From Casetext: Smarter Legal Research

New York v. Scalia

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK
Sep 8, 2020
1:20-cv-1689-GHW (S.D.N.Y. Sep. 8, 2020)

Opinion

1:20-cv-1689-GHW

09-08-2020

STATE OF NEW YORK, COMMONWEALTH OF PENNSYLVANIA, STATE OF CALIFORNIA, STATE OF COLORADO, STATE OF DELAWARE, DISTRICT OF COLUMBIA, STATE OF ILLINOIS, STATE OF MARYLAND, COMMONWEALTH OF MASSACHUSETTS, STATE OF MICHIGAN, STATE OF MINNESOTA, STATE OF NEW JERSEY, STATE OF NEW MEXICO, STATE OF OREGON, STATE OF RHODE ISLAND, STATE OF WASHINGTON, STATE OF VERMONT, and COMMONWEALTH OF VIRGINIA, Plaintiffs, v. EUGENE SCALIA, Secretary of the United States Department Of Labor, UNITED STATES DEPARTMENT OF LABOR, and UNITED STATES OF AMERICA, Defendants, -and- INTERNATIONAL FRANCHISE ASSOCIATION, THE CHAMBER OF COMMERCE OF THE UNITED STATES OF AMERICA, HR POLICY ASSOCIATION, NATIONAL RETAIL FEDERATION, ASSOCIATED BUILDERS AND CONTRACTORS, and AMERICAN LODGING AND HOTEL ASSOCIATION, Intervenor-Defendants.


MEMORANDUM OPINION AND ORDER :

For more than eighty years, the Department of Labor (the "Department") has recognized that multiple employers may qualify as "joint employers" under the Fair Labor Standards Act (the "FLSA"). Suppose, for instance, that an employee works for a subcontractor and that a primary contractor hires the subcontractor. If both the primary contractor and the subcontractor meet the FLSA's definition of an "employer," they are joint employers. Joint employers are jointly and severally liable for damages for FLSA violations.

Earlier this year, the Department issued a final rule (the "Final Rule") that narrows the definition of joint employment under the FLSA. Eighteen States (the "States") sued, arguing that the Final Rule is invalid.

The Final Rule violates the Administrative Procedure Act (the "APA"). It conflicts with the FLSA because it ignores the statute's broad definitions. And the Department failed to adequately justify its departure from its prior interpretations and to account for some of the Final Rule's important costs. So the Final Rule is also arbitrary and capricious. But one part of the Final Rule is severable from the portions that are legally infirm. For those and other reasons discussed below, the parties' cross motions for summary judgment are GRANTED in part and DENIED in part.

I. BACKGROUND

A. Statutory and Regulatory Background

The Court's prior opinion provides additional background. See New York v. Scalia (Scalia I), No. 1:20-cv-1689 (GHW), 2020 WL 2857207, at *1-4 (S.D.N.Y. June 1, 2020).

1. The FLSA

"The principal congressional purpose in enacting the FLSA was to protect all covered workers from substandard wages and oppressive working hours[.]" Scalia I, 2020 WL 2857207, at *1 (quoting Mei Xing Yu v. Hasaki Rest., Inc., 944 F.3d 395, 402 (2d Cir. 2019)). "Consistent with [that] 'remedial and humanitarian' purpose, Congress adopted definitions of 'employ,' 'employee,' and 'employer' that brought a broad swath of workers within the statute's protection." Id. (quoting Salinas v. Com. Interiors, Inc., 848 F.3d 125, 133 (4th Cir. 2017)). "The FLSA defines an 'employee' as 'any individual employed by an employer,'" an "'employer' to include 'any person acting directly or indirectly in the interest of an employer in relation to an employee[,]'" and "the term 'employ'" to include "'to suffer or permit to work.'" Id. (quoting 29 U.S.C. §§ 203(e)(1), 203(d), 203(g)).

These definitions are broad. In 1945, the Supreme Court noted that "the term 'employee'" in the FLSA has "'the broadest definition . . . ever . . . included in any one act.'" United States v. Rosenwasser, 323 U.S. 360, 363 n.3 (1945) (quoting 81 Cong. Rec. 7657 (1937) (statement of Sen. Hugo Black)). Two years later, the Court observed that the FLSA's "definition of 'employ' is broad." Rutherford Food Corp. v. McComb, 331 U.S. 722, 728 (1947). Indeed, the FLSA's definitions are so "comprehensive" that they apply "to many persons and working relationships" that did not historically "fall within an employer-employee category." Walling v. Portland Terminal Co., 330 U.S. 148, 150-51 (1947) (citation omitted). Decades later, the Court again noted the "striking breadth" of the FLSA's definition of "employ[.]" Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 326 (1992). That definition "stretches the meaning of 'employee' to cover some parties who might not qualify as such" under "traditional agency law principles." Id.

The test for an employment relationship under the FLSA rests on "economic reality." In Rutherford, the Supreme Court observed that there is "no definition" that precisely delimits the scope "of the employer-employee relationship under the [FLSA]." 331 U.S. at 728. Whether an employment relationship exists "does not depend on . . . isolated factors but rather upon the circumstances of the whole activity." Id. at 730. The Court later distilled Rutherford into an "economic reality" test: "[T]he 'economic reality' rather than 'technical concepts'" determines employment under the FLSA. Goldberg v. Whitaker House Co-op., Inc., 366 U.S. 28, 33 (1961) (citing United States v. Silk, 331 U.S. 704, 713 (1947); Rutherford, 331 U.S. at 729).

2. Joint Employer Doctrine

The joint employer doctrine is longstanding. The Department has recognized joint employment since 1939. Scalia I, 2020 WL 2857207, at *2. That year, the Department issued an "interpretative bulletin" establishing that multiple employers could simultaneously employ an employee. See id. (citing Interpretative Bulletin No. 13, "Hours Worked: Determination of Hours for Which Employees are Entitled to Compensation under the Fair Labor Standards Act of 1938," at 16-17 (U.S. Dep't of Labor July 1939)).

"In 1958, the Department first codified the joint employment standard." Id. (citing 23 Fed. Reg. 5905 (Aug. 5, 1958)). Like the 1939 bulletin, the Department's 1958 regulations "recognized that 'a single individual may'" simultaneously have "'two or more employers'" under the FLSA. Id. at *3 (quoting Salinas, 848 F.3d at 133, in turn quoting former 29 C.F.R. § 791.2(a)) (alterations omitted). "The Department's 1958 regulations distinguished 'separate and distinct employment' and 'joint employment.'" Id. (quoting Salinas, 848 F.3d at 133, in turn quoting former 29 C.F.R. § 791.2(a)) (brackets omitted). "[J]oint employment exists when 'the facts establish that employment by one employer is not completely disassociated from employment by the other employer.'" Id. (quoting Salinas, 848 F.3d at 133, in turn quoting former 29 C.F.R. § 791.2(a)).

The Supreme Court first explicitly recognized joint employer liability in Falk v. Brennan, 414 U.S. 190 (1973). Given the FLSA's "expansive[] . . . definition of 'employer[,]'" the Court recognized that an employee could have multiple employers for a single set of hours worked. Id. at 195; see, e.g., Bonnette v. California Health & Welfare Agency, 704 F.2d 1465, 1469 (9th Cir. 1983) (citing Falk for the proposition that "[t]wo or more employers may jointly employ someone for purposes of the FLSA").

Congress enacted the Migrant and Seasonal Agricultural Workers Protection Act (the "MSPA") in 1983. See 29 U.S.C. §§ 1801 et seq. The MSPA "uses the same definition of 'employ' as the FLSA." Scalia I, 2020 WL 2857207, at *3 n.2 (quoting Salinas, 848 F.3d at 135); see also 29 U.S.C. § 1802(5) ("The term 'employ' has the meaning given such term under section 3(g) of the [FLSA] (29 U.S.C. [§] 203(g))[.]"). The MSPA does not define "employer."

The Department issued regulations defining joint employment under the MSPA (the "MSPA Regulations"). See 29 C.F.R. § 500.20(h). The regulations adopted the FLSA's definitions of employ, employer, and employee. See id. § 500.20(h)(1)-(3). Thus, "employ" under the MSPA Regulations "includes to suffer or permit to work." Id. § 500.20(h)(1). The Department explained that "Congress[] incorporat[ed]" the term "employ" into the MSPA from the FLSA to "adopt[] the FLSA joint employer doctrine as the 'central foundation' of MSPA and 'the best means'" to fulfill its purposes. Id. § 500.20(h)(5)(ii) (quoting legislative history of the MSPA). In sum, "[j]oint employment under the [FLSA] is joint employment under the MSPA." Id. § 500.20(h)(5)(i).

The Department issued new guidance to clarify the "definition of 'joint employment' under the MSPA" in 1997 (the "1997 Guidance"). 62 Fed. Reg. 11734, 11734 (Mar. 12, 1997). The 1997 Guidance observed that "the concept of 'joint employment'" in the MSPA depends on "[t]he MSPA statutory definition of 'employ[,]'" which is also "the FLSA statutory definition of 'employ[.]'" Id. Courts have "expansive[ly] interpret[ed] . . . the statutory definition of employ under the FLSA[.]" Id. And courts "interpreting the FLSA definition of employ" have rejected "the traditional common law 'right to control' test[.]" Id. The 1997 Guidance noted that the "test of an employment relationship under the FLSA is 'economic dependence,' which requires" a court to examine the "relationships among the employee(s) and the putative employer(s) to determine upon whom the employee is economically dependent." Id.

In 2014, the Department issued an Administrator's Interpretation (the "2014 AI"), which discussed joint employment under the FLSA. Administrator's Interpretation No. 2014-2, "Joint employment of home care workers in consumer-directed, Medicaid-funded programs by public entities under the Fair Labor Standards Act," 2014 WL 2816951 (June 19, 2014). The 2014 AI explained that the definitions of "employee," "employ," and "employer" in the FLSA "are exceedingly broad." Id. at *2. To determine "whether an employer-employee relationship exists" under the FLSA, a court must look not to "'isolated factors but rather [to] the circumstances of the whole activity[.]'" Id. (quoting Rutherford, 331 U.S. at 730). "[T]he touchstone is 'economic reality.'" Id. (quoting Whitaker, 366 U.S. at 33). "[T]he 'economic realities' test examines [multiple] factors to determine whether a worker . . . economically depend[s] on a purported employer, thus creating an employment relationship." Id.

The Department again rejected a test that addressed only "the potential joint employer's control" over employees. Id. at *2 n.5. A test that "focuses solely on the formal right to control the physical performance of another's work" is "unduly narrow[.]" Id. (quoting Zheng v. Liberty Apparel Co., 355 F.3d 61, 69 (2d Cir. 2003)). And a control-based test conflicts "with the 'suffer or permit' language in the FLSA, which . . . reaches beyond traditional agency law." Id. (quoting Zheng, 355 F.3d at 69). The 2014 AI concluded that a test that "addresses only control" conflicts "with the breadth of employment under the FLSA." Id. The Final Rule rescinds the 2014 AI. See 29 C.F.R. § 791.1 (rescinding "prior administrative rulings, interpretations, practices, or enforcement policies [about] joint employer status" that "conflict with" the Final Rule).

The Department issued another Administrator's Interpretation about joint employer liability in 2016 (the "2016 AI"). Administrator's Interpretation No. 2016-1, "Joint employment under the Fair Labor Standards Act and Migrant and Seasonal Agricultural Worker Protection Act," 2016 WL 284582 (Jan. 20, 2016). The Department again explained that "[t]he concepts of employment and joint employment under the FLSA and MSPA are notably broader than the common law concepts of employment and joint employment, which look to the amount of control that an employer exercises over an employee." Id. at *3 (citing Antenor v. D & S Farms, 88 F.3d 925, 933 (11th Cir. 1996)). That is because "the 'suffer or permit' standard broadens the scope of employment relationships covered by the FLSA." Id. (citing Walling, 330 U.S. at 150-51; Darden, 503 U.S. at 326). And the MSPA and the FLSA "define[] 'employ'" identically, so "the scope of employment relationships under MSPA is" identical to the FLSA. Id. at *2 (citing 29 U.S.C. § 1802(5); 29 C.F.R. § 500.20(h)(1)-(3)).

The 2016 AI distinguished between "vertical" and "horizontal" joint employment. See id. at *2. "Horizontal joint employment exists where the employee has employment relationships with two or more employers and the employers are sufficiently associated . . . with respect to the employee [so] that they jointly employ the employee." Id.

Vertical joint employment exists where the employee has an employment relationship with one employer (typically a staffing agency, subcontractor, labor provider, or other intermediary employer) and the economic realities show that he or she is economically dependent on, and thus employed by, another entity involved in the work. This other employer, who typically contracts with the intermediary employer to receive the benefit of the employee's labor, would be the potential joint employer. Where there is potential vertical joint employment, the analysis focuses on the economic realities of the working relationship between the employee and the potential joint employer.

The Department reiterated that "the Supreme Court and the Circuit Courts of Appeals apply an economic realities analysis to determine" whether an employment relationship exists under the FLSA and MSPA. Id. at *9 (citing, among others, 2014 AI; Tony & Susan Alamo Found. v. Sec'y of Labor, 471 U.S. 290, 301 (1985); Whitaker, 366 U.S. at 33). This test "is not a control test." Id. The Department noted that "[s]ome courts . . . apply factors that address only or primarily the potential joint employer's control (power to hire and fire, supervision and control of conditions or work schedules, determination of rate and method of pay, and maintenance of employment records)." Id. at *11 (citing Baystate Alt. Staffing Inc. v. Herman, 163 F.3d 668, 675 (1st Cir. 1998); In re Enter. Rent-A-Car Wage & Hour Emp't Pracs. Litig. (Enterprise), 683 F.3d 462, 468-69 (3d Cir. 2012)). But "[t]his approach is not consistent with the breadth of employment under the FLSA." Id.

That is because "the FLSA rejected control as the standard for determining employment[.]" Id. So "any vertical joint employment analysis must [examine] more than the potential joint employer's control over the employee." Id. "As with all aspects of the employment relationship under the FLSA and MSPA," courts must consider "the expansive definition of 'employ' as including 'to suffer or permit to work'" to "determin[e] joint employment[.]" Id. at *13. The Department withdrew the 2016 AI in 2017. See News Release, U.S. Dep't of Labor, "U.S. Secretary of Labor Withdraws Joint Employment, Independent Contractor Informal Guidance" (June 7, 2017).

https://www.dol.gov/newsroom/releases/opa/opa20170607.

B. The Final Rule

"Against this backdrop, the Department published a Notice of Proposed Rulemaking ('NPRM') setting forth proposed revisions to the 'joint employer' regulation." Scalia I, 2020 WL 2857207, at *4 (citing 84 Fed. Reg. 14043-02 (Apr. 9, 2019)); see also Local Rule 56.1 Statement ("56.1"), Dkt No. 105, ¶ 1. "The Department issued the Final Rule after the receipt and review" of almost 57,000 comments. Scalia I, 2020 WL 2857207, at *5 (citing 85 Fed. Reg. 2820 (Jan. 16, 2020), codified at 29 C.F.R. §§ 791.1-3); see also 56.1 ¶¶ 2-3. The Final Rule took effect on March 16, 2020. 56.1 ¶ 3.

The Final Rule maintains the distinction between vertical and horizontal employment. See 29 C.F.R. § 791.2 ("There are two joint employer scenarios under the FLSA."). This lawsuit focuses on the Final Rule's revisions to the standard for vertical joint employment. Unless otherwise noted, "joint employment" in this opinion means vertical joint employment.

Here is an example of vertical joint employment. Imagine that an employee works for a contractor. A corporation hires the contractor. The contractor fails to pay the employee the minimum wage, as required by the FLSA. If the contractor and the corporation are the employee's joint employers, the employee can sue both the contractor and the corporation for back wages. In other words, the contractor and the corporation are both on the hook for the employee's damages. But if the contractor and the corporation are separate employers, then the employee can sue only the contractor. Note that the wages due to the employee are the same in either scenario. The joint employment doctrine addresses only from whom the employee may collect damages. Imagine, for example, that the contractor goes bankrupt. If the corporation is her joint employer, the employee can still recover. If not, the employee is out of luck.

The Final Rule revises the standard for vertical joint employer liability. In that scenario, "the employee has an employer who suffers, permits, or otherwise employs the employee to work, but another person simultaneously benefits from that work." 29 C.F.R. § 791.2(a)(1) (citing 29 U.S.C. §§ 203(e)(1), 203(g)). "The other person is the employee's joint employer only if that person is acting directly or indirectly in the interest of the employer in relation to the employee." Id. (citing 29 U.S.C. § 203(d)). The Final Rule's "primary purpose . . . is to offer guidance explaining how to determine joint employer status" in this scenario. 85 Fed. Reg. at 2823.

The Final Rule "adopt[s] a four-factor balancing test derived from Bonnette v. California Health & Welfare Agency." Id. at 2820 (citing 704 F.2d 1465 (9th Cir. 1983)). The factors are whether the putative joint employer "(i) hires or fires the employee; (ii) supervises and controls the employee's work schedule or conditions of employment to a substantial degree; (iii) determines the employee's rate and method of payment; and (iv) maintains the employee's employment records." 29 C.F.R. § 791.2(a)(1)(i)-(iv) (capitalization altered). "[T]he appropriate weight to give each factor will vary depending on . . . how that factor . . . suggest[s] control in [a given] case." Id. § 791.2(a)(3)(i).

But if an entity satisfies only the fourth factor, it is not a joint employer. Id. § 791.2(a)(2).

Although the Final Rule's test generally tracks the factors enumerated in Bonnette, the Department "narrowed" the first factor to "consider only whether the potential joint employer hires or fires the employee, rather than whether the potential joint employer has the 'power' to hire or fire the employee." 85 Fed. Reg. at 2830. This change reflects the Final Rule's focus on control as the core of the joint employer inquiry. Id.

Thus, control is the touchstone of the joint employer analysis under the Final Rule. The four factors signify the putative joint employer's control. See id. And "[t]he potential joint employer must actually exercise—directly or indirectly—one or more of these indicia of control to be jointly liable under the [FLSA]." Id. While an entity's "reserved right to act can play some role in determining joint employer status," the Final Rule requires " actual . . . control." 85 Fed. Reg. at 2821. In other words, although an "ability, power, or reserved right to act in relation to the employee may be relevant for determining joint employer status," these cannot "demonstrate joint employer status" unless the entity "actual[ly] exercise[s] . . . control." 29 C.F.R. § 791.2(a)(1)(i). Indeed, the four factors "were intended to focus on the economic realities of the potential joint employer's . . . control over the terms and conditions of the employee's work." 85 Fed. Reg. at 2828.

The joint employer inquiry under the Final Rule purports to be holistic. "No single factor is dispositive in determining joint employer status[.]" 29 C.F.R. § 791.2(a)(3)(i). And although "the four factors should determine joint employer status in most cases . . . the Department recognize[d]. . . that additional factors may be relevant for determining joint employer status." 85 Fed. Reg. at 2821. But other facts are relevant only if they reflect control. So "[a]dditional factors may be relevant for determining joint employer status" under the Final Rule, "but only if they . . . indic[ate] whether the potential joint employer . . . significant[ly] control[s] . . . the terms and conditions of the employee's work." 29 C.F.R. § 791.2(b).

Indirect control can be sufficient for joint employer status. "[T]he potential joint employer" exercises indirect control "through mandatory directions to another employer that directly controls the employee." Id. § 791(a)(3)(ii). But these directions must be mandatory. So if "the direct employer[] voluntar[il]y deci[des] to grant the potential joint employer's request, recommendation, or suggestion[,]" that "does not constitute indirect control[.]" Id. And "[a]cts that incidentally impact the employee also do not indicate joint employer status." Id.

Economic dependence is irrelevant under the Final Rule. Id. § 791.2(c). ("Whether the employee . . . economically depend[s] on the potential joint employer is not relevant[.]"). Thus, "no factors should be used to assess economic dependence" in the joint employer calculus. Id. The Final Rule offers four examples of "economic dependence" factors that are irrelevant including "[w]hether the employee is in a specialty job or a job that otherwise requires special skill, initiative, judgment, or foresight[.]" Id. The Final Rule clarifies that "[e]conomic dependence is relevant when . . . determining whether a worker is an employee under the [FLSA.]" 85 Fed. Reg. at 2821. But "determining whether a worker who is an employee under the [FLSA] has a joint employer . . . is a different analysis[.]" Id.

The Final Rule excludes certain employer characteristics as irrelevant to the joint employer analysis. See 29 C.F.R. § 791(d)(2)-(5). In short, the Final Rule renders irrelevant to the joint employer inquiry "certain business models (such as a franchise model), certain business practices (such as allowing the operation of a store on one's premises), and certain contractual agreements (such as requiring a party in a contract to institute sexual harassment policies)[.]" 85 Fed. Reg. at 2821.

The Final Rule reaffirms that joint employers are jointly and severally liable for damages under the FLSA. "For each workweek that a person is a joint employer of an employee, that joint employer is jointly and severally liable with the employer and any other joint employers for compliance with" the FLSA. 29 C.F.R. § 791.2(f).

The Final Rule states that the FLSA's definition of "employer" in section 3(d) is the "sole textual basis" for joint employer liability in the statute. 85 Fed. Reg. at 2825. Recall that the FLSA defines employer in section 3(d) to "include[] any person acting directly or indirectly in the interest of an employer in relation to an employee." 29 U.S.C. § 203(d). Under the Final Rule, "[t]hat language alone provides the textual basis for determining joint employer status[.]" 85 Fed. Reg. at 2820. The Department "explicitly tether[ed] the joint employer standard . . . to section 3(d)[.]" Id. at 2825 (quotation and brackets omitted). Recall also that the FLSA defines "employee" in section 3(e)(1) to "mean[] any individual employed by an employer" and "employ" in section 3(g) to "include[] to suffer or permit to work." 29 U.S.C. §§ 203(e)(1), 203(g). The Final Rule says that these definitions are irrelevant to the joint employer analysis.

The Department ignored sections 3(g) and 3(e)(1) because section 3(d)'s "plain terms[] contemplate[] an employment relationship between an employer and an employee, as well as another person who may be an employer too[.]" 85 Fed. Reg. at 2827. This scenario "fits the . . . joint employer scenario[.]" Id. The Department explained that section 3(d)'s text "makes sense only if there is an employer and employee with an existing employment relationship and the issue is whether another person is an employer." Id. "Indeed," the Department observed, "among the Act's definitions, only . . . section 3(d) contemplates" multiple employers. Id. On the other hand, section 3(e)'s "plain terms[] focus[] on the individual's status as an employee[.]" Id. But in a joint employer scenario, "the individual's status as an employee is unquestioned." Id. "[T]he individual is an employee of one employer" and her "work for that employer happens to simultaneously benefit another person[.]" Id. In the Department's view, the only issue in a joint employer scenario "is whether th[e] other person is also the employee's employer." Id.

The Department also criticized some existing joint employment tests because they "are expressly grounded in the principle that the FLSA should be read broadly[.]" Id. at 2824. In the Department's view, the Supreme Court's decision in Encino Motorcars, LLC v. Navarro (Encino II), casts doubt on the "continued viability of that principle." Id. (citing 138 S. Ct. 1134 (2018)). Encino II rejected the argument that "exemptions to the FLSA should be construed narrowly." 138 S. Ct. at 1142. "Because the FLSA gives no 'textual indication' that its exemptions should be construed narrowly, 'there is no reason to give them anything other than a fair (rather than a "narrow") interpretation.'" Id. (quoting A. Scalia & B. Garner, Reading Law 363 (2012)). The joint employer doctrine does not depend on an FLSA exemption, but Department cited Encino II as authority for its narrower construction of the standard for joint employer liability.

The Department emphasized the benefits of a uniform joint employment standard. "[C]ircuit courts currently use a variety of multi-factor tests to determine joint employer status[.]" 85 Fed. Reg. at 2823. As a result, workers have been treated "inconsistent[ly,]" and employers face "uncertainty . . . and increased compliance and litigation costs." Id. The Department also recognized that "given the divergent views of joint employment in the circuit courts, it would not be possible to provide detailed guidance that is consistent with all of them." Id. at 2824. "Indeed, this variance across the country is [a] primary reason[]" for implementing the Final Rule. Id. at 2831. "[B]y promulgating a clear and straightforward regulation, the Department hope[d] to encourage greater consistency for stakeholders." Id. In other words, the Department adopted the four-factor test "[t]o promote greater uniformity in court decisions and predictability for organizations and employees[.]" Id. at 2824. The Department similarly touted the benefits of increased "clarity." Id. at 2853. By promoting clarity, the Final Rule argues that it "promote[s] innovation and certainty in business relationships[.]" Id.

The Final Rule acknowledges the 2014 and 2016 AIs. See id. at 2822-23. The Department noted that the 2014 AI "opined that 'a set of joint employer factors that addresses only control'" conflicts with "section 3(g)'s 'suffer or permit' language[,] which governs FLSA joint employer status." Id. at 2822 (quoting 2014 AI, 2014 WL 2816951, at *2 n.5) (brackets omitted). The Department also observed that 2016 AI "rejected the common law control standard" because of "the expansive definition of 'employ' in both the FLSA and MSPA[.]" Id. (quoting 2016 AI, 2016 WL 284582, at *3). The Department also acknowledged that the 2016 AI concluded that "joint employment, like employment generally, should be defined expansively." Id. (quoting 2016 AI, 2016 WL 284582, at *3) (some quotation marks omitted). The Department did not explain why the Final Rule departs from these prior interpretations.

The Department also recognized that the Final Rule makes it harder for employees to recover back wages. The Department observed that employees "are not likely to see a change in the wages owed them under the FLSA" because of the Final Rule. Id. at 2853. That is because the primary employer "is liable to the employee for all wages due under the [FLSA] for the hours worked." Id. Thus, an employee could still "collect the entire wages due" from the primary employer. Id. Although in certain circumstances "[t]he employee would no longer have a legal right to collect the wages due" from a putative joint employer under the Final Rule's narrower standard, she could still collect those wages from the primary employer. Id. In the NPRM, the Department "assumed that employers always fulfill their legal obligations under the [FLSA] and pay their employees in full." Id. Yet commenters noted that if that "were true, there would be no successful FLSA investigations or cases." Id. In the Final Rule, the Department agreed that the new standard might "reduce the number of businesses currently found to be joint employers from which employees [can] collect back wages due to them under the FLSA[.]" Id. The Department thus conceded that the Final Rule might "reduce the amount of back wages that employees [can] collect when" their primary employer "does not comply" with the FLSA "and, for example, . . . is or becomes insolvent." Id.

But the Department did not account for these costs to employees because it did not believe they could be quantified. The Economic Policy Institute (the "EPI") submitted a comment that included a "quantitative analysis of transfers" between employers and employees under the Final Rule. Id. But the Department disregarded "EPI's quantitative analysis" because it did "not believe there are data to accurately quantify" the Final Rule's effect on workers' ability to collect back wages. Id. "The Department lack[ed] data on the current number of businesses that are in a joint employment relationship, or to estimate the financial capabilities (or lack thereof) of these businesses[.]" Id. Thus, it was "unable to estimate the magnitude of a decrease" in how many employers would be "liable as joint employers" under the Final Rule. Id. For that reason, although "the Department acknowledge[d] that there may be transfers from employees to employers" because of the Final Rule, it did not account for these transfers. Id. at 2821.

C. Procedural History

The States "sued 'to vacate the Final Rule and enjoin its implementation' because the Final Rule" violates the APA. Scalia I, 2020 WL 2857207, at *5 (quoting Complaint, Dkt No. 1, ¶ 9). "Defendants moved to dismiss for lack of subject-matter jurisdiction." Id. at *7 (citing Dkt Nos. 62-63). The Court denied the motion to dismiss in Scalia I.

Five trade associations (the "Associations") then moved to intervene. Dkt Nos. 76-78. The Court granted that motion. See New York v. Scalia, No. 1:20-cv-1689 (GHW), 2020 WL 3498755 (S.D.N.Y. June 29, 2020).

The parties now cross move for summary judgment. Dkt Nos. 91-94, 103-10. The Restaurant Law Center and the Society of Human Resource Management moved to file amicus briefs. Dkt Nos. 111-13. The Court granted those motions. Dkt No. 114. The States filed a consolidated opposition and reply to Defendants' cross-motion for summary judgment. Dkt Nos. 119-20. Defendants replied. Dkt Nos. 121-22.

II. LEGAL STANDARD

A movant is entitled to summary judgment if she "shows that there is no genuine dispute as to any material fact[.]" Fed. R. Civ. Pro. 56(a); see, e.g., Celotex Corp. v. Catrett, 477 U.S. 317, 322-23 (1986). When "a party seeks review of agency action under the APA . . . the entire case on review is a question of law." Ass'n of Proprietary Colls. v. Duncan, 107 F. Supp. 3d 332, 344 (S.D.N.Y. 2015) (quoting Am. Bioscience, Inc. v. Thompson, 269 F.3d 1077, 1083 (D.C. Cir. 2001)) (alterations omitted); see also New York v. United States Dep't of Health & Human Servs. (HHS), 414 F. Supp. 3d 475, 516 (S.D.N.Y. 2019). That is because whether an agency action violated the APA is a "legal question[]" that is "amenable to summary disposition." HHS, 414 F. Supp. 3d at 516 (quoting Ass'n of Proprietary Colls., 107 F. Supp. 3d at 344).

"The APA 'sets forth the procedures by which federal agencies are accountable to the public and their actions subject to review by the courts.'" Dep't of Homeland Sec. v. Regents of the Univ. of Cal. (DACA), 140 S. Ct. 1891, 1905 (2020) (quoting Franklin v. Massachusetts, 505 U.S. 788, 796 (1992)). "It requires agencies to engage in 'reasoned decisionmaking[.]'" Id. (quoting Michigan v. EPA, 576 U.S. 743, 750 (2015)). The APA "directs that agency actions be 'set aside' if they are 'arbitrary' or 'capricious.'" Id. (quoting 5 U.S.C. § 706(2)(A)); see also Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 414 (1971) ("[A]gency action must be set aside if the action was 'arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law[.]'" (quoting 5 U.S.C. § 706(2)(A))).

"The 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.'" Encino Motorcars, LLC v. Navarro (Encino I), 136 S. Ct. 2117, 2125 (2016) (quoting Motor Vehicle Mfrs. Ass'n of United States, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). "This requirement allows courts to assess whether the agency has promulgated an arbitrary and capricious rule by 'entirely failing to consider an important aspect of the problem or offering an explanation for its decision that runs counter to the evidence before it.'" Little Sisters of the Poor Saints Peter & Paul Home v. Pennsylvania, 140 S. Ct. 2367, 2383-84 (2020) (quoting State Farm, 463 U.S. at 43) (brackets omitted). Reviewing courts must also consider whether the agency's decision "is so implausible that it could not be ascribed to a difference in view or the product of agency expertise." Nat'l Ass'n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 658 (2007) (quoting State Farm, 463 U.S. at 43).

Courts reviewing agency action under the APA apply a "narrow standard of review[.]" DACA, 140 S. Ct. at 1905. A court must not "substitute its judgment for that of the agency[.]" Id. (quoting FCC v. Fox Television Stations, Inc. (Fox), 556 U.S. 502, 513 (2009)) (ellipsis omitted). Review is limited to "whether the decision was 'based on a consideration of the relevant factors and whether there has been a clear error of judgment.'" Id. (quoting Overton Park, 401 U.S. at 416). And a court "should 'uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned.'" Fox, 556 U.S. at 513-14 (quoting Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974)).

"Generally, a court 'reviewing an agency decision is confined to the administrative record compiled by the agency when it made the decision.'" HHS, 414 F. Supp. 3d at 517 (quoting Nat'l Audubon Soc'y v. Hoffman, 132 F.3d 7, 14 (2d Cir. 1997)); see also Fla. Power & Light Co. v. Lorion, 470 U.S. 729, 743-44 (1985). A district court assesses only "whether or not as a matter of law the evidence in the administrative record permitted the agency to make the decision it did." HHS, 414 F. Supp. 3d at 517 (quoting Roberts v. United States, 883 F. Supp. 2d 56, 62 (D.D.C. 2012)). There is an exception to this rule for evidence that establishes a plaintiff's standing under Article III, which is discussed below.

III. DISCUSSION

A. Threshold Issues

1. Standing

The Court begins, as it must, with jurisdiction. Scalia I held the States had plausibly alleged that they had standing because they alleged that "the Final Rule will reduce their tax revenue and increase their administrative and enforcement costs." 2020 WL 2857207, at *1. But "[e]ach element [of standing] must be supported in the same way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and degree of evidence required at the successive stages of the litigation." Susan B. Anthony List v. Driehaus, 573 U.S. 149, 158 (2014) (quoting Lujan v. Defs. of Wildlife, 504 U.S. 555, 561 (1992)); see also New York v. United States Dep't of Commerce (Census I), 351 F. Supp. 3d 502, 573 (S.D.N.Y.), aff'd in part, rev'd in part and remanded sub nom. Dep't of Commerce v. New York (Census II), 139 S. Ct. 2551 (2019).

The Court described the legal standard for a plaintiff to establish Article III standing in Scalia I. 2020 WL 2857207, at *7-8.

Because the case is now on summary judgment, the Court must examine the evidence put forward by the States to see if they have established, and not merely alleged, standing. "[A]lthough judicial review of agency action is typically confined to the administrative record, where there is insufficient evidence of standing in the record because the question was not before the agency, plaintiffs may submit extra-record evidence to establish standing." WildEarth Guardians v. Zinke, 368 F. Supp. 3d 41, 61 (D.D.C. 2019) (citing Sierra Club v. EPA, 292 F.3d 895, 899 (D.C. Cir. 2002)); see also Census I, 351 F. Supp. 3d at 572-73 & n.30.

a. Administrative Costs

The States have established that the Final Rule will increase their administrative costs. The Final Rule itself acknowledges that it "will impose direct costs on private businesses and state and local government entities by requiring them to review the new regulation." 85 Fed. Reg. at 2851. The States have submitted evidence to show the Final Rule's administrative burden on them. For example, Colorado estimates that the Final Rule will cause it to spend "several hundred hours of staff time for (1) temporary and permanent rulemaking" on the Colorado state law FLSA analogue; "(2) additional analysis and investigations of joint employment matters; (3) new guidance for employers and employees; and (4) new training of staff members on the Final Rule." 56.1 ¶ 22 (citing Declaration of Scott Moss, Dkt No. 68-2, ¶ 8). The District of Columbia notes that it will "review and revise its analysis of joint employment," "retract current guidance that incorporates prior FLSA jurisprudence," implement new guidance, train its employees to enforce the new guidance, and generally "undertake efforts to educate the public about the newly distinct analyses for joint employment under the District law and federal law[.]" Id. ¶¶ 24-25 (quoting Declaration of Unique Morris-Hughes, Dkt No. 68-4, ¶¶ 14, 16-17).

Minnesota will "develop guidance and conduct an outreach and education campaign" to "educate both employers and employees about the difference between the application of the joint employer doctrine under Minnesota and federal law[.]" Id. ¶ 28 (quoting Declaration of Nicole Blissenbach, Dkt No. 68-9, ¶¶ 15-16). Rhode Island will need to "revis[e]" its "guidance on joint employment" because of the Final Rule. Id. ¶ 32 (quoting Declaration of Joseph Degnan ("Degnan Dec."), Dkt No. 68-13, ¶ 16). "Updating the guidance will" be costly because Rhode Island must "research[], draft[], approv[e], and publish[] new guidance." Id. (quoting Degnan Dec. ¶ 17. The same is true for Virginia and Washington. Id. ¶¶ 33-34 (quoting Declaration of C. Ray Davenport, Dkt No. 68-15, ¶ 14; Declaration of Margaret Leland ("Leland Dec."), Dkt No. 68-16, ¶¶ 19-20).

This undisputed evidence is adequate to establish constitutional standing. "Monetary expenditures to mitigate and recover from harms that could have been prevented absent an agency action are precisely the kind of 'pocketbook' injury that constitute an injury to a proprietary interest for standing purposes." Scalia I, 2020 WL 2857207, at *11 (quoting New York v. United States Dep't of Labor (DOL), 363 F. Supp. 3d 109, 126 (D.D.C. 2019)) (brackets omitted); see also Air All. Houston v. EPA, 906 F.3d 1049, 1059-60 (D.C. Cir. 2018). Indeed, "governmental administrative costs caused by changes in federal policy" can be "cognizable injuries." Scalia I, 2020 WL 2857207, at *11 (quoting City & Cnty. of San Francisco v. U.S. Citizenship & Immigration Servs. (USCIS), 408 F. Supp. 3d 1057, 1123 (N.D. Cal. 2019)) (brackets omitted). So "the States have standing based on the Final Rule's direct imposition of an increased regulatory burden on them." Scalia I, 2020 WL 2857207, at *11 (quoting DOL, 363 F. Supp. 3d at 126) (brackets omitted). The Court need not examine the rest of the evidence submitted by the States because this evidence is enough to show that at least one State has Article III standing. See Census II, 139 S. Ct. at 2565 ("For a legal dispute to qualify as a genuine case or controversy, at least one plaintiff must have standing to sue.").

See, e.g., Massachusetts v. United States Dep't of Health & Human Servs., 923 F.3d 209, 222 (1st Cir. 2019) ("We hold that the Commonwealth has demonstrated Article III standing for its substantive claim based on an imminent fiscal injury that is fairly traceable to the federal regulations and redressable by a favorable decision."); Texas v. United States, 945 F.3d 355, 386 (5th Cir. 2019), cert. granted sub nom. California v. Texas, 140 S. Ct. 1262 (2020) ("[W]e [have] held that the state of Texas had standing to challenge the federal government's DAPA program because it stood to 'have a major effect on the states' fisc.'" (quoting Texas v. United States (DAPA), 809 F.3d 134, 152 (5th Cir. 2015), aff'd by an equally divided Court sub nom. United States v. Texas, 136 S. Ct. 2271 (2016)).

The Department challenges the States' evidence as "vague, speculative, and conclusory." See generally 56.1. But ironically, the Department's denials are vague and conclusory. "Conclusory allegations or denials . . . are not evidence and cannot by themselves create a genuine issue of material fact where none would otherwise exist." FTC v. Moses, 913 F.3d 297, 305 (2d Cir. 2019) (quoting Fletcher v. Atex, Inc., 68 F.3d 1451, 1456 (2d Cir. 1995)) (brackets omitted). Thus, the Department's bald assertions that the States' evidence is "vague, speculative, and conclusory" does not create an issue of fact about the States' standing. In any event, the States have offered specific evidence that the Final Rule will increase their administrative costs. See, e.g., Leland Dec. ¶¶ 21-22 (estimating that Washington will spend about $500,000 because of the Final Rule). Thus, the States have standing to challenge the Final Rule because it will increase their administrative costs.

The Final Rule itself touts increased uniformity as one of its benefits. See 85 Fed. Reg. at 2823. It makes sense that the States would need to incur administrative costs, such as issuing new guidance, to realize this benefit. So the Final Rule itself contemplates that the States will expend administrative resources to respond to its promulgation.

The Associations renew the argument that the States' increased administrative costs are "self-inflicted." Clapper v. Amnesty Int'l USA, 568 U.S. 398, 418 (2013). Scalia I rejected that argument, reasoning that "[t]he States have plausibly alleged that they will incur administrative costs by reviewing current guidance on joint employer liability and either retracting or issuing new or revised guidance." 2020 WL 2857207, at *11. "[T]hese increased administrative costs are 'predictable, likely, and imminent.'" Id. (quoting USCIS, 408 F. Supp. 3d at 1124); see also Cal. v. Azar, 911 F.3d 558, 573-74 (9th Cir. 2018), cert. denied sub nom. Little Sisters of the Poor Jeanne Jugan Residence v. Cal., 139 S. Ct. 2716 (2019) (holding that States' injuries were not self-inflicted). The States have provided evidentiary support for those allegations, so the Court adheres to its conclusion in Scalia I.

Oneida Indian Nation v. United States DOI is not to the contrary. 789 F. App'x 271 (2d Cir. 2019). The Associations argue that in Oneida, "the Second Circuit . . . held on analogous facts that standing is not available to plaintiffs who claim only a 'hypothetical future harm.'" Memorandum of Law in Support of Cross Motion for Summary Judgment ("Associations' Mem."), Dkt No. 107, at 22 (quoting Oneida, 789 F. App'x at 276). The facts of Oneida—a non-precedential summary order—are not even remotely "analogous" to the facts here. Oneida involved an Indian tribe's "petition[ to] the Trademark Trial and Appeal Board of the United States Patent and Trademark Office." 789 F. App'x at 274. The issue was whether the name of one Indian tribe was confusingly similar to the name of a different tribe. See id. In that context, Oneida held that "[i]ncurring costs in anticipation of potential future harm that is not concrete or imminent is insufficient to create an injury that will confer standing." Id. at 276 (citing Amnesty Int'l, 568 U.S. at 416). That is an unremarkable proposition. But the Associations' argument that this statement is relevant here because Oneida is factually "analogous" to this case is unconvincing.

The Associations also fail to distinguish the cases cited in Scalia I. The Associations note that DOL was decided on a motion to dismiss. Associations' Mem. at 22 n.25. True enough. But they fail to explain why that distinction is relevant. The Associations do not argue that the States' evidence does not support the allegations in their complaint. Nor could they. To be sure, Defendants dispute the legal significance of the States' evidence. But that is a rehash of Defendants' prior argument that the States' injuries are self-inflicted. It has nothing to do with the different legal standards applicable on a motion to dismiss and on summary judgment.

The Associations also note that USCIS held that while some States and counties had "submitted evidence of cognizable, irreparable costs," other States' evidence was "vague or speculative[.]" 408 F. Supp. 3d at 1124. But the States' evidence is neither vague nor speculative. Indeed, the increase in their administrative costs caused by the Final Rule is "predictable, likely, and imminent" and is probably occurring already. Id. The Associations' attempts to distinguish DOL and USCIS are unpersuasive, so the States have standing because the Final Rule will increase their administrative costs.

b. Enforcement Costs

The States have also shown that the Final Rule will increase their enforcement costs for state-level analogues of the FLSA. The Final Rule narrows the standard for joint employer liability under the FLSA. But many States will continue to enforce a broader standard for joint employer liability under analogous state laws. So to maintain the same level of protection for their workers, the States must increase the resources they devote to state-level analogues of the FLSA. For example, "Massachusetts will be required to allocate 'additional resources for enforcement of the state's wage and hour laws to ensure workers receive their earned wages.'" 56.1 ¶ 39 (quoting Declaration of Heather Rowe, Dkt No. 68-7, ¶ 10) (ellipsis omitted). "Michigan will have to fill in the gap in federal enforcement created by the Final Rule" to "ensure the current level of protections." Id. ¶ 40 (quoting Declaration of Sean Egan, Dkt No. 68-8, ¶ 23). To compensate for the decline in federal enforcement, "Vermont would have to 'increase its claim examiner staffing by approximately 50%.'" Id. ¶ 45 (quoting Declaration of Dirk Anderson, Dkt No. 68-14, ¶ 9). "Washington estimates spending $460,000-575,000 in hiring costs to address additional complaints and investigations that Washington will have to undertake because of the Final Rule." Id. ¶ 46 (citing Leland Dec. ¶¶ 21-22).

The States' increased enforcement costs are not self-inflicted. That is because the States' decision to devote increased resources to enforce state laws to protect their workers is a reasonable response to the Final Rule. A predictable effect of the Final Rule is that fewer employees will be able to recover back wages legally owed to them under the FLSA. Faced with that predictable effect, the States must increase their enforcement of state-level FLSA analogues to ensure the same level of protection for their workers. Thus, the Final Rule puts the States to a "forced choice": They can permit their workers to suffer a decreased level of protection or they can increase the resources they devote to enforce state-level analogues of the FLSA. Texas v. United States, 787 F.3d 733, 749 (5th Cir. 2015). "By forcing the" States "to make this . . . choice, the" Final Rule "results in a constitutional injury sufficient to establish standing[.]" Cnty. of Santa Clara v. Trump, 250 F. Supp. 3d 497, 538 (N.D. Cal. 2017). Thus, the States have standing to challenge the Final Rule because it will increase their enforcement costs.

c. Tax Revenue

The Court does not decide whether the States' evidence that the Final Rule will decrease their tax revenue is another ground on which they have established constitutional standing. The States have presented evidence that the Final Rule will reduce their tax revenue. See 56.1 ¶¶ 4-20 (citing exhibits to the Declaration of Fiona Kaye, Dkt No. 68). The States "will lose tax revenue as a result of decreased collections of judgments due to the Final Rule, including judgments for unpaid wages and unpaid workers' compensation as the Rule reduces the number of businesses" that qualify as joint employers under the FLSA. Id. ¶ 5 (citing Declaration of Heidi Shierholz ("Shierholz Dec."), Dkt No. 68-1, ¶¶ 6, 15, 18, 19, 23, 27). The States' expert, Dr. Shierholz, contends that each State will see an average decrease of about $20 million in its income tax revenue because of the Final Rule. Shierholz Dec. ¶¶ 18-19.

Dr. Shierholz also states that it is "likely that many low-wage workers who suffer a loss of wages due to wage theft and workplace fissuring[] may be forced to rely even more heavily on public assistance programs[,]" raising the States' "spending on safety net programs[.]" Id. ¶ 21. That undisputed evidence is perhaps another ground on which the States have established standing. See, e.g., Pennsylvania v. President United States, 930 F.3d 543, 562 (3d Cir. 2019), rev'd and remanded sub nom. on other grounds Little Sisters, 140 S. Ct. at 2367 ("[T]he States will suffer a concrete financial injury from the increased use of state-funded services."); DAPA, 809 F.3d at 155 (holding that Texas showed injury in fact "by demonstrating that it would incur significant costs in issuing driver's licenses to DAPA beneficiaries"). The Court does not decide whether this is an independent ground on which the States have established standing because the States have established standing on other grounds.

But the Associations have presented evidence that the opposite is true—that the Final Rule will increase the States' tax revenue. See Declaration of Ronald Bird ("Bird Dec."), Dkt No. 110-7, ¶ 8 (arguing that the Final Rule promotes "clarity and certainty" which will "foster job growth . . . increase jobs and, accordingly, the taxable wage base" in the States). This evidence may create an issue of fact about the effect of the Final Rule on the States' tax revenues.

The Court cannot resolve disputed issues of fact on a motion for summary judgment. See Pub. Citizen, Inc. v. Trump, 361 F. Supp. 3d 60, 64 (D.D.C. 2019) ("To carry the more onerous burden applicable on summary judgment, Plaintiffs must show that there is no genuine dispute of material fact regarding their standing to sue."). Thus, if this were the only ground on which the States argued they had standing, the Court might need to hold a trial to resolve the issue. See Census I, 351 F. Supp. 3d at 572-73 ("Defendants now concede that trial was necessary to resolve the issue of standing.").

But the Associations' evidence may not create a disputed issue of fact. For one, the "evidence" that the Final Rule will increase the States' tax revenue is paper thin. Dr. Bird, the Associations' expert, contended that "the clarity and certainty contained in the Final Rule is more likely to foster job growth, and thus increase jobs and, accordingly, the taxable wage base in states." Bird Dec. ¶ 8. That sentence is the full extent of Dr. Bird's analysis of the issue. "The object of [summary judgment] is not to replace conclusory allegations of the complaint or answer with conclusory allegations of an affidavit." Lujan, 497 U.S. at 888 (citing Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249 (1986)). As against the detailed evidence submitted by the States, Dr. Bird's statement may be a conclusory assertion that is inadequate at summary judgment.

Even accepting Dr. Bird's assertion would not necessarily defeat the States' standing based on decreased tax revenue. That is because even if the Final Rule increases certain State revenue streams through enhanced "clarity and certainty," it might also decrease other revenue streams. Bird Dec. ¶ 8. The Associations point to the States' "taxable wage base" in the aggregate. Id. But the States collect multiple different taxes on workers' wages. Dr. Shierholz distinguishes between state income taxes and taxes on wages to fund workers' compensation funds and unemployment insurance. See Shierholz Dec. ¶¶ 18-28. These taxes have different bases, so the revenue derived from each could move independently from the others. So, for example, a State might see its income tax receipts increase because of increased economic growth while contributions to its workers' compensation fund simultaneously decrease because the Final Rule imposes a narrower standard for joint employer liability.

The distinction between different revenue streams is important because "[t]he fact that an injury may be outweighed by other benefits does not negate standing." New York v. United States Dep't of Homeland Sec. (DHS), 969 F.3d 42, 60 (2d Cir. 2020) (quoting Denney v. Deutsche Bank AG, 443 F.3d 253, 265 (2d Cir. 2006)) (alterations omitted). The "standing analysis is not an accounting exercise." DAPA, 809 F.3d at 156 (quoting Nat'l Collegiate Athletic Ass'n v. Governor of New Jersey, 730 F.3d 208, 223 (3d Cir. 2013), abrogated on other grounds by Murphy v. Nat'l Collegiate Athletic Ass'n, 138 S. Ct. 1461 (2018)); see also Census I, 351 F. Supp. 3d at 609. So even if the Final Rule increases certain tax revenue streams—indeed, even if the Final Rule increases state tax revenue overall—that would not necessarily defeat the States' standing based on decreased tax revenues. For instance, an increase in a State's income tax revenue would not necessarily offset a decrease in contributions to its unemployment insurance fund for standing purposes. Because Dr. Bird did not distinguish between these different revenue streams, his declaration arguably does not create a disputed issue of fact on this issue.

There might also be a temporal mismatch between the decrease in tax revenue predicted by Dr. Shierholz and the increase predicted by Dr. Bird. Even if the Final Rule would eventually juice economic growth to the point that the States' tax revenue increased, it might also be true that the Final Rule would decrease State tax revenues in the short run. If true, the long-run increase would not necessarily offset the short-run decrease.

The Second Circuit's decision in XY Planning Network, LLC v. United States Securities & Exchange Commission does not preclude the States from establishing standing based on decreased tax revenue. 963 F.3d 244 (2d Cir. 2020). There, the Second Circuit held that a group of States did not have standing to challenge a regulation that "impose[d] a new 'best-interest obligation' on broker-dealers." Id. at 248. XY Planning reasoned that "[a] 'fairly direct link'" between a federal regulation and state tax revenue "is required because 'the unavoidable economic repercussions of virtually all federal policies suggest . . . that impairment of state tax revenues should not, in general, be recognized as sufficient injury in fact to support state standing.'" Id. at 252 (quoting Pennsylvania v. Kleppe, 533 F.2d 668, 672 (D.C. Cir. 1976)) (ellipsis omitted). The Circuit held that "[t]he ultimate annual pool of taxable capital gains in a state is driven by countless variables, from the performance of the broader economy to the composition of individual investor portfolios in the state." Id. at 253. Thus, the States did not have standing because their "theory of injury rest[ed] too heavily on 'conclusory statements and speculative economic data' concerning the long-term effects of [the new regulation] on state budgets[.]" Id. (quoting Wyoming v. U.S. Dep't of Interior, 674 F.3d 1220, 1232 (10th Cir. 2012)).

This case is distinguishable from XY Planning because the States' theory of injury here is less conjectural and hypothetical than was the States' theory in that case. "The States' theory for why the Final Rule will limit their tax revenue is simple. The States collect taxes on wages paid to employees in their states. And the States allege that the Final Rule will reduce aggregate wages paid to employees in their jurisdictions." Scalia I, 2020 WL 2857207, at *9 (citations omitted); see also New York v. Mnuchin, 408 F. Supp. 3d 399, 409-10 (S.D.N.Y. 2019). Indeed, "[t]he Department itself concede[d] that" the Final Rule "'may reduce the number of businesses currently found to be joint employers' and 'may reduce the amount of back wages that employees are able to collect when their employer does not comply with the Act and, for example, their employer is or becomes insolvent.'" Scalia I, 2020 WL 2857207, at *9 (quoting 85 Fed. Reg. at 2853). For that reason, the Court held that "the States ha[d] identified a specific revenue stream that they plausibly link directly to the Final Rule." Id. Because the link between the Final Rule and a reduction in the States' tax revenue is more direct than in XY Planning, that case does not defeat the States' standing to challenge the Final Rule based on decreased tax revenue. But the Court need not resolve whether the States' evidence is adequate to establish standing on this ground because the States have shown that they have standing on other grounds.

The Court also need not address whether the States have parens patriae standing. See Scalia I, 2020 WL 2857207, at *12-13 (declining to decide that issue).

2. Ripeness

This case is ripe. The Associations argue the opposite, albeit in a two-sentence paragraph. Associations' Mem. at 17. "Ripeness reflects constitutional considerations that implicate 'Article III limitations on judicial power,' as well as 'prudential reasons for refusing to exercise jurisdiction.'" Stolt-Nielsen S.A. v. AnimalFeeds Int'l Corp., 559 U.S. 662, 670 n.2 (2010) (quoting Reno v. Catholic Soc. Servs., Inc., 509 U.S. 43, 57 & n.18 (1993)); see also Suitum v. Tahoe Reg'l Planning Agency, 520 U.S. 725, 733 n.7 (1997) (distinguishing between cases that are unripe because they do not "properly present[] a genuine 'case or controversy' sufficient to satisfy Article III" and those that merely "fail[] to satisfy [the] prudential ripeness requirements").

The prudential ripeness "doctrine 'is designed to . . . avoid[] . . . premature adjudication[.]" Census I, 351 F. Supp. 3d at 626 (quoting Ohio Forestry Ass'n, Inc. v. Sierra Club, 523 U.S. 726, 732 (1998)); see also Abbott Labs. v. Gardner, 387 U.S. 136, 148-49 (1967), abrogated on other grounds by Califano v. Sanders, 430 U.S. 99 (1977). It "prevent[s] the courts . . . from entangling themselves in abstract disagreements over administrative policies[.]" Census I, 351 F. Supp. 3d at 626 (quoting Ohio Forestry, 523 U.S. at 732-33). And it "protect[s] the agencies from judicial interference until an administrative decision has been formalized and its effects felt in a concrete way by the challenging parties." Id. (quoting Ohio Forestry, 523 U.S. at 733).

To evaluate a claim's ripeness for judicial review . . . a court must consider three factors: "(1) whether delayed review would cause hardship to the plaintiffs; (2) whether judicial intervention would inappropriately interfere with further administrative action; and (3) whether the courts would benefit from further factual development of the issues presented."
Id. (quoting Ohio Forestry, 523 U.S. at 733).

The States have satisfied the three prudential ripeness requirements. The Final Rule is in effect, so any delay in review would cause hardship to the States. Because the Final Rule is final, judicial intervention would not interfere with any ongoing administrative action. And there can be no further factual development because the Court must decide whether the Final Rule is valid based on the administrative record.

The States have also satisfied the jurisdictional ripeness requirements. "The jurisdictional, or 'constitutional,' ripeness inquiry 'essentially mirrors that governing Article III standing[.]'" Id. at 628 (quoting United States v. Santana, 761 F. Supp. 2d 131, 138 (S.D.N.Y. 2011)). It "asks whether the question presents a 'genuine case or controversy.'" Id. Because the States have standing, this case satisfies Article III's requirements. The case is both prudentially and jurisdictionally ripe.

3. Zone of Interests

The Associations also argue that the States' claims are not within the zone of interests protected by the FLSA and the APA. The zone-of-interests test asks "whether the interest sought to be protected by the complainant is arguably within the zone of interests to be protected or regulated by the statute or constitutional guarantee in question." Scalia I, 2020 WL 2857207, at *13 (quoting Ass'n of Data Processing Serv. Orgs., Inc. v. Camp, 397 U.S. 150, 153 (1970)).

The Associations argue that the States have failed to identify rights "falling within the zone of interests encompassed by the FLSA, as channeled through the APA." Associations' Mem. at 23. Not so.

Because the States collect taxes on wages, the States' interest in protecting their tax base perfectly coincides with their interest in ensuring workers in their jurisdictions are compensated fairly. And that interest in protecting workers is exactly the sort of interest that the FLSA was enacted to protect. So the States' interest in protecting their tax base—and their precisely overlapping interest in ensuring that workers in their jurisdictions receive their wages—falls squarely within the "zone of interests to be protected or regulated by the" FLSA.
Scalia I, 2020 WL 2857207, at *14 (quoting Match-E-Be-Nash-She-Wish Band of Pottawatomi Indians v. Patchak, 567 U.S. 209, 224 (2012)). That the States are suing under the APA bolsters this conclusion because "'Congress's evident intent when enacting the APA' was 'to make agency action presumptively reviewable.'" Id. (quoting Patchak, 567 U.S. at 225).

The States have also satisfied the zone-of-interests test because they have established that they will suffer "secondary economic injuries" under the Final Rule. Id. The States have shown that "the Final Rule will reduce their tax revenue and increase their administrative and enforcement costs." Id. And "'[c]laims of financial injury,' including 'lost tax revenue and extra municipal expenses satisfy the "cause-of-action" (or "prudential standing") requirement.'" Id. (quoting Bank of Am. Corp. v. City of Miami, Fla., 137 S. Ct. 1296, 1303 (2017)) (alterations omitted). The Associations ignore Scalia I's holding that the States satisfied the zone-of-interests test based on secondary economic injuries. The States' claims satisfy the zone-of-interests test.

B. Contrary to Law

In the complaint, the States argue that the Final Rule is "not in accordance with law" and is "arbitrary and capricious" as separate causes of action. See Compl. ¶¶ 187-97. "Courts sometimes analyze the APA issue of whether a Rule is 'not in accordance with law' distinctly from the APA issue of whether it is 'arbitrary and capricious,' and sometimes combine these inquiries." HHS, 414 F. Supp. 3d at 535 n.38 (citing Henley v. FDA, 77 F.3d 616, 621 (2d Cir. 1996); FCC v. NextWave Pers. Commc'ns Inc., 537 U.S. 293, 304 (2003); Nat. Res. Def. Council v. U.S. EPA, 808 F.3d 556, 580-84 (2d Cir. 2015)). The Court chooses to analyze these questions separately. See DHS, 969 F.3d at 63-86 (considering these inquiries separately and concluding that a rule was "contrary to the [Immigration and Nationality Act]" and was "arbitrary and capricious").

The Final Rule conflicts with the FLSA. The Final Rule has two major flaws. First, the Department relied on the FLSA's definition of "employer" as the sole textual basis for joint employment liability. Second, the Department distinguished the test for whether an entity is an "employer" from the test for whether the entity is a "joint employer." Both those decisions contradict the text of the FLSA, prior Department interpretations of the FLSA and the MSPA, and caselaw from the Supreme Court and lower courts.

1. Section 3(d) as the Sole Textual Basis for Joint Employer Liability

The Final Rule contradicts the text of the FLSA. The Final Rule is an "interpretive rule." Perez v. Mortg. Bankers Ass'n, 575 U.S. 92, 97 (2015). Interpretive rules "are 'issued by an agency to advise the public of the agency's construction of the statutes and rules which it administers.'" Id. (quoting Shalala v. Guernsey Mem'l Hosp., 514 U.S. 87, 99 (1995)). As the Department acknowledged, "[i]t is axiomatic that any Department interpretation of the FLSA must begin with the text of the statute, following well-settled principles of statutory construction by 'reading the whole statutory text, considering the purpose and context of the statute, and consulting any precedents or authorities that inform the analysis.'" 84 Fed. Reg. at 14047 (quoting Kasten v. Saint-Gobain Performance Plastics Corp., 563 U.S. 1, 7 (2011)); see also Merit Mgmt. Grp., LP v. FTI Consulting, Inc., 138 S. Ct. 883, 893 (2018) (holding that statutory interpretation "begins with the text" of the statute).

a. Statutory Definitions

Start with the statutory definitions. In section 3(d), the FLSA defines an employer to "include[] any person acting directly or indirectly in the interest of an employer in relation to an employee[.]" 29 U.S.C. § 203(d). The FLSA defines "employee" to "mean[] any individual employed by an employer" and "employ" to "include[] to suffer or permit to work." Id. §§ 203(e)(1), (g).

The Final Rule adopts its test for joint employment based solely on the FLSA's definition of "employer." The Department reasoned that section 3(d) is the "sole textual basis for determining joint employer status" under the FLSA. 85 Fed. Reg. at 2825 (capitalization altered). The Department thus determined that the FLSA's definition of "employ" and "employee" are irrelevant to the joint employment analysis. See id. The Department's "statutory interpretation separat[es] sections 3(e) and (g) from section 3(d)." Id. at 2827.

This interpretation stumbles out of the starting gate. The first problem is that FLSA's definition of "employer" includes its definition of "employee." An employer "includes any person acting directly or indirectly in the interest of an employer in relation to an employee[.]" 29 U.S.C. § 203(d) (emphasis added). And the definition of "employee," in turn, includes the FLSA's definition of "employ." An employee is "any individual employed by an employer." Id. § 203(e)(1) (emphasis added). So all three definitions are relevant to determining joint employer status under the FLSA.

In other words, the FLSA's definition of "employer" cannot be read untethered from its related definitions of "employee" and "employ." That is why courts have looked to all three definitions to determine whether an entity qualifies as a joint employer. See, e.g., Barfield v. N.Y.C. Health & Hosps. Corp., 537 F.3d 132, 140-42 (2d Cir. 2008) (citing all three definitions in joint employer analysis and holding that "[i]n identifying the persons or entities who qualify as 'employers' subject to this provision, the statutory definitions sweep broadly").

This opinion calls a putative employer an "entity" for brevity. "A joint employer may be an individual, partnership, association, corporation, business trust, legal representative, public agency, or any organized group of persons, excluding any labor organization (other than when acting as an employer) or anyone acting in the capacity of officer or agent of such a labor organization." 29 C.F.R. § 791.2(d)(1) (citing 29 U.S.C. §§ 203(a), 203(d)).

See also Baystate, 163 F.3d at 675 (citing all three definitions in joint employer analysis and holding that the "remedial purposes of the FLSA require courts to define "'employer" more broadly than the term would be interpreted in traditional common law applications'" (quoting Dole v. Elliott Travel & Tours, Inc., 942 F.2d 962, 965 (6th Cir. 1991)); Torres-Lopez v. May, 111 F.3d 633, 638 (9th Cir. 1997) (citing all three definitions in joint employer analysis and holding that "[t]he FLSA broadly defines the 'employer-employee relationships' subject to its reach" (quoting Rutherford, 331 U.S. at 728) (brackets omitted)).

The Department recognized the principle that the three definitions are interrelated in the NPRM. There, the Department wrote that "[s]ections 3(d), 3(e)(1), and 3(g) work in harmony." 84 Fed. Reg at 14051. But the Department then reasoned that "3(e)(1) and 3(g) determine whether there is an employment relationship between the potential employer and the worker for a specific set of hours worked, and 3(d) alone determines another person's joint liability for those hours worked." 84 Fed. Reg. at 14050. This does not make sense because section 3(d) defines employer based on sections 3(e)(1) and 3(g).

b. Primary vs. Joint Employers

There is another problem, too: The Final Rule applies different tests for "primary" and "joint" employment. Under the Final Rule, "sections 3(e) and 3(g) determine whether an individual worker is an employee under the [FLSA]." 85 Fed. Reg. at 2827. If the worker is an employee, she has an employer. Call this employer the "primary employer." The Department reasoned that section "3(d) alone determines" if an employee has a joint employer. Id. at 2825 (quoting 84 Fed. Reg. at 14050). So, again, the Department's "interpretation separat[es] sections 3(e) and (g) from section 3(d)[.]" Id. at 2827.

At the outset, this argument is strange because it implies that the FLSA's definition of "employer" is irrelevant to whether an entity qualifies as a primary employer. "[W]hether one entity is a worker's 'employer' under the FLSA" is the same inquiry, "framed in reverse[,]" as "whether an individual is an entity's 'employee.'" Salinas, 848 F.3d at 139. So asking "whether an individual worker is an employee under the Act," id., is equivalent to asking whether an entity qualifies as an employer. In the Department's view, only the FLSA's definitions of "employ" and "employee" are relevant to whether an entity is an employer. That is odd.

Setting that oddity aside, the FLSA's text also does not support the Department's interpretation. The FLSA does not separately define a "joint employer." Joint employment arises because multiple entities may simultaneously satisfy the FLSA's definition of "employer." In other words, an employee has joint employers when she has multiple employers for the same set of hours worked. See, e.g., Falk, 414 U.S. at 195 (holding that two employers were joint employers "under the statutory definition").

There is thus no independent test for joint employment under the FLSA. An entity is an employer if it meets the FLSA's definition. It is a joint employer if it meets the definition and another entity also meets the definition. See Keith Cunningham-Parmeter, From Amazon to Uber: Defining Employment in the Modern Economy, 96 B.U. L. Rev. 1673, 1700-01 (2016) ("[T]he FLSA can hold multiple businesses liable as 'joint employers,' as long as each firm satisfies the statute's employment definition." (citation omitted)). For this reason, the Department's decision to "separat[e] sections 3(e) and (g) from section 3(d)" runs aground on the FLSA's text. 85 Fed. Reg. at 2827.

Consider an example. Imagine an employee works for a staffing agency. The staffing agency is the employee's employer. Then imagine that a corporation hires the staffing agency to provide temporary employees. The staffing agency hires out the employee to the corporation. Assume that the corporation also meets the FLSA's definition of employer. Both the staffing agency and the corporation become joint employers. The staffing agency was the employee's "employer." Now it is her "joint employer." Why? Because the corporation has also become the employee's employer. Now assume that the corporation hires the employee full time. The staffing agency stops qualifying as the employee's employer. The corporation was the employee's "joint employer." Now it is just her "employer." Why? Only because the staffing agency is no longer her employer. Put simply, the tests for primary and joint employment must be the same.

This does not mean that the employment analysis must (or should) evaluate a worker's employment relationships in isolation. Take the standard for joint employment under the MSPA. Before the Department promulgated the Final Rule, that test tracked the test for joint employment under the FLSA. See 29 C.F.R. § 500.20(h)(5)(i). Under the MSPA, the standard is whether the worker economically depends on a putative employer. Id. § 500(h)(5)(iii). Now return to the example outlined above. Imagine that the worker's paychecks come from the staffing agency. That fact is relevant to whether the worker economically depends on the corporation. So a court should consider the worker's employment relationship with the staffing agency in evaluating whether the corporation is her employer. But the test for whether the corporation is the employee's employer remains whether the worker is economically dependent on the corporation.

The Department's two tests create interpretive riddles. Imagine that it is unclear which entity is an employee's primary employer and which is her joint employer. To which entity should we apply which test? What if both putative employers argue that the other employer is the primary employer? What if both employers qualify as a primary employer but neither qualifies as a joint employer? The Final Rule assumes that it will always be obvious which entity is the primary employer. But the Department did not explain why that would be true.

The MSPA Regulations also undermine the Department's two-tests approach. Those regulations explain that "[t]he definition of the term 'employ'" under the MSPA "includes the [FLSA's] joint employment principles." 29 C.F.R. § 500.20(h)(5). And "[t]he term joint employment means a condition in which a single individual" is "an employee to two or more persons at the same time." Id. The MSPA Regulations do not contemplate independent tests for primary and joint employment. Thus, the Final Rule contradicts the Department's past understanding.

That inconsistency was not necessarily fatal to the Department's position because "[a]gencies are free to change their existing policies as long as they provide a reasoned explanation for the change." Encino I, 136 S. Ct. at 2125 (citing Nat'l Cable & Telecomms. Ass'n v. Brand X Internet Servs., 545 U.S. 967, 981-982 (2005); Chevron U.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 863-64 (1984)). But "the agency must at least 'display awareness that it is changing position' and 'show that there are good reasons for the new policy.'" Id. (quoting Fox, 556 U.S. at 515). The Final Rule does not do so. Indeed, the Department failed to address the MSPA Regulations—even though it acknowledged a comment that highlighted the disconnect between those regulations and the Final Rule. See 85 Fed. Reg. at 2833. The Department did not respond to the comment. That does not satisfy the requirement that the Department "provide a reasoned explanation for" its departure from the understanding reflected in the MSPA Regulations. Encino I, 136 S. Ct. at 2125 (citations omitted).

c. "Includes" vs. "Means"

The Department also disregarded that the FLSA introduces its definition of "employer" with the verb "includes." Recall that the FLSA says that the term "employer includes any person acting directly or indirectly in the interest of an employer in relation to an employee[.]" 29 U.S.C. § 203(d) (emphasis added). As the Supreme Court reasoned about a different provision of the FLSA, "the definition is introduced with the verb 'includes' instead of 'means.' This word choice is significant because it makes clear that the examples enumerated in the text are intended to be illustrative, not exhaustive." Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 162 (2012) (citation omitted). When a "statutory definition declares what [a term] 'includes[,]'" it is more capacious "than whe[n] the definition declares what a term 'means.'" Id. (quoting Burgess v. United States, 553 U.S. 124, 131 n.3 (2008)) (some alterations omitted). "Indeed, Congress used the narrower word 'means' in other provisions of the FLSA when it wanted to cabin a definition to a specific list of enumerated items." Id. (citing 29 U.S.C. § 203(a)); see also 29 U.S.C. § 203(e)(1) ("[T]he term 'employee' means any individual employed by an employer." (emphasis added)).

The Final Rule ignores the congressional choice to use "includes" in section 3(d). By introducing the definition of employer with "includes," Congress clarified that it is sufficient for employer status if an entity "act[s] directly or indirectly in the interest of an employer in relation to an employee." 29 U.S.C. § 203(d). But the Final Rule converts this into a necessary condition: A second employer "is the employee's joint employer only if that person is acting directly or indirectly in the interest of the employer in relation to the employee." 29 C.F.R. § 791.2(a)(1) (citing 29 U.S.C. § 203(d)) (emphasis added). That cannot be right because the FLSA's definition of employer is "illustrative, not exhaustive." Christopher, 567 U.S. at 162.

With this background, consider again why the Department argued that section 3(d) is the sole textual basis for joint employer liability. The Final Rule reasons that section 3(d) "contemplates an employment relationship between an employer and an employee, [and] another person who may be an employer[,] too[.]" 85 Fed. Reg. at 2827. On the other hand, section 3(e) "focuses on the individual's status as an employee . . . under the [FLSA]." Id. But in a joint employment scenario, "the individual's status as an employee is unquestioned." Id. Thus, "among the Act's definitions, only . . . section 3(d) contemplates" joint employment, according to the Department. Id.

The Department failed to recognize that the FLSA's definition of "employer" is illustrative. The Final Rule argues that section 3(d) "makes sense only if there is an employer and employee with an existing employment relationship[.]" Id. If the FLSA's definition of "employer" were exhaustive, that argument might be defensible. But the definition is not exhaustive. It clarifies that an entity can be an employer even if the employee is also in an employment relationship with a different employer. In other words, it clarifies that the FLSA recognizes joint employment. And that is all the definition does. The Department's interpretation thus overreads section 3(d).

d. The "Suffer or Permit To Work" Standard

The FLSA's definition of "employ" must be relevant to the joint employer inquiry. Recall that the FLSA says that "'[e]mploy' includes to suffer or permit to work." 29 U.S.C. § 203(g). The history of this definition informs its meaning. When a statutory term "is . . . transplanted from another legal source, whether the common law or other legislation, it brings the old soil with it." Hall v. Hall, 138 S. Ct. 1118, 1128 (2018) (quoting Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527, 537 (1947)); see also New York v. Nat'l Highway Traffic Safety Admin., No. 19-2395-AG, 2020 WL 5103860, at *7 (2d Cir. Aug. 31, 2020) (slip op.) (quoting Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. at 537). So the source of the "suffer or permit to work" standard clarifies its meaning.

The Supreme Court has recognized that "[t]he 'suffer or permit to work' standard" in the FLSA's definition of employ "derives from state child-labor laws designed to reach businesses that used middlemen to illegally hire and supervise children." Antenor, 88 F.3d at 929 n.5 (citing Rutherford, 331 U.S. at 728 n.7; People ex rel. Price v. Sheffield Farms-Slawson-Decker Co., 225 N.Y. 25, 29-30 (1918) (Cardozo, J.)); see also Darden, 503 U.S. at 326. This is a joint employment scenario. All agreed that the "middlemen" who directly employed children were their employers. The only question was whether businesses that "used" middlemen were also (joint) employers.

The Department explained the provenance of the "suffer or permit to work" standard in its brief in Rutherford. See Brief for the Administrator in Rutherford Food Corp. v. McComb, 1947 WL 43939, at *27. "The words 'suffer or permit to work' were . . . designed to comprehend . . . [the] relationship[s] which" entities had used to "avoid[]" being classified as employers. Id. "This language was not new, and" Congress knew about its "broad scope." Id. Congress adopted the "suffer or permit to work" standard "verbatim from [State] child labor statutes" and "from the proposed Uniform Child Labor Law." Id.. See also Cunningham-Parmeter, From Amazon to Uber, 96 B.U. L. Rev. at 1694 ("[There is] a near-continuous line of judicial authority that broadly construed the 'suffer or permit' terminology prior to the FLSA's passage. . . . [B]y placing the same wording in the nation's wage and hour law, Congress repudiated more restrictive common law definitions of employment and instead embraced an expansive vision of employer-employee relationships." (footnote and citations omitted)). See generally Bruce Goldstein et al., Enforcing Fair Labor Standards in the Modern American Sweatshop: Rediscovering the Statutory Definition of Employment, 46 UCLA L. Rev. 983, 1015-1103 (1999) (documenting the history of the "suffer or permit to work" standard).

See Cunningham-Parmeter, From Amazon to Uber, 96 B.U. L. Rev. at 1711 ("Just as child labor laws defined employment broadly to prevent businesses from claiming ignorance of unlawful practices, the FLSA extends liability to employers that control working conditions but look the other way when violations occur.").

The Final Rule ignores this history. The Department noted that some courts had rejected its four-factor test as irreconcilable "with the broad 'suffer or permit' standard[.]" 85 Fed. Reg. at 2831 n.58. But "[b]ecause . . . the Department believe[d] that section 3(d), not section 3(g), is the touchstone for joint employer status," it disagreed. Id. That reasoning ignores the source of the "suffer or permit to work" standard.

Consider also that Congress adopted section 3(g) to "disrupt the nation's 'sweating' system[.]" Cunningham-Parmeter, From Amazon to Uber, 96 B.U. L. Rev. at 1693 (citations omitted). "Under this scheme, . . . clothing manufacturers contracted with sweatshops to produce their wares." Id. The "sweatshops exposed workers to oppressive working conditions[.]" Id. The manufacturers "distanced themselves from these" practices to "protect[] their brands from reputational harm." Id. In the FLSA, Congress "extend[ed] liability to parties that 'permitted' wage violations" to target these abusive practices. Id. This, too, is a joint employer scenario. All agreed that the sweatshops employed the workers. The only question was whether the manufacturer was also their (joint) employer. Thus, a key purpose of section 3(g) was to expand joint employer liability.

See also Goldstein, Enforcing Fair Labor Standards, 46 UCLA L. Rev. at 984 ("Reformers who promoted the 'suffer or permit' standard sought to ameliorate abuses of the 'sweating system.' The 'sweater' was the labor intermediary who 'sweated' a profit out of his workers by depressing their wages[.] Factory owners benefited and workers suffered from ruthless competition among these labor contractors. The reformers sought to impose responsibility on the parties with the economic power to improve working conditions, i.e., the manufacturers hiring contractors. One of the reformers' tools was the 'suffer or permit' standard.").

The Final Rule flouts this purpose. The Department concluded that section 3(g) was irrelevant to a joint employment analysis. But Congress adopted section 3(g) to expand joint employer liability. So, by ignoring section 3(g), the Final Rule defies congressional intent.

The FLSA's legislative history confirms this conclusion. "[A] central theme . . . of the FLSA's legislative history" was congressional intent "to cover businesses [that] allow work to be done on their behalf and" have the power to "prevent wage and hour abuses, regardless of indirect business relationships and business formalities." Kati L. Griffith, The Fair Labor Standards Act at 80: Everything Old Is New Again, 104 Cornell L. Rev. 557, 571 (2019). This legislative history reinforces the conclusion that Congress adopted section 3(g) to broaden the scope of joint employer liability. So the Department should have considered that provision.

With this background, it becomes clear that an entity is an "employer" under the FLSA if it "suffers or permits" an employee to work. 29 U.S.C. § 203(g). The FLSA's definitions of "employer" and "employee" are "circular or vacuous[.]" Goldstein, Enforcing Fair Labor Standards, 46 UCLA L. Rev. at 1005. The term "employer includes any person acting directly or indirectly in the interest of an employer in relation to an employee[.]" 29 U.S.C. § 203(d) (emphasis added). This "definition" "relies on the very word it seeks to define[.]" Irizarry v. Catsimatidis, 722 F.3d 99, 103 (2d Cir. 2013). And "[t]he statute nowhere defines 'employer' in the first instance." Id. The definition of "employee" is no more useful. An employee is "any individual employed by an employer." 29 U.S.C. § 203(e)(1). Construing the same definition of "employee" in the Employee Retirement Income Security Act, the Supreme Court commented that the definition "is completely circular and explains nothing." Darden, 503 U.S. at 323. The "suffer or permit to work" standard gives "substance and enormous breadth" to these otherwise unhelpful definitions. Goldstein, Enforcing Fair Labor Standards, 46 UCLA L. Rev. at 1005. That is why Rutherford and other cases since have cited all three definitions in their joint employer analyses. 331 U.S. at 728 n.6; see, e.g., Salinas, 848 F.3d at 133. The Department itself noted the relevance of all three definitions in the 2014 and 2016 AIs.

The statutory definition of a joint employer follows from this understanding. An employer is a joint employer if it suffers or permits an employee to work while another employer simultaneously suffers or permits the same employee to work. But the Final Rule says section 3(g) is irrelevant to the joint employer analysis. That contradicts the plain text of the FLSA.

Return to section 3(d). "[T]he words of a statute must be read in their context and with a view to their place in the overall statutory scheme." Parker Drilling Mgmt. Servs. v. Newton, 139 S. Ct. 1881, 1888 (2019) (quoting Roberts v. Sea-Land Servs., Inc., 566 U.S. 93, 101 (2012)). Indeed, the Department acknowledged that the FLSA's definitions of "employer," "employee," and "employ" "work in harmony." 84 Fed. Reg at 14051. So the sweeping definition of employ in section 3(g) informs the interpretation of section 3(d).

Read alongside section 3(g), the most persuasive interpretation of section 3(d) is that it buttresses the "suffer or permit to work" standard. Recall that under the FLSA, an "'[e]mployer' includes any person acting directly or indirectly in the interest of an employer in relation to an employee[.]" 29 U.S.C. § 203(d) (emphasis added). This definition clarifies that an entity acting only "indirectly" is an employer. Id. That fits with the text, history, and purpose of the "suffer or permit to work" standard. Recall also that section 3(d) is illustrative and not exhaustive. This is why: The FLSA applies to indirect employers that suffer or permit employees to work. The FLSA's definition of "employer" does not narrow the scope of joint employer liability, as the Final Rule argues. To the contrary, section 3(d) bolsters section 3(g) by removing any doubt that even indirect employers are liable for FLSA violations. Because the Department failed to read the FLSA's definition of "employer" in light of its definition of "employ," its interpretation of section 3(d) conflicts with the statutory text.

e. The MSPA

The MSPA confirms this conclusion. Congress imported the MSPA's definition of employ from the FLSA. See 29 U.S.C. § 1802(5). The Department explained that "[t]he definition of the term employ includes the joint employment principles applicable under the [FLSA]." 29 C.F.R. § 500.20(h)(5). Indeed, "Congress[] incorporat[ed]" the FLSA's definition of "employ" in the MSPA to "adopt[] the FLSA joint employer doctrine" into the MSPA. Id. § 500.20(h)(5)(ii) (quoting legislative history of the MSPA).

But the MSPA does not define "employer." So the MSPA imports joint employment principles from the FLSA by adopting its definition of "employ," not "employer." It follows that the FLSA's definition of "employ" in section 3(g) is the textual basis for joint employment doctrine under the FLSA. If the FLSA's definition of "employer" in section 3(d) were the sole textual basis for joint employment doctrine, there would be a textual analogue to section 3(d) in the MSPA. Because there is not, the "suffer or permit to work" standard is the textual basis for joint employer doctrine in both the FLSA and the MSPA. The Department's contrary conclusion contradicts congressional intent.

f. FLSA Caselaw

The Final Rule argues that the Supreme Court's decisions in Rutherford and Falk and the Ninth Circuit's decision in Bonnette support its interpretation. See 85 Fed. Reg. at 2827. The Department is mistaken.

Rather than support the Department's interpretation, Rutherford undermines it. In Rutherford, a meatpacking company contracted with a meat boner. 331 U.S. at 725. The "head boner" hired boners "to be his employees" in the company's slaughterhouse. Id. The Department argued that Rutherford was about whether the boners were employees or independent contractors. That is misleading. In Rutherford, it was undisputed that the "boners were, first and foremost, employed by" the head boner. Zheng, 355 F.3d at 70. So true, the question was whether the boners were the company's employees or independent contractors. But because the Supreme Court held that the boners were the company's employees, the company was the boners' joint employer. See id. ("In Rutherford, the Supreme Court held that a slaughterhouse jointly employed workers who de-boned meat on its premises, [even though] a boning supervisor . . . directly controlled the terms and conditions of the meat boners' employment."). Thus, "Rutherford was a joint employment case[.]" Id.; see also, e.g., Layton v. DHL Exp. (USA), Inc., 686 F.3d 1172, 1180 (11th Cir. 2012) (citing Rutherford as a joint employment case); Torres-Lopez, 111 F.3d at 640 (same).

The Court respectfully disagrees with the Salinas court's conclusion that Rutherford was not a joint employment case. In adopting that view, the Fourth Circuit held that it is wrong to "view joint employment as a question of economic dependency." 848 F.3d at 139. According to Salinas, a test that "focus[es] on whether 'as a matter of economic reality, the individual is dependent' on a putative joint employer . . . reflects a failure to distinguish the joint employment inquiry from the separate, employee-independent contractor inquiry." Id. at 138 (quoting Layton, 686 F.3d at 1175). Salinas recognized that it was breaking with other circuits that have held that economic dependence is the touchstone of joint employer liability. See id. (citing Torres-Lopez, 111 F.3d at 639-40; Antenor, 88 F.3d at 932; Bonnette, 704 F.2d at 1469). In accord with the weight of circuit authority other than Salinas, the Court does not agree that "employee-independent contractor inquiry" is "separate" from the "joint employment inquiry[.]" Id. Fundamentally, these inquiries seek to answer the same question: Whether a worker and an entity have formed an employment relationship. Take the example of a worker working for a subcontractor. Assume that the subcontractor employs the worker as an employee. Now imagine that the worker argues that the primary contractor is also her employer. Salinas says that there are two questions here. The first is whether the worker is the contractor's employee or an independent contractor. The second is whether the contractor is the employee's joint employer. But the answer to both questions is the same. If the worker is the primary contractor's employee (and not an independent contractor), the contractor is her joint employer because the subcontractor is also her employer. It cannot be true that the worker is the primary contractor's employee and that the primary contractor is not her joint employer. And conversely, it cannot be true that the worker is not the primary contractor's employee (and is thus an independent contractor) but that the primary contractor is her joint employer. That is because the two supposedly separate tests track the same thing: Whether the worker is the primary contractor's employee. Now return to Rutherford. Salinas held that "Rutherford Food embraced economic dependency as a vehicle for distinguishing employees from independent contractors—not for determining whether two entities jointly employ a putative employee for purposes of the FLSA." Id. But as noted above, "Rutherford was a joint employment case" because "the boners were, first and foremost, employed by the boning supervisor who had entered into a contract with the slaughterhouse." Zheng, 355 F.3d at 70. So it was precisely because the boners were the slaughterhouse's employees that the slaughterhouse was their joint employer. Salinas reached a contrary conclusion based on the Department's former joint employment regulations. Recall that those regulations said that two employers were joint employers if they were "'not completely disassociated' with respect to establishing the terms and conditions of a worker's employment." Salinas, 848 F.3d at 138 (quoting former 29 C.F.R. § 791.2(a)). Keying off the "not completely disassociated" standard, Salinas held that other circuits have "improper[ly] focus[ed] on the relationship between a putative joint employer and a worker, rather than the relationship between putative joint employers." Id. at 139. But the joint employer inquiry asks whether two employers are sufficiently associated with respect to a given set of hours worked by an employee. If the worker works a set of hours that benefits multiple businesses, the question is whether each business suffers or permits the employee to work for that set of hours. Thus, the Court believes the joint employer inquiry should focus on the employee's relationship with each putative employer. The "not completely disassociated" test makes sense because the degree of association between two employers is relevant to this inquiry. If two employers are "completely disassociated," they cannot simultaneously suffer or permit an employee to work for the same set of hours. Thus, the degree of association between two employers is a proxy for whether both employers simultaneously suffered or permitted the employee to work. But the ultimate inquiry is whether the worker and the business have formed an employment relationship. See, e.g., Torres-Lopez, 111 F.3d at 641 ("The issue is not whether a farmworker is more dependent upon the farm labor contractor or the grower. Rather, the inquiry must focus on the economic reality of the particular relationship between the farmworker and the alleged joint employer." (citation omitted)); see also Goldstein, Enforcing Fair Labor Standards, 46 UCLA L. Rev. at 1132 ("The multiplicity of employers, rather than their jointness, is the key. Indeed, the confusion that courts have created in trying to analyze jointness suggests that the doctrine is misleadingly named and would be better renamed the Multiple Employer doctrine or presumption."). That is why the great weight of authority has concluded that economic dependence is relevant both to whether a worker is an employee or an independent contractor and whether two employers are joint employers. Although the presence of two employers may affect how a court applies the standard, the standard itself is the same for both inquiries.

The Department also selectively quoted Rutherford in the Final Rule. The Final Rule reasons that Rutherford "relied only on section 3(g)" to "determin[e] the workers' status as employees or independent contractors[.]" 85 Fed. Reg. at 2827. As support, the Department quoted the Supreme Court's statement that "[t]he definition of 'employ' is broad. It evidently derives from the child labor statutes[.]" Id. (quoting Rutherford, 331 U.S. at 728). But the Department left out the two sentences that directly precede the quoted statement: "There is in the Fair Labor Standards Act no definition that solves problems as to the limits of the employer-employee relationship under the Act. Provisions which have some bearing appear in the margin." 331 U.S. at 728 (emphasis added). And in a footnote, the Court quoted the FLSA's definitions of "employ," "employer," and "employee." See id. at 728 n.6. Thus, Rutherford says that all three definitions "have some bearing" on the "employer-employee relationship" under the FLSA. Id. at 728. It does not support the Department's conclusion that section 3(d) is the sole textual basis for joint employer liability.

Nor does Falk. Falk considered whether "maintenance workers employed at" an apartment complex who were "managed by" the petitioners in that case were "employees of the apartment owner or of the petitioners[.]" Falk, 414 U.S. at 195. It was undisputed that the "maintenance workers [were] employees of the building owners." Id. But the Supreme Court reasoned that the petitioners were "also an 'employer' of the maintenance workers under [section] 3(d) of the Act[.]" Id. After quoting the FLSA's definitions of "employer" and "employee," the Court noted that the FLSA's definition of "employer" is "expansive[.]" Id. And the petitioners had extensive "managerial responsibilities at . . . the buildings," so they "substantial[ly] control[led] . . . the terms and conditions" of the maintenance workers' employment. Id. For those reasons, the petitioners were the maintenance workers' "'employer'. . . under the statutory definition[.]" Id.

Falk cuts against the Department's argument that section 3(d) is the sole textual basis for joint employer liability. It affirmed that an employee can have more than one employer under the FLSA. See, e.g., Bonnette, 704 F.2d at 1469 (citing Falk for the proposition that "[t]wo or more employers may jointly employ someone" under the FLSA). It said that the FLSA's definition of "employer" is "expansive[.]" Falk, 414 U.S. at 195. And it cited the definition of "employee" to construe the FLSA's definition of "employer." The Department itself acknowledged that the FLSA's definition of "employee" "incorporates the [FLSA]'s definition (in section 3(g)) of 'employ' as including 'to suffer or permit to work.'" 85 Fed. Reg. at 2827. So the Department agreed that the FLSA's definition of "employ" is relevant to its definition of "employee." And Falk says that its definition of "employee" is relevant to its definition of "employer." Thus, the FLSA's definition of "employ" must also be relevant to its definition of "employer." For those reasons, Falk also undermines the Department's conclusion that section 3(d) is the sole textual basis for joint employer liability.

Bonnette likewise does not support the Department's interpretation. True, as the Department argues, Bonnette cited only section 3(d)—and so did not cite sections 3(g) or 3(e)—in discussing joint employer liability. 704 F.2d at 1469. But it also held that "[t]he determination of whether an employer-employee relationship exists does not depend on 'isolated factors but rather upon the circumstances of the whole activity.'" Id. (quoting Rutherford, 331 U.S. at 730) (emphasis added). And Bonnette noted that the FLSA's definition of "employer" "is not limited by the common law concept" of that term and should be interpreted "expansive[ly]" to "effectuate the FLSA's broad remedial purposes." Id. (citation omitted). So Bonnette does not support the Department's attempt to separate the FLSA's definition of "employer" from its definitions of "employ" and "employee."

The Department's novel interpretation that the FLSA's definitions of "employ" and "employee" are irrelevant to the joint employer analysis also led it to ignore that the Supreme Court has repeatedly held that the FLSA's definitions are exceedingly broad. See, e.g., Walling, 330 U.S. at 150-51 (holding that the FLSA's definitions are "comprehensive enough" that they apply "to many persons and working relationships" that did not historically "fall within an employer-employee category"); Rosenwasser, 323 U.S. at 363 n.3 (noting that the FLSA's definition of "employee" is "the broadest definition that has ever been included in any one act" (quotation omitted)).

Indeed, the Department relied heavily on its conclusion that sections 3(e) and 3(g) are irrelevant to the joint employer analysis. For example, the Department noted that "[t]he Second and Fourth Circuits rejected the Bonnette test" as the exclusive test for joint employer liability "because they did not believe it could be reconciled with the broad 'suffer or permit' [to work] standard[.]" 85 Fed. Reg. at 2831 n.58. The Department rejected those interpretations because, in its view, "section 3(d), not section 3(g), is the touchstone for joint employer status[.]" Id. "[A] Bonnette-based four-factor balancing test[,]" the Department reasoned, "is preferable and consistent with the text of that statutory provision." Id. Even if it were true that section 3(d) is the sole textual basis for joint employer liability, section 3(g) informs the correct interpretation of section 3(d). And in any event, section 3(d) is not the sole textual basis for joint employer liability. So the Department's explanation for ignoring authority holding that the Bonnette factors are "unduly narrow" is unconvincing. Zheng, 355 F.3d at 69.

The Final Rule also relies on the Supreme Court's decision in Encino II, but that decision cannot bear the weight the Department puts on it. Encino II was about an FLSA exemption from the overtime-pay requirement for "'any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles' at a covered dealership." 138 S. Ct. at 1138 (quoting 29 U.S.C. § 213(b)(10)(A)). The Supreme Court held that "service advisors—employees at car dealerships who consult with customers about their servicing needs and sell them servicing solutions"—fall within this exemption. Id. The joint employer doctrine does not rest on an FLSA exemption, so Encino II is not directly on point.

The Department relied on Encino II because that decision rejected "the principle that exemptions to the FLSA should be construed narrowly." Id. at 1142. The Supreme Court reasoned that "[b]ecause the FLSA gives no 'textual indication' that its exemptions should be construed narrowly, there is no reason to give them anything other than a fair (rather than a 'narrow') interpretation." Id. (citation and brackets omitted). "The narrow-construction principle relies on the flawed premise that the FLSA 'pursues' its remedial purpose 'at all costs.'" Id. (quoting Am. Express Co. v. Italian Colors Rest., 570 U.S. 228, 234 (2013)) (some quotation marks omitted). Relying on Encino II, the Department rejected circuit decisions that cited the FLSA's "remedial and humanitarian . . . purpose" in fashioning joint employment tests. 85 Fed. Reg. at 2824 (quoting Salinas, 848 F.3d at 140). The Department reasoned that Encino II casts doubt on "the continued viability" of the principle that this purpose is a "'useful guidepost for interpreting the FLSA[.]'" Id. (quoting Encino II, 138 S. Ct. at 1142).

Encino II does not support the Department's interpretation. To begin with, the Supreme Court has repeatedly held that the FLSA should be construed "liberally" because "broad coverage is essential to accomplish [its] goal[s.]" Alamo Found., 471 U.S. at 296 (quoting Mitchell v. Lublin, McGaughy & Assocs., 358 U.S. 207, 211 (1959) and citing Powell v. United States Cartridge Co., 339 U.S. 497, 516 (1950)). Although Encino II rejected the principle that exemptions to the FLSA "should be construed narrowly," it did not purport to overrule this prior caselaw. 138 S. Ct. at 1142.

Still, it is true that statutory purpose often carries little weight when interpreting a specific provision. For good reason, the Supreme Court has instructed that interpretation begins with the text of a statute. Purpose becomes relevant only if the text is ambiguous. See, e.g., Nat'l Ass'n of Mfrs. v. DOD, 138 S. Ct. 617, 631 (2018) (holding that when a statute is "unambiguous," a court's "inquiry begins with the statutory text, and ends there as well" (quoting BedRoc Ltd., LLC v. United States, 541 U.S. 176, 183 (2004) (plurality opinion)); see also Markle Ints., L.L.C. v. United States Fish & Wildlife Serv., 848 F.3d 635, 649 (5th Cir. 2017) (Jones, J., dissenting from the denial of rehearing en banc) ("As Justice Kagan has recently declared, 'We are all textualists now.'"). So if the States relied exclusively on the FLSA's "remedial and humanitarian . . . purpose," they would face an uphill battle. Tenn. Coal, Iron & R.R. Co. v. Muscoda Local No. 123, 321 U.S. 590, 597 (1944). But it is the Department's crabbed interpretation that runs afoul of the FLSA's text. Congress expressed the FLSA's remedial purpose by writing broad definitions in sections 3(d), (e), and (g). The joint employer doctrine must reflect the breadth of those provisions.

Encino II is not to the contrary. That decision rejected the principle that the FLSA's exemptions "should be construed narrowly" because that principle is unmoored from the statute's text. 138 S. Ct. at 1142. Thus, Encino II follows from the maxim that text drives statutory interpretation. But that maxim cuts against the Department's interpretation because the FLSA's definitions of "employ," "employee," and "employer" are "exceedingly broad." 2014 AI, 2014 WL 2816951, at *2. So Encino II does not support the Department's interpretation; that decision undermines it.

The Department also cited zero cases holding that section 3(d) is the sole textual basis for joint employer liability. Even the Department did not argue that Rutherford, Falk, or Bonnette explicitly adopted the "textual delineation" it now claims to discern "clear[ly]" in the FLSA's text. 85 Fed. Reg. at 2825 (citation omitted). It has been eight decades since Congress passed the FLSA and the Department recognized joint employer liability. And the best the Department can do is to draw a negative inference from Falk and Bonnette. Because those courts did not cite sections 3(e)(1) and 3(g), the Department argued that section 3(d) is separable from the FLSA's other statutory definitions.

That is thin gruel. If the Department's interpretation were "clear" (or even permissible), some court would have probably adopted its rationale. But the Department has found not a one. Over eighty years later, this dog has yet to bark. Cf. A. Doyle, Silver Blaze, in The Complete Sherlock Holmes 335 (1927).

The Department's novel interpretation that section 3(d) is the sole textual basis for joint employer liability conflicts with the FLSA. That is reason enough to conclude that the Final Rule must be set aside.

2. Control as the Touchstone of Joint Employer Liability

The Final Rule also has other flaws. For one, the Department's test for joint employer liability is impermissibly narrow. As noted, the Final Rule's test for joint employer liability balances four factors. These are whether a putative joint employer: "(i) hires or fires the employee; (ii) supervises and controls the employee's work schedule or conditions of employment to a substantial degree; (iii) determines the employee's rate and method of payment; and (iv) maintains the employee's employment records." 29 C.F.R. § 791.2(a)(1) (capitalization altered)

These factors are a proxy for control. "[T]o be a joint employer under the [Final Rule], the [putative joint employer] must actually exercise—directly or indirectly—one or more of the four control factors." 85 Fed. Reg. at 2821. Although "the appropriate weight to give each factor will vary depending on the circumstances," an "actual exercise of control" is a prerequisite for joint employer liability under the Final Rule. Id. at 2820, 2833-34.

That standard follows the common-law employment standard, which focuses on control. The most important factor in the common-law employment inquiry is whether an entity has the "right to control the manner and means by which the product is accomplished[.]" Darden, 503 U.S. at 323; see also Restatement (Second) of Agency § 220(1) (1958) (defining "servant" as someone "employed to perform services in the affairs of another and who . . . is subject to the other's control or right to control"). If so, the entity is an employer.

But the FLSA "expressly reject[s] the common-law definition of employment, which is based on limiting concepts of control and supervision." Antenor, 88 F.3d at 929 (citations omitted). The FLSA "defines the verb 'employ' expansively to mean 'suffer or permit to work.'" Darden, 503 U.S. at 326 (citing 29 U.S.C. §§ 203(e), 203(g)). Darden noted this definition's "striking breadth[,]" which "stretches the meaning of 'employee' to cover some parties who might not qualify as such under a strict application of traditional agency law principles." Id. Even the Final Rule concedes that "the [FLSA]'s definition of 'employ' . . . reject[ed] . . . the common law standard for determining who is an employee[.]" 85 Fed. Reg. at 2827. The FLSA opted instead for a "broader scope of coverage." Id. Thus, the FLSA's definition of "employer" is broader than its common law counterpart.

See Goldstein et al., Enforcing Fair Labor Standards, 46 U.C.L.A. L. Rev. at 984 ("To employ at common law predominantly meant to engage to work when the engaging party had the right to control the manner in which the work was performed. To permit to work was broader. It did not require the affirmative act of engaging a person to work, but only a decision to allow the work to take place. . . . [T]o suffer to work was broader still. To suffer in this context meant to tolerate or to acquiesce in. It required only that the business owner have the reasonable ability to know that the work was being performed and the power to prevent it. Thus, work performed as a necessary step [to produce] a product was almost always suffered or permitted by the business owner."); see also Zheng, 355 F.3d at 69 ("[T]he 'suffer or permit' language in the statute . . . reaches beyond traditional agency law[.]"); cf. Sec'y of Labor, United States Dep't of Labor v. Lauritzen, 835 F.2d 1529, 1543 (7th Cir. 1987) (Easterbrook, J., concurring) ("The definition [of employ], written in the passive, sweeps in almost any work done on the employer's premises, potentially any work done for the employer's benefit or with the employer's acquiescence.").

Courts have criticized the Bonnette factors as "unduly narrow" because they "focus[] solely on the formal right to control the physical performance of another's work." Zheng, 355 F.3d at 69. As noted, the right to control "is central to the common-law employment relationship[.]" Id. (citing Restatement of Agency § 220(1) (1933)). So the four Bonnette factors may "approximate the common-law test for identifying joint employers." Id. And the four factors may be "sufficient to establish employer status." Id. But a test that requires "a positive finding on those four factors [as] necessary to establish an employment relationship" contradicts the FLSA. Id.; see also Salinas, 848 F.3d at 137 ("Bonnette's reliance on common-law agency principles does not square with Congress's intent that the FLSA's definition of 'employee' encompass a broader swath of workers than would constitute employees at common law.").

Of course, "control" is a vague concept. But "the physical control at issue in suffering someone to work is not common-law control in the sense of standing over her and telling her what to do and how to do it. Rather, it is the much broader control the owner exercises over his business premises and his business that enables him to prevent the work." Goldstein, Enforcing Fair Labor Standards, 46 UCLA L. Rev. at 1138.

To be clear, all agree that the Final Rule's four factors can be relevant to the joint employer inquiry. Indeed, the four factors may constitute sufficient conditions for joint employer status. But the Final Rule says that a joint employer "must actually exercise—directly or indirectly—one or more of the four control factors." 85 Fed. Reg. at 2821 (emphasis added). In other words, the conclusion that an employer satisfies "one or more of the control factors" is a necessary condition for an entity to qualify as a joint employer. Id. That conflicts with the FLSA.

The Department's reading of Falk also confused necessary and sufficient conditions. It is true that Falk found that the petitioners "substantial[ly] control[led] . . . the terms and conditions" of the employees' work. 414 U.S. at 195. In part for that reason, Falk concluded that the petitioners were joint employers. Id. The Department argued that because Falk noted the petitioners' "substantial control," that case supports a control-based test for joint employer liability. Id. But the Department failed to distinguish necessary and sufficient conditions. No one disputes that substantial control may be sufficient to support joint employer liability, as it was in Falk. But Falk does not suggest that control is necessary for joint employer status, as it is under the Final Rule.

The Department did not deny that the Final Rule's test for joint employer liability mirrors the common-law standard. It was "not the Department's intent" for the Final Rule to be "narrower than the common law[.]" 85 Fed. Reg. at 2834 (emphasis added). Even if the Final Rule is not narrower than the common law, the Department did not argue that its four-factor test for joint employment is broader than the common law's. And the Department conceded that the FLSA rejects the common-law definition of employment as overly cramped. Thus, the Final Rule is impermissibly narrow.

It is unclear whether the Final Rule is narrower than the common law. At common law, the joint employer inquiry asks whether the principal "controls or has the right to control the physical conduct of the other in the performance of the service." Restatement (Second) of Agency § 2 (1958) (emphasis added). But the Final Rule requires an actual exercise of control. See 85 Fed. Reg. at 2821 ("The . . . ability, power, or reserved right to act in relation to the employee may be relevant for determining joint employer status[.] [B]ut such ability, power, or right alone does not demonstrate joint employer status without some actual exercise of control.").

Indeed, the Department's test is even narrower than the Bonnette factors, which courts have rejected as "unduly narrow." Zheng, 355 F.3d at 69. Bonnette held that "the power to hire and fire the employees" was relevant to the joint employment analysis. 704 F.2d at 1470 (emphasis added). But the Final Rule requires that the putative employer "[h]ires or fires the employee[.]" 29 C.F.R. § 791.2(a)(1)(i). If Bonnette's four factors are impermissibly narrow, a narrower test is also inadequate.

That the Final Rule permits "the consideration of additional factors" does not cure the problem because it is limited "to those that indicate control." 85 Fed. Reg. at 2836. "[T]he purpose of the Department's four-factor balancing test" is to "evaluat[e] . . . the potential joint employer['s control] over the employee[.]" Id. So other factors are relevant "only if they [indicate] whether the potential joint employer . . . significant[ly] control[s] . . . the terms and conditions of the employee's work." 29 C.F.R. § 791.2(b). Thus, the Final Rule unabashedly adopts a control-based test. That test is unlawfully narrow.

The cases the Department cited do not support its narrow test. Begin with Bonnette. Although it enumerated four factors, Bonnette did not purport to announce an exhaustive test. To the contrary, Bonnette noted that the four factors it enumerated "are not etched in stone[.]" 704 F.2d at 1470. Bonnette applied those factors because they were "relevant to th[e] particular situation" presented in that case. Id. And Bonnette cautioned that courts should not "appl[y]" the factors "blindly" because the joint employer analysis "must be based 'upon the circumstances of the whole activity.'" Id. (quoting Rutherford, 331 U.S. at 730). So the Department is wrong that "Bonnette adopted a similar four-factor test to determine whether a potential joint employer is liable." 85 Fed. Reg. at 2830.

The Final Rule argues "[t]he First and Fifth Circuits apply the Bonnette test, which is very close to the Department's proposed test[.]" Id. at 2831. Not true. In Baystate, the First Circuit held that the FLSA's "remedial purposes . . . require courts to define 'employer' more broadly than" at "common law." 163 F.3d at 675 (quotation omitted). That contradicts the Department's interpretation of the FLSA's definition of "employer." In any event, Baystate held that the Bonnette factors "provide a useful framework." Id. It applied the factors to conclude that multiple entities were joint employers. Id. at 675-76. So Baystate held that the Bonnette factors can be sufficient to establish joint employment. But it did not hold that they are necessary.

The Fifth Circuit case the Department relied on, Gray v. Powers, is closer to the mark, but even it does not support the Department's view. 673 F.3d 352, 355-57 (5th Cir. 2012). Gray relied on an earlier Fifth Circuit case that cited the four Bonnette factors. See Watson v. Graves, 909 F.2d 1549, 1553 (5th Cir. 1990). But as the Fifth Circuit later summarized, Watson found the Bonnette factors were not "dispositive." Williams v. Henagan, 595 F.3d 610, 620 (5th Cir. 2010). So "the court turned to the FLSA's twin purposes to maintain a minimum living standard and to reduce unfair competition among firms" and workers. Id. at 620-21 (citations omitted). Thus, Fifth Circuit law does not support applying the Bonnette factors as the exclusive test for joint employer liability.

The Final Rule argues that the "Third Circuit also applies a similar four-factor test" for joint employer liability. 85 Fed. Reg. at 2831. Again, that is not so. The Third Circuit held that because the FLSA is "unique[,]" a test for "joint employment 'must be based on a consideration of the total employment situation and the economic realities of the work relationship.'" Enterprise, 683 F.3d at 469 (quoting Bonnette, 704 F.2d at 1470). Enterprise enumerated four factors resembling the Bonnette factors. See id. But the Third Circuit took pains to "emphasize . . . that these factors do not constitute an exhaustive list of all potentially relevant facts[] and should not be 'blindly applied.'" Id. (quoting Bonnette, 704 F.2d at 1469-70). Indeed, the Third Circuit observed that applying the Bonnette factors "would only find joint employment where an employer had direct control over the employee[.]" Id. "[B]ut the FLSA [also] designates" as joint employers "entities with sufficient indirect control[.]" Id. That cuts against the Department's position that "an entity must actually exercise—directly or indirectly—one or more of the four control factors" to be a joint employer. 85 Fed. Reg. at 2821. So Enterprise also undermines the Department's position.

The Final Rule also claims that "[t]he Seventh Circuit has . . . suggested that joint employment depends on the measure of control exercised over the employee and that the Bonnette factors are relevant when assessing control." 85 Fed. Reg. at 2831. Yet again, the Department misread the case it cited. In Moldenhauer v. Tazewell-Pekin Consol. Commc'ns Ctr., the Seventh Circuit laid out a four-factor test for joint employment for the Family and Medical Leave Act. See 536 F.3d 640, 644 (7th Cir. 2008). But Moldenhauer also "decline[d] to . . . limit [its] review" to those four factors "in this case or subsequent cases." Id. "Although these factors are . . . relevant . . . [to] whether an employer-employee relationship exists, it would be foolhardy to suggest that these are the only relevant factors, or even the most important." Id. Moldenhauer cuts against the Department's interpretation.

At bottom, the cases cited by the Department do not support its statutory interpretation separating section 3(d) from sections 3(g) and 3(e) as the sole textual basis for joint employer doctrine. The Final Rule's emphasis on control as the touchstone of joint employer liability flows from that interpretive error. Because a control-based test for joint employer liability is unduly narrow, the Final Rule must be set aside.

3. Prohibition on Considering Additional Factors

The Final Rule must also be vacated because it unlawfully limits the factors the Department will consider in the joint employer inquiry. The Final Rule says that "[w]hether the employee is economically dependent on the potential joint employer is not relevant" to the joint employer inquiry. 29 C.F.R. § 791.2(c). The Final Rule also gives "[e]xamples of factors that are not relevant because they assess economic dependence[,]" including

(1) whether the employee is in a specialty job or a job that otherwise requires special skill, initiative, judgment, or foresight; (2) whether the employee has the opportunity for profit or loss based on his or her managerial skill; (3) whether the employee invests in equipment or materials required for work or the employment of helpers; and (4) the number of contractual relationships, other than with the employer, that the potential joint employer has entered into to receive similar services.
Id. (capitalization altered).

Excluding economic dependence as irrelevant to joint employer status contradicts caselaw and the Department's own views. The Department conceded that "[e]conomic dependence is relevant when applying section 3(g) [to] determin[e] whether a worker is an employee under the [FLSA.]" 85 Fed. Reg. at 2821. But the FLSA does not distinguish between employers and joint employers. Any factor that is relevant to whether an entity is an employer is also relevant to whether the entity is a joint employer. That is why courts have focused on economic dependence as the crux of the joint employment inquiry. See, e.g., Layton, 686 F.3d at 1178 ("[I]n considering a joint- employment relationship, we must not allow common-law concepts of employment to distract our focus from economic dependency." (quoting Antenor, 88 F.3d at 933)); Baystate, 163 F.3d at 675 ("[T]o determine whether an employment relationship exists for the purposes of federal welfare legislation, courts look not to the common law conceptions of that relationship, but rather to the 'economic reality' of the totality of the circumstances bearing on whether the putative employee is economically dependent on the alleged employer." (citing Aimable v. Long & Scott Farms, 20 F.3d 434, 439 (11th Cir. 1994)).

Judge Easterbrook has criticized both the "economic dependence" and the related "economic reality" tests as unhelpful:

A reference to "economic reality" tells the court to disregard economic fantasy but does not say which aspects of "reality" have what legal consequences. "'Reality' encompasses millions of facts, and unless we have a legal rule with which to sift the material from the immaterial, we might as well examine these facts through a kaleidoscope." The reference to "economic dependence" is scarcely more helpful. Thousands of employers compete for the services of agricultural workers; when deciding whether to accept Zarate's offer rather than someone else's, these workers were not "dependent" on Remington. Once they arrived in Indiana they were "dependent" in the sense that they relied on Zarate (and perhaps on Remington) to fulfill the deal (as both the travel and the work precede payment). They had sunk costs that may have led them to accept substandard housing and toilet facilities rather than endure transitional unemployment during a search for other work. This difference between the ex ante and ex post situations is true of all labor, however; how can it help us to determine whether to disregard a given independent contractor?

The Final Rule's enumeration of specific economic dependence factors as irrelevant also contravenes Rutherford. For example, the Final Rule prohibits courts from considering "whether the employee is in a specialty job" in the joint employer inquiry. 29 C.F.R. § 791.2(c)(1). But Rutherford held that it was relevant that the workers "did a specialty job on the production line." 331 U.S. at 730. Because "Rutherford was a joint employment case," the Final Rule's limitation contradicts Supreme Court precedent. Zheng, 355 F.3d at 70.

The Final Rule also conflicts with Supreme Court caselaw in another way. The Final Rule asserts that "there must be limits on the . . . factors" that the Department used to "determin[e] joint employer status[.]" 85 Fed. Reg. at 2836. Thus, four factors are "irrelevant to the joint employer inquiry" under the Final Rule. Id. at 2839. These include "certain business models (such as a franchise model), certain business practices (such as allowing the operation of a store on one's premises), and certain contractual agreements (such as requiring a party in a contract to institute sexual harassment policies)[.]" Id. at 2821; see 29 C.F.R. § 791.2(d)(2)-(5).

These limitations are also out of step with Supreme Court caselaw. To begin, Rutherford noted that the slaughterhouse "furnish[ed] a room in its plant for the work, known as the boning vestibule" and that the boning was "done at one place and under one roof." 331 U.S. at 725-26. Thus, the boners were "part of the integrated unit of production[.]" Id. at 729. This suggested that the slaughterhouse was the boners' (joint) employer. Id. Yet the Final Rule says that whether the "potential joint employer[] . . . allow[s] [a second] employer to operate a business on its premises" is irrelevant. 29 C.F.R. § 791.2(d)(5). That limitation conflicts with Rutherford.

The Rutherford Court also held that the employment "relationship does not depend on . . . isolated factors but rather upon the circumstances of the whole activity." 331 U.S. at 730. Excluding any factors as irrelevant is in tension with this holistic inquiry. The Final Rule does not say that these factors are insufficient to confer joint employer status. That limitation might comply with Rutherford. But the Final Rule sweeps too broadly by excluding these factors as categorically irrelevant to the joint employer inquiry.

The Final Rule's limitations also do not make sense. Consider again a putative joint employer that permits a second employer to operate a store on its premises. It is common sense that this arrangement could be relevant to whether the putative joint employer "suffer[s] or permit[s]" the second employer's employees to work. 29 U.S.C. § 203(g). So that limitation is both unlawful and illogical.

The Court is mindful that the Final Rule is entitled to a "measure of respect." Alaska Dep't of Envt'l Conservation v. E.P.A., 540 U.S. 461, 488 (2004). "[T]he weight accorded to such interpretations depends on their 'thoroughness,' 'validity,' 'consistency,' and 'power to persuade.'" De La Mota v. United States Dep't of Educ., 412 F.3d 71, 80 (2d Cir. 2005) (quoting Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944)); see also United States v. Mead Corp., 533 U.S. 218, 235 (2001) (holding that the "respect" owed to an agency interpretation is "proportional" to its "'power to persuade'" (quoting Skidmore, 323 U.S. at 140)). The Department's interpretation is unpersuasive. It also conflicts with prior Department interpretations. Compare, e.g., 2016 AI, 2016 WL 284582, at *3 (contending that the test for joint employer liability "is not a control test") with 85 Fed. Reg. at 2821 (requiring "some actual exercise of control" for joint employer liability).

The Department agrees that the Court should not accord Chevron deference to the Final Rule. Department's Memorandum of Law, Dkt No. 104, at 10-11. --------

In any event, the Department's interpretation contradicts the FLSA. And "[i]f the intent of Congress is clear, that is the end of the matter[.]" Pereira v. Sessions, 138 S. Ct. 2105, 2113 (2018) (quoting Chevron, 467 U.S. at 842-43). Both the Court and the Department "must give effect to the unambiguously expressed intent of Congress." Id. (quotation omitted). Because the Final Rule conflicts with the FLSA, it must be set aside.

C. Arbitrary and Capricious

1. Unexplained Inconsistency

The Final Rule is also arbitrary and capricious for at least three reasons. The first is that the Department did not adequately explain why it departed from its prior interpretations. As noted above, "[a]gencies are free to change their existing policies as long as they provide a reasoned explanation for the change." Encino I, 136 S. Ct. at 2125 (citations omitted). An agency "need not demonstrate . . . that the reasons for the new policy are better than the reasons for the old one[.]" Fox, 556 U.S. at 515 (emphasis omitted). "[I]t suffices that the new policy is permissible under the statute, that there are good reasons for it, and that the agency believes it to be better, which the conscious change . . . adequately indicates." Id. (emphasis omitted). But "the agency must at least 'display awareness that it is changing position[.]'" Encino I, 136 S. Ct. at 2125 (quoting Fox, 556 U.S. at 515). "[A]n 'unexplained inconsistency' in agency policy is 'a reason for holding [that] an interpretation [is] an arbitrary and capricious change from agency practice.'" Id. (quoting Brand X, 545 U.S. at 981).

The Department did not adequately explain why it departed from its prior interpretations. To begin, the Final Rule is a volte-face from the Department's 1997 Guidance in multiple respects. There, the Department explained that "the concept of 'joint employment'" in the MSPA rests on "[t]he MSPA statutory definition of 'employ[,]'" which is the same as "the FLSA statutory definition of 'employ[.]'" 62 Fed. Reg. at 11734. And the Department said that "the FLSA definition of employ" rejects "the traditional common law 'right to control' test[.]" Id. at 11745. The 1997 Guidance also applied the same test for primary and joint employment relationships. Id.

The Final Rule departs from these positions, but it does not explain why. Indeed, the Final Rule does not cite the 1997 Guidance, except to quote a commenter who argued that "the Department[] . . . revers[ed] [its] position [from] the last time it engaged in rulemaking [about] joint employer status." 85 Fed. Reg. at 2833. That comment is spot on, yet the Department did not respond to it. The Department failed to acknowledge that it had shifted its position from the 1997 Guidance, much less to explain why.

Nor did the Department satisfactorily explain why it departed from the 2014 and 2016 AIs. The 2014 AI stated that for joint employment, "a test that 'focuses solely on the formal right to control the physical performance of another's work'" is "'unduly narrow[.]'" 2014 WL 2816951, at *2 n.5 (quoting Zheng, 355 F.3d at 69). The 2016 AI observed that the test for joint employment in the FLSA "is not a control test." 2016 WL 284582, at *9. But the Final Rule says that a joint employer must "actual[ly] exercise . . . control." 85 Fed. Reg. at 2821.

That inconsistency demands an explanation, but the Final Rule gives none. The Final Rule discusses the 2014 and 2016 AIs in the "regulatory and judicial history" section. 85 Fed. Reg. at 2821-23 (capitalization altered). Yet the Department did not acknowledge that it was departing from those interpretations. Nor did it explain why it now believes the 2014 and 2016 AIs were wrong. The Final Rule is arbitrary and capricious because of this unexplained inconsistency.

2. Conflict Between the FLSA and the MSPA

The Final Rule is also arbitrary and capricious because the Department did not consider the conflict between it and the MSPA Regulations. The standard for joint employer liability under the MSPA is whether, as a matter of "economic reality," "the worker . . . economically depend[s]" on the putative joint employer. 29 C.F.R. § 500.20(h)(5)(iii). The MSPA Regulations state that "[j]oint employment under the [FLSA] is joint employment under the MSPA." Id. § 500.20(h)(5)(i). And the Final Rule acknowledges that the Department "will continue to use the standards in its MSPA joint employer regulation to determine joint employer status under MSPA[.]" 85 Fed. Reg. at 2828 n.55 (citing 29 C.F.R. § 500.20(h)(5)).

Thus, the Department now applies different standards for joint employer liability under the FLSA and the MSPA. Leaving aside that Congress intended the standards to be uniform, this could lead to increased costs for employers subject to both standards. The Final Rule does not acknowledge those costs or explain why the other benefits of the Final Rule outweigh them. That is arbitrary and capricious.

3. Cost to Workers

The Final Rule is arbitrary and capricious for a third reason: It does not adequately consider the Final Rule's cost to workers. In the NPRM, the Department contended that the Final Rule would not impose any such costs. 84 Fed. Reg. at 14054. There, the Department explained that even if the Final Rule "reduces the number of . . . joint employers . . .," it would not affect "the wages due the employee under the [FLSA] nor the [primary] employer's liability for [those] wages due[.]" Id. In other words, an employee could still "collect the entire wages due from [her primary] employer." Id. Thus, "assuming that all employers always fulfill their legal obligations under the [FLSA]," the Final Rule would not reduce employee wages. Id. at 14054-55.

This is silly. True, if we "assum[e] that all employers always fulfill their legal obligations[,]" the Final Rule will not make employees worse off. Id. But employers do not uniformly fulfill their legal obligations. That is the whole point of joint employer liability: Workers can recover from a joint employer when their primary employer flakes on its legal obligations. If employers always fulfill their legal obligations, the Final Rule serves no purpose. Cf. Reyes, 495 F.3d at 409 ("If everyone abides by the law, treating a firm . . . as a joint employer will not increase its costs.").

In the Final Rule, the Department retreated from this untenable assumption. It acknowledged that the Final Rule "may reduce the number of businesses . . . found to be joint employers[.]" 85 Fed. Reg. at 2853. That might "reduce the amount of back wages that employees [can] collect when their [primary] employer does not comply with the [FLSA] and, for example, . . . is or becomes insolvent." Id.

But the Department did not try to account for this effect. The Final Rule notes that EPI "submitted a quantitative analysis of transfers, estimating that transfers will result from both an increase in workplace fissuring and increased losses due to wage theft by employers." Id.; see also EPI Comment, Dkt No. 94-3. The Department "appreciate[d] EPI's quantitative analysis[.]" 85 Fed. Reg. at 2853. "[B]ut [it] d[id] not believe there are data to accurately quantify" these costs. Id. "The Department lack[ed] data on" how many joint employers currently exist. Id. And it could not "estimate the[ir] financial capabilities (or lack thereof)[.]" Id. So the Department was "unable to estimate the magnitude of a decrease" in joint employers because of the Final Rule. Id.

This is not a satisfactory explanation. To be sure, an agency is required to "give only a 'reasoned explanation' for its decision, not necessarily a quantitative one." XY Planning, 963 F.3d at 256 n.10 (citations omitted). And agencies are "not required to parse costs and benefits precisely." USCIS, 408 F. Supp. 3d at 1106. But even if the "exact harms are . . . difficult to predict," the agency may not "'disregard[] [an] effect entirely.'" Id. (quoting Pub. Citizen v. Fed. Motor Carrier Safety Admin., 374 F.3d 1209, 1219 (D.C. Cir. 2004)). "The mere fact that the magnitude of [a policy's] effect[] is uncertain" does not "justif[y]" ignoring that effect. Pub. Citizen, 374 F.3d at 1219 (emphasis omitted).

Yet that is what the Department did. The Final Rule concedes that it might reduce employees' ability to collect back wages owed to them. But the Department did not account for that effect because it was "unable to estimate the magnitude of" the decrease in joint employers under the Final Rule. 85 Fed. Reg. at 2853. In doing so, the Department almost "entirely failed to consider" the cost to workers, an "important aspect of the problem." DACA, 140 S. Ct. at 1913 (quoting State Farm, 463 U.S. at 43).

The Department's inability-to-quantify rationale is especially unpersuasive because the Department failed to quantify the supposed benefits of the Final Rule. The Department promulgated the Final Rule "to promote certainty for employers and employees, reduce litigation, promote greater uniformity among court decisions, and encourage innovation in the economy." 85 Fed. Reg. at 2820. It quantified none of those benefits. So why was the Department entitled to count unquantifiable and uncertain benefits on one side of the ledger but ignore them on the other? That double standard is unreasonable. And it suggests that the Department's inability to quantify the Final Rule's cost to workers was not the true reason that it ignored those costs.

The Department also failed to explain why it believed EPI's estimate was wrong. EPI estimated that the Final Rule would cost workers $1 billion per year. EPI Comment at 10. The Department did not have to agree with that estimate. It rests on reasonably debatable assumptions. Nor was the Department required to respond to every comment or precisely quantify every cost and benefit of the Final Rule. But by ignoring EPI's estimate, the Department effectively assumed that the Final Rule would cost workers nothing—an obviously unreasonable assumption. The Department should have explained why EPI's estimate was so bad that it preferred to use no estimate at all.

To be clear, the Department's justifications for engaging in rulemaking are valid. Promoting uniformity and clarity given the (at least superficially) widely divergent tests for joint employer liability in different circuits is a worthwhile objective. The Court is sympathetic to the Department's concern that putative joint employers face uncertainty, and that this uncertainty is costly. This opinion does not imply that the Department cannot engage in rulemaking to try to harmonize joint employer standards.

But the Department must do better than this. Any future rulemaking must adhere to the text of the FLSA and Supreme Court precedent. If the Department departs from its prior interpretation, it must explain why. And it must make more than a perfunctory attempt to consider important costs, including costs to workers, and explain why the benefits of the new rule outweigh those costs. Because the Final Rule does none of these things, it is legally infirm.

D. Severability

Although the Final Rule's test for "vertical" joint employer liability must be set aside under the APA, the Court will permit the Department's revisions to the "horizontal" joint employment scenario to remain in effect. See 2016 AI, 2016 WL 284582, at *2 (distinguishing between "vertical" and "horizontal" joint employment scenarios); see also n.2, supra.

A court may sever an unlawful portion of a rule under the APA. "The APA requires courts to 'hold unlawful and set aside agency action' that is not in accordance with law[.] . . . 'Agency action' may include 'the whole or a part of an agency rule.'" New York v. United States Dep't of Labor, No. 20-cv-3020 (JPO), 2020 WL 4462260, at *13 (S.D.N.Y. Aug. 3, 2020) (quoting 5 U.S.C. §§ 551(13), 706(2)). "Thus, the APA permits a court to sever a rule by setting aside only the offending parts[.]" Id. (quoting Carlson v. Postal Regulatory Comm'n, 938 F.3d 337, 351 (D.C. Cir. 2019)). The "invalid part of a . . . regulation may be dropped if what is left is fully operative as a law, absent evidence that the agency would not have enacted" the lawful portions "independently" of those that are unlawful. Id. (quoting United States v. Smith, 945 F.3d 729, 738 (2d Cir. 2019)) (alterations omitted). And the Final Rule has a severability clause:

If any provision of this part is held to be invalid or unenforceable by its terms, or as applied to any person or circumstance, or stayed pending further agency action, the provision shall be construed so as to continue to give the maximum effect to the provision permitted by law, unless such holding shall be one of utter invalidity or unenforceability, in which event the provision shall be severable from part 791 and shall not affect the remainder thereof.
29 C.F.R. § 791.3.

The Court vacates the Final Rule's novel standard for vertical joint employer liability. For the reasons set forth above, the Final Rule's revisions to that scenario are flawed in just about every respect. "The APA violations that the Court has found . . . are numerous, fundamental, and far-reaching." HHS, 414 F. Supp. 3d at 577. "[A] decision to leave standing isolated shards of the Rule that [are not] specifically infirm would ignore the big picture: that [it] was . . . shot through with glaring legal defects[.]" Id.

But the Final Rule's changes to horizontal joint employer liability are severable. The Final Rule makes only "non-substantive revisions" to existing law for horizontal joint employer liability. 85 Fed. Reg. at 2844. These changes are codified at 29 C.F.R. § 791.2(e). They can function independently from the changes to vertical joint employer liability. And there is no evidence that the Department would not have promulgated these revisions independently of the revisions to vertical joint employment. Thus, the revised 29 C.F.R. § 791.2(e) remains in effect.

IV. CONCLUSION

The parties' cross motions for summary judgment are GRANTED in part and DENIED in part. The Department's novel interpretation for vertical joint employer liability conflicts with the FLSA and is arbitrary and capricious. But the Department's non-substantive revisions to horizontal joint employer liability are severable, so 29 C.F.R. § 791.2(e) remains in effect. The Court vacates the rest of the revised 29 C.F.R. § 791.2.

The Clerk of Court is directed to terminate the motions pending at Dkt Nos. 91, 103, and 106, to grant judgment to Plaintiffs in part and Defendants in part, and to close this case.

SO ORDERED. Dated: September 8, 2020
New York, New York

/s/_________

GREGORY H. WOODS

United States District Judge

Id.; see also Guaraca v. Cafetasia Inc., No. 17-cv-1516 (VSB), 2018 WL 4538894, at *5 (S.D.N.Y. Sept. 20, 2018) ("Joint employers may either be vertical joint employers or horizontal joint employers." (citing Murphy v. Heartshare Human Servs. of N.Y., 254 F. Supp. 3d 392, 396 (E.D.N.Y. 2017))).

Reyes v. Remington Hybrid Seed Co., 495 F.3d 403, 407 (7th Cir. 2007) (quoting Lauritzen, 835 F.2d at 1539). These criticisms are apt, but the Court must apply the "economic reality" test because it is binding Supreme Court precedent. See Whitaker, 366 U.S. at 33.


Summaries of

New York v. Scalia

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK
Sep 8, 2020
1:20-cv-1689-GHW (S.D.N.Y. Sep. 8, 2020)
Case details for

New York v. Scalia

Case Details

Full title:STATE OF NEW YORK, COMMONWEALTH OF PENNSYLVANIA, STATE OF CALIFORNIA…

Court:UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK

Date published: Sep 8, 2020

Citations

1:20-cv-1689-GHW (S.D.N.Y. Sep. 8, 2020)