking ointment. As noted above, the DOE “must justify the assumptions essential to its actions, notwithstanding a party‘s failure to challenge those assumptions before the agency, as part of its affirmative duty to engage in rational decisionmaking.” Am. Mar. Ass‘n, 766 F.2d at 567 n.30. The DOE assumed that programs would be able to effectively use the alternate-earnings appeal process. See generally 79 Fed. Reg. 64,890. It also assumed that the required response rates - 50 percent for state-sponsored data systems and nearly 100 percent for school-specific surveys - had some statistical significance. See 79 Fed. Reg. at 64,995-96. But the DOE did not explain these assumptions.7 See 79 Fed. Reg. at 64,995-96;
3. The DOE Did Not Act Arbitrarily or Capriciously in Rejecting the Alternative Identified by AACS
AACS further contends that the DOE failed to adequately consider reasonable alternatives presented during the notice-and-comment period. See Pl.‘s Mot. Summ. J. at 20. It specifically points to Dr. Bettinger‘s suggestion that the DOE could use average-income figures from the Bureau of Labor Statistics, which derives its data from surveys, rather than from income reported to the IRS. The DOE responds that it adequately justified its re-
The arbitrary-and-capricious standard described above also applies to an agency‘s consideration of regulatory alternatives. See Pillai v. Civil Aeronautics Bd., 485 F.2d 1018, 1027 (D.C. Cir. 1973) (finding the “artificial narrowing of options to be arbitrary and capricious“); accord Int‘l Ladies’ Garment Workers’ Union, 722 F.2d at 817. This does not mean that an agency must consider all possible policy alternatives in reaching its decision. State Farm, 463 U.S. at 51. But the agency must address obviously germane alternatives proposed by commenters during the notice-and-comment period. See Int‘l Ladies’ Garment Workers’ Union, 722 F.2d at 817-18. The agency must give rational consideration to each of these alternatives and an “adequate explanation” for each rejection. See id. (citing Action on Smoking and Health v. Civil Aeronautics Bd., 699 F.2d 1209 (D.C. Cir. 1983), supplemented by 713 F.2d 795 (D.C. Cir. 1983)).
The DOE adequately explained why it did not use the Bureau of Labor Statistics data as opposed to SSA data. Indeed, a court in this district has already held that to be the case. APSCU II, 110 F. Supp. 3d
Plaintiffs suggest that this problem could be alleviated by adding an adjustment factor to the SSA data for each school. Pl.‘s Mot. at 9-10. Thus, schools with graduates who have lower reported income will have proportionately low levels of unreported income added to their numbers. Pl.‘s Mot. at 9-10. As the DOE concluded in its response to notice and comment, this methodology does not necessarily solve the problem. This methodology assumes that each program‘s graduates are employed in the industry for which they are trained at proportionate rates. See 79 Fed. Reg. at 64,955-56. Thus, if a cosmetology program‘s graduates cannot find work as cosmetologists and therefore must work in other fields with higher rates of income reporting, that program would still benefit from the average rate of underreporting in the cosmetology industry if the adjustment-factor methodology is used. See
Whether the DOE‘s decision to reject the Bureau of Labor Statistics approach suggested by Dr. Bettinger was correct is not at issue here. The Court merely asks whether the decision was reasoned. Because it was rational, based on a consideration of the need to individualize its review of schools, and its reasoning was explained, the Court finds that the DOE did not act arbitrarily and capriciously with respect to the alternative identified by AACS.
C. Remedy
Given that the DOE acted arbitrarily and capriciously with respect to the problem of underreporting, the Court now must fashion a remedy. In general, courts narrowly tailor remedies to APA violations. Nat‘l Treasury Emps. Union v. Chertoff, 394 F. Supp. 2d 137, 145 (D.D.C. 2005), rev‘d in part on other grounds, 452 F.3d 839 (D.C. Cir. 2006) (citing Nat‘l Treasury Emps. Union v. Yeutter, 918 F.2d 968, 977 (D.C. Cir. 1990)). Thus, when possible, a court should strike only offending portions of regulations. See id.; Gulf Oil Corp. v. Brock, 778 F.2d 834, 842 (D.C. Cir. 1985).
To remedy the DOE‘s arbitrarily and capriciously rigid use of SSA data to calculate D/E rates, the Court orders that the DOE may not enforce the numerical survey requirements currently in effect for alternate earnings appeals against AACS member schools. In effect, this removes the arbitrary and capricious reasoning behind the otherwise-valid premise that alternate earnings appeals justify the presumptive use of SSA data. It also avoids upending the entire GE regulatory scheme and means that AACS member institutions need not secure any specific amount of survey responses or state-sponsored data
VI. CONCLUSION
For the foregoing reasons, the Court grants AACS‘s motion for summary judgment, denies the DOE‘s motion for summary judgment, and denies AACS‘s motion for a preliminary injunction. An order consistent with this Memorandum Opinion is separately and contemporaneously issued.
Abraham Ron FRAENKEL, et al., Plaintiffs,
v.
ISLAMIC REPUBLIC OF IRAN, et al., Defendants.
Civil Action No. 15-1080 (RMC)
United States District Court, District of Columbia.
Signed 06/28/2017
RUDOLPH CONTRERAS
UNITED STATES DISTRICT JUDGE