Midwest Gas Users Ass'n v. State Corporation Commission

3 Kan. App. 2d 376 (1979) 595 P.2d 735

MIDWEST GAS USERS ASSOCIATION and SEYMOUR FOODS, INC., Applicants/Appellants,
v.
STATE CORPORATION COMMISSION OF KANSAS, Respondent/Appellee.

Nos. 50,380, 50,670

Court of Appeals of Kansas.

Opinion filed June 1, 1979. Petition for review denied September 11, 1979.

W.H. Bates, of Lathrop, Koontz, Righter, Clagett, Parker & Norquist, of Kansas City, Missouri, and Thomas L. Theis, of Sloan, Listrom, Eisenbarth, Sloan & Glassman, of Topeka, for applicants/appellants.

Walker A. Hendrix, special counsel, and Bruce E. Miller, general counsel of the state corporation commission, for the respondent/appellee.

Donal D. Guffey, of Kansas City, Missouri, and Richard C. Byrd, of Anderson, Byrd & Richeson, of Ottawa, for intervenor The Gas Service Company.

Before FOTH, C.J., SPENCER and MEYER, JJ.

FOTH, C.J.:

In this case we are asked to find unreasonable a rate schedule for the retail sale of natural gas proposed by the Gas Service Company and approved by the Kansas Corporation Commission. We are unable to so find.

The issues in this case were first raised on December 9, 1976, when the Company filed an application for a rate increase with the KCC. That application was assigned Docket No. 110,038-U. Appellant Midwest Gas Users Association intervened in that proceeding. Midwest is an association, many of whose members are customers of the Company who are assigned to an "interruptible" status. When total demand on the Company's system exceeds its capacity, service to interruptible customers may be curtailed or cut off. Interruptible customers who wish to continue operations during foreseeable periods when their gas supply is cut off necessarily have available alternative sources of fuel and the physical capacity to utilize such alternatives. Appellant Seymour Foods, Inc., is an industrial "interruptible" customer who separately intervened, but who joined Midwest in all previous proceedings and in this court. We thus have no issue as to Midwest's standing, and all references to "Midwest" are equally applicable to Seymour.

Midwest's objection to the Company's application went not to the amount of the requested increase but to the proposed rate structure. Its claim was that the rate schedule placed too much of the burden of the increase on "interruptible" customers and too little on the Company's "firm" customers, such as residential users, who have a guaranteed supply regardless of demand. The KCC approved the rate schedule and Midwest sought review (under the then provisions of K.S.A. 66-118c) in the district court of Jefferson County. That court affirmed the KCC in November, 1978, and Midwest appealed to this court.

In the meantime, on January 30, 1978, the Company filed a second application for a rate increase, including a new rate schedule. That proceeding was assigned KCC Docket No. 113,728-U. Midwest (and Seymour) again intervened making the same objections to the proposed rate structure. The KCC authorized part of the rate increase sought, and in its order again approved a rate schedule containing the same features deemed objectionable by Midwest in the first proceeding. Midwest again sought judicial review, this time in this court under the 1978 amendment to K.S.A. 66-118a giving us exclusive jurisdiction of such proceedings, in lieu of the district court. (L. 1978, ch. 265, § 1, effective July 1, 1978.) The review proceeding in this court was commenced while the earlier proceeding was still pending in the district court.

The two proceedings were consolidated in this court. The KCC and the Company moved to dismiss the appeal in the earlier proceeding on the grounds of mootness, citing Six Cities v. State Corporation Commission, 213 Kan. 413, 516 P.2d 596 (1973). That position may technically be correct, since the later rate schedule supersedes the earlier one and, since no stay bond was posted, revenues collected under the earlier schedule are not subject to refund. We nevertheless decline to dismiss because of the close interrelationship of the two proceedings, including an identity of issues and the fact that the order entered and much of the testimony taken in the earlier proceeding were relied on and incorporated by reference in the later one. While our discussion is largely in terms of the later proceeding, it applies in principle to both. Since we affirm in both, the result is the same.

I.

It should be noted at the outset that neither of the rate increases sought was based on the increased cost of gas to the Company. Higher prices from the Company's wholesale supplier (Cities Service Gas Pipeline Company), when authorized by the federal regulatory agency, are passed on to the Company's retail customers by "purchased gas adjustments." Those price increases are preauthorized by the KCC and do not require KCC approval each time they are encountered, although their effect will later be incorporated in a proposed rate which is submitted for approval. Rather, the increases sought here are the result of increased operating expenses.

As previously noted, Midwest's objection is not to the amount of the total rate increase allowed, but to the manner in which the increase is to be spread among the Company's customers. That proposal is to increase the rates charged to each customer by a uniform amount (the proposal was $.0435) per thousand cubic feet (Mcf) of gas consumed. This makes the increase proportionately higher for "interruptible" customers because they have always paid lower rates than the Company's firm customers. Although the rate schedule covers a number of different classes of customers, the following incomplete and highly condensed table from the latest application will illustrate the result:

percentage
base rate    increase   new rate    increase
Domestic & small            $2.9589      .0435     $3.0024      1.47
commercial/industrial
(firm customers)
Large commercial            $1.1476      .0435     $1.1911      3.79
(interruptible)
Large industrial            $1.1449      .0435     $1.1884      3.80
(interruptible)

The "base rate" is the charge per Mcf for the first "block" of gas consumed each month. Rates decrease as usage increases, with lower rates prescribed for each additional block.

The historical justification for the rate discrepancy demonstrated has been this: A utility must have the plant and distribution capacity to meet its firm customers' peak demand — for a gas company, the winter months. During the slack season there will necessarily be excess capacity; the interruptible customers take the product during the slack season with only a minimal cost to the utility for additional distribution lines, thus purchasing a commodity that would otherwise go unsold. Any revenue from such sales over the direct cost of making them helps meet fixed costs. Thus such sales, even at reduced rates, are seen as benefiting the firm customers by permitting the reduction of their rates by the amount of "profit" derived from the interruptibles, while preserving the utility's overall return on its rate base. The rates set seek to achieve a balance between the elements of demand and the quantity of the commodity consumed.

Application of this rationale in the past has resulted in the above schedule, whereby domestic and other firm customers pay something over 2 1/2 times as much for gas as large commercial and industrial users. The proposed increase, at a flat rate per Mcf for all customers, will have a tendency to reduce the disparity, although obviously it is far from eliminating the price concession made to interruptible customers.

II.

K.S.A. 1978 Supp. 66-118d limits judicial review of an order by the commission to determining whether the order is "lawful" or "reasonable." Kansas Gas & Electric Co. v. State Corporation Commission, 218 Kan. 670, Syl. ¶ 1, 544 P.2d 1396 (1976). A court has no power to set aside such an order unless it finds that the commission acted unlawfully or unreasonably. Jones v. Kansas Gas and Electric Co., 222 Kan. 390, 396-7, 565 P.2d 597 (1977). An order is "lawful" if it is within the statutory authority of the commission, and if the prescribed statutory and procedural rules are followed in making the order. Central Kansas Power Co. v. State Corporation Commission, 221 Kan. 505, Syl. ¶ 1, 561 P.2d 779 (1977). An order is generally considered "reasonable" if it is based on substantial competent evidence. Jones v. Kansas Gas and Electric Co., 222 Kan. 390, Syl. ¶ 2.

The legislature has vested the commission with wide discretion and its findings have a presumption of validity on review. Central Kansas Power Co. v. State Corporation Commission, 221 Kan. at 511. Since discretionary authority has been delegated to the commission, not to the courts, the power of review does not give the courts authority to substitute their judgment for that of the commission. Central Kansas Power Co. v. State Corporation Commission, 206 Kan. 670, 675, 482 P.2d 1 (1971). The commission's decisions involve the difficult problems of policy, accounting, economics and other special knowledge that go into fixing utility rates. It is aided by a staff of assistants with experience as statisticians, accountants and engineers, while courts have no comparable facilities for making the necessary determinations. Southwestern Bell Tel. Co. v. State Corporation Commission, 192 Kan. 39, 48-9, 386 P.2d 515 (1963). Hence a court may not set aside an order of the commission merely on the ground that it would have arrived at a different conclusion had it beun the trier of fact. It is only when the commission's determination is so wide of the mark as to be outside the realm of fair debate that the court may nullify it. Kansas-Nebraska Natural Gas Co. v. State Corporation Commission, 217 Kan. 604, 617, 538 P.2d 702 (1975); Graves Truck Line, Inc. v. State Corporation Commission, 215 Kan. 565, Syl. ¶ 5, 527 P.2d 1065 (1974).

III.

The primary issue in this case is whether the KCC could lawfully adopt as it did a "volumetric" rate structure, or whether it was required to apply a uniform percentage increase to the existing base rates. In a case such as this a volumetric rate design and a uniform percentage rate design differ not only in their effect, but also in the factors which are relevant in their application. When a uniform percentage approach is adopted, the primary factor is the cost of service — the amount of money expended by a utility in providing service to each of its customers — which is built into the existing rate design to which the percentage is applied. When a volumetric approach is adopted, the primary factor is the value of service — the capacity and willingness of different customer groups to bear increased costs, and the intrinsic value of the commodity furnished. A uniform percentage approach looks to conditions on the supply side of a utility's market, while a volumetric approach focuses on the demand side of the market.

At the hearing before the KCC there were two primary witnesses on this issue. William R. Chaney, for the Company, testified in favor of the volumetric approach embodied in the rate schedule proposed by the Company and approved by the KCC. Stanley C. Whiteaker, for Midwest, presented his cost analysis of the Company's business. He concluded that the interruptible customers were already subsidizing the firm customers, and that any percentage increase in interruptible rates over the percentage increase in total revenue approved would result in an even greater subsidy.

The KCC rejected Mr. Whiteaker's cost study and in its findings on this issue adopted Mr. Chaney's testimony virtually verbatim:

"VII. RATE DESIGN

"28. The rate design testimony by W.R. Chaney, Applicant's witness and Stanley C. Whiteaker, the witness for Midwest Gas Users Association and Seymour Foods, Inc., was incorporated in its entirety from the Applicant's previous rate case in Docket No. 110,038-U, and administrative notice was taken of all rate design exhibits therein. Both witnesses appeared, but only for the purposes of additional cross-examination by the parties including that by intervenor, Rhodina Montgomery who is represented by Kansas Legal Services Corporation (KLSC) and was not a party to the last rate proceeding involving this Applicant.

"The Applicant's proposal is to increase rates over the four basic service classifications, general service, small commercial/industrial service, large commercial service and large industrial service on a uniform cents per MCF basis. Intervenors Midwest and Seymour again recommended that rate increases be spread among the various classes on a uniform percentage basis consistent with a cost of service rationale. Intervenor Rhodina Montgomery suggested the large industrial users be curtailed from nonessential uses of natural gas and that sales for other industrial uses be at the margin, or replacement cost.

"The four basic service classifications historically employed by the Applicant have had separate rate schedules within each class. This has served to recognize differences in load and cost characteristics associated with the service afforded to the various customers. These classifications recognize differences in costs to provide services according to the classes which have been historically determined. Continued recognition of these differences is consistent with the principle that the customers should be billed on a rate schedule applicable to all customers with similar load and cost characteristics. However, the significance of the historical differentials in rates which have existed in the past is becoming less important and should be further reduced at this time consistent with the Commission's orders in Docket No. 110,038-U. Consideration must be given to the economic impact on the company and its consumers resulting from the ever-deteriorating reserves of natural gas. As existing supplies of gas are consumed it will be necessary to replace them at higher prices in order to maintain an adequate reserve for the future load requirements of the company. We have recognized in Docket No. 106,210-U that industrial sales are of a lower priority use for natural gas, and we conclude that current charges for lower priority services should recognize the effect of making lower priority sales at the present time. This necessitates the imposition of higher current prices in order to replenish existing supplies of gas at higher levels of cost. We recognize the fact that natural gas prices for additions to reserves are constantly increasing. Spreading the rate increase we have granted in this docket on a volumetric basis is consistent with our treatment of this issue in Docket No. 110,038-U, and to a certain extent has the effect of replenishing existing gas supplies at the higher levels of cost.[*]

"Prior to the time that the natural gas shortage became apparent, the incremental unit cost to the company of acquiring new gas supplies was ordinarily equal to or only slightly higher than the average unit cost of the existing supply. In that situation the average unit revenues derived from the sale of gas exceeded the incremental unit cost of the gas, and each sale of natural gas produced some allowance to meet fixed and operating costs of the system. The current level of costs required to obtain new gas supplies presents a situation where the differential in price between new gas and flowing gas could exceed the historical differential provided in commercial and industrial rates to meet the fixed and operating costs of the system, and it appears to us that the incremental unit cost for new gas might exceed the average unit revenues for the lower priced rate schedule customers. Rates must be fair and reasonable between customer classes. It is essential that current rate design consider the prices which are required to cover the incremental cost of acquiring new gas supplies.

"Also to be considered is the fact that the cost of new gas should not necessarily and in all cases be equally assessed to all customer classes. The cost of new gas in this case should apply to those customers who would have been less likely to receive service in the future had the new supply not been made available. The pricing design proposed by the company and adopted by the Commission will provide that the lowest service priority customer, who would be curtailed first due to gas supply deficiencies, be priced at least in part on the basis of the current cost of the gas for the service provided to him. If we did not recognize this principle, it is possible that low priority users could receive gas at prices below the per MCF incremental cost thereof, and this would consequently result in their subsidization by customers served under the higher rate schedules.

"A consideration of costs and rate design cannot be complete without a recognition of the relative risks to the company involved in serving the different customer classes. It appears to us that the risk of serving industrial customers is greater than the risk associated with service of domestic customers. One factor involved in this phenomenon is the ability of industrial customers to convert to alternate fuels, and thus leave the company without a portion of their planned market.

"We must reject the proposal of Intervenor, Rhodina Montgomery, that industrial customers who are not curtailed should only be allowed to purchase gas on a marginal or incremental cost pricing basis since this consideration is not adequately supported by the record. We further find that the proposal for a uniform percentage increase to preserve historical costs of service differential is inappropriate with the continuing diminution of the natural gas reserves. We therefore find that the rate design proposed by the company, which spreads the rate increase on a uniform cents per MCF basis, meets the statutory requirements of K.S.A. 66-107 and is hereby approved. The Applicant shall, therefore, file rate schedules and tariffs to recover the increase in revenues granted consistent with this finding."

The sentence marked by the asterisk was the subject of a clarifying statement in the KCC's order denying rehearing:

"The last portion of this sentence does not accurately reflect what the Commission intended since in and of itself, volumetric rates will not help to replenish supplies. What the volumetric rate will do, however, is to place the burden of replenishing supplies on the class of customers, i.e., interruptible, which necessitate the purchase of new gas."

It is Midwest's position that the volumetric rate increase is unreasonable because it ignores the cost of service to each class of customers. Only by a cost of service analysis, it says, can just and nondiscriminatory rates be determined.

IV.

Our reading indicates that the debate over cost-of-service versus value-of-service has been the concern of utility executives, regulators, engineers, and economists. See e.g., Garfield and Lovejoy, Public Utility Economics, ch. 10 (1964); Bonbright, Principles of Public Utility Rates, ch. IV-V (1961); Brown, Value of the Service as a Ceiling on Public Utility Rates, 33 Calif. L. Rev. 283 (1945); Edgerton, Value of the Service as a Factor in Rate Making, 32 Harv. L. Rev. 516-556 (1919); Bauer, Transforming Public Utility Regulation, ch. XIII (1950); Lander, Public Utility Rate Design: The Cost of Service Method of Pricing, 19 St. Louis U.L.J. 36-55 (1974); Ileo and Parcell, Economic Objectives of Regulation — The Trend in Virginia, 14 Wm. & Mary L. Rev. 547-566 (1973).

As affecting rate design, the competing arguments have been directed to regulatory agencies with varying results; when directed to courts, the idea that cost of service data is legally required has been universally rejected. A few examples will illustrate.

A. Electric utilities. In Globe Metallurgical v. Pub. Util. Comm., 40 Ohio St. 2d 40, 319 N.E.2d 360 (1974), interruptible customers challenged an increase in their rates as part of a general rate increase because there had been no cost study justifying their share of the increase. The court rejected the challenge, saying, "We find no statutory requirement ... as to the necessity of the commission making cost of service determinations relative to specific rate classifications." (p. 41.)

Similarly, in Cleveland Elec. Illuminating Co. v. Pub. Util. Comm., 42 Ohio St. 2d 403, 330 N.E.2d 1 (1975), the court approved a rate structure designed without the benefit of a cost study. While cost studies could be required by the commission under Ohio statutes, whether to do so was discretionary. The court concluded: "Thus, the Commission was not required to make the cost of service allocation for classes of customers as contended by the City." (p. 428.)

In General Motors Corp. v. Pub. Util. Comm., 47 Ohio St.2d 58, 351 N.E.2d 183 (1976), the challenged rate structure decreased charges for small users, but increased rates as use increased — that is, the customary decline in block rates was "flattened." The justification offered by testimony in support of the rates was the need, in the light of the energy shortage, to discourage rather than encourage large use as had been past practice. The court, conceding that the changes were not based on cost of service, observed:

"This testimony also points out the fact that cost-relatedness, although one of many factors to be considered in rate making, is not the sole criteria for allocating allowable gross revenue requirements among various classes of utility users. Clearly, the record and this court's examination of outside sources establish the fact that it is impossible to allocate the fixed costs of a utility to particular customers. Any such attempt to recover costs from customers must ultimately require judgment based upon a determination of what is a just and reasonable rate.

"Appellant's contention that certain types of evidence are required to support a modification of rate structures is rejected by this court. Globe Metallurgical v. Pub. Util. Comm. (1974), 40 Ohio St. 2d 40, [319 N.E.2d 360]." (p. 75.)

In Apartment Hse. Coun. of Met. W. v. Public Serv. Com'n, 332 A.2d 53 (D.C. 1975), the rate increase fell heaviest on apartment houses and lightest on residential users. The court approved despite the absence of cost data, saying:

"Finally as to cost factors, AHC's reliance on PEPCO's failure to present specific cost data for each customer class is misplaced. It is not necessary that differences in rate of return be specifically and quantitatively supported by customer class-cost considerations. The Supreme Court has stated that ``[a]llocation of costs is not a matter for the slide-rule. It involves judgment on a myriad of facts. It has no claim to an exact science.' Colorado Interstate Gas Co. v. Federal Power Commission, 324 U.S. 581, 589, 65 S. Ct. 829, 833, 89 L. Ed. 1206 (1945). Moreover, equal return from customer classes is not required. Differences can be based not only on quantity, but also on the nature, time, and pattern of use so as to achieve reasonable efficiency and economic operation. A rule of thumb is that a rate structure must achieve the lowest price for the largest possible number of users." (p. 57.)

B. Water. In West Allis v. Public Service Comm., 42 Wis. 2d 569, 167 N.W.2d 401 (1969), the objection to a water rate increase was that it was merely applied to a prior cost-based rate schedule without a new cost study. The court there held:

"Determining the total revenue requirements of a utility to give it a reasonable return and designing a rate, i.e., by pricing the product to a particular class of customers or to customers within a class to permit the utility to recover the revenue to which it is entitled, are separate processes which entail different and distinct considerations."

"The commission in designing a rate structure to recover the revenue to which a utility is entitled, as shown by a cost analysis, has wide discretion in determining the factors upon which it may base its precise rate schedule, and is not required to apply a cost-of-service formula to each class of customer or to each customer within a class."

"Subjective factors which make a strict cost analysis approach to rate design unrealistic, both in attainment and in terms of operation, include (a) the unfeasibility and impracticability of measuring and incorporating specific serving costs to customers in a rate schedule; (b) the fact that the sum of differential or marginal costs would not equal the total costs of the utility's operations (i.e., the revenues equaling the sum of the costs of all of a utility's services to individual customers would not be sufficient to recapture the costs of doing business); and (c) the further fact that costs relevant to total revenue requirements are different from those that are sought to be applied in determining specific rates."

"Extrapolated from the preceding rule are the following principles: (a) Operation by which rates are designed to recover revenues is not reducible to a simple mathematical exercise; (b) there is no rigid formula which must be followed by rate-making agencies in setting rate schedules; and (c) structure of rate schedule is within the province of the Public Service Commission." Syl. ¶¶ 2-5.

In reaching these conclusions the Wisconsin court reviewed the leading texts on rate making, and also found its conclusions supported by ample judicial authority:

"It is well established that the Commission in designing a rate structure to recover the revenue to which it is entitled, as shown by a cost analysis, has wide discretion in determining the factors upon which it may base its precise rate schedule. It is not required to apply a cost-of-service formula to each class of customer or to each customer within a class.

"The practicalities of rate making were heavily relied upon in Permian Basin Area Rate Cases (1968), 390 U.S. 747, 776, 777, 88 Sup. Ct. 1344, 20 L. Ed. 2d 312:

"``The Court has said that the "legislative discretion implied in the rate making power necessarily extends to the entire legislative process, embracing the method used in reaching the legislative determination as well as that determination itself." ... It follows that rate-making agencies are not bound to the service of any single regulatory formula; they are permitted, unless their statutory authority otherwise plainly indicates, "to make the pragmatic adjustments which may be called for by particular circumstances.'"

"Other cases following the same well accepted principles are Ayrshire Collieries Corp. v. United States (1949), 335 U.S. 573, 593, 69 Sup. Ct. 278, 93 L. Ed. 243 (there is no rigid formula which must be followed by rate-making agencies in setting rate schedules); Mississippi Valley Barge Line Co. v. United States (1934), 292 U.S. 282, 286, 287, 54 Sup. Ct. 692, 78 L. Ed. 1260 (structure of rate schedule is within the province of commission); Fuels Research Council, Inc. v. Federal Power Comm. (7th Cir.1967), 374 Fed.2d 842, 846 (operation by which rates are designed to recover revenues is not reducible to a simple mathematical exercise); Pittsburgh v. Pennsylvania Public Utility Comm. (1952), 171 Pa. Super. 187, 215, 90 A.2d 607 (rejected argument by party challenging rate that rate charged should be based on cost of supplying service to the consumer and that cost allocation studies should be made to determine this cost).

"It seems clear that no responsibility rests upon the Commission to make the exacting type of cost study that is urged by the appellants. It is sufficient that there be, as there is here, substantial evidence in the record to support the rate as a whole." (pp. 577-8.)

C. Telephone. Rate structures of telephone companies have traditionally been based on value of service to the almost total exclusion of cost of service analysis. In Florida Retail Federation, Inc. v. Mayo, 331 So. 2d 308 (Fla. 1976), the rate structure was contested on the ground that "the ``value of service' principle on which that structure was based is arbitrary, unfair, and generally less desirable as a rate-making basis than a system based on ``cost of service' as the principal criterion." (p. 310.) The court, taking note that two leading text writers take "exactly contrary" positions on the subject, concluded:

"Obviously, there is a divergence of expert opinion as to the policy of including ``cost of service' as an essential element in designing a rate structure. Even were we persuaded to one policy or the other (``cost of service' or ``value of service' as the essential element) it is not our prerogative to impose that policy upon the Commission. So long as the policy adopted by the Commission comports with the essential requirements of law we may not meddle. The Legislature has reposed in the Commission the responsibility to make just the kind of choice between competing policies in its area of expertise as it has done here." (pp. 312-13.)

On the other hand, in Mountain States Tel. v. New Mexico State Corp., 90 N.M. 325, 563 P.2d 588 (1977), we find a regulatory commission rejecting a rate schedule because the utility put forward value of service evidence but not the cost of service data the commission required. The court remanded to the commission, finding that the value of service evidence offered was sufficient for the fixing of rates, and that if cost data were to be required the commission would have to define with particularity the kind of data that would be acceptable. As part of the basis for its decision the court noted the complex nature of the estimates and calculations required by a cost of service study, and went on to say:

"The end result of the machinations is not a determination of the actual cost of any given service. Use of the term ``cost of service' is an over-simplification. At best, the result represents an educated guess as to what the costs may be in the test year. It cannot be dignified by being considered a factual determination. It is tenous expert opinion, or informed-judgment evidence, based upon extremely complex and elusive information." (p. 338.)

A similar situation occurred in New England Telephone & Telegraph Co. v. Dept. of Public Utilities, 371 Mass. 67, 354 N.E.2d 860 (1976), and the same result was reached. There the utility's proposed rate schedule was also rejected by the commission for lack of a supporting cost study. The court reversed, saying that the value of service standard traditionally employed by the commission should have been applied in that case, absent fair warning that a different type of proof would be required.

The difficulty in producing reliable cost of service data was one factor relied on in Granite State Alarm Inc. v. New England Tel. & Tel., 111 N.H. 235, 279 A.2d 595 (1971), where the court approved a rate increase allocated only to some users, and particularly private line users. The evidence justifying the allocation was testimony that basic rates had increased over the years but that private lines "have not kept pace" in the opinion of the witness. Bringing private lines "up to pace" and "various values to users" were sufficient bases for the commission's approval; a cost study was not required.

D. Gas. In Arkansas La. Gas Co. v. Sun Oil Co. of Pa., 554 P.2d 14 (Okla. 1976) our sister state to the south approved a retail gas rate structure where all consumers paid the same rate, with a declining price per Mcf for large amounts consumed. On a "cost" basis, large commercial and industrial users paid lower prices but yielded much higher profits (25%, compared to the 7.5% overall return allowed). The order approved recited the following rationale:

"In arriving at such a rate structure, consideration must be given to two competing principles: first, the cost of service declines on the average with an increasing amount of gas purchased by an individual customer, thus justifying a reduced price with increased purchases; and second, that each unit of natural gas has the same energy value regardless of the amount a customer uses, thus suggesting a similar price per unit without regard to amount of usage." (p. 15.)

In approving a structure which balanced those two principles the court reiterated the accepted principle that "not all discrimination between customers is unlawful with the prohibition applying only to those differences in treatment which are unjust or unreasonable." (p. 16.)

Finally, in Application of Arkansas Louisiana Gas Company, 558 P.2d 376 (Okla. 1976), the Oklahoma court again considered a rate increase which put most of the increase on high volume users, "flattening" the rate structure. The court approved the order despite the fact that the company's rate of return from high volume users was 2 1/2 times its rate of return from residential consumers. The commission's rationale, implicitly approved by the court, was characterized as follows:

"From the language of Order No. 108922, there appears to be essentially three reasons for adjusting rate structures to shift substantial portions of the rate increase to industrial and high volume customers: (1) a shift from traditionally accepted criteria for valuing utility service known as the ``cost of service' rationale to an ``intrinsic value' of service rationale, (2) a desire to protect residential rate payers from absorbing more than a fair proportional share of the rate increase, and (3) a desire to channel greater quantities of an admittedly dwindling resource to ``human needs' uses such as home heating, hospitals, nursing homes, etc., thereby discouraging inefficient and wasteful over consumption of natural gas reserves where alternative fuels can be substituted." (p. 378.)

Once again the judicial pronouncement was:

"Commission is not bound by a single formula or a combination of formulas in fixing rates and none is exclusive or more favored than the others. Lone Star Gas Company v. Corporation Commission, 170 Okl. 292, 39 P.2d 547 (1935). There is no precise statutory or court announced basis for determining the justness or reasonableness of class rate level structures or relationships, the Court generally holding that rate making is the responsibility of a regulatory commission effectively exercising its discretion upon sufficient evidence before it. Arkansas Louisiana Gas Co. v. Sun Oil Co. of Pennsylvania, supra." (p. 379.)

V.

From the foregoing discussion it may be seen that as a matter of general utility rate-fixing law the order of the commission was one within its competence to make. Its order accepted the testimony of one expert witness and rejected the conclusions of another, and was thus based on substantial evidence. It made a decision to flatten the rate differential between large and small consumers based on accepted policy considerations engendered by the changing times. Under well established principles of judicial review we must affirm unless the order is contrary to Kansas law. Two Kansas cases are forcefully urged upon us by Midwest as indicating that Kansas somehow takes a different approach to rate design than that of the rest of the country.

The first of these is Central Kansas Power Co. v. State Corporation Commission, 221 Kan. 505, 561 P.2d 779 (1977). There the commission, in passing on a requested rate increase, found it necessary to determine whether one generating plant should be segregated from the rest of the system, or whether it should be considered as an integral part of the system for the purpose of determining the company's rate base and operating expenses. In essence the court held that the question was one of fair debate, in which the courts must yield to the discretion and expertise of the commission.

The court did say that "[i]n public utility rate cases a proper method of allocation should recognize the value of property devoted to a particular service and should result in a rate base which will bring a just and reasonable return on the property used in furnishing that particular service." (Syl. ¶ 4.) Midwest suggests that this language requires a cost analysis of each service rendered, and a rate based on the costs so determined. Read in context, it will be seen that the court was looking to an allocation of rate base and expenses between Kansas jurisdictional customers and federal jurisdictional customers, and those were the "services" under consideration. Costs are of course necessary elements of a rate base and a determination of overall revenues required; that was what was being established as to Kansas jurisdictional customers. As the Wisconsin court said in West Allis, 42 Wis.2d at 575:

"A petitioner seeking a rate change, such as the water utility here, does have the duty to show that its total return on its investment is inadequate. It is its responsibility to prove its cost of services as a whole and to show the Commission what total revenue or rate will give it a reasonable return.... No such duty lies in connection with ``pricing' the product to a particular class of customers or to customers within a class. The function of absolute obeisance to the cost-of-service principle ends when the rate level of the utility as an entity is determined."

As we see it, Central Kansas has no applicability to a rate design case such as we have here, where the objection is not to the total rate allowed but only to the pricing of the commodity to the various classes of customers. It does reflect judicial deference to the wide discretion vested in the KCC.

The second case, of more concern, is Jones v. Kansas Gas and Electric Co., 222 Kan. 390, 565 P.2d 597 (1977). That was a consumer challenge to utilities' "late payment" charge. The argument made and accepted by the court was that it is unfair to charge a uniform penalty for all customers who are delinquent in paying their bills, regardless of whether the company is required to expend time and effort in collection. Second notices and in-person collection efforts obviously cost the company money; these expenses should not be shared by those who pay their bills voluntarily although late, and thus do not create those expenses. In so holding the court said:

"The touchstone of public utility law is the rule that one class of consumers shall not be burdened with costs created by another class." (Syl. ¶ 10.)

Midwest argues that under that principle a study allocating costs to each class of consumers is essential. Without a cost of service study, the argument goes, there is nothing to which the touchstone may be applied.

We think Midwest reads too much into an unexceptionable but abstract statement of law. In Jones the court dealt with two commodities, credit and collection expense, which were quite distinct from the utilities' basic service, which was the sale of electricity. As to those collateral commodities costs were readily available from the companies' books, and were relatively easy to allocate to the classes of customers who created them. That is quite a different situation from attempts to allocate system-wide costs of operating an entire utility. As illustrated by the cases discussed above, the latter task is universally regarded as impossible to perform with any degree of rehability. We cannot believe the law requires, or that the KCC must require, studies which are expensive to make and unreliable when made.

For rate design purposes cost studies may be required by the KCC, or may be offered by a party as was done here. See Secretary of the Army v. State Corporation Commission, 206 Kan. 139, 476 P.2d 629 (1970). However, the weight to be given the resulting data when offered is peculiarly within the domain of the KCC. If the KCC is convinced or the evidence indisputably demonstrates that a rate structure in fact imposes on one class costs created by another, the rate structure cannot withstand the test of Jones. We do not, however, read Jones as requiring a cost of service analysis in every rate design case. A rate design fair on its face, with substantial evidence to support it, may be approved without a cost of service study absent a convincing showing of a Jones violation.

VI.

Although what has been said covers the broad issues raised, we shall deal briefly with some of Midwest's specific points.

A. The claim that interruptible are required to bear costs created by firm customers is not clearly demonstrated by the record. The KCC simply did not give full credence to Mr. Whiteaker's cost study, as was its prerogative. Mr. Chaney testified that increasing the lower priced schedules as much or more than the higher priced schedules could be justified on a cost basis. He explained that "return" was part of costs, and that a higher return should be allowed for the higher risk of serving interruptible who could easily convert to alternative fuels. Midwest asks us to read the testimony of the two opposing experts and determine for ourselves which is more persuasive. We are not permitted to do so; Midwest's arguments based on the weight of the evidence are made in the wrong forum.

B. The fact, if it be so, that the Company will derive a higher rate of return from its sales to interruptible than its sales to firm customers does not make the rate schedule unjust or unreasonable. See Apartment Hse. Coun. of Met. W. v. Public Serv. Com'n, 332 A.2d at 57; Arkansas La. Gas Co. v. Sun Oil Co. of Pa., 554 P.2d 14 (Okla. 1976); Application of Arkansas Louisiana Gas Company, 558 P.2d 376 (Okla. 1976); cf. Nor. Pac. Ry. v. North Dakota, 236 U.S. 585, 598-9, 59 L. Ed. 735, 35 S. Ct. 429 (1915); Secretary of the Army v. State Corporation Commission, 206 Kan. 139, 144-5, 476 P.2d 629 (1970); Capital Transit Co. v. Bosley, 191 Md. 502, 513, 62 A.2d 267 (1948). Midwest concedes that the rate of return may vary, but says that the discrepancy here is far greater than in other cases where variances have been approved. Again, this contention is based on its own cost study which the KCC chose not to accept. Even if Midwest's figures were accepted, the extent of variance is a question for the KCC, not the courts.

C. One of the justifications for the uniform increase in rate per Mcf is the increasing cost of new gas. The KCC reasons that large use now by interruptibles brings ever closer the day when they will convert entirely to other fuels, leaving the firm customer to bear the full brunt of the anticipated higher gas prices. The adopted rate structure partially alleviates the present need to increase the firm rates, and places part of the anticipated future increase on the interruptibles. Midwest argues they thus will pay twice for future increases, once through the present schedule and again through purchased gas adjustments.

Insofar as paying "twice" for gas is concerned, we do not see how the interruptibles are in any different position than the firm customers. Each class will pay a current rate increased by the same amount per Mcf; each will pay a purchased gas adjustment as the Company's cost of gas increases. Midwest's real contention is that the firm customers should bear a greater proportion of the present increase. That contention is answered by the KCC's decision to weigh the value of service more heavily than cost.

D. Midwest objects to the KCC's consideration of evidence indicating that the cost of gas is rising, and that the supply of natural gas is finite. It argues that such evidence is not "competent" evidence upon which to rely in a rate proceeding.

It is difficult for this court to see why the economic and physical facts of life are not competent or why they should not be considered. Of course, how they are to be considered and the conclusions to be drawn from those facts raise different questions. Arguments on those issues, again, are to be addressed to the KCC and not to the courts. The proper resolution of those issues being in the realm of fair debate, we cannot substitute our judgment for that of the commission.

E. Two technical objections are made to the KCC's order: that its findings are not specific enough and that the amendment made could only come after a full rehearing. The purpose of requiring findings of fact by the KCC, as in the case of any administrative agency, is to "convey to the parties, as well as to the courts, an adequate statement of the facts which persuaded the Commission to arrive at its decision." Kansas Public Service Co. v. State Corporation Commission, 199 Kan. 736, 745, 433 P.2d 572 (1967). We have encountered no difficulty in ascertaining the Commission's reasoning from its order, and judging from its exhaustive briefs, Midwest likewise has been able to glean the commission's meaning. If we were to remand for new findings we would be hard put to specify what we wanted, unless it might be the kind of cost data demanded by Midwest. What has been said above disposes of any such requirement. Although the findings might have been more detailed, we conclude they were adequate.

The clarification added in the order denying a rehearing was just that; it did not change the substance of the original order but only its language. A rehearing was not required for that purpose any more than a new trial is required before a court may enter an order nunc pro tunc. Cf. Autry v. Walls I.G.A. Foodliner, Inc., 209 Kan. 424, 497 P.2d 303 (1972), modified by opinion denying rehearing, 209 Kan. 747, 497 P.2d 303 (1972).

VII.

We conclude, along with the other courts which have considered the question, that cost of service studies are not legally required for rate design purposes. We note a growing feeling among academicians and regulators that such studies are desirable, despite their acknowledged shortcomings, but desirability is not an issue upon which judicial action in this area may be based. The cost of service testimony presented by Midwest was discounted by the commission. We cannot say that it compels a finding that the rate schedules approved by the KCC are unlawful or unreasonable. Accordingly, both the KCC order in Docket No. 113,728-U and the judgment of the district court upholding the order in Docket No. 110,038-U are affirmed.