OPINION
Gary Vander Boegh applied for a job with EnergySolutions. He alleges that the prospective employer did not hire him because he engaged in protected whistle-blower activity at a prior job. The district court held that Vander Boegh lacked statutory standing as an applicant — not employee — and granted summary judgment in favor of EnergySolutions. Because we agree that Vander Boegh lacks statutory standing under the Energy Reorganization Act and False Claims Act, and we lack subject-matter jurisdiction over the remaining claims, we affirm.
I.
The U.S. Department of Energy (“DOE”) hired Vander Boegh in 1992 to work at the Paducah Gaseous Diffusion Plant (“PGDP”) as a landfill manager. In 1998, DOE awarded the PGDP contract to Bechtel Jacobs Company, LLC (“BJC”). BJC subcontracted with WESKEM for waste management services. Throughout this time, Vander Boegh continued as the landfill manager. The parties agree that Vander Boegh engaged in a range of protected activity as landfill manager, including reporting environmental violations.
In 2005, after soliciting new bids, DOE awarded the PGDP contract to Paducah Remediation Services, LLC (“PRS”). EnergySolutions provided waste management services to PRS by subcontract.1 Beginning in January 2006, PGDP operations transitioned from BJC-WESKEM to PRS-EnergySolutions. Vander Boegh applied to be the new landfill manager, but EnergySolutions hired another candidate. In April, the transition was complete, and Vander Boegh’s employment terminated.
Vander Boegh subsequently filed an employment discrimination complaint with the Department of Labor (“DOL”). He removed the action to federal district court under 42 U.S.C. § 5851(b)(4), alleging retaliation for prior protected conduct in violation of six federal statutes: The Energy Reorganization Act (“ERA”), 42 U.S.C. § 5851 (Count I); the False Claims Act (“FCA”), 31 U.S.C. § 3730(h)(1) (Count II); and the retaliation provisions of four federal environmental statutes — the Safe Drinking Water Act (“SDWA”), 42 U.S.C. § 300j-9(i); Clean Water Act (“CWA”), 33 U.S.C. § 1367; Toxic Substances Control Act (“TSCA”), 15 U.S.C. § 2622; and Solid Waste Disposal Act (“SWDA”), 42 U.S.C. § 6971 (Count III). The district court granted summary judgment in favor of all defendants. We affirmed the grant of *1059summary judgment in favor of BJC and PRS but reversed with respect to Energy-Solutions. Vander Boegh v. EnergySolutions, Inc., 536 Fed.Appx. 522 (6th Cir.2013). On remand, the district court again granted summary judgment in favor of EnergySolutions, this time holding that Vander Boegh lacked statutory standing because he was an applicant, not an employee. Vander Boegh timely appealed.
Vander Boegh challenges the district court’s holding that he lacks statutory standing, and further argues that the district court erred in declining to reach his alternate argument that he had standing as an employee because he had a. contractual right of first refusal.
II.
In this case, we consider for the first time whether the term “employee” extends to applicants for employment under the ERA, FCA, or four federal environmental statutes — the SDWA, CWA, TSCA, or SWDA. It appears that no federal court of appeals has considered these questions, although at least one has assumed, without deciding, that applicants are employees under the ERA. See Doyle v. Sec’y of Labor, 285 F.3d 243, 251 n. 13 (3d Cir.2002).
We review de novo a district court’s grant of summary judgment. Longaberger Co. v. Kolt, 586 F.3d 459, 465 (6th Cir.2009). “A matter requiring statutory interpretation is a question of law requiring de novo review, and the starting point for interpretation is the language of the statute itself.” Roberts v. Hamer, 655 F.3d 578, 582 (6th Cir.2011) (internal quotation marks omitted).
Vander Boegh argues that “employee” is ambiguous, and asks this court to afford Chevron deference to the DOL’s interpretation of the ERA. See Samodurov v. Gen. Physics Corp., No. 89-ERA-20, 1993 WL 832030, at *3 (Dep’t of Labor Nov. 16, 1993) (interpreting “employee” in the ERA to include applicants). He further asks us to extend that interpretation to the FCA and the four environmental statutes.
“In reviewing an agency’s interpretation of a statute that it is charged with administering, we apply the familiar two-step process announced by the Supreme Court in Chevron U.S.A., Inc. v. NRDC, [467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) ].” Chao v. Occupational Safety & Health Review Comm’n, 540 F.3d 519, 523 (6th Cir.2008). “First, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842-43, 104 S.Ct. 2778. “If the intent of Congress on a matter of statutory meaning is ambiguous, however, the court is to proceed to ‘step two’ of the Chevron inquiry: whether the agency’s interpretation is a ‘permissible construction of the statute.’ ” Mid-Am. Care Found. v. N.L.R.B., 148 F.3d 638, 642 (6th Cir.1998) (quoting Chevron, 467 U.S. at 843, 104 S.Ct. 2778).
“Chevron all too often is taken to mean simply that administrative agencies win in any dispute involving a question of statutory construction. Plainly, this is incorrect; both the Supreme Court and our court have made clear that there are numerous instances in which an agency’s interpretation of an ambiguous statute is not entitled to the broad deference envisioned by Chevron.” Id. “An agency’s interpretation is not entitled to Chevron deference, for example, if the apparent statutory ambiguity can be resolved using ‘traditional tools of statutory construction.’ ” Id. (citing INS v. Cardoza-Fonseca, 480 U.S. 421, 446, 107 S.Ct. 1207, 94 L.Ed.2d 434 (1987)). “[I]n *1060interpreting a statute a court should always turn first to one, cardinal canon before all others'. We have stated time and again that courts must presume that a legislature says in a statute what it means and means in a statute what is says there.” Conn. Nat’l Bank v. Germain, 503 U.S. 249, 253-54, 112 S.Ct. 1146, 117 L.Ed.2d 391 (1992). “When the words of a statute are unambiguous, then, this first canon is also the last: judicial inquiry is complete.” Id. (internal quotation marks omitted).
A.
Vander Boegh first disputes the district court’s conclusion that he lacks statutory standing under the ERA. The threshold question is whether “employee” is ambiguous. The ERA provides that “[n]o employer may discharge any employee or otherwise discriminate against any employee with respect to his compensation, terms, conditions, or privileges of employment” because the employee engaged in an activity protected by the statute. 42 U.S.C. § 5851(a)(1). Although the ERA defines the term “employer,” it does not define “employee.” Id. § 5851(a)(2). Nor do the relevant regulations define “employee.” See 29 C.F.R. §§ 24100-05.
Where no statutory definition exists, a court may consult a dictionary definition for guidance in discerning the plain meaning of a statute’s language. See, e.g., Nat’l Cotton Council of Am. v. U.S. E.P.A., 553 F.3d 927, 937 (6th Cir.2009). An “employee” is “[sjomeone who works in the service of another person (the employer) under an express or implied contract of hire, under which the employer has the right to control the details of work performance.” Black’s Law Dictionary 639 (10th ed.2014). An “employee” is also commonly defined as “[a] person working for another person or a business firm for pay.” Random House Webster’s Unabridged Dictionary 638 (2001).
Applied to this case, the plain meaning of “employee” does not plausibly extend to Vander Boegh because he never workfed for EnergySolutions. As a mere applicant, he was not an “employee” under the plain language. Cf. Old Colony R.R. Co. v. Comm’r of Internal Revenue, 284 U.S. 552, 560, 52 S.Ct. 211, 76 L.Ed. 484 (1932) (“[T]he plain, obvious and rational meaning of a statute is always to be preferred to any curious, narrow, hidden sense that nothing but the exigency of a hard case and the ingenuity and study of an acute and powerful intellect would discover.”).
Moreover, the statutory text itself demonstrates that Congress knew how to define “employee” to include applicants, if it had so intended. In the ERA, Congress defined “employer” to include seven categories of covered employers, including “an applicant for a [Nuclear Regulatory Commission] license.” 42 U.S.C. § 5851(a)(2)(B) (emphasis added). Although it explicitly defined “employer” to include an “applicant,” Congress nonetheless omitted any definition of “employee.” This omission further supports that Congress did not intend “employee” to include applicants. See, e.g., Russello v. United States, 464 U.S. 16, 23, 104 S.Ct. 296, 78 L.Ed.2d 17 (1983) (“[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.”) (internal quotation marks omitted).
The plain meaning alone is sufficient to end the inquiry. However, the Supreme Court has also instructed us on how to interpret “employee” when Congress does not define it. “It is ... well established that where Congress uses terms that have *1061accumulated settled meaning under the common law, a court must infer, unless the statute otherwise dictates, that Congress means to incorporate the established meaning of these terms.” Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 739,109 S.Ct. 2166, 104 L.Ed.2d 811 (1989) (internal quotation marks and alterations omitted); see also N.L.R.B. v. Amax Coal Co., 453 U.S. 322, 329, 101 S.Ct. 2789, 69 L.Ed.2d 672 (1981); Perrin v. United States, 444 U.S. 37, 42, 100 S.Ct. 311, 62 L.Ed.2d 199 (1979). “[W]hen Congress has used the term ‘employee’ without defining it, we have concluded that Congress intended to describe the conventional master-servant relationship as understood by common-law agency doctrine.” Reid, 490 U.S. at 739-40, 109 S.Ct. 2166; see also Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 323, 112 S.Ct. 1344, 117 L.Ed.2d 581 (1992). The phrase “master and servant” means “[t]he relation between two persons, one of whom (the master) has authority over the other (the servant), with the power to direct the time, manner, and place of the services provided.” Black’s Law Dictionary 1123 (10th ed.2014). Under the common law, the primary indicium of a master-servant relationship is a master’s ability to control the manner and means by which production is accomplished. See Reid, 490 U.S. at 751, 109 S.Ct. 2166 (citing Restatement (Second) of Agency § 220).
In Nationwide Mutual Insurance Co. v. Darden, the Supreme Court considered whether a claimant was an “employee” under the Employee Retirement Income Security Act (“ERISA”), 29 U.S.C. §§ 1001 et seq., which nominally defines an “employee” as “any individual employed by an employer.” Id. at § 1002; 503 U.S. at 322-27, 112 S.Ct. 1344. There, the Court applied Reid’s “well established principle” that Congress intended “employee” to connote a conventional master-servant relationship under the common law, and adopted a common-law test for distinguishing between an employee and an independent contractor. 503 U.S. at 322-27, 112 S.Ct. 1344. The Court explained that Reid signaled “[the Supreme Court’s] abandonment of [United States v. Silk ]’s [331 U.S. 704, 67 S.Ct. 1463, 91 L.Ed. 1757 (1947) ] emphasis on construing [a] term in the light of the mischief to be corrected and the end to be attained.” Darden, 503 U.S. at 325, 112 S.Ct. 1344 (internal quotation marks omitted). In other words, the Darden court confirmed that courts should presume Congress intended a term to have its settled, common-law definition “unless [the statute] clearly indicates otherwise,” even over the argument that a broad reading is necessary to “advance ... [an act’s] remedial purposes.” Id. at 325, 112 S.Ct. 1344.
In this case, the statutory text does not “otherwise dictate” that the settled, common-law meaning of “employee” should not apply. See Reid, 490 U.S. at 739, 109 S.Ct. 2166. The Supreme Court thus instructs us to look to the common-law master-servant relationship to define “employee.” Vander Boegh never worked for Energy-Solutions, and it never controlled the manner or means of Vander Boegh’s production. Accordingly, to adopt Vander Boegh’s position that he was an employee when he was merely an applicant would require a strained and unnatural reading of “employee.” We decline to read “employee” in such a peculiar fashion.
Our decision in Demski v. U.S. Department of Labor, 419 F.3d 488 (6th Cir.2005), does not compel a different result. In Demski, this court considered a distinct legal question. There, the president of a company that provided contract services to an ERA-covered employer alleged that the covered employer had cancelled her company’s contract in retaliation for her pro*1062tected conduct. Id. at 490; Demski v. Ind. Mich. Power Co., l-ERA-36, 2004 WL 785547 (Dep’t of Labor Apr. 9, 2004). We considered whether Demski, the president, was an “employee” of the ERA-covered employer with whom her company had contracted. Demski, 419 F.3d at 490. Unlike Vander Boegh, Demski was not an applicant for employment with the covered employer. See Demski, 2004 WL 785547. The agency ruled that Demski was not an employee because, applying common-law principles of agency, she was never a “hired party.” Id. at *5 (citing Reid, 490 U.S. at 751-52, 109 S.Ct. 2166, and Darden, 503 U.S. at 322-23, 112 S.Ct. 1344). Affording the agency’s interpretation Chevron deference, this court held that the agency’s construction was permissible and affirmed its grant of summary judgment in favor of the defendant ERA-covered employer. Demski, 419 F.3d at 491-92.
The instant case is distinguishable. In Demski, there was a contractual relationship between Demski’s corporations and the ERA-covered employer, and Demski herself was an employer of independent contractors who worked on site for the covered employer. The court was tasked with distinguishing, in that three-way relationship, whether Demski was an employee or independent contractor- — or neither — of the covered employer. In this case, by contrast, there was never any contractual relationship. This case thus presents a distinct legal question that is readily resolved by the plain language.
Because the plain meaning of “employee” does not extend to applicants, Vander Boegh lacks statutory standing under the ERA. EnergySolutions is entitled to summary judgment in its favor on this claim.
B.
Vander Boegh also appeals the district court’s ruling that he lacked statutory standing under the FCA. Applying the above plain meaning interpretation of “employee” to the FCA, we conclude that the term “employee” does not extend to applicants. - Moreover, the FCA’s legislative history and case law in other circuits reinforce our reading of the plain language.
As originally enacted, the FCA’s anti-retaliation provision provided that “[a]ny 'employee who is discharged, demoted, suspended, threatened, harassed, or in any other manner discriminated against in the terms and conditions of his employment by his or her employer because of [a protected activity] shall be entitled to all relief necessary to make the employee whole.” 31 U.S.C. § 3730(h); False Claim Amendments Act, Pub.L. 99-562, 100 Stat. 3153 (1986) (emphasis added). Congress amended the FCA in 2009 to expand its scope to “[a]ny employee, contractor, or agent.” Fraud Enforcement and Recovery Act of 2009, Pub.L. 111-21, 123 Stat. 1617, 1624-25 (emphasis added). Although there is no binding precedent in our circuit, other courts have overwhelmingly concluded that the term “employee” in the FCA did not extend to persons outside the employer-employee relationship before the amendment. See, e.g., Mruz v. Caring, Inc., 991 F.Supp. 701, 708-09 (D.N.J.1998) (limiting “employee” to the “employment relationship” under the plain language); U.S. ex rel. Morgan v. Sci. Applications Int’l Co., 604 F.Supp.2d 245, 250 (D.D.C.2009) (observing that liability does not extend to non-employers); Palladino ex rel. U.S. v. VNA of S. New Jersey, Inc., 68 F.Supp.2d 455, 464-65 (D.N.J.1999) (same).
Further, after Congress amended the FCA to include contractors and agents, at least one court has persuasively held that Congress still intended to limit the FCA to employment-like relationships. In U.S. ex rel. Abou-Hussein v. Science Applications *1063 International Corp., a plaintiff brought FCA claims against his employer and two non-employer corporations. Civ. No. 2:09—1858-RMG, 2012 WL 6892716, at *3-4 (D.S.C. May 3, 2012), aff'd 475 Fed.Appx. 851 (4th Cir.2012) (per curiam). The court dismissed the claims against the two corporations because the plaintiff was not an employee, independent contractor, or agent of those defendants. Id. at *2-4. Examining the legislative history, the court reasoned that Congress amended the FCA to correct recent court decisions that denied FCA retaliation protection to persons in employment-like relationships that were not technically “employees” because Congress found the decisions unduly narrow. Id. at *3 (citing Vessell v. DPS Assocs., 148 F.3d 407, 411 (4th Cir.1998) and U.S. ex rel. Watson v. Conn. Gen. Life Ins. Co., 87 Fed.Appx. 257, 261 (3d Cir.2004)). The court reasoned that Congress intended to “ ‘correct this loophole’ ” and extend protection to “‘individuals who [a]re not technically employees within the typical employer[-]employee relationship, but nonetheless have a contractual or agent relationship with an employer.’ ” Id. at *3 (quoting S.Rep. No. 110-507,. 110th Cong., 2d Session (Sept. 25, 2008), 2008 WL 4415147, at *26-27). Additional legislative history supports this interpretation. See 155 Cong. Rec. E1295-03, 2009 WL 1544226 (June 3, 2009) (statement of Rep. Howard L. Berman) (stating, as the House sponsor of the amendment, that the purpose was to “cover[ ] ... retaliation against contractors and agents of the discriminating party who have been denied relief by some courts because they are not technically ‘employees’ ” and to “protect persons who seek to stop [FCA violations] regardless of whether the person is a salaried employee, an employee hired as an independent contractor, or an employee hired in an agency relationship.”) (emphasis added).
This body of case law and legislative history reinforces our conclusion that the FCA does not extend to non-employee applicants. Vander Boegh concedes that he was never an EnergySolutions employee, contractor, or agent. He simply argues that we should broadly construe the term “employee” because (1) the legislative history mentions “blacklisted” workers, and (2) this court has previously interpreted “employee” to include applicants under the Fair Labor Standards Act (“FLSA”), see Dunlop v. Carriage Carpet Co., 548 F.2d 139, 144-47 (6th Cir.1977).
Vander Boegh’s arguments are unpersuasive. First, regarding the FCA’s legislative history, the term “blacklisted” is couched in the phrase “[temporary, blacklisted or discharged workers,” which merely suggests that “employee” extends to former employees, as well as present employees. See S. Rep. 99-345, 1986 WL 31937, at *34 (1986); see also Dole v. United Steelworkers of Am., 494 U.S. 26, 36, 110 S.Ct. 929, 108 L.Ed.2d 23 (1990) (“The traditional canon of construction, noscitur a sociis, dictates that words grouped in a list should be given related meaning.”) (internal quotation marks omitted). The phrase does not suggest that “employee” should be read so broadly to include applicants with no prior employment relationship to the prospective employer.
Second, regarding the FLSA, Vander Boegh fails to fully develop this argument and therefore has abandoned it. See Gen. Star Nat’l Ins. Co. v. Administratia Asigurarilor de Stat, 289 F.3d 434, 441 (6th Cir.2002) (observing that “perfunctory and undeveloped arguments” are waived on appeal) (internal quotation marks omitted). He has not, for example, explained why this court should look to the FLSA as a model. And we are hesitant to do so given that Congress did not include the FLSA in the list of statutes on which it modeled the *1064FCA retaliation provision. See S. Rep. 99-345, 1986 WL 31937, at *34 (1986). Finally, even if we were to consider this argument, we would not be persuaded. In Dunlop, we simply extended FLSA protections to a former employee who was not “technically employed” at the time of the alleged discrimination. 548 F.2d at 146-47. In this case, by contrast, Vander Boegh never had an employment relationship with EnergySolutions. Because the plain meaning of “employee” does not include applicants and the FCA’s legislative history and case law from other courts reinforce that “employee”, is limited to employment-like relationships, we agree with the district court that Vander Boegh lacks statutory standing under the FCA. Accordingly, EnergySolutions is also entitled to summary judgment in its favor on this claim.
C.
Next, Vander Boegh appeals the district court’s ruling that he lacks statutory standing under four federal environmental statutes: the SDWA, CWA, TSCA, and SWDA. However, because we lack subject-matter jurisdiction, we must dismiss these claims.
Federal courts are courts of limited jurisdiction and the law “presume[s] that a cause lies outside this limited jurisdiction.” Kokkonen v. Guardian Life Ins. Co. of Am., 511 U.S. 375, 377, 114 S.Ct. 1673, 128 L.Ed.2d 391 (1994). The burden of establishing the contrary rests upon the party asserting jurisdiction. Id. Because subject-matter jurisdiction is “an [Article III] as well as a statutory requirement ... no action of the parties can confer subject-matter jurisdiction upon a federal court.” Ins. Corp. of Ireland, Ltd. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702, 102 S.Ct. 2099, 72 L.Ed.2d 492 (1982). “Subject-matter jurisdiction can never be waived or forfeited,” and courts are obligated to consider sua sponte whether they have such jurisdiction. Gonzalez v. Thaler, — U.S. -, 132 S.Ct. 641, 648, 181 L.Ed.2d 619 (2012).
In this case, the environmental statutes provide an exclusive set of remedies: administrative review in the DOL, followed by judicial review of a final agency order in the “Court of Appeals for the circuit in which the violation allegedly occurred or the circuit in which the complainant resided on the date of the violation.” 29 C.F.R. § 24.112(a); see also id. § 24.100; 42 U.S.C. § 300j-9(i)(3)(A)-(B) (SDWA); 15 U.S.C. § 2622(c) (TSCA).2 The statutes do not grant jurisdiction to the federal district courts to hear claims filed by claimants under the retaliation provisions.3 *1065Although the ERA — which shares implementing regulations with the environmental statutes — contains an administrative by-pass option that allows claimants to remove administrative complaints to district court a year after filing with the agency, none of the environmental statutes contains an administrative by-pass option. See 42 U.S.C. § 5851(b)(4); 29 C.F.R. § 24.114. Rather, they require a claimant to fully exhaust his administrative remedies before seeking judicial review in the courts of appeals. See 29 C.F.R. § 24.106 (“If no timely objection is filed ... the finding and order will become the final decision of the Secretary, not subject to judicial review.”), § 24.110 (“If no timely petition for review is filed, the resulting final order is not subject to judicial review.”), and § 24.112(a) (“A final order of the ARB is not subject to judicial review in any criminal or other civil proceeding.”).
In this case, Vander Boegh filed a retaliation complaint with the DOL on April 18, 2006 (case no. 2006-ERA-26). He alleged that his complaint had been pending for more than one year without an agency decision. In February 2010, he exercised his statutory right to bring an ERA action in federal district court. It appears from the parties’ representations that Vander Boegh does not have a pending claim before the agency, and he never filed an appeal in this court of the agency’s final order dismissing his claims. The question is therefore whether .the statutory review procedures are exclusive or whether Vander Boegh always retained the right to bring his four environmental claims in district court.
“It is a well settled principle that where Congress establishes a special statutory review procedure for administrative action, that procedure is generally the exclusive means of review for those actions.” Greater Detroit Res. Recovery Auth. v. U.S. E.P.A., 916 F.2d 317, 321 (6th Cir.1990). When a statute provides a scheme of administrative review followed by judicial review in a federal appellate court, district court jurisdiction over a plaintiffs statutory claim is precluded if “Congress’s intent to preclude district court jurisdiction [is] fairly discernible in the statutory scheme.” Elgin v. Dep’t of Treasury, — U.S. -, 132 S.Ct. 2126, 2132, 183 L.Ed.2d 1 (2012) (internal quotation marks omitted). ‘Whether a statute is intended to preclude initial judicial review is determined from the statute’s language, structure, and purpose, its legislative history, and whether the claims can be afforded meaningful review.” Thunder Basin Coal Co. v. Reich, 510 U.S. 200, 207, 114 S.Ct. 771, 127 L.Ed.2d 29 (1994).
Here, Congress’s intent to provide a single avenue of review is easily discernible from the statutory text. There is nothing about the statutes’ language, structure, purpose, or legislative history that suggests Congress intended claimants to have a concurrent review process in district court, and judicial review by the courts of appeals affords claimants meaningful review. We thus conclude that a claimant may not bring a claim under the environmental statutes’ retaliation provisions directly in- federal district court.
Several courts have considered whether Congress intended district courts to have *1066subject-matter jurisdiction over such claims. These cases provide additional support for our construction of the statutes. For example, in Birnschein v. Anderson Machining Service, Inc., No. 09-cv-706-bbc, 2009 WL 5197185, at *2 (W.D.Wis. Dec. 18, 2009), a district court dismissed claims under all four environmental statutes for lack of subject-matter jurisdiction because “all establish a specific procedure that an employee must follow in seeking review of an employer’s adverse employment decision.” In that case, the plaintiff had not initiated proceedings in the DOL. Id. Thus, similar to this case, he did not have a final agency order before filing his complaint in federal district court. The court further observed that, even if plaintiff had exhausted his administrative remedies, the court still would have lacked jurisdiction because “[t]he only judicial review provided for by statute is by the federal courts of appeals.” Id.; see also Anael v. Interstate Brands Corporation, No. 00 C 6765, 2002 WL 31109451 (N.D.Ill. Sept. 20, 2002) (TSCA); Rhode Island Dep’t of Environ. Mgmt. v. United States, 304 F.3d 31, 43 (1st Cir.2002) (“The SWDA provides a mechanism for facilitating judicial review of final determinations of whether an employer has unlawfully retaliated against an employee ... and such review procedures are normally considered exclusive.”); Rockefeller v. Abraham, 58 Fed.Appx. 425, 427 (10th Cir.2003) (affirming the district court’s grant of summary judgment in favor of defendant on plaintiffs SDWA and TSCA claims because plaintiff never filed a complaint with the DOL).
The Seventh Circuit has examined the similar issue of whether a claimant may seek district court review of a final agency order. In Rhode v. City of W. Lafayette, 21 F.3d 430 (7th Cir.1994), the court considered whether a claimant could challenge in district court the Secretary of Labor’s approval of a settlement agreement of his SWDA and CWA claims. The court observed that “original jurisdiction for review of the Secretary’s final decision under the Acts is vested exclusively in the federal appellate courts.” Id. “Because the congressional intent to vest exclusive jurisdiction in the federal appellate courts is clear, and because nothing suggests that the proscribed remedy is inadequate, the district court properly dismissed the present action for lack of subject matter jurisdiction.” Id. (internal citation omitted) (citing Gen. Fin. Corp. v. FTC, 700 F.2d 366, 368 (7th Cir.1983) (a litigant “may not bypass the specific method that Congress has provided for reviewing adverse agency action simply by suing the agency in federal district court under [28 U.S.C.] § 1331 or 1337; the specific statutory method, if adequate, is exclusive.”)).
Although the parties ostensibly concede that the district court lacked jurisdiction to hear these claims under 28 U.S.C. § 1331, they nonetheless seek our review on the merits, arguing that (1) the “exhaustion requirement” is non-jurisdictional, and (2) we should exercise pendent appellate jurisdiction. We disagree.
First, the statutory scheme is squarely jurisdictional. We have previously described this issue as a question of whether a statutory limitation goes to a plaintiffs right to relief or his right to enter the federal courts. Hoogerheide v. I.R.S., 637 F.3d 634, 638 (6th Cir.2011). The Supreme Court has drawn an “administrable bright line” between the claims-processing requirements of a cause of action and jurisdiction over that cause of action. See id. at 636 (quoting Arbaugh v. Y & H Corp., 546 U.S. 500, 516, 126 S.Ct. 1235, 163 L.Ed.2d 1097 (2006)). It has explained:
*1067If the Legislature clearly states that a threshold limitation on a statute’s scope shall count as jurisdictional, then courts and litigants will be duly instructed and will not be left to wrestle with the issue---- But when Congress does not rank a statutory limitation on coverage as jurisdictional, courts should treat the restriction as nonjurisdictional in character.
Arbaugh, 546 U.S. at 515-16, 126 S.Ct. 1235. In other words, “[u]nder Arbaugh, we look to see if there is any ‘clear’ indication that Congress wanted the rule to be ‘jurisdictional.’ ” Henderson v. Shinseki, 562 U.S. 428, 131 S.Ct. 1197, 1203, 179 L.Ed.2d 159 (2011). “Congress, of course, need not use magic words in order to speak clearly on this point.” Id. “Rather, the jurisdictional analysis must focus on the legal character of the requirement, which we discern by looking to the condition’s text, context, and relevant historical treatment.” Reed Elsevier, Inc. v. Muchnick, 559 U.S. 154, 166, 130 S.Ct. 1237, 176 L.Ed.2d 18 (2010) (internal quotation marks and citation omitted).
Since Arbaugh, this court has assessed the line between claims-processing requirements and jurisdictional limits in several settings. See Hoogerheide, 637 F.3d at 636 (surveying cases). Most relevant here are the cases in which we considered whether administrative exhaustion requirements were jurisdictional. In Allen v. Highlands Hospital Corp., for example, we addressed the administrative filing requirements of the Age Discrimination in Employment Act (“ADEA”), 29 U.S.C. § 626(d). 545 F.3d 387 (6th Cir.2008). In light of the statutory text and context, we found that the requirements — located in a separate provision from the jurisdictional provision — were not written in jurisdictional terms. Id. at 401-02. And the jurisdictional provision provided that “[a]ny person aggrieved may bring a civil action in any court of competent jurisdiction____” Id. (quoting 29 U.S.C. § 626(c)(1)). Accordingly, we held that Congress did not clearly state that filing a charge with the EEOC was a precondition to the district court’s jurisdiction over ADEA claims. Id. at 402-03.
Similarly, in Abraitis v. United States, we held that the administrative exhaustion requirements for judicial review of an IRS jeopardy levy were not jurisdictional. 709 F.3d 641, 645 (6th Cir.2013). In that case, the exhaustion provision detailed the administrative filing deadlines but expressly referenced the separate jurisdictional provision for purposes of determining jurisdiction. Id. (citing 26 U.S.C. § 7429(b)(1)). The jurisdiction provision broadly provided that “the district courts of the United States shall have exclusive jurisdiction over any civil action for a determination under this subsection.” Id. (citing 26 U.S.C. § 7429(b)(2)).
However, those statutes are fundamentally different from the environmental statutes now before us. In the ADEA and IRS levy statutes, Congress granted jurisdiction to the district courts. We were therefore tasked with determining whether Congress had clearly communicated an intent to strip district courts of jurisdiction when a party fails to exhaust his administrative remedies before filing in district court. In considerable contrast, none of the statutes before us grant jurisdiction to the district courts in the first instance. See 42 U.S.C. § 300j-9(i)(3)(A)-(B) (SDWA); 33 U.S.C. § 1367(b)(CWA); 15 U.S.C. § 2622(c) (TSCA); 42 U.S.C. § 6971(b) (SWDA); see also 29 C.F.R. § 24.112(a). In other words, there is no private right of action in district court. See Anael, 2002 WL 31109451, at *6 (“[T]he unambiguous language of the [TSCAj’s retaliation provision clearly *1068evinces Congress’ intent not to create a private right of action.”). No district court review is available to a claimant either as an original claim or on judicial review of an agency order (unless filed by the Secretary as an enforcement action), nor is judicial review available in the courts of appeals until the secretary has issued a final order. See 29 C.F.R. §§ 24.106, 24.110. And unless administrative review is exhausted, judicial review is expressly precluded in the courts of appeals. Id. § 24.112(a). Accordingly, because the statutes clearly communicate Congress’s intent to limit jurisdiction to the DOL and courts of appeals, this case falls squarely on the other side of the claims processing-versus-jurisdictional line.
Context also corroborates this reading. Most telling is the expressly jurisdictional language of the ERA, which shares implementing regulations with the environmental statutes. See 29 C.F.R. §§ 24.100-14. The ERA provides the only avenue by which a claimant may bring his original claim to district court. See 42 U.S.C. § 5851(b)(4) (providing that if there is no final agency order after one year, the district court “shall have jurisdiction ... without regard to the amount in controversy” over a claimant’s ERA action filed in district court). Likewise, the regulations distinguish the ERA from the environmental statutes in jurisdictional terms. 29 C.F.R. § 24.114 (detailing a district court’s jurisdiction over ERA retaliation claims). The language of the ERA is instructive because it demonstrates that Congress knew how to grant a claimant a statutory by-pass to district court — and expressly vest jurisdiction in district court — but it elected not to do so in the four environmental statutes.
Other enforcement provisions in the environmental statutes also provide a useful comparison. • For example, two of the four environmental statutes state in express jurisdictional terms:
Whenever a person has failed to comply with [a final order of the Secretary], the Secretary shall file a civil action in the United States district court for the district in which the violation was found to occur to enforce such order. In actions brought under this subsection, the district courts shall ( have jurisdiction to grant all appropriate relief, including injunctive relief and compensatory and exemplary damages. •
42 U.S.C. § 300j-9(i)(4) (SDWA); 15 U.S.C. § 2622(d) (TSCA); see also 29 C.F.R. § 24.113 (providing that the Secretary may file a civil action in district court to enforce final agency orders issued under any of the environmental statutes). These provisions stand in striking contrast to the retaliation provisions of the environmental statutes, which limit review to the DOL, followed by exclusive judicial review in the courts of appeals. This context reinforces our conclusion that the district court lacked jurisdiction to hear any such claims because there is no private right of action in district court under the four environmental statutes. Unlike the exhaustion cases cited by EnergySolutions, the four environmental statutes do not contain any grant of jurisdiction to the district court over a claimant’s action filed in district court, regardless of exhaustion. We thus find the statutory review procedures jurisdictional.
Vander Boegh asks us to exercise pendent appellate jurisdiction. For support, he cites cases in which appellate courts have exercised jurisdiction over Comprehensive Environmental Response, Compensation, and Liability Act (“CERCLA”) claims while reviewing other environmental claims properly before them. See, e.g., Kaufman v. Perez, 745 F.3d 521, 527 (D.C.Cir.2014), reh’g en banc denied May *106919, 2014; Carpenter v. Solis, 439 Fed.Appx. 480, 483-84 (6th Cir.2011). These cases are distinguishable in at least two respects. First, the regulation allowing for pendent appellate jurisdiction is unique to CERCLA and does not encompass the other environmental statutes. See 29 C.F.R. § 24.112(d) (providing that a claimant may appeal a final agency order issued under CERCLA directly to the court of appeals when the final order is also issued under another environmental statute listed in 29 C.F.R. § 24.100(a)). Second, this case presents the inverse procedural posture of the CERCLA cases, which undercuts Vander Boegh’s position. In the CERCLA cases, all claims had been presented to the DOL, and all but the CERCLA claims were properly before the court of appeals under the applicable statutory review procedures. By exercising pendent appellate jurisdiction over the CERCLA claims, the courts allowed the claimants to by-pass review in district court on the theory that the courts may review an agency’s order issued under two or more statutes when at least one of the statutes provides for direct review in the court of appeals. In this case, Vander Boegh asks us to flip the statutory scheme on its head, allowing a claimant to file in federal district court — -a court without jurisdiction in the first instance or on review. - Because the statutes provide for exclusive review in the DOL and courts of appeals, to read the statutes as suggested would contravene Congress’s intent. Pendent appellate jurisdiction is inappropriate here.4
We hold that the district court lacked subject-matter jurisdiction over these environmental claims. We thus lack jurisdiction over these claims that were .filed in district court, rather than appealed directly from the agency to this court. For these reasons, we dismiss Vander Boegh’s four environmental claims.
III.
Finally, Vander Boegh argues that the district court abused its discretion in declining to reach his alternate argument that he was an EnergySolutions employee by virtue of a contractual right of first refusal in a contract between the DOE and PRS. Applying the law of the case doctrine, the district court held that Vander Boegh’s contractual status had already been considered and rejected by the district court and this court. We review a district court’s decision to apply the law of the case doctrine for abuse-of discretion. Rouse v. DaimlerChrysler Corp., 300 F.3d 711, 715 (6th Cir.2002). Finding no abuse, we affirm.
Vander Boegh raised his contractual right of first refusal as circumstantial evidence of retaliatory motive by all defendants, arguing that their failure to recognize his right of first refusal was' “use of irregular procedure” that supported an inference of retaliatory motive. The district *1070court considered Vander Boegh’s ,argument under the second element of his prima facie case — whether EnergySolutions knew of his protected activity. It observed:
Plaintiff next alleges that BJC failed to identify him as a grandfathered employee to PRS. BJC has admitted that Plaintiff was a grandfathered employee during the 1998 contract transition from Lockheed Martin to BJC. In the employee information list BJC provided to PRS, Plaintiff is specifically identified as grandfathered (“GF”) for pension-related purposes. Schifferer Declaration, DN 47-12 at p. 17. Kelli Schifferer, the former BJC Human Resource Manager and the current custodian of BJC records relating to personnel and employee relations, stated that the employee information list was provided to PRS on April 6, 2006. Schifferer Declaration, DN 47-12 at ¶ 5. Thus, BJC did inform PRS of Plaintiffs grandfathered status. Nevertheless, this does not necessarily mean that Plaintiff was a grandfathered employee under the contract between the DOE and PRS. As proof that Plaintiff was a grandfathered employee under the contract between the DOE and PRS, Plaintiff has presented a DOE letter concerning an Employee Concern of another individual by the name of John Bobo. That letter states that, under the PRS contract, employees are eligible to receive a preference in hiring for vacancies for nonmanagerial positions if they are grandfathered employees who are on the rolls of BJC’s first and second tier subcontractors at the Paducah site at contract transition. DN 48-25 at p. 1. Even if Plaintiff qualified as a grandfathered employee, it was not BJC’s responsibility to recognize this right or to ensure that Duratek-Energy did so. Furthermore, even if PRS and Duratek-EnergySolutions failed to recognize Plaintiffis grandfathered status, their failure to do so has no bearing on whether Kelly had knowledge of Plaintiffs protected activities.
It appears that the court assumed for purposes of argument that Vander Boegh was a grandfathered employee, but concluded that the fact would not support an inference of knowledge of Vander Boegh’s protected activities.
On appeal, Vander Boegh argued that the following actions were evidence of retaliatory motive: (1) BJC failed to provide a list of grandfathered employees to PRS until the late date of April 6, 2006, and (2) PRS and EnergySolutions departed from normal practice by failing to honor his right of first refusal. EnergySolutions responded that Vander Boegh was not a grandfathered employee with respect to a right of first refusal, citing PRS contract language (“ ‘Grandfathered employees’ are’ non-managerial employees (i.e., those below the first level of supervision)”). It further argued that even if he was a grandfathered employee, it would be irrelevant to show pretext. This court rejected Vander Boegh’s argument that he was a grandfathered employee. We stated:
Vander Boegh also argues that an employer’s failure to follow its normal procedure can provide circumstantial evidence of a retaliatory motive and, thus, actual knowledge of .protected activity. See DeFord v. Sec’y of Labor, 700 F.2d 281, 287 (6th Cir.1983). In particular, Vander Boegh claims that the contract provision granting “grandfathered employees” a hiring preference for vacancies, including a .right of first refusal in the job they currently perform, was not followed. The PGDP contract entered into by PRS does, in fact, contain a provision providing this hiring preference, and Vander Boegh maintained his position through all previous contract *1071transitions. Further, one could reasonably infer that Kelly read the contract as part of his due diligence responsibility during the transition. The flaw in the argument, however, is that Vander Boegh has failed to present any evidence beyond his personal belief that he was a grandfathered employee under the new contract. Only “non-managerial” employees are considered “grandfathered” under the contract, and the landfill manager position is specifically denoted as “managerial.”
Vander Boegh, 536 Fed.Appx. at 531.
On remand, the district court declined to address Vander Boegh’s argument that he was an employee by virtue of his contractual right of first refusal, opining that
[t]his argument has been considered and rejected both by this Court, (see Docket No. 52, at 25), and by the Sixth Circuit, (see Docket No. 74, at 15-16, 19). Consequently, it need not be addressed again here.
Vander Boegh now argues that the district court erred by declining to reach this issue because the question decided on appeal was limited to whether defendants had knowledge of his protected activity. EnergySolutions responds that the law of the case doctrine governs. “The law-of-the-case doctrine ‘posits that when a court decides upon a rule of law, that decision should continue to govern the same issues in subsequent stages in the same case.’” Scott v. Churchill, 377 F.3d 565, 569-70 (6th Cir.2004) (quoting Arizona v. California, 460 U.S. 605, 618, 103 S.Ct. 1382, 75 L.Ed.2d 318 (1983)). “The doctrine precludes a court from reconsideration of issues decided at an early stage of litigation, either explicitly or by necessary inference from the disposition.” Westside Mothers v. Olszewski 454 F.3d 532, 538 (6th Cir.2006) (internal quotation marks omitted). Application of this doctrine is “limited to those questions necessarily decided in the earlier appeal.” Hanover Ins. Co. v. Am. Eng’g Co., 105 F.3d 306, 312 (6th Cir.1997). “[T]he phrase ‘necessarily decided’ .:. describes all issues that were ‘fully briefed and squarely decided’ in an earlier appeal.” Perkins v. Am. Elec. Power Fuel Supply, Inc., 91 Fed.Appx. 370, 374 (6th Cir.2004) (quoting 1B James Wm. Moore, Moore’s Federal Practice ¶0.404[1], at II — 5 (2d ed.1996)). We have recognized three exceptions to the law of the case doctrine: “(1) where substantially different evidence is raised on subsequent trial; (2) where a subsequent contrary view of the law is decided by the controlling authority; or (3) where a decision is clearly erroneous and would work a manifest injustice.” Hanover, 105 F.3d at 312.
In this case, Vander Boegh had a full and fair opportunity to litigate whether he was a grandfathered employee. In deciding whether EnergySolutions’ failure to honor his alleged status was proof of knowledge of protected activity, the district court assumed that Vander Boegh was a grandfathered employee but found it irrelevant to EnergySolutions’ knowledge. Rather than assuming he was a grandfathered employee, this court held that Vander Boegh had “failed to present any evidence beyond his personal belief that he was a grandfathered employee under the new contract.” Vander Boegh, 536 Fed.Appx. at 531. We thus explicitly decided the issue of whether Vander Boegh had established that he was a grandfathered employee, and none of the doctrine’s exceptions applies. It is of no consequence that we considered this issue with respect to his prima facie case, even though Vander Boegh now argues the point to overcome his lack of statutory standing. The district court did not abuse its discretion in *1072interpreting our opinion as having decided the issue.
IV.
For these reasons, we affirm the judgment of the district court.