269 S.E.2d 595 (1980)
300 N.C. 474
STATE of North Carolina ex rel. COMMISSIONER OF INSURANCE
v.
NORTH CAROLINA RATE BUREAU, Central Mutual Insurance Company, Great American Insurance Company, Iowa National Mutual Insurance Company, State Capital Insurance Company, and United States Fire Insurance Company.
In the Matter of a FILING DATED JUNE 30, 1978, BY the NORTH CAROLINA RATE BUREAU FOR A PREMIUM LEVEL REVISION ON the HOMEOWNER'S PROGRAM, DOCKET NO. 281.
No. 54.
Supreme Court of North Carolina.
July 15, 1980.
Atty. Gen., Rufus L. Edmisten by Asst. Atty. Gen., Isham B. Hudson, Jr., for plaintiff-appellant.
Young, Moore, Henderson & Alvis by Charles H. Young and William M. Trott, Raleigh, for defendants-appellants.
CARLTON, Justice.
I.
On 30 June 1978 the North Carolina Rate Bureau, on its own behalf and on behalf of its member companies writing homeowners' insurance in North Carolina, filed with the Commissioner of Insurance a proposed revised premium rate schedule for homeowners' insurance. The filing stated that statistical information substantiated the need for an average increase of " + 9.1%" in premiums. Included in the filing were proposed changes in the amount of insurance, form and protection/construction relativities, as well as rate changes by territory based upon a review of experience by territory. Changes were also proposed in the optional coverages.
The Commissioner gave notice of public hearing, contending that the filing failed to comply with statutory and other requirements and was otherwise incomplete in a number of respects. After the hearing, the Commissioner made extensive findings of fact and conclusions of law and disapproved the filing in its entirety. From the Commissioner's disapproval order, the Rate Bureau appealed to the North Carolina Court of Appeals. That Court, in a brief opinion by Judge Vaughn with Judge Hill concurring and Judge Erwin dissenting, vacated and set aside the Commissioner's order. Judge Vaughn wrote:
The dissent in this case makes it reasonably certain that the final disposition of the appeal will be determined by the Supreme Court. We will not, therefore, attempt to recapitulate the evidence or set out a detailed statement of the reasoning that leads us to the conclusion that the order is so affected by errors of law that it must be vacated.
44 N.C.App. at 76, 259 S.E.2d at 927.
In light of this rather cursory treatment of a complicated and important case to the citizens of North Carolina, we deem it necessary to review all assignments of error and arguments presented to the Court of Appeals. However, in summary, we hold that because of the magnitude of error in the Commissioner's order, we affirm the decision of the Court of Appeals. We also declare the Commissioner's order void and further order that the escrowed premium funds representing this proposed increase be remitted to the member insurers pursuant to G.S. 58-124.22(b).
Other facts important to an understanding of our decision are noted below.
II.
The Court of Appeals held that the Commissioner erred as a matter of law in concluding that unaudited data submitted in a filing of this nature is not reliable. We affirm. This portion of our decision is controlled by Section II. of our decision in Case No. 85, State ex rel. Commissioner of Insurance v. North Carolina Rate Bureau, ___ N.C. ___, 269 S.E.2d 547, filed today.
III.
The Court of Appeals held that the Commissioner erred as a matter of law in concluding that underwriting profit be reduced by an amount for theoretical investment income on unearned premium reserves and loss reserves. We affirm. This portion of our holding is controlled by Section V. of our opinion in Case No. 85 filed today.
IV.
The Commissioner's order concluded that investment income was not properly taken into account in this ratemaking. We reverse. This portion of our decision is controlled by Section IV. of our opinion in Case No. 85 filed today.
V.
The Commissioner concluded that:
[T]he determination of underwriting profit margins should be calculated in accord with contemporary concepts of risk and return as understood in financial theory, specifically the capital asset pricing model as testified to by expert witness Dr. William Bishop Fairley . . . the use of which theory and methodology in insurance ratemaking has been upheld by the Supreme Judicial Court of Massachusetts.
We reverse. This portion of our decision is controlled by Section V. of our opinion in Case No. 85 filed today.
VI.
Appellees raise again on this appeal the question whether the burden of proof in a ratemaking hearing has been shifted to the Commissioner by virtue of changes made by the 1977 Legislature. We reaffirm our holding in Section VI. of our opinion in Case No. 85 filed today. The burden of proof, as that term is ordinarily understood in civil litigation, rests with the Rate Bureau in a ratemaking hearing of this nature.
VII.
In his order, the Commissioner concluded that "official notice was taken of the methodology of Dr. William Fairley as adopted by the Commissioner in his order of February 27, 1978 . . . pursuant to North Carolina General Statute 150A-30." The earlier order was then incorporated by reference into the order in the instant case.
Dr. William Fairley was not present at the hearings in the instant case. He had testified at the 1977 automobile rate filing hearing, the subject of our opinion in Case No. 85 filed today. During the course of the earlier hearing, this witness had presented a paper setting forth his theory with respect to automobile insurance ratemaking. That paper was attached to and made a part of the 27 February 1978 order disapproving the 1977 filing. Appellees contended to the Court of Appeals and again on this appeal, that this manner of allowing Dr. Fairley's testimony resulted in the Rate Bureau having no opportunity to cross-examine the witness. It correctly cites the general rule that:
Ordinarily, testimony given by a witness in a preliminary hearing, or former trial, will not be admitted as substantive evidence in a trial unless it is impossible to produce the witness. The witness himself, if available, must be produced and testify de novo.
State v. Cope, 240 N.C. 244, 248-49, 81 S.E.2d 773, 777 (1954). See also, Smith v. Moore, 149 N.C. 185, 62 S.E. 892 (1908).
While we agree with the rules cited by appellees for trials and believe it the better practice to produce a witness in an administrative hearing of this nature, in lieu of relying on past exhibits furnished in earlier hearings, we do not find prejudicial error in the Commissioner's action in this respect in this hearing. G.S. 150A-30 provides that:
Official notice may be taken of all facts of which judicial notice may be taken and of other facts within the specialized knowledge of the agency. The noticed fact and its source shall be stated and made known to affected parties at the earliest practicable time, and any party shall on timely request be afforded an opportunity to dispute the noticed fact through submission of evidence and argument.
The record reveals that the Commissioner was sensitive to this provision of our Administrative Procedure Act. In his Notice of Public Hearing dated 28 July 1978, the Commissioner stated in part:
You are hereby notified that pursuant to North Carolina General Statute 150A-30 the Commissioner takes official notice of the methodology of Dr. William Fairley as adopted by the Commissioner in his order of February 27, 1978 to the North Carolina Rate Bureau. You are hereby directed to furnish revised rate calculations in conformity with the methodology of Dr. Fairley no later than August 23, 1978.
Clearly, the Rate Bureau had adequate notice that Dr. Fairley's paper would be relied upon by the Commissioner in the hearing, and the Rate Bureau certainly had sufficient time to make a "timely request [for] an opportunity to dispute the noticed fact through submission of evidence and argument." Indeed, appellees could have attempted to have Dr. Fairley present had they desired to cross-examine him. We therefore find no error in this respect with the Commissioner's actions.
VIII.
In his findings of fact, the Commissioner found that in several respects the filing was not based upon the data of all member companies of the Rate Bureau. In his conclusions of law, the Commissioner concluded that rates based on the experience of less than all companies writing homeowners' insurance in North Carolina could be excessive or inadequate and further concluded that basing relativities and other factors on the experience of less than all companies writing homeowners' insurance is improper.
In this connection, it is important to note that the filing which is the subject of this appeal is composed of two separate and distinct parts: (1) A rate level adjustment, and (2) an adjustment in the "relativities" used in homeowners' insurance.
With respect to the statewide rate level adjustment, we note that the record reveals that the Rate Bureau's witness Murphy testified that all companies' data was utilized. Moreover, the evidence established that the data was collected in accordance with generally accepted methods and procedures for the collection of such data and that the filing, methods and calculations contained in the filing were actuarially sound. Hence, the Commissioner erred in finding and concluding that less than all company data was employed with respect to the filing for a rate level increase.
We next turn to the contention that it is improper to base relativities on the experience of less than 100% of all companies writing homeowners' insurance in the State. The number of variables involved in the writing of a homeowners' insurance policy precludes the possibility of establishing a rate which would be applicable to a single policy having every combination of the variables involved. Therefore, to simplify pricing, the homeowners' premium rate structure utilizes what is commonly referred to as "relativities." For example, the amount of insurance being written on a particular home is one of the variables involved. A specific type of homeowners' insurance is not written at a fixed dollar premium rate per $1,000.00 of coverage, but each amount of insurance is written at a premium rate that, based upon statistical experience, is appropriate for that amount of insurance. Hence, the premium, for a $20,000.00 policy on a given risk would not necessarily be twice the premium for a $10,000.00 policy on the identical risk. The amount would depend upon the statistical experience involved. The record reveals that, in establishing the "policy amount relativities," a certain sum of insurance was used as the base or standard and was assigned a "relativity" of 1.00. Factors were then developed to determine the comparable premium for other amounts of insurance. Using $30,000.00 as the base amount of insurance, some of the "policy amount relativities" in effect prior to the filing were as follows:
Policy Amount Relativity
$20,000.00 .65
$25,000.00 .82
$30,000.00 1.00
$35,000.00 1.21
$40,000.00 1.42
In the same manner, relativities are established for the particular form of insurance being written, the type of construction, the type of protection available to the home, and the type of occupancy.
The Commissioner concluded that it would be improper to base these relativities and other factors on the experience of less than 100% of all companies writing homeowners' insurance in North Carolina. Appellees concede that, in establishing the relativities, less than 100% of company data was utilized. In some instances, countrywide data was employed in lieu of using North Carolina data only.
We think it unnecessary to our decision to include a detailed discussion of the complicated factors involved in establishing "relativities." Suffice it to say that expert witnesses on behalf of the appellees testified that the procedures utilized in establishing the relativities were conventional and actuarially sound methods of determining homeowners' rates and that the entire filing and rates proposed were actuarially sound and fully justified. We find no evidence in the record to the contrary. Hence, we find the Commissioner's findings and conclusions that less than 100% data utilization is improper in establishing relativities to be unsupported by material and substantial evidence in view of the entire record as submitted. G.S. 58-9.6(b)(5) and G.S. 150A-51(5). Moreover, we note that there is no requirement in Chapter 58 of our General Statutes requiring that data from all companies be utilized in a filing. G.S. 58-124.20(c) requires the Bureau to maintain "reasonable records . . . of the experience of its members and of the data, statistics or information collected or used by it in connection with the rates . . . made or used by it." We do not believe that "reasonable records" require, absent evidence of possible error, that all company data be presented. Hence, we find that the Commissioner erred as a matter of law in concluding that all company data is required. G.S. 58-9.6(b)(4); G.S. 150A-51(4).
IX.
An exhibit submitted with the filing indicated that the yearly rate level loss ratios for the five years of experience data were weighted as follows: The earliest year of experience, 1972, was weighted by a factor of .10; 1973, by .15; 1974, by .20; 1975, by .25; and 1976, by .30. As a result of multiplying these loss ratios by their weights, a composite rate level loss ratio was developed. In one of his findings of fact, the Commissioner found that the weighting of the rate level loss ratios by years was arbitrary and that the Rate Bureau's witness was unable to explain the derivation of the weights. The Commissioner argues that "the losses used in the filing have already been adjusted and trended for inflation." He contends that the weighting procedure, whereby more weight is attached to experience for 1975 and 1976, means that more weight is being given to the frequency and severity of losses in those years. Also, he contends there is no explanation in the record why experience for the years 1975 and 1976 better reflect the severity and frequency of losses for policies to be issued in 1979 than do the years of experience for 1972, 1973 and 1974.
We first note that, except for his general conclusion of law that the evidence presented in support of the filing was not credible, we find no specific conclusion of law supported by the finding of fact mentioned. Moreover, the finding must simply fall. The record clearly shows that the reason for the weighting procedure was to give more recognition or emphasis to recent years of experience, and that the weights were not arbitrarily selected but were standard weights used by experts on a nationwide basis. The record also established that the methods and procedures utilized in this respect were in common usage throughout the country and were appropriate and reliable for ratemaking and that such methods and calculations were actuarially sound. There is no evidence to the contrary to support the Commissioner's finding of fact in this respect.
Finally, we note that the Commissioner's Notice of Public Hearing made no mention of this alleged deficiency. G.S. 58-124.21(a) provides in part that "the Commissioner may give written notice to the Bureau specifying in what respect and to what extent he contends such filing fails to comply with the requirements of this Article . . . ." Here, the Commissioner gave no notice of his intention to challenge the weighting process utilized in this filing which was set forth clearly and prominently in the filing. Such omission clearly violates the quoted portion of G.S. 58-124.21(a).
X.
In his findings of fact the Commissioner found that the Rate Bureau failed to provide complete data regarding number of paid claims, number of claims with cash reserves, average paid claim and average reserved claim, and a showing as to how weights attributed to the Boeckh Residential Index (BRI) and the Modified Consumer Price Index and a county-by-county breakdown of premium and loss experience with unadjusted or trended loss ratios and also found that the Rate Bureau did not maintain reasonable records of the experience of its members and that county-by-county experience is necessary to properly review the territorial groupings and rates. From these findings, the Commissioner concluded that the failure to furnish the aforementioned data was a dilatory action and constituted bad faith on the part of the Bureau, that the Bureau did not carry its burden of proving that the territorial rating was not unfairly discriminatory by its failure to provide county-by-county experience, and that the Rate Bureau did not maintain reasonable records in compliance with North Carolina General Statute 58-124.20(c). Suffice it to say that the Commissioner's findings in these respects are unsupported by material and substantial evidence in view of the entire record and that the findings do not support his conclusions of law. This portion of our holding is also controlled by Section IX. of our opinion in Case No. 85 filed today.
XI.
Other issues were raised by the Rate Bureau in its appeal to the Court of Appeals but these issues were not specifically discussed by the Court of Appeals and have not been raised in the Commissioner-appellant's brief to this Court. All such issues are deemed abandoned. Rule 28(a), North Carolina Rules of Appellate Procedure.
XII.
In accordance with our discussion above, the decision of the Court of Appeals is
Modified and affirmed.
As in Cases No. 85, 300 N.C. 547, 269 S.E.2d 547, and No. 86, 300 N.C. 460, 269 S.E.2d 538 filed today, it is apparent that the errors committed by the Commissioner in the order before us are of such magnitude as to make remand for further proceedings futile. The order of the Commissioner dated 21 September 1978 is therefore
Reversed and declared null and void.
Accordingly, all escrowed premium funds representing this proposed rate increase pursuant to G.S. 58-124.22(b) shall be remitted to the member insurers forthwith.
It is so ordered.
BROCK, J., took no part in the consideration or decision of this case.