Railroad Commission of Texas v. Lone Star Gas Co.

SHANNON, Justice. The Railroad Commission of Texas has appealed from the judgment of the district court of Travis County setting aside the order of the Commission in Gas Utilities docket number 1669 that had set the residential and commercial rates for natural gas within the Grand Prairie Distribution System. Appellants are the Railroad Commission and the City of Grand Prairie, and appellee is Lone Star Gas Company. Appellants attack the judgment by eight points of error. The heart of the appeal, however, is stated in point of error two: the district court erred in concluding that the Commission’s use of the discounted cash flow formula was not supported by substantial evidence and constituted a decision based upon unlawful procedure. A utility company’s return on its investment is, of course, the product of the rate base multiplied by a fair rate of return. Railroad Comm’n of Texas v. Lone Star Gas Co., 599 S.W.2d 659 (Tex.Civ.App.1980, writ ref’d n. r. e.). In point of error two the computation of the rate base is not placed in issue; instead, the fair rate of return is questioned. The rate of return is the sum of money that a utility is allowed an opportunity to earn, over and above operating expenses, depreciation, and taxes. The rate of return is expressed as a percentage of the utility’s rate base. Butler, The Rate of Return in Texas —The Neglected Issue, 28 Baylor L.Rev. 937, 938 (1976). To achieve the rate of return that a utility should be allowed to earn, the regulatory agency considers the cost to the utility of its capital expressed as (1) interest on long-term debt; (2) dividends on preferred stock; and (3) earnings on common stock. Each of the elements of the capital structure of the utility is given a weighting based upon the company’s capital structure to arrive at a composite rate of return. The composite rate of return is then applied to the utility’s rate base to arrive at the proper amount of return. Garfield & Lovejoy, Public Utility Economics 116 (1964). In calculating the rate of return, the costs of long-term debt capital and preferred stock capital may be ascertained with relative ease. The cost of debt capital is the interest rate contractually agreed upon by the investor and the utility. Likewise, the cost of preferred stock is the dividend percentage agreed upon by the utility and the investor. Both such costs are predetermined contractually and, accordingly, do not fluctuate. The cost of common stock capital, however, is more controversial because with common stock, there is no contractual interest rate or fixed-dividend rate. The determination of cost of common stock capital is a matter of judgment subject to varying expert opinion. Railroad Comm’n of Texas v. Lone Star Gas Co., supra. Several methods are used currently by rate of return experts to estimate the cost of common stock capital. Two methods are: (1) comparable rates of earnings of other companies and (2) the past price earnings ratio of the utility’s common stock. During the hearing, appellee called Gloria L. Ramsey to testify as to the cost of common stock capital. Ramsey is a graduate economist, and at the time of hearing was employed by appellee as Director of Rate Research. Neither counsel for Grand Prairie nor the examiner challenged Ramsey’s qualifications as an expert. Ramsey estimated the cost of appellee’s common stock capital based upon comparable rates of earnings of other gas companies to be 18 percent. There was no other testimony as to appel-lee’s cost of common stock capital, since the *910 testimony tendered by Grand Prairie’s witness was not admitted by the examiner. A study of the Examiner’s Proposal for Decision shows that the examiner rejected the testimony of appellee’s expert witness relating to the cost of common stock capital based upon comparable rates of earnings of other gas companies. Instead, the examiner employed the “Gordon Model,” a discounted cash flow method to calculate the cost of common stock capital. The examiner’s proposal for decision states the “Gordon Model” in the following equation: (D/P)l.l + G where: D = Dividends per share; P = Current price per share; G = The growth factor; and 1.1 = A factor to account for flotation costs and market pressure. The examiner’s calculation of the cost of common stock capital is found in several pages of complicated computations in the appendix attached to the Proposal for Decision. The examiner’s calculations, using the “Gordon Model,” resulted in a cost of common stock capital of 13.65 percent. The record is silent as to exactly what the “Gordon Model” is or how it is to be properly employed, although the results of the formula are shown in the proposals for decision. This Court has concluded from argument and independent research that the “Gordon Model” is one variation of the discounted cash flow method sometimes used to calculate the cost of common stock capital. Under the discounted cash flow method, the average growth per year of a utility’s book value per common share over a past period of time is projected forward into the next few years. This projected dollar amount is applied against the current yield from the stock to determine the exact dollar return per share of common stock. The discounted cash flow theory looks to the cost of common stock capital for any given company as being the sum of the current yield of the stock plus the experience of its growth. The underlying basis for the discounted cash flow theory is that the investor considers both growth in book value of his stock as well as its yield. Butler, supra, 940. Under point of error two, the Commission admits that its use of the discounted cash flow method to ascertain the cost of common stock capital is not supported by any evidence. The Commission’s vaulting claim is, however, that its use of that particular methodology need not be supported by evidence. The Commission contends in this connection that it is not bound by “incorrect opinions” offered by interested expert witnesses. The Commission’s claimed right to rely upon methodology unsupported by record proof is contrary to basic notions of a fair hearing requiring that a party be apprised of the evidence contrary to his position so that he may refute, test, and explain that evidence. Richardson v. City of Pasadena, 513 S.W.2d 1 (Tex.1974); English v. City of Long Beach, 35 Cal.2d 155, 217 P.2d 22 (Cal.1950); La Prade v. Department of Water & Power, 27 Cal.2d 47, 162 P.2d 13 (Cal.1945); Elizabeth Fed. Sav. & Loan Ass’n v. Howell, 24 N.J. 488, 132 A.2d 779 (1957). Moreover, the Commission’s position is counter to the Administrative Procedure and Texas Register Act, Tex.Rev.Civ. Stat.Ann. art. 6252-13a §§ 13(d) and 14(p) requiring in a contested case that “... opportunity must be afforded all parties to respond and present evidence and argument on all issues involved,” and “. .. a party may conduct cross-examination required for a full and true disclosure of the facts.” To justify noncompliance with the requirements of a fair hearing and art. 6252-13a §§ 13(d) and 14(p), the Commission argues that the discounted cash formula is not itself evidence, but instead a formula applied to facts in evidence to produce a rate, and therefore was not the proper subject for inquiry or cross-examination. This argument is without merit. One of the most controversial parts of a general rate case is the establishment of the fair rate of return. The most difficult element in establishing the fair rate of return is the ascertainment of the overall cost of capital. The really critical problem in the determination of the overall cost of capital in cases *911 such as this one, is that of estimating the cost of the common stock capital component. Bonbright, Principles of Public Utility Rates 246 (1961). The determination of that component of the cost of capital is not an exact science so as to establish that cost as a matter of law, but instead is the proper subject of expert opinion. Railroad Comm’n of Texas v. Lone Star Gas Co., supra. The method to estimate the cost of common stock capital sponsored by the gas company’s expert witness was the so-called “comparable rates of earnings” method. The Commission rejected outright that method as being “inaccurate and questionable.” Nonetheless, the discounted cash flow formula, embraced ex parte by the Commission, is not without its detractors. See Butler, The Rate of Return in Texas — The Neglected Issue, supra. The point is that none of the methods used by the experts to estimate the cost of common stock capital possess any particular messianic properties, and, at best, the ascertainment of that element of the cost of capital, by whatever method employed, is no more than a measured estimate. See Bonbright, supra, 242. The Commission is not bound to accept the testimony of witnesses, expert or non-expert. City of Frisco v. Texas Water Rights Comm’n, 579 S.W.2d 66, 69 (Tex.Civ.App.1979, writ ref’d n. r. e.). In the context of this case, the Commission is free to adopt any recognized method it judges best suited to estimate cost of common stock capital, provided that whichever method adopted must find record support and be subject to traditional means of inquiry and examination and to those afforded by the Administrative Procedure and Texas Register Act to those opposing its adoption. In connection with its argument that its methodology need not be supported by proof, the Commission states that it possesses the discretion and responsibility to use its own expertise in reaching an ultimate decision that fulfills its obligation under the Administrative Procedure and Texas Register Act. In the process of decision, as distinguished from the process of proof, agency officials may employ their own expert knowledge and experience in evaluating the evidence contained in the record and in drawing conclusions from that evidence. Such use of agency expertise is not only permissible, but is certainly desirable. This, of course, is quite a different thing from the utilization of agency expertise as a substitute for evidence and as a basis for making factual findings as to matters not supported by record evidence. I. Cooper, State Administrative Law 419 (1965). Stated differently, the Commission’s expertise cannot be a substitute for proof. A valid exercise of agency expertise, like other agency action, must find ultimate support upon evidence taken at the hearing or upon facts officially noticed by the hearings officer in the record of such hearing. Texas Health Fac. Comm’n v. Nueces County Hospital Dist., 581 S.W.2d 768 (Tex.Civ.App.1979, writ ref’d n. r. e.). Likewise, the discretion conferred upon the Commission by the Administrative Procedure and Texas Register Act to set rates is not unbridled; its findings may not be arbitrary or capricious, but instead must find support in substantial evidence. Gerst v. Cain, 388 S.W.2d 168 (Tex.1965). The judgment is affirmed. POWERS, J., not sitting.