The Indiana Municipal Electric Association and its 24 municipal electric system members (Association)1 have asked us to review two orders of the Federal Energy Regulatory Commission.2 The Public Service Company of Indiana, Inc. (the Company) became an intervenor in these proceedings on September 19, 1979. The Association has asked us to reverse two of the Commission's orders and remand the cause for further consideration, whereas the Company has asked us to affirm the orders. We affirm.

In September 1975, the Company applied to the Commission for an increase in its wholesale electric rates from February 24, 1976,3 to January 27, 1979. Pursuant to the applicable regulations,4 the Company filed actual cost of service data for Period I and projected cost of service data for Period II.5

To understand the problem presented by this case, it should be remembered that during periods of high demand, a utility may have to purchase short-term power from other utilities to meet additional demands of its own customers. This cost is treated as an operating expense for which the Company is entitled to compensation. At other times, a utility may have generated more power than its own customers need and consequently will sell the excess power to other utilities. The revenues from these sales are used to reduce the overall operating costs of the Company by crediting the amount of revenues received to purchased power expense. Of course, as the Administrative Law Judge (ALJ) realized, a utility may be a net importer of power in one year and a net exporter in the next year. Therefore it is rather tricky for a utility to project the exact position in which it is likely to be in terms of necessary net purchases. But to be successful in securing an adjustment, the party challenging the proposed utility rates has the burden of proving the projections "substantially in error." As the ALJ put it, the "substantiality" requirement has been imposed because "a certain degree of latitude is required in deference to the fact that unanticipated subsequent events normally cut both ways, i. e., overstated estimates will (as here) almost certainly be balanced by other offsetting understatements by the company" (Joint App. 30, quoting from Commission Opinion No. 783, Ass'n App. B-2 at 13).

In its September 1975 rate application, the Company projected revenue from wholesale short-term sales of power for Period II at approximately $11.5 million (Joint App. 3 n. 6). This figure was used to offset the $11,475,697 costs that the Company expected to incur in making short-term power purchases during high demand intervals (Joint App. 95). When this administrative hearing commenced in October 1976, the actual data for part of Period II was available. At that time, the Association intervened to show that during the first nine months of Period II, the Company already received $13,755,064 from utilities to which it made short-term sales of power. Therefore, the Association sought an adjustment of $3,273,406 in the Company's short-term purchased power expense estimate (see infra ).

The Company introduced rebuttal evidence on this subject. Its comptroller testified that these larger than projected revenues resulted from the sale of short-term power to the Tennessee Valley Authority (TVA) for $2,896,000 (Joint App. 3). Because two of TVA's generating units had been damaged by fire, as he noted, such sales were atypical and not a true reflection of normal operations. The Company also introduced actual overall wholesale cost of service data for Period II. This material showed that on the basis of the Company's rates which were in effect subject to refund,6 estimated overall wholesale operating revenues for Period II had been $48,303,000 and actual operating revenues were $44,263,000. The actual adjusted revenues for that period were therefore about $4,040,000 less than estimated, while actual pre-tax expenses were about $4,400,000 less than estimated (Joint App. 97).

In his decision of June 27, 1978, the ALJ found that the Company's projected purchased power expense should be used as a basis for determining its rate of return for the period beginning February 24, 1976, to January 27, 1979. He concluded that the Company had proved that its purchased power expense estimate was a reasonable projection upon which to base rates for the time period they would be in effect, and that this estimate was not substantially in error because the Company "attempted to balance the fluctuation(s) and reflect costs that are typical of the Company's operating costs over a span of time longer than the (Period II) test year" (Joint App. 30).

Before the Commission, the Association excepted to the ALJ's decision on the ground that the disparity between the actual and projected purchased power expense required an adjustment. On June 28, 1979, the Commission handed down Opinion No. 44 rejecting the Association's claim (Joint App. 42-71). In that opinion, the Commission indicated that valid test period projections would not be considered substantially in error unless the use of such projections would yield unreasonable results (Joint App. 46). It found that the Company had established that its estimated purchased power expense for Period II was reasonable when made (Joint App. 4-5) and that the Company had presented evidence of actual total Period II expenses which supported the use of its projections as yielding a reasonable rate result (Joint App. 49). The Association's petition for rehearing was denied by order of August 27, 1979 (Joint App. 88-99). The Association has asked us to reverse the Commission's June 28, 1979, order accompanying Opinion No. 44 and to reverse the Commission's order of August 27, 1979, denying a rehearing of Opinion No. 44. We hold that the Commission's order of June 1979 resulted in just and reasonable rates and therefore rehearing was properly denied.

I. The Commission's Determination of Wholesale Electric Rates on Projected Cost of Service Data for a One-Year Test Period Was Proper.

The opinion under attack was based on test year rate-making concepts. Under the applicable regulations (n. 4 supra ), a utility wishing to raise its wholesale electric rates must file cost of service data for two time periods. Period I is to contain actual data for the most recent 12 months available. Period II, the test year, is to contain estimated cost of service data for any 12 consecutive months beginning after the end of Period I but no later than the proposed effective date of the rate filing, here February 24, 1976 (Ass'n App. A 12). The test year rule embodied in Period II was initiated by a December 14, 1972, notice from the Commission and was incorporated in the regulations in July 1973 because the historic test period exclusively used in the past was too rigid to result in the truly just and reasonable rates mandated by Section 205(a) of the Federal Power Act (16 U.S.C. § 824d(a)).

As noted, the Association has challenged the projected Period II rates submitted by the Company regarding its short-term purchased power expense because the actual figures for that period differed from the projections. In effect it contends that the Federal Power Act requires available actual cost ratemaking, but this argument was rejected in American Public Power Association v. Federal Power Commission, 522 F.2d 142 (D.C.Cir.1975), upholding these regulations. We agree with that decision that Congress did not compel the Commission to set rates solely on the basis of historic test period costs but gave the Commission "broad discretion in regard to the methodology of testing the reasonableness of rates." 522 F.2d at 146. To require a reworking of a utility's estimated costs in light of subsequent actual costs not only would result in interminable delays in already lengthy rate proceedings but would encourage dilatoriness in challengers in the hope that history would spoil the utility's estimated cost of service.

Since rate proceedings often are unfortunately of long duration, actual Period II data sometimes becomes available during the administrative hearings. However, as the Commission properly noted, actual costs are not really more reliable than projected costs because, as here, the actual costs for Period II may reflect a unique situation. Therefore the Commission requires a utility to "substantiate the Period II figure in terms of its typicality not only for the test period but also for the projected effective term of the tendered rates."7 However, if a utility always had to adjust its Period II projections because of actual experience as the Association contends, the Commission would be forced to return to historic cost even though Congress did not so intend. American Public Power Association v. Federal Power Commission, supra, 522 F.2d at 146. In accord with the D.C. Circuit, we hold that public utilities may file for wholesale electric power rates "based on estimates of future costs as well as past actual costs, rather than upon past actual costs alone" (522 F.2d at 143) and that the test year method employed here was reasonable as a rate-fixing device.

II. The Company's Protected Short-Term Public Power Costs Were Typical and Its Actual Purchased Power Costs Were Atypical.

The Association submits that during the first nine months of Period II the Commission must consider the Company's $13,755,064 gain from sales of power to other electric utilities instead of its projected $6,475,000 gain.8 The Association, after certain adjustments, therefore proposed that the Company's projected minus $6,475,000 purchased power 9-month expense be decreased by $3,273,406, thus decreasing the Company's proposed rate increase by $614,690 (Br. 4-5; Joint App. 46).9 Nevertheless, the Association has not shown that the Company's short-term purchased power expense projection was unreasonable when made or that the Company did not prove it was a reasonable basis upon which to base rates.

In rejecting the proffered adjustment, the Commission used two tests. First, it concluded that the challenged estimates were not clearly unreasonable when made. It noted that the Company's purchased power expense for Period I (the year ending June 30, 1975) was $3,542,060 whereas its projected purchased power expense for the following year was minus $6,475,000. The Commission stated that it was reasonable for the Company to predict a Period II expense figure $10 million lower than its actual expense in Period I because its two huge generating plants had been put into operation, changing the Company from a net importer of purchased power to a net exporter. Because the Company's Period II minus $6,475,000 expense estimate therefore did not "deviate unaccountably from a typical pattern of experience" (Joint App. 47), it was sustained as not clearly unreasonable when made. We cannot fault this conclusion, particularly since the actual purchased power expense for the full Period II was atypical in that it included almost $3,000,000 short-term power sales to TVA because of fire damage to two of its units (Joint App. 3). Therefore the Commission properly took the position that "sound estimates * * * may well comprise a far more reliable * * * basis for rational ratemaking" when actual date for a year, as here, would "present a mere aberration from the norm" (Joint App. 49).

Second, the Commission concluded that this estimate was acceptable because subsequent events did not "indicate that to use the challenged estimate as a basis for future projections would yield unreasonable results" (Joint App. 46). It noted that the placing of the two new generating units in operation exacerbated the Company's difficulty in estimating future sales of excess power. The Commission also noted that, as shown by the Company's actual costs and expenses, its $4,040,000 underestimate in wholesale revenues for the entire Period II was almost totally offset by an underestimate of wholesale operating expenses of $4,400,000. These were both wholesale revenue and expense figures, so that the Commission did not offset unrelated accounts or use retail and wholesale figures as the Association has contended. This offset was considered corroborative evidence of the reasonableness of the Company's proffered estimates, so that their use as a basis for future projections would "evidently yield reasonable ratemaking results" (Joint App. 49).

The Company's projected Period II net operating income was $9,850,000 and the actual was $10,021,000. The fact that there was only a spread of $168,000 attests to the reasonableness of the estimates. If we were to accept the Association's view, its members would benefit both from the estimated data in the Company's underestimated categories and the actual data in its overestimated categories! As the Company has capsulated it, such a method would produce "one-way streets both running in their direction" (Company Br. 5), a result surely not intended by Congress and producing a wholesale rate at a level lower than projected to be needed and, as time proved, lower than actually needed.

III. Because of the Differing Records the Commission's Opinion No. 44 Is Consistent with Its Opinion No. 783-A.

In Opinion No. 783-A, which we affirmed in large measure in 1978 (575 F.2d 1204), the Commission had rejected the Company's projection of $15,113,000 purchased power costs for the first nine months of 1974, whereas the actual experience for early 1974 showed that net purchased power expense would probably be zero. The Commission rejected the $15,113,000 estimate because at the time it was made "there was no reasonable expectancy that the company would have substantial off-system resales" (Association App. B-2 at 14). Nevertheless, at the same time, the Commission recognized that actual cost data would not be dispositive where, as in the present case, there were short-lived abnormal conditions (ibid. at 15) such as the unexpected TVA purchases.

Also in Opinion No. 783-A, the Association sought an adjustment of $1,051,000 because the Company's estimated operating expenses, exclusive of fuel and purchased power expense, were assertedly greater than actual operating expenses, also exclusive of fuel and purchased power expense. However, the Commission refused to make such an adjustment because although "other operations" expense was below estimates, "maintenance expense" was above estimates so that the Association had not shown that "the estimated cost of service as a whole would become substantially excessive" (ibid. at 16). Similarly, the Commission considered offsets in the present case (see p. 484 supra ).

Consistently with the theory of Opinion No. 783-A, in the present proceeding the Company's estimate was not rejected because here the Association has not shown that "subsequent events indicate that to use the challenged estimate as a basis for future projections would yield unreasonable results" (Joint App. 46) and because the Company's rebuttal comprehensive updated study (completely absent at the close of the record in the No. 783-A rate case) employing actual figures for Period II (Joint App. 97) showed that the Company's forecasts did not yield unreasonable results. If there had been such rebuttal testimony in the earlier case, the Commission and this Court would not have disturbed the Company's estimated purchase power costs for 1974.

IV. The Commission's Use of the Company's Projections Does Not Result in Unjust and Unreasonable Rates.

The Commission does not make hindsight adjustments in test year cost of service data based only on actual test year variance as the Association would like it to do here. Indeed the Commission has permissibly "indicated its preference to rely exclusively on the test year data unless it can be demonstrated that the estimates were either unreasonable when made, or if reasonable when made, subsequent events indicate that to use them as a basis for future projections would yield unreasonable results." Commission Opinion No. 55 (August 1, 1979) at 7, reproduced in Commission App. A; see also Opinion No. 783-A supra refusing an adjustment unless the estimated cost of service as a whole would be excessive (Ass'n App. B-2 at 12, 13, 14, 15). Therefore, to secure an adjustment, the Association had to show that the use of the Company's projections would yield an unreasonable result, but the Commission correctly found that use of the projections would not yield unreasonable rates here partly because the actual Period II figures submitted by the Company on rebuttal showed that total operating expenses for the wholesale class revenues were $4,400,000 less than projected while revenues from wholesale customers were lower by $4,040,000 (Joint App. 46, 48 n. 16, 49).

In reaching the conclusion that use of the projections would not yield unreasonable rates, the Commission applied its expertise to the various components of the cost of service data contained in the evidence of record. The Commission consequently granted the Company a 12.5% return on common equity instead of the requested 13.2%, so that the Company's overall rate of return became 9.14% instead of 9.75% requested by the Company.

The Commission did not operate only on a "balancing of errors" approach despite such characterization by the Association (Br. 22-25). The Commission used the Company's overall cost of service projections, simply noting that any errors in the Company's Period II expense estimates were almost totally offset by errors in its revenue estimates (Joint App. 49). The Commission did not solely balance the purchased power projection against any other account but merely stated that the Company's schedule of overall actual expenses and revenues was relevant to show the reasonableness of the Company's proffered estimates (idem ). Such nonconclusive use of offsets is in tune with the "end result" test established in Federal Power Commission v. Hope Natural Gas,10 320 U.S. 591, 602, 64 S. Ct. 281, 288, 88 L. Ed. 333, where the Supreme Court set forth the following precepts to govern ratemaking:

"Under the statutory standard of 'just and reasonable' it is the result reached not the method employed which is controlling. * * * It is not theory but the impact of the rate order which counts. If the total effect of the rate order cannot be said to be unjust and unreasonable, judicial inquiry under the Act is at an end. The fact that the method employed to reach that result may contain infirmities is not then important. Moreover, the Commission's order does not become suspect by reason of the fact that it is challenged. It is the product of expert judgment which carries a presumption of validity." (Citations omitted.)

Instead of balancing wholesale against retail costs as the Association suggests it did (Br. 23, 25), the Commission considered the Company's actual Period II figures for wholesale revenues and wholesale expenses in determining whether the use of the estimates would result in unreasonable rates. Consistently with Opinion No. 783-A (Ass'n App. B-2) the Commission adhered to the standard that "particular items of expense, if challenged as excessive, must be demonstrated to have been substantially in error because of subsequent events which were not reasonably foreseeable at the time such estimate(s) were developed" (Ass'n App. B-2 at 13). As noted previously, the reason for requiring substantiality is that a certain degree of latitude has to be permitted since overstated estimates would almost certainly be balanced by other offsetting understatements. Idem. In sum, the Association has been unable to satisfy the Commission or us that the Company's Period II projections are "self-serving misprojections" (Ass'n Br. 36) indicative of mala fides.

Since the Association has not shown that the rates being established by the Commission in this proceeding are unjust and unreasonable and since the Commission properly allocated burdens of proof, the Commission's rate increase order of June 28, 1979, and its order of August 27, 1979, denying rehearing are affirmed.