dissenting.
I am unable at several points to agree with the majority opinion. I would reverse the trial court judgment and would reinstate the order of the Commission.
The validity of separate classification of heat pump users for gas tariff purposes. Scope of judicial review of Public Service Commission order.
I think the first question which must be decided here, and which the majority opinion leaves open, is whether the imposition of a surcharge upon a heat pump user is permitted at all. If that question is not decided, the parties are left in doubt, even if competent and substantial evidence were to establish (as the majority opinion says it does not) the proper changeover point, whether the establishment of a different rate structure for heat pump users is a permissible classification. The question is presented here. It should be decided.
This is a tremendously expensive case, involving hundreds of hours’ time and corresponding expense. Must the parties start over, present their case again to the Commission, and go through the appeal process again — only to find out on a second trip to this court that the classification was arbitrary and capricious and therefore impermissible?
So I would, first of all, decide that most basic point.
Second, it should be decided to approve the classification, for the following reasons:
The Commission decided that the separate classification for heat pump users was a permissible classification, against the contention of the heat pump side of this case that the surcharge constituted a discriminatory and unreasonable classification. I would hold that the classification is permissible, and that the order of the Commission could not be set aside as unlawful on the ground that the classification was arbitrary and unreasonable.
If we had been sitting as a Commission deciding this case originally, we might have decided that as a matter of regulatory policy, the heat pump surcharge should not be allowed. But we may not, sitting as a reviewing court, substitute our judgment for that of the Commission. We may not in our reviewing role so limit the Commission’s discretion. The business of utility regulation is a highly complex business. A reasoned decision on the question of the heat pump surcharge calls for a balancing of a wide range of economic and social factors, done by arbiters whose experience and learning equip them for that task. Ours is a more limited role; we review only for “unlawfulness” or “unreasonableness”. Sec. 386.510, RSMo 1978. It has become trite to say, and yet we need continually to remind ourselves, that we may not substitute our judgment for that of the Commission in matters which have been entrusted to it by law. State ex rel. Utility Consumers Council of Missouri, Inc. v. Public Service Commission, 585 S.W.2d 41, 47 (Mo. banc 1979); State ex rel. Utility Consumers Council of Missouri, Inc. v. Public Service Commission, 606 S.W.2d 222, 223 (Mo.App.1980), cert. denied, 450 U.S. 1042, 101 S.Ct. 1761, 68 L.Ed.2d 240 (1981).
On this record I am unable to say that there is anything unlawful or unreasonable about the classification which places heat *223pump users in a separate category for purposes of the surcharge. It is not an arbitrary or a capricious classification; it is based upon realistic differences between the gas consumer with heat pump and the gas consumer without heat pump.1 The former uses less gas and what he uses is more expensive gas at the wholesale level. The heat pump user begins to use gas in quantity only when the temperature goes below 34 degrees Fahrenheit, which is the point at which Laclede begins to draw upon its more expensive gas. The “peak demand factor” for heat pump users is 30 percent higher than for non-heat pump users. (The “peak demand factor” is arrived at by dividing amount used in January, the coldest month, by Vn of the amount used during the other 11 months.)
There was no dispute that the gas company’s revenue for gas sales to heat pump customers was insufficient. Only the amount of the shortfall was debated. The Commission found, upon ample evidence, that that shortfall was $90 per annum. The surcharge was fixed to compensate for that. If the utility’s cost of the gas is evenly distributed over the more expensive and the less expensive gas, and evenly distributed (per unit) over the entire customer base, then the customer who uses both the more expensive and the less expensive gas (i.e., the non-heat pump user) is subsidizing the customer who uses only the more expensive gas (i.e., the heat pump user) — or else the gas company is receiving less in rates than it is entitled to receive. The surcharge is a way of re-allocating the cost more equitably among residential gas users. The wisdom of the surcharge may be debatable — but that is only to say that it is not arbitrary, and therefore that it is not to be overturned by the reviewing court.
The majority opinion takes seriously the surcharge opponents’ contention that heat pumps are not unlike other auxiliary heat sources, such as stoves, burning wood, coal or kerosene, fireplaces or solar devices. The Commission, however, plausibly and correctly found that the heat pump principle, in its effect upon Laclede’s cost of supplying gas, was inherently different from that of such other auxiliary heating devices. Unlike the predictable and measurable impact of the heat pump, the impact of the other devices is entirely random. Their use depends upon many variables other than the outside temperature, including the user’s supply of fuel and individual preference or taste. They typically are used in only one room, not in the whole house. Of course, when these devices save gas they are as likely to save it at the coldest temperatures (when Laclede’s gas costs are at the highest) as at the warmer, adding nothing at all to the peaking factor which is the real mischief-maker from Lac-lede’s standpoint. There is nothing arbitrary or capricious about treating heat pumps differently than those devices.
Competence and substantiality of evidence to support Commission order.
Passing over the question whether Mr. Davis’s testimony alone would have been sufficient to support the Commission’s finding (as I believe it was and as the majority opinion holds it was not), the use of the 34-degree-Fahrenheit changeover point as falling within a reasonable range was supported by other evidence in the record. Witnesses for the surcharge opponents who were undoubtedly qualified as experts in technical operation of heat pumps gave testimony which would support the range of 44 degrees Fahrenheit on the high side and 27 degrees Fahrenheit on the low side. *224This testimony was given to support a higher changeover temperature than 34 degrees Fahrenheit, if there should be any surcharge at all. The reason for such testimony from the surcharge opponents was that the higher changeover point would result in a smaller revenue deficiency and therefore a smaller surcharge. Mr. Wagner, an electrical engineer and special projects manager for Union Electric, opted for a 40-degree-Fahrenheit changeover point which would result, he calculated, in a revenue deficiency of $77.78 per heat pump customer per annum as opposed to $95.70 at the 34-degree level. His studies included only those two temperatures. Electrical Engineer and Energy consultant Jurgen Worthing, testifying for surcharge opponent Marco Sales Co., computed customer heating costs and savings based upon only three different hypothetical changeover temperatures — 43 degrees, 32 degrees, and 27 degrees. A Union Electric advertisement in the record uses a changeover point of “30 degrees or so”.
This is an abundance of evidence to show that the 34-degree changeover point is a reasonable one, particularly when one takes it into account that the trier of fact is itself an expert in the field in which the witness is testifying. It may use its own expert knowledge, not to supply missing evidence, but to evaluate evidence. 73A C.J.S. Public Administrative Law and Procedure §§ 127(b), 130 (1983); Gaddy v. State Board of Registration for Healing Arts, 397 S.W.2d 347, 355 (Mo.App.1965).
The majority opinion seems to proceed on the erroneous assumption that various units have changeover points ranging from a high of 44 degrees Fahrenheit down to -2 degrees Fahrenheit. It points out that there is no evidence as to how many of each are in use. The idea seems to be that such evidence is necessary to enable the Commission to determine with a degree of mathematical nicety the mean or average changeover point.
Any such study would be wholly useless. While the various units may be capable of operating between the temperatures of 44 degrees Fahrenheit and -2 degrees Fahrenheit, the actual changeover points fall within a much narrower range. They are set by the owner at the desired temperature. The two factors which go into his selection of a changeover temperature are bodily comfort and economics. These considerations indicate a changeover point falling within a range of 43 degrees Fahrenheit to 27 degrees Fahrenheit. This is established by the expert witnesses of the opponents of the heat pump surcharge, earlier referred to.
It was not necessary for the purpose of tariff design to establish with mathematical exactitude the changeover temperature at which the heat pump customers with a higher changeover point and those with a lower changeover point would be in exact equilibrium. All in all, the selection of the 34-degree temperature as the changeover point for the purpose of fixing the amount of the surcharge was reasonable.
Remand instead of outright reversal
If we are to reverse this decision on the evidence point, it should not be reversed outright. It should be remanded to the Commission for rehearing on the limited issue of the correct changeover point. It should be reversed outright only if we decide the basic issue — the permissibility of the classification — adversely to the Commission’s decision.
Davis’s opinion that 34 degrees Fahrenheit was a proper changeover point — which the majority opinion says he was not qualified to give — was in the record, before the Commission, with no objection. The first objection comes on appeal. If we are to exclude this evidence, and hold that its absence leaves Laclede without a submissi-ble case, we should remand.
We have decided, however, that the case should be remanded for a new trial rather than to remand with directions to enter judgment for defendant. While we cannot point to any available evidence that might enable plaintiff to make a submissible case, we can understand that plaintiff’s counsel, having the positive testimony of a witness which, if accept*225ed, would have been sufficient to supply the essential proof, may not have presented all available evidence (particularly of an expert nature) which might aid plaintiff in making a stronger showing.
Prince v. Bennett, 322 S.W.2d 886, 891 (Mo.1959).
I would hold that the Commission’s decision is abundantly supported by the evidence. But if we reverse on the evidence point, I believe the case should be remanded to the Commission for rehearing on the limited issue of the correct changeover point.
. Compare State ex rel. City of St. Louis v. Public Service Commission, 327 Mo. 318, 36 S.W.2d 947 (1931), where Southwestern Bell converted its telephone system from a manually operated switchboard to a dial or "automatic" system. In doing so, it equipped with a dial at no charge (1) individual line subscriber telephones and (2) the switchboard of private branch exchange (PBX) subscribers. In contrast, Southwestern Bell charged 25 cents per telephone to equip with a dial any individual station telephones connected to the PBX subscriber’s switchboard. The Commission found and the court held not unjustly discriminatory (1) the differing rate treatment under the new system of individual line subscribers and PBX subscribers and (2) the 25 cents per station telephone charge to PBX subscribers under the new system when under the old system PBX users received the same service without the added charge.