concurring:
I agree with almost everything in the court’s opinion, but I concur only reluctantly in the ultimate disposition, which permits FERC to reconsider the rate of return issue on a second remand.
In its Opinion No. 190, FERC established a methodology for calculating Tennessee Gas’s return on equity and applied the methodology to the period under consideration in that case. See Tennessee Gas Pipeline Co., 25 F.E.R.C. 1161,020, at 61,091-97 (1983). In setting Tennessee’s return on equity for the immediately following period, therefore, FERC was obliged either to apply the Opinion 190 methodology or else provide a reasoned explanation for not applying it. See, e.g., Public Serv. Comm’n of New York v. FERC, 813 F.2d 448, 451 (D.C.Cir.1987) (PSCNY); cf. Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41-42, 103 S.Ct. 2856, 2865-66, 77 L.Ed.2d 443 (1983) (rule-making context). Twice already, FERC has sought to deviate from Opinion 190 in order to account for falling interest rates during the later period, and twice this court has found its proffered explanations wholly inadequate. The first time, before performing the appropriate calculation, FERC selectively replaced certain data from the later period with obsolete data from the earlier one. See Tennessee Gas Pipeline Co., 32 F.E.R.C. ¶ 61,086, at 61,223-25 (1985) (Opinion 240); Tennessee Gas Pipeline Co., 33 F.E.R.C. U 61,005, at 61,009 (1985) (denying rehearing of Opinion 240). We set aside this “result-oriented manipulation of an objective ratemaking calculation,” but did not foreclose the Commission on remand from justifying its desired result in a more principled fashion. See PSCNY, 813 F.2d at 465. The second time, FERC posited directly — with only the flimsiest of support — its theory that investors need over six months to ascertain and take into account the market interest rates prevailing at any given time. See Tennessee Gas Pipeline Co., 46 F.E.R.C. 11 61,089, at 61,-383-84 (1989) (Order on Remand); Tennessee Gas Pipeline Co., 49 F.E.R.C. 11 61,392, at 62,419-23 (denying rehearing of Order on Remand).
I heartily agree with the court’s rejection of the Commission’s more recent explanation. At the very least, FERC was obliged to offer some convincing evidence in support of its facially implausible economic assumption, cf. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986) (holding that for summary judgment purposes, “if the claim is one that simply makes no economic sense — [plaintiffs] must come forward with more persuasive evidence to support their claim than would otherwise be necessary”), and some explanation why accepting that assumption would not render worthless any market-based measure of value — including the DCF measure incorporated into Opinion 190 itself. As Judge Williams convincingly *341demonstrates, FERC has failed to meet these burdens, and its order must therefore be set aside. See ante at 336-338.
Given this unusual background, Tennessee seeks an unusual remedy — an order instructing FERC on remand to set the return on equity at 15.9%. Reasoning that “ratemaking is for the Commission and not for us,” ante at 335, the court decides to remand without instructions. Were this a case in which Tennessee argued to this court that a 15.9% rate of return is appropriate because it is “just and reasonable,” 15 U.S.C. § 717c(a), the court’s analysis undoubtedly would be correct. Tennessee, however, argues that 15.9% is appropriate because it is required by FERC’s own prior precedents — specifically, by Opinion 190. FERC concedes that Opinion 190, if applied, would compel this result.1 The issue, then, is not whether this court may set a rate of return in the first instance when FERC has repeatedly failed to justify the rate it selects; rather, the issue is whether this court may order FERC to set the rate of return compelled by its own precedents when FERC has repeatedly failed to justify a deviation from those precedents.
Greyhound Corp. v. ICC, 668 F.2d 1354 (D.C.Cir.1981), involved an analogous situation. In Greyhound, the Interstate Commerce Commission repeatedly sought to exercise securities jurisdiction over a particular holding company of a common carrier, despite administrative precedents that precluded it from doing so. Twice, we concluded that the agency had failed to provide sufficient justification for the deviation. See id. at 1358-61; Greyhound Corp. v. ICC, 551 F.2d 414, 417-18 (D.C. Cir.1977). The second time, we remanded with directions to release the holding company from the agency’s jurisdiction, finding “no useful purpose to be served by allowing the Commission” a third opportunity to proffer a satisfactory explanation. See 668 F.2d at 1364. Greyhound thus indicates that at some point, the reviewing court may bind an agency to its own precedents.
In many respects, this is an appealing case for application of Greyhound. FERC seeks not just a second, but a third bite at the apple (or a fifth bite, counting the two denials of rehearing). Moreover, this case involves proceedings that have dragged on now for about eight years, see Tennessee Gas Pipeline Co., 25 F.E.R.C. ¶ 63,052, at 65,166-70 (1983) (ALJ’s initial decision on rate of return issue) — a factor we considered relevant in Greyhound itself, see 668 F.2d at 1364. Finally, the woeful inadequacy of FERC’s explanations, in both the original and the remanded proceedings, suggest the sort of agency intransigence against which Greyhound is designed to protect.
Ultimately, I conclude that this is not an appropriate case in which to apply Greyhound. Except in rare situations where, unlike here, the reviewing court may confidently conclude that no explanation would justify a deviation from administrative precedent, application of Greyhound inevitably entails some judicial usurpation of agency authority. For that reason, I believe, Greyhound must be reserved for truly extraordinary situations: legitimate concerns about judicial overreaching always militate in favor of affording the agency just one more chance to explain its decision. Thus, in the decade since Greyhound was decided, we have not once invoked it to afford the remedy it authorizes. Tennessee points to no other cases during that time affording the same remedy, and I have been unable to find any.
FERC’s conduct, though bad, has not yet reached the level of egregiousness necessary to trigger a once-in-a-decade sort of remedy. In my judgment, however, FERC has come perilously close.
. In fact, FERC concluded that application of Opinion 190 would require a rate of return of 15.99% — higher than the rate sought by Tennessee. See 49 F.E.R.C. at 62,419-20.