Chamber of Commerce of the United States of America

• Court: District Court, E.D. Texas
• Decided: March 18, 2024
• Docket: 6:23-cv-00553
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT EASTERN DISTRICT OF TEXAS

No. 6:23-cv-00553

Chamber of Commerce of the United States of America et al., Plaintiffs, v. National Labor Relations Board et al., Defendants.

OPINION AND ORDER Plaintiffs sue the National Labor Relations Board and its mem- bers, seeking relief from final agency action rescinding a regula- tion and replacing it with a new one governing status as a “joint employer” under the National Labor Relations Act. Now pending before the court are the parties’ cross-motions for summary judg- ment (Docs. 10, 34) and defendants’ motion to transfer the case (Doc. 25). For the reasons given below, plaintiffs’ motion for sum- mary judgment is granted, and defendants’ motions to transfer and for summary judgment are denied. Background 1. Almost 90 years ago, Congress enacted the National Labor Relations Act (NLRA), “encouraging the practice and procedure of collective bargaining” to resolve “industrial disputes arising out of differences as to wages, hours, or other working condi- tions.” 29 U.S.C. § 151. The National Labor Relations Board is charged with administering the Act. Id. § 153; NLRB v. SW Gen., Inc., 580 U.S. 288, 297 (2017). Section 7 of the Act defines employees’ rights to self-organi- zation and concerted action for the purpose of collective bargain- ing. 29 U.S.C. § 157. Section 8 of the Act, in turn, makes it an “unfair labor practice” for employers to interfere with employees’ exercise of their § 7 rights or to refuse to bargain collectively with employees’ representatives. Id. § 158(a). The obligation to bargain collectively is mutual, applying both to an employer and to em- ployees’ representative. Id. § 158(d). They must “meet at reason- able times and confer in good faith with respect to wages, hours, and other terms and conditions of employment, or the negotiation of an agreement.” Id. That duty also requires adherence to speci- fied procedures if a collective-bargaining agreement exists. Id. Section 9 of the Act entrusts to the Board certain employee- representation matters, such as the appropriate employer unit for collective bargaining. Id. § 159(b). The Board is similarly charged with deciding petitions regarding the identity of employees’ rep- resentative. Id. § 159(c). Section 10 of the Act, in turn, empowers the Board to issue orders preventing and curing unfair labor practices after receiving an allegation, taking evidence, and holding a hearing. Id. § 160(a)– (c). Section 10 also provides for judicial review to enforce or dis- pute such a Board order. Id. § 160(e)–(f). 2. Many rights and obligations flow from status as an “em- ployer” or “employee” under the Act. Determining who is and is not an “employer” of an individual is thus of great importance to workers, businesses, and labor unions. At first blush, that inquiry could seem straightforward. But it has been the subject of litigation since the early days of the Act. In 1944, the Supreme Court reasoned in NLRB v. Hearst Publica- tions, Inc., that “the broad language of the Act’s definitions . . . re- ject conventional limitations,” such that the employment classifi- cation should be “determined broadly, in doubtful situations, by underlying economic facts rather than technically and exclusively by previously established legal classifications.” 322 U.S. 111, 129 (1944). Hearst rejected an interpretation limited by “technical concepts” sounding in the common law of agency. Id.; see id. at 128 n.27 (“Control of ‘physical conduct in the performance of the service’ is the traditional test of the ‘employee relationship’ at common law.”) (citing Restatement (First) of the Law of Agency § 220(1) (1933)). Instead, Hearst held, “Where all the conditions of the relation require protection, protection ought to be given.” Id. at 129 (quotation marks omitted). Under that economic-reali- ties test, the Act’s reach was “not confined exclusively to ‘em- ployees’ within the traditional legal distinctions separating them from ‘independent contractors.’” Id. at 126. Congress responded adversely to Hearst’s economic-realities test by amending the Act in 1947. In the Taft–Hartley Act, Pub. L. No. 80-101, 61 Stat. 136 (1947), Congress first amended the statutory definition of “employer.” It previously covered persons “acting in the interest of any employer,” but Congress changed the definition to those “acting as an agent of an employer.” 29 U.S.C. § 152(2). Congress then changed the definition of “em- ployee” to specifically exclude an independent contractor. Id. § 152(3). As the Supreme Court later explained, “The obvious purpose of this amendment was to have the Board and the courts apply general agency principles in distinguishing between employees and independent contractors under the Act.” NLRB v. United Ins. Co. of Am., 390 U.S. 254, 256 (1968). Courts must therefore “ap- ply the common-law agency test here in distinguishing an em- ployee from an independent contractor.” Id.; accord NLRB v. Town & Country Elec., Inc., 516 U.S. 85, 94 (1995) (explaining that Con- gress “intended to describe the conventional master-servant rela- tionship as understood by common-law agency doctrine”) (quo- tation marks omitted); Local 777, Democratic Union Org. Comm. v. NLRB, 603 F.2d 862, 880 (D.C. Cir. 1978) (noting that the Taft– Hartley Act’s legislative history provides “clear evidence that Congress did not intend that an unusually expansive meaning should be given to the term ‘employee’ for the purpose of the Act”). 3. The Act does not expressly address whether a given em- ployee may have more than one employer. But the Board and the courts have held that two entities may be joint employers of the same employees—and thus must each collectively bargain. See, e.g., Franklin Simon & Co., 94 N.L.R.B. 576, 579 (1951); Boire v. Greyhound Corp., 376 U.S. 473, 481 (1964). In Boire, for instance, a corporation (Floors, Inc.) provided cleaning and maintenance services to various customers and en- tered into a contract with Greyhound to service four bus stations. 376 U.S. at 475. Floors hired, paid, disciplined, transferred, pro- moted, and discharged the individuals who performed that work— and was thus their employer. Id. But the Supreme Court recog- nized open fact questions about whether Greyhound exercised so much control over the workers as to also qualify as their employer, jointly with Floors. Id. at 474–76, 481. On remand the Fifth Cir- cuit held that Greyhound and Floors were indeed “joint employ- ers” of the workers in question based on findings that Greyhound sufficiently shared or codetermined essential terms and condi- tions of their employment. NLRB v. Greyhound Corp., 368 F.2d 778, 780–81 (5th Cir. 1966). Of course, the Act itself does not create a special status as a “joint employer.” Rather, the Act’s classification that triggers col- lective-bargaining duties and rights is either “employer” or “em- ployee” as opposed to “independent contractor.” The label “joint employer” arises from case law and simply means that multiple entities each qualify as an “employer” of the same group of “em- ployees” within the statute’s agency-law meaning of the terms. Courts also widely treat the “joint employer” doctrine as dif- ferent than the “single employer” doctrine. The former applies when two entities, otherwise independent, share or codetermine sufficient control over the same group of workers as to each qual- ify as the workers’ employer. Boire, 376 U.S. at 481. The latter ap- plies when nominally separate companies are so intertwined as to form a single, integrated entity—as with a group of radio stations each owned and operated by one broadcasting service. Radio & Television Broad. Technicians Loc. Union 1264 v. Broad. Serv. of Mo- bile, Inc., 380 U.S. 255, 256 (1965). In making that single-employer determination, the Board examines four criteria: “interrelation of operations, common management, centralized control of labor re- lations[,] and common ownership.” Id. Not every circuit clearly treats the two doctrines as distinct. The Eighth Circuit has applied the four-factor “single employer” test to determine “joint employer” status as well. See Pulitzer Publ’g Co. v. NLRB, 618 F.2d 1275, 1278–79 (8th Cir. 1980); Misc. Drivers & Helpers Union, Local No. 610 v. NLRB, 624 F.2d 831, 833 (8th Cir. 1980) (declining to modify that court’s Pulitzer Publish- ing test); NLRB v. C.R. Adams Trucking, Inc., 718 F.2d 869, 870 (8th Cir. 1983) (reciting the four-factor Pulitzer Publishing test for joint-employer status, although finding dispositive only the sec- ond factor of “control over the employment conditions”). But a leading Third Circuit case has persuaded many courts that the “‘joint employer’ and ‘single employer’ concepts are dis- tinct.” NLRB v. Browning-Ferris Indus. of Pa., Inc., 691 F.2d 1117, 1122 (3d Cir. 1982) (BFI I). In that case, the Third Circuit upheld the Board’s conclusion that waste-management company BFI was, along with trucking brokers used by BFI, a “joint employer” of certain truck drivers. Id. at 1119. BFI argued that “joint employer” status should be decided us- ing the same, four-factor test used to decide “single employer” status. Id. at 1122. The Third Circuit disagreed. “Single em- ployer” status, the court held, “is characterized as an absence of an arm’s length relationship found among unintegrated compa- nies” and ultimately turns on all the circumstances of the case. Id. (quotation marks omitted). But “joint employer” status “assumes in the first instance that companies are what they appear to be— independent legal entities that have merely historically chosen to handle jointly important aspects of their employer-employee rela- tionship.” Id. (quotation and ellipsis marks omitted). Accordingly, the court held, “joint employer” status does not require unity of ownership or closer than an arm’s length relation- ship between two companies. Id. Instead, the test focuses only on each alleged employer’s practices regarding the workers in ques- tion: Where “two or more employers exert significant control over the same employees—where from the evidence it can be shown that they share or co-determine those matters governing essential terms and conditions of employment—they constitute ‘joint em- ployers.’” Id. at 1124. The Third Circuit then upheld the Board’s application of that test based on the agency’s specific findings in the case. Id. After that decision, other circuits and the Board have largely settled on a “joint employer” test focusing, not on whether two alleged employers of the same workers share corporate ownership or management, but exclusively on whether each alleged employer “meaningfully affects matters relating to the employment rela- tionship such as hiring, firing, discipline, supervision, and direc- tion.” E.g., Teamsters Local Unions Nos. 75 & 200 v. Barry Truck- ing, Inc., 176 F.3d 1004, 1008–09 (7th Cir. 1999) (quoting Osco Drug v. Truck Drivers, Oil Drivers, Local Union 705, 294 N.L.R.B. 779, 785 (1989)). And, over time, two limiting principles on that test have solidified: (a) the potential-control limit and (b) the indi- rect-control limit. a. First, under the potential-control limit, an entity cannot be deemed a joint employer simply because it had a contractual right to control workers, where the entity did not actually exercise that contractual right of control. Actual control, not potential control, is required for employer status—including joint-employer status. TLI, Inc., 271 N.L.R.B. 798, 798 (1984). That limit follows from the Third Circuit’s focus in BFI I on whether an alleged joint em- ployer actually “exerted” sufficient control over the workers in question. BFI I, 691 F.2d at 1125. The joint-employer test thus “does not rely merely on the existence of such contractual provi- sions [granting a right to control essential employment terms], but rather looks to the actual practice of the parties.” AM Prop. Hold- ing Corp., 350 N.L.R.B. 998, 1000 (2007). b. Second, under the indirect-control limit, an entity cannot be deemed a joint employer based on mere indirect control of a worker’s terms and conditions of employment. Rather, “direct and immediate” control is required for joint-employer status. Air- borne Freight Co., 338 N.L.R.B. 597, 597 n.1 (2002) (holding that a putative second employer must have “control over employment matters” that is “direct and immediate”). That limit sounds in BFI I’s requirement of “significant” con- trol over the relevant employees. BFI I, 691 F.2d at 1125. Because it focuses not on counting how many intermediaries may convey a direction, but on who significantly causes adherence, that indi- rect-control limit is akin to tort law’s proximate-causation limit. In an interconnected market, many third-party needs may have a cause-in-fact relationship to how an employer sets an employee’s terms and conditions of employment. See Airborne Freight, 338 N.L.R.B. at 606 (A.L.J. opinion) (rejecting a “de facto control” test that, “if extended to its logical conclusion, would mean that in virtually all contractor–subcontractor relationships, the two companies involved should necessarily be construed as joint em- ployers”). But just as tort liability generally requires proximate causation as well as causation in fact, proximate control is required to qualify as an employer. The Board thus held for decades that the “essential element in this analysis is whether a putative joint employer’s control over employment matters is direct and imme- diate.” Id. at 597 n.1. Customer preferences that do in fact influ- ence an employer’s control over its employees due to market forces (i.e., “economic realities”) are not enough. The indirect-control limit also means that a customer cannot be deemed a joint employer based only on “limited and routine” supervision of the employees of an independent company serving the customer. TLI, 271 N.L.R.B. at 799. The Board has recognized that a customer necessarily exercises some oversight to ensure that a company delivers the agreed-upon services in a way that does not unduly interfere with the customer’s pursuits. Thus, a customer’s limited and routine participation in the “daily opera- tional activities” of a provider company’s undisputed employees does not “constitute sufficient control to support a joint employer finding.” Id. (citing Laerco Transp., 269 N.L.R.B. 324 (1984)); ac- cord So. Cal Gas Co., 302 N.L.R.B. 456, 461 (1991). 4. That was the landscape before the Board pivoted in a 2015 order. But that order was reversed on judicial review. And a rule- making by the Board followed. a. That pivot occurred in the adjudication of Browning-Ferris Industries of California, Inc., 362 N.L.R.B. 1599 (2015) (BFI II). There, the Board stated the issue in remarkably similar terms to Hearst’s economic-realities test: whether to update the joint-em- ployer standard to reflect “the current economic landscape.” Id. at 1599. By a 3–2 vote, the Board decided that the then-existing joint-employer test was “increasingly out of step with changing economic circumstances,” id., providing “reason enough” to re- visit it despite Congress not modifying the relevant statutory lan- guage since the 1947 amendments, id. at 1609. The Board began by arguing that the Third Circuit’s review of the law in BFI I reflected a two-part test: In determining whether a putative joint employer meets this standard, the initial inquiry is whether there is a com- mon-law employment relationship with the employees in question. If this common-law employment relationship ex- ists, the inquiry then turns to whether the putative joint employer possesses sufficient control over employees’ es- sential terms and conditions of employment to permit meaningful collective bargaining. Id. at 1600. The Board then eliminated the requirement that a joint employer’s relationship to another company’s employees must include proximate (not indirect) control, exercised in prac- tice (not merely reserved): [The Board] will no longer require that a joint employer not only possess the authority to control employees’ terms and conditions of employment, but must also exercise that authority, and do so directly, immediately, and not in a “limited and routine” manner. Accordingly, we overrule Laerco, TLI, A&M Property, and Airborne Express, supra, and other Board decisions, to the extent that they are in- consistent with our decision today. Id. at 1613–14. The dissenting Board members read the common law to re- quire actual, proximate control over workers to create an employ- ment relationship, even if evidence of lesser influence over work- ers could be considered as reinforcing that proof of an employ- ment relationship: Our fundamental disagreement with the majority’s test is not just that they view indicia of indirect, and even poten- tial, control to be probative of employer status, they hold such indicia can be dispositive without any evidence of di- rect control. Under the common law, in our view, evidence of indirect control is probative only to the extent that it supplements and reinforces evidence of direct control. Id. at 1620. The Board’s reformulated test also included a second step: even after an alleged joint employer of another company’s em- ployees qualifies as their common-law employer, the alleged joint employer must also “share or codetermine those matters govern- ing the essential terms and conditions of employment.” Id. at 1613. The Board did not explain how the second step adds any- thing to the first. Nor is it clear that the Third Circuit understood the “share or codetermine” test as anything but a restatement of the common-law employment test in the context of an alleged sec- ond employer. See BFI I, 691 F.2d at 1123 (presenting the share-or- codetermine test as requiring separate “business entities”—not separate common-law “employers”—taking that action to qualify as an employer). Finally, the Board applied its revised standard to a “supplier firm” (Leadpoint) that contracted to sort materials and clean and maintain equipment and premises for a “user firm” (BFI) that op- erated a recycling plant. BFI II, 362 N.L.R.B. at 1600. After a re- view of BFI’s contractual rights and actual practices regarding Leadpoint’s employees, the Board concluded that BFI was also their employer, jointly with Leadpoint. Id. at 1617–18. b. The D.C. Circuit then reviewed the Board’s order. Brown- ing-Ferris Indus. of Cal., Inc. v. NLRB, 911 F.3d 1195 (D.C. Cir. 2018) (BFI III). While that review was pending, the Board an- nounced that it planned to undertake rulemaking on the joint-em- ployer standard. Id. at 1206. Because any new rule would be pro- spective only, the D.C. Circuit proceeded to review the Board’s order that Browning-Ferris engaged in an unfair labor practice. Id. The D.C. Circuit began by holding that Chevron deference is not due to the Board’s view of common-law principles defining status as an “employer” or “employee,” as those are pure ques- tions of law that courts review de novo. Id. at 1206–08. The court explained: the Board “must color within the common-law lines identified by the judiciary.” Id. at 1208. On the merits, the court first rejected the argument that the employees of a customer’s “independent contractor” necessarily cannot be that customer’s “employees” because the two statuses are mutually exclusive. Id. at 1213–16. The court analogized the statuses to different hand tools whose capabilities are not mutu- ally exclusive, even though “independent contractor” is in fact an exclusion from the statutory term that triggers collective-bargain- ing duties—“employee.” Id. at 1215. The court then blessed “consideration” of two types of evi- dence as “relevant” to joint-employment status: (1) unexercised, retained control over workers and (2) indirect control over work- ers, at least insofar as that means control exercised through an in- termediary. Id. at 1210–18. But the court did not decide whether either is sufficient to show joint-employer status. And the court acknowledged that “whether indirect control can be dispositive is not at issue.” Id. at 1218 (quotation marks omitted). The court reversed the Board, however, for failing to articulate any “blueprint for what counts as ‘indirect’ control.” Id. at 1220. The court accepted that the joint-employer test will turn on the facts of each distinct case. But because the Board’s order began by stating a “governing common-law test,” the court required the Board to “erect some legal scaffolding that keeps the inquiry within traditional common-law bounds.” Id. The court made es- sentially a due-process point: If the Board gives no specifics on how indirect-control evidence can show status as a joint employer, a court “cannot tell” how to review the Board’s work in a given case. Id. at 1221. See generally LeMoyne-Owen Coll. v. NLRB, 357 F.3d 55, 61 (D.C. Cir. 2004) (Roberts, J.) (“Lack of definite stand- ards creates a void into which attempts to influence are bound to rush; legal vacuums are quite like physical ones in that respect.” (quoting Henry J. Friendly, Benchmarks 104 (1967)). And that lack of clarity was material. The Board’s broad inclu- sion of indirect control failed to distinguish and exclude “quotid- ian aspects of common-law third-party contract relationships,” which do not show an employment relationship. BFI III, 911 F.3d at 1220. But the court recognized and approved the limited-and- routine-supervision limit on acts creating joint-employer status: By contrast [to a user’s control that codetermines essential terms and conditions of employment,] decisions that set the objectives, basic ground rules, and expectations for a third-party contractor cast no meaningful light on joint- employer status. Id. For example, held the court, a “cost-plus contract” setting the costs for which a customer will reimburse a service provider is ex- cluded from analysis as merely routine, indirect control over the provider’s determination of its employees’ terms and conditions of employment. Id.; see, e.g., Goodyear Tire & Rubber Co., 312 N.L.R.B. 674, 677 (1993) (ruling that a contract that “sets forth the wage reimbursement schedules under which Goodyear paid TU for the cost of labor which TU provided” is “nothing more than a ‘cost-plus’ contract” and is not “what the Board or the courts had in mind when using the phrase ‘share and codeter- mine’ essential terms such as wages”). As a parting note, the D.C. Circuit criticized the Board’s two- step framing of the joint-employer test. In fact, the court held that the action of “sharing or codetermining” essential terms and con- ditions of employment is what makes two companies each an employer of the same workers in the first place. BFI III, 911 F.3d at 1201 (stating that “separate business entities” are joint employ- ers if they each exert significant control over the same employees “in that” they share or codetermine essential employment condi- tions). In contrast, the Board’s order on review framed the act of sharing or codetermining essential terms as an additional joint- employer requirement, for companies that already qualify as em- ployers of the same workers under the common law. Even then, the Board “did not meaningfully apply the second step of its test.” Id. at 1221. The D.C. Circuit expected some ex- planation of what terms and conditions of employment are essen- tial under the Board’s second step for an employer to control, alt- hough not essential under the common-law test. Id. at 1222. And if such a narrowed list were defined by reference to a need for “meaningful” collective bargaining, the court expected an expla- nation of what that concept means and how it works. Id. Judge Randolph dissented, opining that the majority should not have issued a merits opinion given the pending rulemaking. Id. at 1223. He also opined that the majority “misstates the com- mon law, misframes the questions in the case, and adds to the un- certainty the [Board] has generated.” Id. Specifically, he con- cluded that the Board “overturned decades of settled law” be- cause “[d]irect and immediate control of employees, not just in- direct control or potential control, had been required before a company could be deemed a joint employer of another company’s employees for the purposes of collective bargaining.” Id. c. In February 2020, before issuing a new order on remand after BFI III, the Board’s rulemaking yielded a final rule on joint- employer status. Joint Employer Status Under the National Labor Relations Act, 85 Fed. Reg. 11,184 (Feb. 26, 2020), codified at 29 C.F.R. § 103.40. That rule provides that an entity is “a joint em- ployer of a separate employer’s employees only if the two employ- ers share or codetermine the employees’ essential terms and con- ditions of employment.” 29 C.F.R. § 103.40(a) (2020). To meet that test as to another employer’s employees, the second entity “must possess and exercise such substantial direct and immediate control over one or more essential terms or conditions of their employment as would warrant finding that the entity meaningfully affects matters relating to the employment relationship with those employees.” Id. And the 2020 Rule provides that indirect control and purely reserved control can be considered but are not, them- selves, sufficient to show joint-employer status: Evidence of the entity’s indirect control over essential terms and conditions of employment of another employer’s employees, the entity’s contractually reserved but never ex- ercised authority over the essential terms and conditions of employment of another employer’s employees, or the entity’s control over mandatory subjects of bargaining other than the essential terms and conditions of employment is probative of joint-employer status, but only to the extent it supplements and reinforces evidence of the entity’s posses- sion or exercise of direct and immediate control over a par- ticular essential term and condition of employment. Id. The rule then defines eight discrete “essential terms and con- ditions of employment”: wages, benefits, hours of work, hiring, discharge, discipline, supervision, and direction. Id. § 103.40(b). For each term of employment, the rule specifies what counts as direct control and what does not. E.g., id. § 103.40(c)(1) (“An en- tity exercises direct and immediate control over wages if it actu- ally determines the wage rates, salary or other rate of pay that is paid to another employer’s individual employees or job classifica- tions. An entity does not exercise direct and immediate control over wages by entering into a cost-plus contract [with or without a maximum reimbursable wage rate].”). That rule is the operative joint-employment standard today (hence, the “Current Rule” or “2020 Rule”). It has been applied by the Board once, resulting in a finding of joint employment. Cog- nizant Tech. Sol’ns U.S. Corp., 372 N.L.R.B. No. 108 (2023) (Case No. 16-CA-326027), appeal docketed, No. 24-1003 (D.C. Cir. Jan. 5, 2024). Administrative law judges within the agency have also applied the rule. See, e.g., 3 Corners, LLC, 2023 WL 6226274 (NLRB Div. of Judges Sept. 25, 2023). 5. In 2023, a newly constituted Board issued a final rule to rescind and replace the 2020 Rule with a new standard for joint- employer status. Standard for Determining Joint Employer Status, 88 Fed. Reg. 73,946 (Oct. 27, 2023), to be codified at 29 C.F.R. § 103.40 (“the New Rule,” “the 2023 Rule,” or “the Rule”). The New Rule first rescinds the Current Rule. Id. at 74,017 (removing 29 C.F.R. part 103, subpart D, consisting of the 2020 Rule). The New Rule then adds a 29 C.F.R. part 103, subpart E, consisting of a new § 103.40. Id. The mechanics of the New Rule are disputed by the parties. But at least four changes are evident: • Reserved control alone meets at least one step (maybe the only step) of the test for joint-employer status, whereas it did not before. New Rule § 103.40(e)(1) • Indirect control meets at least one step (maybe the only step) of the test for joint-employer status, whereas it did not before. New Rule § 103.40(e)(2). • Two broad categories are added to the 2020 Rule’s list of essential terms and conditions of employment: (1) “[w]ork rules and directions governing the manner, means, and methods of the performance,” and (2) “[w]orking condi- tions related to the safety and health of employees.” New Rule § 103.40(d). • The Board eliminated the 2020 Rule’s provision that con- trol over workers “exercised on a sporadic, isolated, or de minimis basis” is not sufficient to establish joint-employer status. 2020 Rule § 103.40(d). Plaintiffs challenge the 2023 Rule on two grounds. First, they argue that it is inconsistent with the common law, which all par- ties agree is the benchmark. Doc. 10 at 10; Doc. 34 at 9. Second, plaintiffs argue that the 2023 Rule is arbitrary and capricious for ignoring serious practical problems and failing to articulate a comprehensible standard, with meaningful guidance to the regu- lated parties. Plaintiffs’ standing is undisputed and apparent, see Doc. 31 at 1, and they move for summary judgment on both grounds. Defendants move to transfer this case to the D.C. Circuit, ar- guing that this court lacks jurisdiction because any challenge to a joint-employer rule belongs only in a court of appeals. Doc. 25. In the alternative, defendants move for summary judgment in their favor. Doc. 34. The Board postponed the Rule’s effective date to February 26, 2024. Press Release, NLRB, Board Extends Effective Date of Joint-Employer Rule to February 26, 2024 (Nov. 16, 2023). This court then further postponed the Rule’s effective date to March 11, 2024. Doc. 43. As the challenged rule has not yet taken effect, the 2020 Rule remains the operative joint-employer regulation, despite any changes already made in the Code of Federal Regula- tions. Analysis The court first denies defendants’ motion to transfer, holding that 29 U.S.C. § 160(f)’s review provision concerns only Board orders adjudicating unfair-labor-practice allegations and, there- fore, does not displace the traditional route of review provided by the Administrative Procedure Act. The court then grants plain- tiffs’ motion for summary judgment and denies defendants’ cross- motion for summary judgment. 1. Motion to transfer for lack of jurisdiction Under 28 U.S.C. § 1331, federal district courts “shall have original jurisdiction of all civil actions” arising under federal law. And APA actions for review of federal agency rulemaking “arise under” federal law. See 5 U.S.C. § 702. That much is undisputed. It is also undisputed that the National Labor Relations Act does not expressly repeal district courts’ § 1331 jurisdiction over APA claims for review of Board rulemaking. But defendants main- tain that the Act implicitly repeals that jurisdiction. They argue that its § 10(f)—which authorizes review in specified courts of appeals, but not district courts, of certain Board “orders”—ap- plies here because the challenged rule is such an order. Implicit jurisdiction-stripping depends on whether it is “fairly discernible” from the statute that Congress silently divested the jurisdiction that § 1331 confers. E.g., Block v. Cmty. Nutrition Inst., 467 U.S. 340, 351 (1984). “To determine whether it is ‘fairly dis- cernible’ that Congress precluded district court jurisdiction over petitioners’ claims, we examine the [Act’s] text, structure, and purpose.” Elgin v. Dep’t of Treasury, 567 U.S. 1, 10 (2012). Here, the Act defines a specialized procedure for accessing courts to enforce or dispute Board adjudications of unfair-labor- practice allegations. But neither the text, structure, nor purpose of the Act supports defendants’ extension of that procedure to cover judicial review of prospective Board rulemaking. a. Section 10 of the Act authorizes the Board to prevent un- fair labor practices affecting commerce. 29 U.S.C. § 160(a). The Board’s authority begins when someone files a charge alleging an “unfair labor practice.” Id. § 160(b). After receiving evidence and holding a hearing, the Board reaches its final decision on whether any person named in the complaint engaged in an unfair labor practice. Id. § 160(c). If so, the Board can order the person to cease and desist from the practice and to take affirmative remedial action, such as reinstatement of employees. Id. Section 10 also provides for judicial review if (1) the Board pe- titions for a temporary injunction or for enforcement of a final or- der remedying unfair labor practices, id. § 160(e), or (2) a person aggrieved by a Board order granting or denying relief from an un- fair labor practice petitions for judicial review, id. § 160(f). In defining a specialized procedure for judicial review, subsec- tions (e) and (f) each allow access to a court “wherein the unfair labor practice in question” occurred (even allegedly) or where an aggrieved person “resides or transacts business.” Id. § 160(e), (f). Those provisions use permissive language, not mandatory language, offering a choice of local venues to the Board or an ag- grieved party. Id. The purpose evident from those provisions is allowing access to a convenient court, local to the parties or practices in question. Nothing about that purpose fairly justifies discernment of a con- gressional intent to displace § 1331 jurisdiction to hear APA chal- lenges to rulemakings rather than adjudications, especially since the venue statute governing APA challenges also affords plaintiffs a choice among convenient options. See 28 U.S.C. § 1391(e)(1) (al- lowing venue where a plaintiff resides or where a substantial part of relevant events occurred). Unsurprisingly, then, § 10’s provisions governing judicial re- view mention—not prospective rulemaking—but only “orders” for temporary or final relief against a person charged with engag- ing in an unfair labor practice: • Section 10(e) allows the Board to petition for judicial en- forcement of “such order,” referring to an order of the Board under § 10(c) “requiring such person [one named in the complaint] to cease and desist from such unfair labor practice, and to take such affirmative action including re- instatement of employees with or without back pay.” 29 U.S.C. § 160(c), (e). • Section 10(f) allows a petition for review of a final “order” of the Board “granting or denying” the relief sought, by a person aggrieved by “such order.” Id. § 160(f). • Section 10(j) allows the Board, only “upon issuance of a complaint as provided in subsection (b) charging that any person has engaged in or is engaging in an unfair labor practice,” to petition a court for “appropriate temporary relief or restraining order.” Id. § 160(j). Not one of those judicial-review provisions mentions rulemaking. And Congress knew how to do so, as § 6 of the Act expressly grants the Board authority to issue “rules and regulations.” Id. § 156. That contrast between § 6’s separate discussion of rule- making and § 10’s specialized scheme for issuance and review of orders adjudicating allegations against specific parties is a strong textual and structural sign that § 10’s judicial-review provisions do not cover Board rulemaking. b. All other signs point to the same conclusion. Section 10(f) defines a review scheme for “a final order of the Board granting or denying in whole or in part the relief sought.” Id. § 160(f) (empha- ses added). That precisely describes the posture of a Board adju- dication of a charge that a person engaged in an unfair labor prac- tice. A final order resolving such a charge either grants or denies relief such as a cease-and-desist order, reinstatement, or back pay. Id. § 160(c). But those statutory qualifiers do not easily apply to rulemak- ing. The Board suggests that, if one squints at it, a rulemaking can be viewed as “granting or denying” the “relief sought” of creating or modifying a rule. Doc. 30 at 8. But that is not a natural reading. See AFL-CIO v. NLRB, 466 F. Supp.3d 68, 83 (D.D.C. 2020), rev’d on other grounds, 57 F.4th 1023 (D.C. Cir. 2023) (“[T]here is no reasonable argument that credibly casts the 2019 Election Rule as an agency action that grants or denies relief to any regulated party, and this problem alone is sufficient to cast doubt on the NLRB’s contentions that section 160(f) applies to the [chal- lenger’s] claims.”). In any event, the Board’s squinting view does not capture the whole picture. To be sure, some agency rules issue after a petition for rulemaking. And some rules issue after an agency receives public comment on a proposed rule. But not all rulemaking occurs that way. An agency can issue rules on its own initiative, without a petition for rulemaking. And an agency can issue rules even if no member of the public responds during the comment period, or if a comment period can be dispensed with. So even if it can some- times be said that a person has “sought” the “relief” of a certain type of agency rule, that will not always be true of a rule. The Board is thus arguing for a jurisdictional rule based on the procedural history of a given rulemaking—whether a rule could be said to have “granted” or “denied” some feature “sought” by a particular member of the public. The Board’s reading would thus deprive litigants and courts “of clear guidance about the proper forum for [a litigant’s] claims at the outset of the case” and would, instead, require production of the administrative record or perhaps discovery into whether the plaintiff had petitioned for or informally sought agency rulemaking. Elgin, 567 U.S. at 15. In con- trast, “a jurisdictional rule based on the type of . . . agency action at issue does not involve such amorphous distinctions” and is thus more likely correct. Id. (relying on that fact to adopt the more co- herent rule). A similar point applies to the Board’s reliance on the state- ment in AFL-CIO v. NLRB, 57 F.4th 1023 (D.C. Cir. 2023), that § 10(f) authorizes initial D.C. Circuit review of Board rules “con- cerning unfair labor practices” but not rules concerning “repre- sentation matters.” Id. at 1032. As an initial matter, that statement is dictum because the Board rule there concerned only “represen- tation matters.” Id. The court’s remark about other types of rules was not necessary to its decision. That statement is also unpersuasive because it drew on differ- ent statutory schemes without any analysis of the textual or struc- tural points discussed above regarding § 10(f). Id. And a closer examination of the statement undermines it further. The line be- tween a rule “concerning” unfair labor practices and a rule “con- cerning” representation matters is “hazy at best and incoherent at worst.” Elgin, 567 U.S. at 15. As the Board admits, the joint- employer rule at issue here “applies in both the representation- case and unfair-labor-practice-case contexts.” Doc. 25 at 10; see 88 Fed. Reg. at 74,017 (defining joint employment “[f]or all pur- poses under the Act”). So the Board’s distinction may involve as- certaining whether a particular plaintiff anticipates future appli- cation of the Rule in a representation matter or in a charge of un- fair labor practices. Again, as with the rule in Elgin, the unpredict- ability created by such a reading cuts against its persuasiveness. The Board’s reading also fits poorly with § 10(f)’s allowance for judicial review “in the circuit wherein the unfair labor practice in question was alleged to have been engaged in.” 29 U.S.C. § 160(f). Section 10(f) review contemplates a Board order resolv- ing a charge that an entity engaged in an unfair labor practice. Pro- spective rulemaking defining a legal standard does not resolve such an allegation and does not meet that description. In sum, although the word “order” may in some contexts in- clude an agency rule, the context here does not support that read- ing. See, e.g., Am. Petroleum Inst. v. SEC, 714 F.3d 1329, 1332–35 (D.C. Cir. 2013) (holding that the provision for review of “a final order of the Commission,” under § 25(a) of the Exchange Act did not cover challenges to rules because that statute’s text, structure, and history showed a contrary congressional intent). If the text of § 10(f) were not enough in isolation to reject the Board’s position, the Act’s overall text, structure, and purpose unambiguously de- feat any inference of an intent to displace § 1331 jurisdiction over challenges to Board rulemaking. That statute-specific conclusion overcomes any presumption said to favor initial review only by a court of appeals. See id. at 1332–35 (declining to follow a presump- tion that the D.C. Circuit recognizes). Defendants’ motion to transfer based on lack of jurisdiction (Doc. 25) is denied. 2. Cross-motions for summary judgment Plaintiffs challenge the 2023 Rule as contrary to law and as ar- bitrary and capricious. Doc. 1. All parties now move for summary judgment, agreeing that there are no genuine issues of material fact and that plaintiffs’ claims are ripe for resolution. Doc. 10 at 16; Doc. 34 at 9. a. The parties disagree, however, on the legal matter of what the Rule does. The interpretive dispute first turns on whether the new rule has a meaningful “step two” filter for qualifying as a joint employer, beyond subsection (a)’s initial requirement that a joint employer be a common-law “employer” of the relevant workers. Subsection (b) of the rule does purport to state an additional re- quirement. But the content of that second test is either coexten- sive with or a superset of the first test (which is the legal boundary within which the Board must color). What is more, subsection (e)’s broad language declares certain showings sufficient to estab- lish joint-employer status, independent of subsection (b)’s test. A walk through the Rule’s new § 103.40 explains the disagree- ment. Subsection (a) starts out well enough. It provides that one is an “employer” of particular “employees” within the Act’s def- initions of those terms if an employment relationship between the two exists under the common law of agency: (a) An employer, as defined by section 2(2) of the National Labor Relations Act (the Act), is an employer of par- ticular employees, as defined by section 2(3) of the Act, if the employer has an employment relationship with those employees under common-law agency prin- ciples. New Rule § 103.40, 88 Fed. Reg. at 74,017. Subsection (b) then states that two things are required to be a “joint employer” of particular employees: (1) being their em- ployer at all (under the common-law test), along with at least one other employer of them, and (2) sharing or codetermining matters governing their essential terms and conditions of employment: (b) For all purposes under the Act, [1] two or more em- ployers of the same particular employees are joint em- ployers of those employees if [2] the employers share or codetermine those matters governing employees’ essential terms and conditions of employment. Id. (brackets added). Subsection (c) states the Board’s position that even controlling indirectly, or possessing unexercised control over, the specified employment terms qualifies as sharing or co- determining those matters: (c) To “share or codetermine those matters governing em- ployees’ essential terms and conditions of employ- ment” means for an employer to possess the authority to control (whether directly, indirectly, or both), or to exercise the power to control (whether directly, indirectly, or both), one or more of the employees’ es- sential terms and conditions of employment. Id. And subsection (d) then lists seven broad categories as essen- tial terms and conditions of employment, including directions on the manner of work and working conditions related to worker safety and health: (d) “Essential terms and conditions of employment” are (1) Wages, benefits, and other compensation; (2) Hours of work and scheduling; (3) The assignment of duties to be performed; (4) The supervision of the performance of duties; (5) Work rules and directions governing the manner, means, and methods of the performance of duties and the grounds for discipline; (6) The tenure of employment, including hiring and discharge; and (7) Working conditions related to the safety and health of employees. Id. at 75,017–18. Given the text of subsection (b), the Board argues as a formal- ist matter that the “regulatory framework involves two steps”: first, an entity must “qualify as a common-law employer of the disputed employees,” and second, “only if the entity is a com- mon-law employer, then it must also have control over one or more essential terms and conditions of employment.” Doc. 34 at 9 (emphasis omitted). But plaintiffs respond that the second test is always met if the first test is met, so the Rule’s joint-employer inquiry has just one step for all practical purposes. That logical relation appears right. An employer of a worker under the common law of agency must have the power to control “the material details of how the work is to be performed.” NLRB v. Town & Country Elec., Inc., 516 U.S. 85, 90 (1995) (quotation marks omitted). And that control meets subsection (c)’s test of determining an essential term and condi- tion of employment listed in subsection (d), namely, “[w]ork rules and directions governing the manner, means and methods of the performance of duties.” New Rule § 103.40(d), 88 Fed. Reg. at 74,017. So it seems that an entity satisfying step one, along with some other entity doing so, will always satisfy step two. Indeed, the Board has not been able to come up with any ex- ample of an entity satisfying step one but not step two. The Board suggests in a footnote that “there may be a rare case where an en- tity is sufficiently involved with a group of workers to qualify as a common-law employer (particularly after an isolated instance of tort or insurance liability) but does not control an enumerated es- sential term.” Doc. 37 at 7 n.8. But that footnote does not explain how an entity could qualify as an employer for purposes of tort or insurance liability (even assuming that matches step one’s com- mon-law test) without having control over matters governing es- sential terms and conditions of employment. At oral argument, the Board was asked if the Rule’s subsec- tions that “define the concept of sharing or codetermining essen- tial terms” are “doing any filtering work to arrive at the set of par- ties who would qualify as joint employers[,] in addition to the fil- tering work already being done by subsection (a).” Doc. 42 (Hr’g Tr.) at 59. The Board replied that “there might be a hypothetical example” where the step-two test had an additional filtering role but was unable to back up that suggestion with any actual exam- ple. Id. at 60. To its credit, the Board candidly admitted that it “struggled with that[,] to come up with an answer to that ques- tion.” Id. The court thus accepts plaintiffs’ position that step two is met whenever step one is met. That makes step two either co- extensive with step one or a superset of step one (e.g., if step two goes beyond the common law in identifying essential employment terms and thus sweeps in more than step one does). Subsection (e) is another focus of interpretive dispute. Its first sentence simply confirms that common-law agency principles control. But its second sentence states two broad propositions about what suffices to establish status “as a joint employer” for purposes of new § 103.40: (e) . . . . For the purposes of this section: (1) Possessing the authority to control one or more es- sential terms and conditions of employment is suf- ficient to establish status as a joint employer, re- gardless of whether control is exercised. (2) Exercising the power to control indirectly (includ- ing through an intermediary) one or more essential terms and conditions of employment is sufficient to establish status as a joint employer, regardless of whether the power is exercised directly. 88 Fed. Reg. at 74,018. The Board argues that subsection (e) “ap- plies only to the second step of the inquiry (focusing on control over essential terms and conditions of employment as a prerequi- site for a joint-employer finding) and merely confirms that re- served or indirect control over an essential term is sufficient to satisfy this step.” Doc. 37 at 6. But that is not what subsections (e)(1) and (2) say. They do not say that the defined power to control (i.e., reserved control and indirect control) suffice to meet only one of two requirements to be a joint employer. They say that the defined power “is suffi- cient to establish status as a joint employer” for “purposes of this section”—the new § 103.40. Under subsections (e)(1) and (2), once the specified power is shown, joint-employer status attaches without any need to additionally demonstrate an employment re- lationship under the common law of agency. To the extent the Board relies on Federal Register passages to cabin the reach of subsections (e)(1) and (2), see 88 Fed. Reg. at 73,981–87, they are unpersuasive in light of the plain text of new § 103.40 itself. And the Board does not cite any authority requir- ing formal deference to an agency’s views in the Supplementary Information portion of a rulemaking about how a regulation is in- terpreted. Doc. 42 (Hr’g Tr.) at 70 (“I will admit Your Honor, I don’t have any authority upfront. I didn’t really anticipate that ques- tion . . . .”). Although the court gives the agency’s view respectful consideration, the cited passages make no good argument for bypassing subsection (e)’s text stating that exercising indirect control over, or possessing reserved authority to control, one or more essential terms is itself sufficient to establish status as a joint employer, making no allowance for a further test under the com- mon law of agency. b. Subsections (d) and (e) broadly classify many aspects of work as essential terms and conditions of employment and broadly classify many entities’ potential influence over those as- pects as sufficient to establish joint-employer status. So if an en- tity exercises or has the power to exercise control (even indirect control) over at least one essential term, the entity is an employer, jointly with workers’ undisputed employer. That would treat vir- tually every entity that contracts for labor as a joint employer be- cause virtually every contract for third-party labor has terms that impact, at least indirectly, at least one of the specified “essential terms and conditions of employment.” Consider an example. IceCo is an ice cream shop and con- tracts with MowCo, a lawn service, to tend to the lawn of the shop on Thursday afternoons (before weekend shopping begins). MowCo assigns A, an employee, to provide the lawn service. MowCo trains A, sets A’s compensation, supplies A with a mower and fertilizer, and ensures that IceCo is satisfied with A’s work. IceCo’s contract with MowCo gives IceCo the right to refuse the use of certain fertilizers for health or safety reasons, but IceCo has never attended to what fertilizer is used. IceCo agrees to pay MowCo the cost of wages up to $18 per hour per MowCo em- ployee plus a 15% markup. That is known as a “cost-plus” con- tract. The New Rule’s list of “essential terms and conditions of em- ployment” include “wages,” “hours of work,” and “working con- ditions related to the health and safety of employees.” 88 Fed. Reg. at 74,017–18. IceCo indirectly controls A’s wages in that IceCo negotiates a contract that creates an economic reality pres- suring MowCo to cap the wage of any employee working for IceCo at $18 per hour. The same economic control may even be present without a cost-plus contract, based simply on a fixed fee agreed upon for the work and the reality of the number of hours needed to complete the work. Either way, the economic realities of the situation mean that IceCo has some degree of factual control, even if indirectly and diffuse, over A’s wage. Moreover, IceCo indirectly controls A’s hours of work be- cause the contract specifies that MowCo must mow the lawn on Thursday afternoons. And IceCo has reserved control over work- ing conditions related to health and safety because, although it has never exercised this right, it has contractual authority to control what fertilizers are used on its lawn. Under subsections (e)(1) and (2) of the new rule, each of those aspects of the relationship would suffice to establish IceCo’s sta- tus as a joint employer of A, regardless of whether such indirect or reserved control tipped the scales under the common-law test and regardless of whether the type of fertilizer used would even be an essential term and condition of employment under the com- mon law. That reach exceeds the bounds of the common law and is thus contrary to law. Member Kaplan in his dissent from the rulemaking here pro- vided another illustration: For example, a widely used standard contract in the con- struction industry includes a provision that makes the gen- eral contractor “responsible for initiating, maintaining, and supervising all safety precautions and programs in connection with the performance of the [c]ontract.” That clause—a routine component of company-to-company contracting in the construction industry—evidences the general contractor’s indirect control (at least) of “working conditions related to the safety and health of employees” of each of its subcontractors, an essential term and condi- tion of employment under § 103.40(d)(7) of the final rule. Id. at 73,991 (footnotes omitted). Treating a general contractor as an employer of a subcontrac- tor’s employees in this way runs headlong into the Supreme Court’s holding in NLRB v. Denver Building & Construction Trades Council, 341 U.S. 675 (1951). There, the Court held that “the fact that the [general] contractor and subcontractor were engaged on the same construction project, and that the contractor had some supervision over the subcontractor’s work, did not eliminate the status of each as an independent contractor or make the employ- ees of one the employees of the other.” Id. at 689–90. The Board’s primary response is that plaintiffs “misrepresent what the Rule is and what it allows.” Doc. 34 at 19 n.58. But the court has explained why it disagrees based on the Rule’s text. It bears noting, however, that even the Board’s preferred interpreta- tion appears arbitrary and capricious for failing to achieve the pur- pose that justified the rulemaking—establishing “a definite, read- ily available standard” that will “assist employers and labor or- ganizations in complying with the Act” and “reduce uncertainty and litigation over the basic parameters of joint-employer status.” 88 Fed. Reg. at 73,957; see DHS v. Regents of Univ. of Cal., 140 S. Ct. 1891, 1909 (2020) (requiring an agency to defend its actions “based on the reasons it gave when it acted”). The “whole point of rulemaking as opposed to adjudication (or of statutory law as opposed to case-by-case common law de- velopment) is to incur a small possibility of inaccuracy in ex- change for a large increase in efficiency and predictability.” Ass’n of Data Processing Serv. Orgs., Inc. v. Bd. of Govs. of Fed. Rsrv. Sys., 745 F.2d 677, 689 (D.C. Cir. 1984) (Scalia, J.). But the Board largely backhanded and thus failed to reasonably address the dis- ruptive impact of the new rule on various industries, see 88 Fed. Reg. at 74,001 (Kaplan, dissenting), resolve ambiguities in a way making the rule more predictable than common-law adjudication, or explain how the rule does anything other than mandate piece- meal bargaining that will likely promote labor strife rather than peace by forcing an underdefined category of entities to take a seat at a bargaining table and negotiate over a multitude of influences that may otherwise be presented (and resolved) only through the invisible hand of the marketplace, id. at 73,999. But because the court rests its conclusion on the unlawfulness of the rule’s sweep beyond common-law limits, a definitive resolution of the arbi- trary-and-capricious point need not be reached here. c. Given the invalidity of new § 103.40, a question remains as to the 2020 version of that regulation. The Board justified its rescission of the 2020 Rule primarily on the ground that it was not lawful. See, e.g., 88 Fed. Reg. at 73,957 (“[T]he 2020 rule must be rescinded because it is contrary to the common-law agency prin- ciples incorporated into the Act when it was adopted and, accord- ingly, is not a permissible interpretation of the Act.”). In the al- ternative, the Board stated that it would rescind the 2020 Rule “even if that rule were valid because it fails to fully promote the policies of the Act.” Id. at 73,957. The Board’s first rationale is legally erroneous. The common law sets the outer limits of a permissible standard for bargaining duties as a joint employer, but the Board has long been permitted to draw on its expertise to erect prudential requirements within those bounds. See id. at 73,988 (Kaplan, dissenting) (collecting ex- amples). Even assuming that the 2020 rule is narrower than com- mon-law bounds, that does not make it unlawful because the Board may still act on its “policy expertise” within those bounds. BFI III, 911 F.3d at 1208. Here, when rescinding the 2020 Rule, the Board “did not appear to appreciate the full scope of [its] dis- cretion.” DHS v. Regents of Univ. Cal., 140 S. Ct. 1891, 1911 (2020). Its resulting rescission is thus arbitrary and capricious within the meaning of APA case law. See id. In any event, the Board’s view about the lawfulness of the 2020 Rule is “internal[ly] inconsisten[t].” Chamber of Commerce v. DOL, 885 F.3d 360, 382 (5th Cir. 2018). Recall that on the Board’s reading, step two of its test functions to limit the class of common-law employers that qualify as joint employers. If a lawful rule defining joint-employer status had to mirror the common law exactly, the Board’s reading of its 2023 Rule would be unlawful. That leaves the Board to rely on its alternative rationale for rescinding the 2020 Rule: policy considerations. To survive arbitrary-and-capricious review, agency action must be “reasona- ble and reasonably explained.” FCC v. Prometheus Radio Project, 592 U.S. 414, 423 (2021). Agencies are not “precluded from re- vising policy,” but a reviewing court must “ensure that the agency has recognized the change, reasoned through it without factual or legal error, and balanced all relevant interests affected by the change.” Louisiana v. DOE, 90 F.4th 461, 469 (5th Cir. 2024). Rescinding the 2020 Rule without issuing the 2023 Rule, which has now been held unlawful, would return the Board to the joint-employment standard set out in its case-by-case adjudica- tions. The Board recognized that point. 88 Fed. Reg. at 73,973. But it did not articulate a good reason (or any reason at all) why it believed a joint-employer standard set in adjudications to be pref- erable to a standard set in rulemaking. Such a reason may exist, but the Board must at least articulate it for it to support its alter- native rationale for the rescission. To be sure, the Rule attempts to justify rescinding the 2020 Rule on policy grounds. But it only does so in the context of stat- ing its preference for the 2023 Rule over the 2020 Rule, not its preference for no rule (only prior decisional case law) over the 2020 Rule. For example, the Board noted that the 2020 Rule “un- dermined the Act’s protections for employees who work in set- tings where multiple firms possess or exercise control over their essential terms or conditions of employment.” Id. at 73,974. That observation was part of a larger point that the 2023 Rule could “particularly benefit vulnerable employees.” Id. Be that as it may, the Board must give “good reasons for the new policy”—here, re- scission of the 2020 Rule independent of the contents of the 2023 Rule. FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009) (emphasis added). And the state of play after rescission would not necessarily be the same as under the 2023 Rule. See, e.g., 88 Fed. Reg. at 73,963–64 (contrasting BFI II’s “nonexhaustive list of es- sential terms and conditions of employment” with the 2023 Rule’s “exhaustive list”). The Board’s alternative rationale for re- scinding the 2020 Rule is thus arbitrary and capricious. d. As to remedy, the Declaratory Judgment Act allows a re- viewing court to “declare the rights and other legal relations of any interested party seeking such declaration.” 28 U.S.C. § 2201(a). Any such declaration “shall have the force and effect of a final judgment or decree.” Id. If necessary, a court may later grant an injunction to enforce its declaratory judgment. Id. § 2202; Powell v. McCormack, 395 U.S. 486, 499 (1969). The court finds it proper to exercise its discretion to issue such relief here. See Sherwin-Williams Co. v. Holmes Cnty., 343 F.3d 383, 388 (5th Cir. 2003). The court will thus issue a final judgment declaring that enforcement of the 2023 Rule against plaintiffs or their members would be contrary to law as to the Rule’s addition of a new 29 C.F.R. § 103.40 and arbitrary and ca- pricious as to the Rule’s removal of the existing 29 C.F.R. § 103.40 (2020). It is “anticipated that [defendants] would respect the de- claratory judgment,” Poe v. Gerstein, 417 U.S. 281, 281 (1974), so the court chooses not to issue an injunction at this time, see Mor- row v. Harwell, 768 F.2d 619, 627 (5th Cir. 1985). Plaintiffs or their members may, of course, seek an injunction should defendants threaten to depart from the declaratory judgment. The final question is whether to vacate the new rule—relief beyond just enjoining its enforcement against the plaintiffs in a given case. Although this court has questioned the principle’s un- derpinnings, “vacatur of an agency action is the default rule” in the Fifth Circuit when it is found to be discordant with the law or arbitrary and capricious. Cargill v. Garland, 57 F.4th 447, 472 (5th Cir.) (en banc), cert. granted, 144 S. Ct. 374 (2023); Franciscan Al- liance, Inc. v. Becerra, 47 F.4th 368, 374–75 (5th Cir. 2022) (“Va- catur is the only statutorily prescribed remedy for a successful APA challenge to a regulation.”). See generally R.J. Reynolds To- bacco Co. v. FDA, 2022 WL 17489170 (E.D. Tex. 2022), appeal pending, No. 23-40076 (5th Cir.). As such, the court will issue a final judgment vacating the 2023 Rule, both insofar as it rescinds the current version of 29 C.F.R. § 103.40 and insofar as it prom- ulgates a new version of that regulation. Conclusion For the reasons given above, plaintiffs’ motion for summary judgment (Doc. 10) is granted, and defendants’ cross-motion for summary judgment (Doc. 34) and motion to transfer for lack of jurisdiction (Doc. 25) are denied. The court will issue a final judg- ment forthwith. So ordered by the court on March 8, 2024. _faoboks BARKER United States District Judge

-31-