concurring:
I concur completely in the result reached by the majority and its basic textual analysis of the Coal Industry Retiree Health Benefit Act of 1992 (“Coal Act”), 26 U.S.C. §§ 9701-22. This analysis alone is sufficient to support the outcome reached by the majority; the remainder of the opinion is obiter dicta. I write separately because I fear the majority’s discussion of the trustees’ final argument — that we should adopt a broad substantial continuity test — may be misleading. The cases from our sister circuits that have “extend[ed] successor liability beyond the textual bounds of a statute” are in no way relevant to our analysis in this case. Opinion at 825. While it is fashionable in some legal circles to deride “hyper-technical reliance upon statutory provisions,” Palm Beach County Canvassing Bd. v. Harris, 772 So.2d 1220, 1227 (Fla.), vacated, 531 U.S. 70, 121 S.Ct. 471, 148 L.Ed.2d 366 (2000), this Court does not — and should not— move in them.
The majority cites the Supreme Court’s decision in Traynor v. Turnage, 485 U.S. 535, 545-46, 108 S.Ct. 1372, 99 L.Ed.2d 618 (1988), for the proposition that we must assume that “Congress’s choice of words reflects a familiarity with judicial treatment of comparable language.” Opinion at 824. In fact, the Traynor Court assumed that by using a specific term in a veterans’ benefits statute, Congress “intended that the term receive the same meaning for *827purposes of that statute as it had received for purposes of other veterans’ benefits statutes.” 485 U.S. at 546, 108 S.Ct. 1372. In other words, the Court stated that statutory terms should be interpreted in the same way in statutes covering similar topics — not in any statute covering any topic. Cf. Del Commercial Props., Inc. v. Commissioner, 251 F.3d 210, 218 (D.C.Cir.2001).
In the present case, this maxim of statutory interpretation suggests that the term “successor in interest” in the Coal Act should be interpreted consistently throughout the Internal Revenue Code, see Commissioner v. Keystone Consol. Indus., 508 U.S. 152, 159, 113 S.Ct. 2006, 124 L.Ed.2d 71 (1993), and that Congress was aware of how the term had been interpreted in that context, see Traynor, 485 U.S. at 546, 108 S.Ct. 1372. Treasury Regulations define “successor in interest” as “an acquiring corporation that succeeds to the tax attributes of an acquired corporation” through the following transactions: the liquidation of a subsidiary, a merger or consolidation, the sale of “substantially all of the properties of another corporation” for voting stock, 26 U.S.C. § 381, or the mere change of identity. 26 C.F.R. § 1.1503-2A(c)(3)(vii)(B). Relatedly, state business association law historically has been used to determine the tax liability of “a successor corporation for the debts of its predecessors.” 15 Mertens Law of Federal INCOME Taxation § 61.17, at 47 (2000); see also United States v. First Dakota Nat’l Bank, 137 F.3d 1077, 1079-80 (8th Cir.1998); Atlas Tool Co. v. Commissioner, 614 F.2d 860, 870 (3d Cir.1980); George v. Commissioner, 2000 WL 1545066 (T.C. Oct. 19, 2000) (mem.). In West Virginia, where both the defendant companies are incorporated, a corporation that purchases the assets of another corporation is not liable for the debts of the seller unless: (1) the purchaser expressly or impliedly assumes the liability, (2) the transaction was fraudulent, (3) “some element of the transaction was not made in good faith,” (4) the purchase effected a consolidation or merger, or (5) “the successor corporation is a mere continuation or reincarnation of its predecessor.” Jordan v. Ravenswood Aluminum Corp., 193 W.Va. 192, 455 S.E.2d 561, 563 (W.Va.1995); see also West Texas Ref. & Dev. Co. v. Commissioner, 68 F.2d 77, 80 (10th Cir.1933) (applying the same factors in a tax dispute). Under either of these formulations, the defendant companies would not be liable under the Coal Act. This analysis is sufficient to affirm the district court’s decision in this case.
If Congress had sought to adopt something similar to the trustees’ articulation of the substantial continuity of operations test it likely would have done so explicitly. For example, the Black Lung Benefits Act holds companies hable for benefit payments to coal-mining employees when the companies are successive operators of a coal mine or acquire substantially all of the assets of the previous operator. See 30 U.S.C. § 932(i)(1). Of course, in the Coal Act, Congress did not adopt this approach.
The majority’s lengthy discussion of cases employing the substantial continuity of interest test does not clarify “judicial treatment of comparable language.” Opinion at 824. As the majority recognizes, courts largely have adopted this test in cases when the statute at issue does not address successorship, much less use the term “successor in interest.” See, e.g., Wheeler v. Snyder Buick, Inc., 794 F.2d 1228 (7th Cir.1986).
The majority rightly traces this analysis to four Supreme Court decisions. Two of these decisions dealt with claims brought under the Labor Management Relations Act (LMRA), and two arose from decisions issued by the National Labor Relations *828Board (NLRB) pursuant to the National Labor Relations Act (NLRA). In the first of the LMRA decisions, John Wiley & Sons, Inc. v. Livingston, 376 U.S. 543, 544, 84 S.Ct. 909, 11 L.Ed.2d 898 (1964), the Supreme Court addressed whether a successor company was required to comply with an arbitration clause in a collective bargaining agreement between a union and the predecessor company. Significantly, the predecessor corporation had merged with the successor corporation, and the Court was confronted with a question of “contractual origin” — and therefore did not interpret or apply any statute. See id. at 545, 546, 550-51, 84 S.Ct. 909. Construing the contract and noting the “substantial continuity of identity in the business enterprise” following the merger, the Court held that the successor corporation was required to comply with the arbitration clause. Id. at 551, 84 S.Ct. 909. Later, the Court explained that the Wiley “holding dealt with a merger occurring against a backdrop of state law that embodied the general rule that in merger situations the surviving corporation is liable for the obligations of the disappearing corporation.” NLRB v. Burns Int’l Sec. Servs., Inc., 406 U.S. 272, 286, 92 S.Ct. 1571, 32 L.Ed.2d 61 (1972).
In the second LMRA case, Howard Johnson Co. v. Detroit Local Joint Executive Board, 417 U.S. 249, 251-52, 94 S.Ct. 2236, 41 L.Ed.2d 46 (1974), a franchiser bought all of the personal property associated with a franchisee’s business operation, but did not retain the franchisee’s employees. The employees’ union claimed that the franchiser was required to “arbitrate the extent of its obligations” to the employees under the collective bargaining agreement between the franchisee and union. Id. at 253, 94 S.Ct. 2236. The Court held that the franchiser was not bound by the agreement because it did not have a “substantial continuity of identity” with the franchisee. Id. at 264, 94 S.Ct. 2236. The Howard Johnson Court contrasted the circumstances of the case with Wiley, emphasizing that Wiley “involved a merger, as a result of which the initial employing entity disappeared.” Id. at 257, 94 S.Ct. 2236. In Howard Johnson, however, “the initial employers remain in existence as viable corporate entities.” Id. Significantly, in neither Wiley nor Howard Johnson did the Supreme Court purport to interpret the LMRA — or any statute — to encompass any successorship doctrine.
Unlike Wiley and Howard Johnson, the Court in Golden State Bottling Co. v. NLRB, 414 U.S. 168, 171-72, 94 S.Ct. 414, 38 L.Ed.2d 388 (1973), and NLRB v. Burns Int’l Sec. Servs., Inc., 406 U.S. at 277, reviewed the Board’s interpretation of its organic statute. When reviewing such decisions, the Court employs a deferential standard, upholding the Board’s interpretation of the NLRA as long as it “adopts a rule that is rational and consistent with the Act.” Fall River Dyeing and Finishing Corp. v. NLRB, 482 U.S. 27, 42, 107 S.Ct. 2225, 96 L.Ed.2d 22 (1987); see Golden State Bottling Co., 414 U.S. at 181, 94 S.Ct. 414 (considering “whether the Board properly exercised its discretion” in ordering a bona fide purchaser of a business to reinstate employees with backpay when the predecessor corporation had engaged in unfair labor practices); Burns Int’l Sec. Servs., 406 U.S. at 278-79, 92 S.Ct. 1571 (holding that “it was not unreasonable” for the NLRB to hold that the successor employer is required to bargain with an existing certified union). Indeed, “[i]n Bums [the Supreme Court] approved the approach taken by the Board and accepted by the courts with respect to determining whether a new company was indeed the successor to the old.” Fall River, 482 U.S. at 43, 107 S.Ct. 2225. That approach, as the majority explains, is the broad totality *829of the circumstances test that the trustees now advocate. See id.
In previous cases, this Court has affirmed the use of the substantial continuity of interest standard to determine the obligations of successor corporations — but only when reviewing an agency decision that had employed it. For example, in Harter Tomato Prods. Co. v. NLRB, 133 F.3d 934, 938 (D.C.Cir.1998), we upheld the NLRB’s conclusion that a company that merely leased assets from a predecessor company could still be a successor required to bargain with an existing union if it met the broad substantial continuity test. See also Pa. Transformer Tech., Inc. v. NLRB, No. 00-1388, slip op. at 4-5 (D.C.Cir. June 29, 2001). Our decision was based on the deference we accord to NLRB rules that are “rational and consistent” with the NLRA. Harter Tomato, 133 F.3d at 937 (internal quotation omitted). Similarly, we deferred to the Federal Mine Safety and Health Review Commission when it used the test to determine successor liability under the Mine Safety and Health Act. See Secretary of Labor v. Mullins, 888 F.2d 1448, 1451 n. 10, 1453-54 & n. 15 (D.C.Cir.1989). Our deference to reasonable statutory interpretations made by agencies to which Congress has specifically delegated authority should not be confused with an adoption of those interpretations or a belief that they are correct. We have never interpreted a statute de novo and concluded that liability under the statute is determined based on a substantial continuity of interest.
In contrast, some courts have adopted the substantial continuity standard when interpreting statutes de novo. In doing so, they have relied on the four Supreme Court decisions discussed above — even though the cases before them did not review an agency decision, did not focus on labor contracts, and did not even deal with statutes that mention successorship. This reliance is mistaken.
The Supreme Court has never adopted any amorphous totality-of-the-circumstances test in cases raising successorship questions not arising in a context requiring deference to an agency. Instead, it has stated that there must be a “substantial continuity of identity in the business enterprise,” which “necessarily includes ... a substantial continuity in the identity of the work force across the change in ownership.” Howard Johnson Co., 417 U.S. at 263, 94 S.Ct. 2236 (citation omitted). Even if this language could be read with the breadth some of our sister circuits suggest, the Supreme Court has only applied it in cases dealing with current employees seeking to establish a company’s obligations under an existing collective bargaining agreement, and then only has found such a continuity when two companies have merged, one of those companies has been extinguished, and all of the predecessor’s employees have been retained by the successor company. Furthermore, nothing in either Howard Johnson or Wiley can be read to extend the reach of their holdings beyond their “contractual origin,” much less beyond statutes governing labor-management relations.
The courts that have morphed the substantial continuity standard of Howard Johnson and Wiley into a sweeping totality-of-the-circumstances standard have allowed rules adopted by the NLRB pursuant to its authority under the National Labor Relations Act, see, e.g., Fall River, 482 U.S. at 43, 107 S.Ct. 2225, to serve as the law for other statutes well outside the NLRB’s reach. See, e.g., Musikiwamba v. ESSI, Inc., 760 F.2d 740, 750 (7th Cir.1985). It is the role of Congress, not the judiciary, to establish when successors should be held liable for the statutory violations of predecessor companies and *830whether successors have obligations to their predecessor’s former employees. See Wheeler v. Snyder Buick, Inc., 794 F.2d 1228, 1237 (7th Cir.1986). Courts simply should not extend liability “to a variety of statutory contexts when the equities have so dictated” no matter how “important” the policy they seek to fulfill. Upholsterers’ Int’l Union Pension Fund v. Artistic Furniture of Pontiac, 920 F.2d 1323, 1327 (7th Cir.1990). When Congress seeks to establish broad rules of successor liability, it will do so on its own. See, e.g., 30 U.S.C. § 932(i)(1). Under the Coal Act, it did not. Insofar as the majority’s opinion by discussion of the “factor[s] motivating courts to extend successor liability beyond the textual bounds” of statutes suggests any legitimacy for that approach, I wish to make plain that that is not the holding of this case, nor the analysis that commands my concurrence.