358 So.2d 418 (1978)
MISSISSIPPI PUBLIC SERVICE COMMISSION, Appellant-Cross Appellee,
v.
MISSISSIPPI VALLEY GAS COMPANY, Appellee-Cross Appellant,
Greenwood Utilities Commission, City of Clarksdale, and Yazoo City Public Service Commission, Appellees.
No. 50254.
Supreme Court of Mississippi.
May 10, 1978.
Bennett E. Smith, Asst. Atty. Gen., Jackson, Hall, Callender & Dantin, Maurice Dantin, Columbia, for Miss. Public Service Comm.
Overstreet & Kuykendall, John M. Kuykendall, Jr., Carson M. Hughes, Jackson, for Miss. Valley Gas Co.
Lott, Sanders & Gwin, Hardy Lott, Greenwood, Sullivan, Smith, Hunt & Vickery, David R. Hunt, Clarksdale, Campbell & Campbell, T.H. Campbell, Jr., Yazoo City, William Thomas Miller, Washington, D.C., for Greenwood Utilities, City of Clarksdale and Yazoo City Public Service Comm.
Before ROBERTSON, LEE and BOWLING, JJ.
ROBERTSON, Presiding Justice, for the Court:
Mississippi Public Service Commission has appealed from that portion of a final decree entered by the Chancery Court of the First Judicial District of Hinds County, Mississippi, on April 12, 1977:
1. Reversing the May 6, 1976, Order of the Commission which fixed the rate of return for Mississippi Valley Gas Company at 8.83%; and
2. Remanding the case to the Commission for further hearing after the Commission has furnished certain information to the Company. Specifically the information required was the data fed into, the questions asked, and the tasks assigned to, a giant computer at the University of Chicago, which computer produced the CRISPE tapes, the tapes used by Dr. James L. Bicksler, the Commission's only witness, as the sole basis for his expert opinion as to what constituted a fair and reasonable rate of return to the Company.
The Company has cross-appealed from that portion of the final decree affirming the Commission's order, which:
1. Required the Company to file a revised rate schedule "for municipal generating facilities Rate Schedule # 110 in order that it conform with the lowest block (of charges) of Schedule # 108";
2. Eliminated the $300 monthly facility charge made against the Greenwood Utilities Commission, the City of Clarksdale, Mississippi, and the Yazoo City Public Service Commission; and
3. Provided that "Rate Schedule # 111 shall hereafter be the same as Rate Schedule # 110, ...".
This appeal presents three major issues for determination by this Court:
(1) The discovery issue;
(2) The rate of return issue; and
(3) The municipal electric generating rate issue.
On November 25, 1975, Mississippi Valley Gas Company filed with the Commission its schedule of proposed gas rate changes, a statement of the necessity for the proposed changes, and detailed supporting data, as required by Mississippi Code Annotated section 77-3-37 (1972). As provided in Section 77-3-39, the Company posted a refunding bond and placed the new rates into effect on December 26, 1975. Later the Company filed a petition to change to systemwide rates and a systemwide purchase gas adjustment provision. The Company also proposed a $300 per month facility charge, and increased gas rates for municipal electric generating plants operated by the Cities of Greenwood, Clarksdale and Yazoo City.
Greenwood, Clarksdale and Yazoo City intervened and were successful before the Commission, and that portion of the Commission's Order affecting these three municipalities was affirmed by the Chancery Court.
This Court is of the opinion, as was the chancery court that the Order of the Commission (a) requiring that the rate level of Rate Schedule 110 (Revised) and Rate Schedule 111 "conform with the lowest block of Rate Schedule # 108"; (b) eliminating the monthly facilities charge of $300.00 from Rate Schedule 110 (Revised); and (c) requiring the Company to keep the Commission apprised of gas available for Category 6 customers and when such gas is available, to furnish it "immediately to municipal generating facilities under rates as hereinabove set forth," is supported by substantial evidence and should be affirmed.
On the rate issue, the Company's witnesses were its president and chief executive officer, Taylor G. Holland, Jr.; James W. Welch, Jr., senior vice-president and treasurer; E.R. Butler, senior vice-president of operations; Richard S. Johnson, vice-president of Stone & Webster Management Consultants, Inc., of New York; W.C. Randolph, the Company's director of rates and pipeline supply; Wayne D. Monteau, manager of Financial Analytical Services for Stone & Webster; William A. Whitney, senior vice-president and trust officer for Deposit Guaranty National Bank; and John H. Geary, vice-president of Paine-Webber.
The Company's officers testified at length and supplied detailed charts and tables as to the assets, facilities, income and increased expenses of the Company. They testified in detail as to the necessity for increased rates in these times of inflation.
The Commission offered one witness, Dr. James L. Bicksler, Professor of Finance and Director of Research at Rutgers University Graduate School of Business.
Professor Bicksler was employed by the Commission:
"(a) To develop and present written testimony on all essential evidence on rate of return.
"(b) To assist Commission staff in the cross-examination of Mississippi Valley Gas Company's witnesses.
"(c) To assist the staff in the preparation of the Commission's final order and to be available for consultation with the Commission staff as requested."
Bicksler's testimony and his opinion as an expert on public utility matters was based entirely on the CRISPE tapes which were produced by a giant computer at the Center for Research and Security Prices at the University of Chicago. It was admitted that no data concerning Mississippi Valley Gas Company was submitted to the computer. Professor Bicksler, in discussing the Capital Asset Pricing Model (CAPM), which was used in the CRISPE tapes, testified:
"Q Would you briefly describe the underlying economic rationale of the capital asset pricing model (CAPM)?
A The basic assumptions underlying the two-parameter capital asset pricing model (CAPM) are:
1. All investors are risk-average and choose portfolios in a manner consistent with maximizing expected utility of single-period terminal wealth.
2. Portfolio investment opportunities can be described solely in terms of means and variances (or standard deviations) of the distribution of expected one-period portfolio returns.
3. Investors have homogeneous expectations regarding means, variances, and covariances of returns for all securities in the investment opportunity set and, in addition, all investors have identical investment opportunity sets.
4. Capital markets are efficient in the sense that borrowing and lending rates are equal. There are no restrictions to short sales, no taxes, no transactions costs, and capital assets are perfectly divisible, etc.
5. The supply of all capital assets is given.
Under these assumptions and other less restrictive conditions, equilibrium relationships for risky capital have been derived. What is important from the viewpoint of public utility regulation is not the assumptions, which are somewhat technical in nature, but the fact that testable hypotheses on risk-return equilibrium can be set forth."
The word "Beta" is used throughout Professor Bicksler's testimony. He defined Beta in this way:
"Q What is beta?
A Beta is the systematic or undiversifiable risk of a security. It measures the security's covariation with market portfolio or index.
Q Would you give an example?
A Yes. If a security has a beta of 1, and the market portfolio had an expected return of 10 percent, then the expected return of the security would be 10 percent."
When asked the question: Has there been any criticism of the capital asset pricing model (CAPM)? Dr. Bicksler replied:
"A There has been criticism of the capital asset pricing model (CAPM). However, almost all of the criticism has been a result of an attempt to mechanically interpret or apply the results. Further, as in all science, the derived inferences are probabilistic in nature with the ever-present possibility that the presentation of new evidence will result in a modification in the nature of a `confirmed' hypothesis.
Q What are the testable empirical implications about risk-return relationships of the capital asset pricing model (CAPM)?
A The major implications for expected returns of the capital asset pricing model (CAPM) are:
1. Expected returns vary linearly with beta risk.
2. Expected returns are not systematically affected by non-beta measures of risk. In other words, beta is a complete measure of the risk of a security.
3. There is a positive expected return-risk-trade-off (e.g., there is a positive expected risk premium).
Q What do empirical tests of the capital asset pricing model (CAPM) show?
A The empirical tests support the three major implications of the capital asset pricing model (CAPM) and its postulated risk-return relationships enumerated above.
Q Does the CAPM explain expected return-risk relationships perfectly?
A No. The CAPM has been shown to be surprisingly empirically robust. However, as always in scientific work, there are grounds for extending the model in a number of directions.
Q Put into perspective what the empirical usefulness of the CAPM is.
A The CAPM is empirically more robust then alternative equilibrium valuation scenarios. Further, the CAPM has been shown to have a high degree of explanatory power with regard to prices and yields of equities of public utilities."
The professor was asked this further question about some of his approaches to "the fair rate of return" problem, and his answer was:
"Q Would you briefly summarize again why your Bayesian approach is relevant for making `good' estimates?
A The Bayesian approach is strictly consistent with the rules of inductive reasoning. Hence, the use of Bayesian framework in the decision problem of estimating the fair rate of return is a rational procedure for systematically and coherently revising one's initial or prior judgment with the subsequent arrival of further information."
Prior to the hearing, the Company, on March 30, 1976, pursuant to Mississippi Code Annotated section 11-1-51 (1972), filed a motion to direct and order Dr. Bicksler to furnish the written material, documents, letters and data fed to the computer, and the questions asked, and tasks assigned, to the computer, for the purpose of cross-examining Dr. Bicksler about the CRISPE tapes. This motion was renewed at the hearing but was denied by the Commission. The questions asked and the answers given by Dr. Bicksler were:
"Q And what did you do? I want to know what you did to get the computer to generate this data.
A I don't really understand the mechanics of the computer in the sense that, you know, once the cards go in and, you know, there's an order to the computer, I'm not really that familiar with the mechanics of the computer memory, etcetera, etcetera, so I can't answer that.
Q What information did you give to the owner or the holder of the computer, the person who performed this?
A Essentially it's what's here on Page 8. Essentially what we wanted to do was derive the risk premium using the methodology we've stated, using specific time periods enunciated, utilizing equation one and utilizing as a basic in-put source CRISPE tapes, so all of that was well defined just as I have here on Page 8.
Q And did you fill out a form and tell them what you wanted to have done? Did you writ a letter? How did you do this, how did you communicate to the individual that performed this function?
A We have worked in the past on various things and this was done over the telephone.
Q Tell us what you told this individual on the telephone.
A I don't have a tape recording of the session but I mean specifically here is the research methodology and, you know, this is the essence of the procedure used to derive the risk premium.
Q Do you have in your possession or is there a document that precisely states what it is that you asked the computer to do that you have?
A No, I don't.
Q Have you ever had such a document?
A I don't really understand what you mean by this document.
Q Notes of what you asked the computer.
A You can't go up to the computer and say, look, I've got a list of notes, go do it, so that wouldn't serve.
Q Well tell us what you said on the telephone to this individual.
BY CHMN DALE:
Let's go on to something else, Mr. Hughes. I think you've been over that three or four times."
Bicksler further testified:
"Q Do you supervise or direct the construction of the CRISPE tape or the maintenance of those tapes?
A No. The CRISPE tape of the Center for Research and Security Prices is the most comprehensive and reliable data source of security returns in the world. It is an indispensable data file for any modern researcher in the economics of capital markets. The reliability of this data is well established and there's no need to check on it. I would like further "
After this illuminating testimony from Professor Bicksler, who was the Commission's only witness, the Commission was able to make this finding in its order of May 6, 1976:
"The Commission witness testified and the Commission finds that the frameworks employed by the company witness (Mr. Monteau) do not provide rigorous and systematic inferences about the fair rate of return on equity.
... The Commission expert (Bicksler) employed the capital asset pricing model which has several well-defined operational hypotheses. The procedure has been subjected to rigorous empirical tests in the light of modern day finance and provided for a specific measurement of risk. There was identification of the relationship between expected returns and risk. Likewise, the Commission expert gave an economic interpretation of the Hope case and found, as this Commission has found, that risk is center-stage of the regulatory rate of return process. The Commission finds that the capital asset pricing model used by the Commission expert is far more reliable than the exercise employed by the company expert and is well recognized in modern finance. The procedure employed by the Commission expert has been subjected to rigorous empirical tests and the witness was thoroughly versed in modern capital market theory and its application and development."
In its opinion, the chancery court said:
"It is obvious that Dr. Bicksler did not make any study of Mississippi Valley Gas Company itself. Instead, he used a new and quite different approach for the determination of reasonable rate of return, a Capital Assets Pricing Model, or CAPM. It is not the purpose of this Court to pass upon the relative merits of the Commission witness' new method as against the more traditional approach of the company's witnesses since the duty, and responsibility, to make such judgments is clearly upon the Mississippi Public Service Commission. But, here, the Commission's witness, Bicksler, was permitted to give his expert opinion as to what a reasonable rate of return should be, which said opinion was clearly based on facts which were not in evidence, which were available to the said witness, and upon the company's timely request should have been made available to the company for examination by the company and for its use in the cross examination of the Commission's witness. The Commission refused to recognize the inherent right on the part of the company to examine the raw data upon which the Commission witness based his expert opinion. Only Dr. Bicksler could know what he used as raw data upon which to base his opinion, what input there was from him to the computer, what the calls were which he made upon the computer for information, and all of the information requested by the company was within the easy order and production by Dr. Bicksler at no personal cost to himself and should have been required by the Commission ... In a situation where the rate expert hired by the Commission and relied upon by the Commission did not even make a study of the company in his determination of what a reasonable rate of return for the company would be makes it even more necessary that the closest sort of scrutiny be permitted of his manner and method of calculations and upon what data he based his opinion or used to arrive at his opinion. An extensive cross examination was permitted by the Commission of the said Commission witness running into many hours; nevertheless, where the Commission shielded its witness from having the raw data upon which he based his opinion examined by the company, there was a deprivation of due process and the company's constitutional rights in this regard were violated. It may very well be that the result which Dr. Bicksler reached is the right and proper one. In the present state of the record there is no way to tell. But insofar as the order of the Commission of May 6, 1976 pertained to the rate of return, and the schedule of rates thereunder, it must be reversed and remanded."
City of Laurel v. Upton, 253 Miss. 380, 175 So.2d 621, 625 (1965), and Butler v. State, 245 So.2d 605, 607 (Miss. 1971), are in accord.
In 1 Pt. 2 Moore's Federal Practice, Pt. 2.717 at 142 (1977), we find this reasoning:
"It is essential that the underlying data used in the analyses, programs and programming method and all relevant computer inputs and outputs be made available to the opposing party far in advance of trial. This procedure is required in the interest of fairness and should facilitate the introduction of admissible computer evidence. This procedure provides the adverse party and the court with an opportunity to test and examine the underlying data on which the machine analysis is based, the program and all outputs prior to trial. The pretrial ruling on objections can then be made by the court. Without agreement among the litigants, introduction of the computer outputs should be feasible if the party or parties who supervised the data processing testify to the validity of the methods used, the reliability of the computer, the accuracy of the inputs, the validity of the programming and the accuracy and completeness of the outputs."
As the chancellor stated in his opinion, Dr. Bicksler might have reached the right result, but in the present state of the record there is no way to tell.
Because of the lack of substantial evidence supporting that portion of the Commission's Order reversed by the chancery court, the decree of the lower court is affirmed on direct and cross appeals, and this cause is remanded to the Commission for discovery of the data requested and any additional evidence that the Commission or Company care to adduce. The Commission should then reevaluate the entire record before rendering its opinion.
AFFIRMED AND REMANDED TO THE COMMISSION.
PATTERSON, C.J., SMITH, P.J., and SUGG, WALKER, BROOM, LEE, BOWLING and COFER, JJ., concur.