Idaho Power Company petitions this court to set aside the Federal Energy Regulatory Commission's decision to license construction of a hydroelectric plant at Horseshoe Bend, a site on the Payette River in Boise County, Idaho. See Boise Cascade Corp., 36 F.E.R.C. p 61,135 (1986) (the "Licensing Order"), request for reh'g denied, sub nom. Horseshoe Bend Hydroelectric Co., 42 F.E.R.C. p 61,072 (1988) (the "Rehearing Order"). The original licensee was Boise Cascade Corporation, but it transferred the license to its wholly-owned subsidiary, Horseshoe Bend Hydroelectric Company. In this opinion we refer to the licensee as Boise, to avoid confusion with our use of Horseshoe Bend for the project or the site. Idaho Power attacks the Commission's findings that there was a "need" for the plant and that it was financially feasible. The grounds of attack do not persuade us, and we dismiss the petition.

The Commission has jurisdiction under Sec. 4(e) of the Federal Power Act, 16 U.S.C. Sec. 797(e) (1982 & Supp. IV 1986), to license dams on navigable streams. Sec. 10(a) of the Act provides that it is to issue a license only on the condition that the project will be

best adapted to ... developing a waterway ... for the use or benefit of interstate or foreign commerce, for the ... utilization of water-power development, for the adequate protection ... of fish and wildlife ... and for other beneficial public uses.

16 U.S.C. Sec. 803(a) (1982 & Supp. IV 1986). The Supreme Court has construed the subsection as creating a "public interest" test, see Udall v. FPC, 387 U.S. 428, 450, 87 S. Ct. 1712, 1724, 18 L. Ed. 2d 869 (1967). The Court included within that concept a consideration of "future power demand and supply, [and] alternate sources of power," id., and the parties appear to agree on interpreting the statute (and the Court's gloss) as requiring a determination of "need." It is not clear just how the Commission relates "need" to demand and supply, which are normally conceived of as depending upon price, but no party raises the issue.

Idaho Power first argues that by considering the energy needs of the entire Pacific Northwest, FERC departed from what it terms FERC's prior policy of restricting need-for-power determinations to the much smaller area to which the applicant expressly proposed to transmit the power. Here that would mean Idaho Power itself, for the application stated that the power would "be sold to the Idaho Power Company." Joint Appendix ("J.A.") at 60. Boise's expectation of sales to Idaho Power was apparently based on its anticipation that the plant would be a "qualifying facility" under Sec. 210 of the Public Utility Regulatory Policies Act ("PURPA"), 16 U.S.C. Sec. 823a (1982), enabling it to require a utility, here Idaho Power, to purchase its output at the purchaser's "avoided cost."

Idaho Power relies on three cases for its characterization of FERC's prior policy: (1) Idaho Power Co., 25 F.E.R.C. p 61,436 (1983), on reh'g, 27 F.E.R.C. p 61,175 (1984), aff'd, Idaho Power Co. v. FERC, 767 F.2d 1359 (9th Cir.1985) ("Idaho Power I "); (2) Joseph M. Keating, 24 F.E.R.C. p 61,343 (1983), 32 F.E.R.C. p 61,290 (1985), rev'd on other grounds, sub nom. Harriet F. LaFlamme v. FERC, 842 F.2d 1063 (9th Cir.1988); and (3) Calaveras County Water District, 18 F.E.R.C. p 61,124, reh'g denied, 20 F.E.R.C. p 61,031 (1982), aff'd sub nom. Friends of the River v. FERC, 720 F.2d 93 (D.C.Cir.1983).

In fact none of these supports Idaho Power's description of FERC's previous policy. In both Keating and Calaveras FERC found a need for power within a narrow region and thus had no cause to look further. See Keating, 24 F.E.R.C. p 61,343 at 61,734; Calaveras, 20 F.E.R.C. p 61,031 at 61,056. In Idaho Power I, the applicant itself, Idaho Power, had represented that the power would not be necessary for a considerable time and that it did not intend to start construction on issuance of a license. See Idaho Power I, 767 F.2d at 1360-61. Idaho Power had in essence rested its case on a contention that the Commission should allow "site banking" by the licensees--i.e., should issue licenses to applicants in anticipation of their delaying construction until the arrival of a better moment. The Commission read the Federal Power Act as precluding such a license, and the Ninth Circuit affirmed its view. Id. at 1362-63. Thus, as FERC explained, see 42 F.E.R.C. p 61,072 at 61,321-22, 61,326 n. 5, there is no inconsistency for FERC to look to the needs of a broader region to find support for an applicant's claim that power is necessary. While it is not clear at whose initiative FERC came to extend its inquiry beyond the area implicit in Boise's application, the cited cases plainly do not bar that extension.

Idaho Power also claims that FERC simply assumed that the Horseshoe Bend power could get to where FERC had concluded it would be needed. It is quite true that FERC did not attempt to pin down either a specific destination or an available route for Horseshoe Bend power, but to characterize its reasoning as mere assumption is unfair. It noted that selecting either in advance would be treacherous, as the Pacific Northwest transmission system featured "major and varying circulating power flows that change direction and magnitude with seasonal changes in contractual agreements." Rehearing Order, 42 F.E.R.C. p 61,072 at p 61,323. But it pointed to evidence of growing interties in the region, not disputed by Idaho Power, and on that evidence (plus its finding of the existence of a potential market in the region), surmised that Boise could find a route for the power. Given the Commission's expertise, we cannot say the surmise was unreasonable.

II. "FINANCIAL FEASIBILITY" AND "LEVELIZED COST"OST"

Idaho Power first claims that in finding "financial feasibility" the Commission abandoned an approach it had employed in Utah Board of Water Resources, 27 F.E.R.C. p 61,437 (1984). There, as here, it was assessing a small power production facility that would in all likelihood be eligible under PURPA to "put" its power to a utility at avoided cost rates.1 It determined feasibility by asking whether the project's annual power costs would exceed the avoided cost of the prospective buyer.

As Idaho Power's avoided costs are now 47 mils/kwh and the project's "levelized cost" is 76.7 mils/kwh, an analysis exactly tracking Utah Board would seem to point to infeasibility. In its initial order, the Commission resolved the issue by suggesting that Idaho Power's avoided costs might rise in the future, and that other utilities in the region might buy for as much as 80 mils/kwh. In its Order on Rehearing the Commission elaborated somewhat, concentrating on a hypothetical coal plant that it projected would likely be necessary by 2002 and that would produce at a "levelized" cost of 96.9 mils/kwh. This fleshed out FERC's earlier suggestion that in later periods the Horseshoe Bend power would sell for more than could be inferred from Idaho Power's present avoided costs. Thus the different facts appear to supply ample reason for adjusting the approach taken in Utah Board of Water Resources.

Next, Idaho Power attacks the Commission's use of the hypothetical coal plant on the ground that it fails to support an assumption that "someone will be willing to purchase Horseshoe Bend power for up to 96.9 mils/kwh in 1992." Petitioner's Brief at 40.

The Commission's Order on Rehearing does not seem to us to claim that anyone would purchase Horseshoe Bend power in 1992 at 96.9 mils/kwh, or even at a price exceeding 76.7 mils/kwh. As we read the Order, its point is that the higher revenues in 2002-2042 are enough to offset losses in 1992-2002. It is not necessarily material who bears them. If indeed the later gains outweigh the early losses, then the plant will be feasible for Boise itself--even if it bears the early losses. At oral argument, Commission counsel suggested that because of the expectation of higher prices in 2002 (or thereabouts), a potential buyer might pay Boise 76.7 mils/kwh even as early as 1992, in order to assure itself the long-run supply. If that is so, so much the better for Boise. But we do not think it necessary to support the Commission's principle that financial feasibility may rest upon late-year gains outweighing early losses.

Of course this outweighing is possible only if the Commission has properly compared costs and benefits in different time periods. In making its comparison, the Commission used what it refers to as "levelized cost" figures. First, it took 76.7 mils/kwh as the "levelized cost" of power from Horseshoe Bend. Then, to calculate the return, it considered two phases, 1992-2002 and 2002-2040. For the first period it used Idaho Power's current avoided cost of 47 mils/kwh, and for the second it used the hypothesized fossil-fuel plant's "levelized cost" of 96.9 mils/kwh. If we assume that Horseshoe Bend's power production is the same each year, and that the units employed are truly the same, the 38 years of gaining 20 mils/kwh would outweigh the 10 years of losing about 30 mils/kwh. The units indeed are the same only if they are all present discounted values (or values discounted to any specific year--the Commission here apparently used 1992). At oral argument counsel for intervenor assured us that the "levelized costs" used were indeed all discounted to present values. Although the record before us is not clear, no one questioned his assertion, and we take it as correct. Accordingly, we can find no error here.

None of this discussion, we note, bears directly on the suitability of licensing a plant that is expected, for 10 years after it starts production, to generate power costing more than the power it displaces (if that is what the Commission expects--it isn't clear). Such a 10-year period of costs exceeding the apparent benefits (equal to alternative costs averted) suggests that FERC might do better to "bank" the site for development 10 years later. By deferring costs and eliminating or reducing a period where the benefits generated are of relatively low value, this strategy would increase the present discounted value of the site's use. Nor would it appear to contradict the Commission's view, approved by the Ninth Circuit in Idaho Power I, that it may not issue licenses under which the licensee banks the site. But no one raises the issue.

DENIED.