State ex rel. Commissioner of Insurance v. Rate Bureau, 40 N.C. App. 85 (1979)

March 6, 1979 · North Carolina Court of Appeals · No. 7810INS238
40 N.C. App. 85

STATE OF NORTH CAROLINA, ex rel. COMMISSIONER OF INSURANCE, Appellee v. NORTH CAROLINA RATE BUREAU, AETNA CASUALTY AND SURETY COMPANY, AMERICAN MOTORISTS INSURANCE COMPANY, AMERICAN MUTUAL LIABILITY INSURANCE COMPANY, CONTINENTAL INSURANCE COMPANY, EMPLOYERS MUTUAL LIABILITY INSURANCE COMPANY, FIDELITY AND GUARANTY INSURANCE UNDERWRITERS, GREAT AMERICAN INSURANCE COMPANY, HARTFORD ACCIDENT AND INDEMNITY COMPANY, HOME INDEMNITY COMPANY, HOME INSURANCE COMPANY, IOWA NATIONAL MUTUAL INSURANCE COMPANY, LIBERTY MUTUAL FIRE INSURANCE COMPANY, LIBERTY MUTUAL INSURANCE COMPANY, LUMBERMENS MUTUAL CASUALTY COMPANY, MARYLAND CASUALTY COMPANY, NATIONWIDE MUTUAL INSURANCE COMPANY, PENNSYLVANIA NATIONAL MUTUAL CASUALTY INSURANCE COMPANY, ST. PAUL FIRE AND MARINE INSURANCE COMPANY, SHELBY MUTUAL INSURANCE COMPANY, STANDARD FIRE INSURANCE COMPANY, TRAVELERS INDEMNITY COMPANY, TRAVELERS INSURANCE COMPANY, TWIN CITY FIRE INSURANCE COMPANY, UNITED STATES FIDELITY AND GUARANTY INSURANCE COMPANY, and UNITED STATES FIRE INSURANCE COMPANY, Appellants.

No. 7810INS238

(Filed 6 March 1979)

1. Master and Servant § 80— workmen’s compensation rates — new statutes — elimination of delay

In enacting a new statutory procedure for workmen’s compensation rate-making in 1977, G.S. Ch. 58, Art. 12B, it was the legislative intent to eliminate unfair and unnecessary delay in the rate-making process.

2. Master and Servant § 80— workmen’s compensation rates —no disapproval by inaction

The Commissioner of Insurance can no longer effectively disapprove a workmen’s compensation rate filing by inaction or a bare assertion that the Rate Bureau has not carried its burden of proof.

3. Master and Servant § 80— workmen’s compensation rates — disapproval of filing-findings required of Commissioner of Insurance

While the new statutory scheme for workmen’s compensation rate-making does not shift the ultimate burden of proof from the Rate Bureau to the Commissioner, it does place on the Commissioner, in disapproving a filing, the burden of affirmatively and specifically showing how the Rate Bureau has not *86carried its burden of proof, and, if the Commissioner fails to do so by substantial evidence, the presumption of prima facie correctness given to an order of the Commissioner by G.S. 58-9.4 and G.S. 58-9.6 is rebutted. G.S. 58-124.21(a).

4. Master and Servant § 80— workmen’s compensation rates — appellate review of order disapproving

If the appellate court finds that an order disapproving a workmen’s compensation rate filing is not supported by material and substantial evidence, the court must then determine whether the filing complies with statutory standards and methods and is supported by substantial evidence. If the appellate court does not find such compliance, the disapproval order will be vacated and the cause remanded for proceedings as directed; however, if the court does find such compliance, the disapproval order will be vacated and the filing approved, and this will constitute a final determination under G.S. 58-124.22.

5. Master and Servant § 80— workmen’s compensation rates — effect of statutory changes — use of national distribution tables

A finding by the Commissioner of Insurance that a workmen’s compensation rate filing violated G.S. 58-124.19 because it relied on national distribution tables in calculating the effect of statutory changes on the rate structure when credible North Carolina data was available was not supported by substantial evidence where the Rate Bureau presented substantial evidence to show that the filing was based on credible North Carolina data when such data was available, and the Rate Bureau only relied on national statistics in distributing injuries and medical cost frequencies; there was no evidence tending to show that the national data used in the filing was not representative of North Carolina experience or that the distribution data had no validity in North Carolina; and there was no evidence that credible North Carolina experience existed.

6. Master and Servant § 80— workmen’s compensation rates —use of countrywide expense data

A finding by the Commissioner of Insurance that a workmen’s compensation rate filing was defective because it relied on countrywide expense data when credible North Carolina experience was available was unsupported by substantial evidence where the Rate Bureau presented evidence that credible North Carolina expense data was not available because of the interstate nature of workmen’s compensation insurance, and there was no evidence to the contrary.

7. Master and Servant § 80— workmen’s compensation rates —use of national credibility factors

A finding by the Commissioner of Insurance that a workmen’s compensation rate filing was defective because national credibility factors were used to supplement North Carolina credibility factors contrary to earlier rate-making procedure was unsupported by substantial evidence where there was testimony that the new procedure did not affect overall rates and reflected the class relativities more accurately than the old procedure, and there was no evidence that credible North Carolina evidence was ignored or given less weight than before or that the new procedure caused excessive or discriminatory rates.

*878. Master and Servant § 80— workmen’s compensation rates — consideration of expenses of stock companies only

A finding by the Commissioner of Insurance that proposed workmen’s compensation rates were excessive because the expense allowance was based solely upon the expenses of stock companies without consideration of the expenses of mutual companies, which generally have lower expenses than stock companies, was unsupported by material and substantial evidence where there was testimony that averaging the expenses of mutual and stock companies would not provide a fair rate of return for either type of company and that the rates for mutual companies were not excessive because dividends paid by the companies to the policyholders compensate for any extra charges, and the Department of Insurance presented no contradictory evidence.

9. Master and Servant § 80— workmen’s compensation rates — no breakdown of incurred losses

A finding by the Commissioner of Insurance that a workmen’s compensation rate filing was inadequate because it did not contain a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported losses was not supported by substantial evidence, although the Commissioner requested that the Rate Bureau submit such a breakdown of incurred losses and there was substantial evidence that such data was available, where there was no evidence that the data requested would establish that the proposed rates were excessive; there was substantial evidence that the loss development factors would compensate for any overstating of reserves; no evidence was presented by the Insurance Department tending to show that the loss development factors used by the Rating Bureau were not sufficient to compensate for potential excessive reserves; and the breakdown requested by the Commissioner was therefore immaterial to a determination of whether the proposed rates were excessive due to overstated reserves.

10.Master and Servant § 80— workmen’s compensation rates — underwriting margin — consideration of investment income

In a workmen’s compensation rate hearing, the Commissioner of Insurance could properly consider investment income in determining whether a 2.5% margin for underwriting was reasonable; however, the Commissioner erred in requiring the investment income to be considered at a risk-free rate of return rather than the rate of return actually experienced by the companies, since such requirement would limit the range of investments by insurance companies contrary to the provisions of G.S. 58-79.1.

Appeal by the North Carolina Rate Bureau and insurance companies from Order of the Commissioner of Insurance issued 7 December 1977. Heard in the Court of Appeals on 9 January 1979.

On 9 September 1977 the Rate Bureau filed for a 28.4% overall increase in the workers’ compensation insurance premium rate with the Commissioner of Insurance. On 10 October 1977, the Commissioner issued a “Notice of Public Hearing” which contend*88ed that the Filing was not in compliance with Article 12B of Chapter 58 of the General Statutes, because (1) it failed to utilize North Carolina evidence when credible evidence was available, (2) due consideration had not been given to past loss experience in North Carolina because the schedules were based on a national Medical Cost Frequency Distribution Study, the Workers’ Compensation Injury Tables and a distribution of wages called the 1973 Standard Wage Distribution Table, (3) the rates were excessive due to basing the rate formula solely on the expenses of stock companies when mutual companies have lower expenses, (4) the proposed rates were excessive due to excessive reserves being included in the incurred losses reported.

At the hearing on 9 November 1977, Roy Kallop, an actuary for the National Counsel on Compensation Insurance, testified for the appellant. Kallop was qualified as an expert in workers’ compensation rate-making. W. Bryon Tatum, Director of Technical Operations for the North Carolina Department of Insurance, who was qualified as an expert in workers’ compensation rate-making, testified for the Department of Insurance. Their testimony will be set out in detail later.

On 7 December 1977 the Commissioner issued an order disapproving the 9 September 1977 Filing by the Rate Bureau. The Commissioner made the following Findings of Fact and Conclusions of Law:

“Findings of Fact

1. That the North Carolina Rate Bureau (hereinafter called the Bureau) made a filing for revised worker’s compensation insurance rates on September 9, 1977.

2. That said filing proposed a 28.4% increase in the overall level of worker’s compensation rates and rating values presently in force in North Carolina.

3. That said filing proposed rate increases based on the alleged effect of legislative changes in benefits under the Workmen’s Compensation Act and further based on the alleged effect of medical fee changes.

4. That the rate level effects of legislation and revised medical fee schedules were based on the medical cost fre*89quency distribution study of the National Council on Compensation Insurance, a distribution of accidents called the Worker’s Compensation Injury Tables, and a distribution of wages called the 1973 Standard Wage Distribution Table, which are tables of countrywide data, when credible North Carolina experience was available on which to base the rate level effects of most of such legislation and revised medical fee schedules.

5. That the expense allowance in the rate-making formula was based on countrywide expenses when credible North Carolina expense experience was available.

6. That national credibility factors have been used to supplement the North Carolina credibility factors instead of the method of supplementing the North Carolina credibility factors used in previous worker’s compensation filings in North Carolina.

7. That the proposed rates are excessive due to basing the expense allowance in the rate-making formula solely on the expense experience of stock companies when stock companies have greater expenses than other companies.

8. That on cross examination, witness .Kallop testified that no audit of the loss experience reported by the companies had been made.

9. That said filing did not contain a breakdown of incurred losses into paid losses, reserves, bulk reserves, and incurred but not reported (IBNR) reserves.

10. That witness Tatum testified that a breakdown of incurred losses into paid losses, reserves, bulk reserves, and incurred but not reported (IBNR) reserves is required in order to properly analyze the incurred losses presented in the filing.

11. That an overstatement of incurred losses will cause an overstatement of the indicated rate level change.

12. That witness Tatum testified that in order to properly audit incurred losses for an indication of overstatement, in*90curred losses must be separated into paid losses, reserves, bulk reserves, and incurred but not reported (IBNR) reserves.

13. That by letter dated November 4, 1977 the Bureau reported that ‘Loss reserves for North Carolina worker’s compensation insurance are not available’ and that ‘Individual state data on net losses unpaid and incurred but not reported losses are also not available’ and that ‘Similarly unavailable are data with regard to bulk reserves.’

14. That the Insurance Expense Exhibit is a report that all casualty insurance companies are required to file with the Department of Insurance of each state in which they are licensed.

15. That the Insurance Expense Exhibit for calendar year 1976 was required to be filed by each company no later than April 1, 1977.

16. That the instructions for Part IV of the Insurance Expense Exhibit (Exhibit of Workmen’s Compensation Earned Premiums and Incurred Losses by States (Direct Business)) provide that ‘The reserves for unpaid losses used in calculating incurred losses should be the company’s individual estimate of outstanding claims, including reserves for claims incurred but not reported.’

17. That in order to report incurred losses on Part IV of the Insurance Expense Exhibit each company must ascertain its paid losses, reserves, bulk reserves, and incurred but not reported reserves.

18. That the elements of North Carolina Incurred Losses (paid losses, reserves, bulk reserves and incurred but not reported reserves) were available to the Bureau from each company on April 1, 1977 for the calendar year 1976.

19. That the Bureau did not make a request of its member companies for a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported (IBNR) reserves.

20. That witness Tatum testified that the Bureau’s failure to provide a breakdown of incurred losses into paid *91losses, reserves, bulk reserves and incurred but not reported (IBNR) reserves when said information was available and requested was a dilatory action and amounts to bad faith by the Bureau.

21. That witness Tatum testified that the filing contained no calculation of anticipated income on investments of unearned premium reserves and of loss reserves.

22. That witness Tatum testified that invested income on investments of unearned premium reserves and of loss reserves at a risk free rate of return of no less than 6% should be anticipated as future earnings.

23. That an understatement of anticipated earnings for the future will cause an overstatement of the indicated rate level change.

Conclusions of Law

1. That due consideration has not been given to past and prospective loss experience within North Carolina.

2. That the proposed rates are excessive due to basing the expense allowance in the rate-making formula solely on the expense experience of stock companies when stock companies have greater expenses than other companies.

3. That the Bureau has not carried the burden of proving that it has not overstated the indicated rate level change by its failure to provide a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported (IBNR) reserves which was available from its member companies and which is necessary in order to properly analyze and audit incurred losses for an indication of overstatement of incurred losses.

4. That the Bureau was dilatory in not providing a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported (IBNR) reserves.

5. That the Bureau has not carried the burden of proving that the proposed rates are not excessive due to its failure to provide for investment income on unearned premium reserves and on loss reserves at a risk free rate of return.”

*92From this order, the Rate Bureau appeals.

Attorney General Edmisten by Assistant Attorney General Isham B. Hudson, Jr. for plaintiff appellee.

Allen, Steed and Allen by Thomas W. Steed, Jr. and Charles D. Case; Young, Moore, Henderson & Alvis by Charles H. Young and George M. Teague for defendant appellants.

CLARK, Judge.

The Rate Bureau has appealed under the provisions of G.S. 58-124.22 and G.S. 58-9.4 from the order of the Commissioner of Insurance disapproving the 9 September 1977 Filing of the Rate Bureau in its entirety. This Court has the authority under G.S. 58-9.6(b) to reverse or modify the order of the Commissioner “if the substantial rights of the appellants have been prejudiced because the Commissioner’s findings, inferences, conclusions or decisions are: . . . [unsupported by material and substantial evidence in view of the entire record as submitted . . . .” The insurers, represented in this case by appellant Rate Bureau, have the right to a premium rate which will assure a fair and reasonable profit and no more. State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 471, 234 S.E. 2d 720 (1977).

The problem of deciding whether the Commissioner has acted responsibly is a delicate one. In this case brevity must yield to the massive record and the need for construction of 1977 rate-making legislation, codified in Article 12B, Chapter 58, General Statutes of North Carolina. Having given some consideration to the record on appeal, the briefs, relevant statutes and opinions, we approach the decision-making with the desire to avoid the tendency to judicialize administrative rate-making procedures.

I. Workers’ Compensation Rate of 1973

The last rate adjustment for workers’ compensation insurance rates went into effect on 1 December 1973. See State ex rel. Commissioner of Insurance v. Attorney General, 19 N.C. App. 263, 198 S.E. 2d 575, cert. denied 284 N.C. 252, 200 S.E. 2d 659 (1973). Under the statutory scheme at that time the Compensation Rating and Inspection Bureau was required to submit its rate proposals to the Commissioner for approval. G.S. 97-100; G.S. 97-102 to -104. The 1973 rate change was based on the 21 September *931972 Filing made by the Bureau. After the 1973 rate became effective, a rate filing was made on 19 March 1974 and was denied by the Commissioner on 14 October 1975. Upon appeal, this Court on 4 August 1976 ordered a remand for appropriate findings. See State ex rel. Commissioner of Insurance v. Rating and Inspection Bureau, 30 N.C. App. 332, 226 S.E. 2d 822 (1976). Thereafter the Commissioner held a series of hearings in 1976 and 1977, issued his “Revised Findings of Fact” on 11 February 1977, and again disapproved the Filing. The Bureau again appealed, and this Court in an opinion filed on 18 April 1978 vacated the order of the Commissioner on the ground that his findings of fact were not supported by material and substantial evidence, but the proceeding was not remanded because of the 9 September 1977 Filing which is the subject of this appeal. State ex rel. Commissioner of Insurance v. Rating and Inspection Bureau, 36 N.C. App. 98, 242 S.E. 2d 887 (1978).

Since the 21 September 1972 Filing, the basis for the 1973 rate, there have been six legislative changes and three Industrial Commission changes which have increased benefit levels and costs for workers’ compensation. These changes constitute substantially the basis for this 9 September 1977 Filing, an overall increase of 28.4% over the 1973 rate. Assuming that the 1973 rate was fair, in view of the nine changes resulting in increased benefit levels and costs, it is reasonable to conclude some increase in the rate is necessary to assure the insurer a fair and reasonable profit.

II. 1977 Legislation

In 1977 the General Assembly modified G.S. 97-100 and repealed G.S. 97-102 to -104 and other statutes making up a patchwork system of rate-making procedures and enacted new and comprehensive legislation for the purpose of regulating workers’ compensation and other types of insurance rate-making. Article 12B, Chapter 58, General Statutes of North Carolina.

The old “prior approval” method of setting workers’ compensation insurance rates was replaced by a new “file and use” procedure which in substance authorized the Rate Bureau to make a rate filing and provided that “[e]ach filing shall become effective immediately on the date specified therein but not earlier than 90 *94days from the date such filing is received by the Commissioner.” G.S. 58424.20(a).

If the Commissioner contends that the Filing fails to comply with the law he must, within 30 days after the date of the filing, give written notice to the Bureau “specifying in what respect . . . he contends such filing fails to comply . . .” and fix a date for hearing. G.S. 58424.21(a).

At such hearing the Commissioner shall consider the factors specified in G.S. 58424.19 as follows:

“Method of rate making; factors considered.— The following standards shall apply to the making and use of rates:
(1) Rates shall not be excessive, inadequate or unfairly discriminatory.
(2) Due consideration shall be given to past and prospective loss experience, within this State, to the hazards of conflagration and catastrophe, to a reasonable margin for underwriting profit and to contingencies, to dividends, savings or unabsorbed premium deposits allowed or returned by insurers to their policyholders, members or subscribers, to past a.nd prospective expenses specially applicable to this State, and to all other relevant factors including judgment factors, deemed relevant, within this State; provided, however, that countrywide expense and loss experience and other countrywide data shall be considered where credible North Carolina experience or data is not available.”

After hearing, if the Commissioner disapproves the filing, wholly or in part, G.S. 58424.21(a) requires that he include in his order “wherein and to what extent such filing is deemed to be improper . . . .” And to insure proper appellate review upon appeal under G.S. 58-9.4 by the Rate Bureau from all or any part of the order, the Commissioner must make findings of fact which specifically point out the absence of, or deficiencies in, the evidence produced in support of the filing. These findings must be supported by material and substantial evidence in view of the entire record as submitted. G.S. 58-9.6(b)(5); State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 70, 231 S.E. 2d 882 (1977).

*95Pending judicial review of the Commissioner’s order of disapproval, the Rate Bureau has the option to continue to use the filing rate, provided that each insurer member of the Bureau shall place in escrow account “the purportedly unfairly discriminatory or excessive portion of the premium collected during such interim period and the court, upon a final determination, shall order the escrowed funds to be distributed appropriately, except that refunds that are de minimus shall not be required.” G.S. 58-124.22(b).

This completes our attempt to present the new statutory scheme for workers’ compensation insurance rate-making. Verbatim quotations from the various relevant statutes have been limited and many statutes have been summarized, hopefully avoiding accusations of voluminosity without sacrificing clarity. The purpose is not to present a detailed account of the new statutory scheme of rate-making but to present those parts of the scheme which are relevant and significant in understanding and interpreting the statutory scheme.

III. Construction of 1977 Legislation

In construing the new statutory procedures for workers’ compensation rate-making our primary function is to insure that legislative purpose is accomplished. In addition to statutory language the court may also consider the circumstances surrounding the adoption which throws light upon the evil sought to be remedied, which involves a consideration of the terms and construction given to repealed statutes. State ex rel. Commissioner of Insurance v. Rate Office, 294 N.C. 60, 241 S.E. 2d 324 (1978); State ex rel. Commissioner of Insurance v. Rate Office, 293 N.C. 365, 239 S.E. 2d 48 (1977).

[1] Clearly, it was the legislative intent to eliminate unfair and unnecessary delay in the rate-making process. Such delay is clearly illustrated by the Filing, referred to heretofore, made under the old statutory scheme by the Compensation Bureau on 19 March 1974, which was disapproved by the Commissioner and on appeal vacated and remanded, again disapproved by the Commissioner, and finally determined by this Court on 18 April 1978, more than four years after the Filing. Further, this Court could not have made a final determination at that time if the 19 March *961974 Filing had not in effect been superseded by the Filing which is the subject matter of this proceeding, because the Court could only vacate the Commissioner’s order of disapproval and remand again to him. It is apparent that under the old statutory rate-making procedure a Commissioner could delay ad infinitum a rate filing, even though the rate filing, wholly or in part, was clearly supported by substantial evidence and the contentions of the Commissioner were feckless.

In enacting the 1977 legislation it was understood by the General Assembly that the new statutory rate-making scheme would “inevitably . . . entail corresponding changes in the functions and operations of the Insurance Commissioner and the Department Staff” as well as those of the predecessor to the Rate Bureau. LEGISLATIVE RESEARCH COMMISSION, REPORTS OF THE 1977 General Assembly of North Carolina: Fire and Casualty Insurance Rate Regulation (1977) at 47, 49.

[2, 3] In construing the new statutory rate-making process, the provisions having particular significance are the requirements that the Commissioner’s disapproval order must determine “wherein and to what extent such filing is deemed to be improper,” G.S. 58-124.21(a), and the “filing shall become effective immediately on the date specified therein,” G.S. 58424.20(a). It is evident that the Commissioner can no longer effectively disapprove a rate filing by inaction or a bare assertion that the Rate Bureau has not carried its burden of proof. Though the new statutory scheme does not shift the ultimate burden of proof from the Rate Bureau to the Commissioner, it does place on the Commissioner, in disapproving a filing, the burden of affirmatively and specifically showing how the Bureau has not carried its burden of proof, and, if the Commissioner fails to do so by substantial evidence, the presumption of prima facie correctness given to an order of the Commissioner by G.S. 58-9.4, -9.6 is rebutted.

[4] If the Commissioner fails to perform the affirmative duties imposed upon him by the 1977 legislation, this Filing shall be deemed to be approved, just as there is a deemed approval upon his failure to give notice of hearing within 30 days under G.S. 58424.21(b). If this Court, on appeal from the Commissioner’s order of disapproval, finds that the order is not supported by *97material and substantial evidence, it is then the duty of the Court to determine whether the Filing complies with the statutory standards and methods and is supported by substantial evidence. If we do not find such compliance the disapproval order will be vacated and the cause remanded for proceedings as directed. If .we do find such compliance, the disapproval order will be vacated and the Filing approved, and this will constitute a final determination under G.S. 58-124.22, which will require an order distributing the escrowed funds to the members of the Rate Bureau.

In State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 471, 234 S.E. 2d 720, it was held that the reviewing court has no inherent authority to fix rates nor to continue them in effect pending a hearing on remand. This decision was based on the pre-1977 legislative scheme of rate-making. Under the 1977 legislative scheme, this Court is not setting a workers’ compensation rate. The rate is set by the Commissioner in failing to carry the burden of showing affirmatively and specifically that the Filing does not comply with statutory standards. Any other conclusion would be ineffectual in carrying out the purpose of the legislature to establish without unnecessary delay rates which “shall not be excessive, inadequate or unfairly discriminatory.” G.S. 58-124.19(1).

We now reach the question of whether the Commissioner’s order of 7 December 1977, disapproving the Filing, was within his authority and terminated the effectiveness of the Filing.

IV. Use of Countrywide Data

The Commissioner found that the Filing by the Rate Bureau was deficient because the Bureau improperly relied on countrywide data in three instances in calculating the proposed rates. The Filing was held to be defective because the Bureau (A) relied on national distribution tables in calculating the effect of statutory changes on the rate structure, (Finding of Fact 4); (B) based its expense allowance on countrywide expenses, (Finding of Fact 5); and (C) supplemented North Carolina credibility factors with national credibility factors, (Finding of Fact 6).

A.

[5] In Finding of Fact 4, the Commissioner found that the Filing was in violation of G.S. 58-124.19 because the Filing relied on the *98medical cost frequency distribution study of the National Council on Compensation Insurance, a distribution of accidents called the Workers’ Compensation Injury Tables and a distribution of wages entitled the 1973 Standard Wage Distribution Table, in calculating the effect of statutory modifications on the rate structure when credible North Carolina data was available. The Rate Bureau contends that the Commissioner’s finding was not supported by substantial evidence.

Since 1 December 1973, the date at which the current rates went into effect, the legislature and the Industrial Commission have provided for numerous increases in Workers’ Compensation benefit levels. Roy Kallop, expert witness for the Rate Bureau, testified that some of the amendments were completely reflected in North Carolina experience and therefore no national data was relied upon in calculating the effect of those modifications on the rate structure. However, not all of the amendments were completely reflected in experience. Since the liability of the companies increased as a result of the amendments, it was necessary to include the anticipated effects of the changes in the Filing. In determining the effect of the statutory changes upon the rate structure the Rate Bureau used the following procedure: First, the Bureau calculated what portion of each statutory change was reflected in experience, applying a weighted average to those amendments which went into effect in the middle of a calendar year. The Filing then calculated the loss experience caused by the statutory changes which were already fully or partially reflected in experience. Second, the Bureau calculated the effect of the benefit level increases and medical fee changes not reflected in experience by (1) generating the old monetary costs and new monetary costs, based solely on North Carolina data such as the minimum weekly benefit and average weekly wage in this State, (2) dividing the new costs by the old costs to determine the percentage of change in costs and (3) distributing the percentage of change among the various types of injuries. Finally, applying North Carolina loss experience by the types of injuries and medical fees, the Bureau calculated the overall cost increase.

Kallop testified that the countrywide data was only used in calculating the effect of statutory changes not yet reflected in experience. The countrywide data was used in calculating distribution since a large pool of experience is necessary in order *99to obtain reliable figures, and North Carolina experience is insufficient to distribute dependency groups and types of injuries such as death or dismemberment.

The Rate Bureau presented substantial evidence to show that the Filing was based on credible North Carolina data wherever such data was available, and that the Rate Bureau only relied on national statistics in distributing injuries and medical cost frequencies. There is no evidence which tends to show that the countrywide data utilized in the Filing was not representative of North Carolina experience or that the distribution tables had no validity in North Carolina. See, State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 70, 231 S.E. 2d 882 (1977). Nor was there any testimony that credible North Carolina experience existed. G.S. 58-124.19 specifically authorizes the Rate Bureau to rely upon countrywide data if there is insufficient experience in North Carolina to provide credible statistics. The Commissioner may not reject as untrustworthy evidence that is uncontradicted or unimpeached. State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 471, 234 S.E. 2d 720 (1977). The Commissioner’s Finding of Fact 4 was not based on substantial evidence and therefore it is apparent that this finding provides no support for the Commissioner’s Conclusion of Law 1.

B.

[6] The Commissioner also found in Finding of Fact 5, that the Filing relied upon countrywide expense data when credible North Carolina experience was available. There was testimony by Kallop that the workers’ compensation business is inherently interstate in nature and that it is difficult and impractical to isolate expenses separately by state. The Department of Insurance presented no testimony to contradict Kallop, nor was there any evidence that credible North Carolina evidence was available or that the countrywide data did not reflect North Carolina expenses. In State ex rel. Insurance Commissioner v. Rating Bureau, 292 N.C. 70, 231 S.E. 2d 882, the court held that the evidence did not support the Commissioner’s finding of fact that the Filing was defective because it relied on countrywide expense data. The uncontradicted evidence established that the countrywide data was representative of interstate experience. In State ex rel. Commissioner of Insurance v. Attorney General, 19 *100N.C. App. 263, 198 S.E. 2d 575, the court noted that most workmen’s compensation insurers do business in many states and that a company employee is not paid more or less while working on a North Carolina transaction than while working on a matter from another state. These expenses tend to be uniform from state to state. In addition, the insurers are often multi-state corporations. Here, the testimony of Kallop tends to show that credible North Carolina expense data is not available because of the interstate nature of workers’ compensation insurance. There is no evidence to the contrary. Thus the record reveals that the Commissioner’s Finding of Fact 2 is not supported by substantial evidence and cannot support the Commissioner’s Conclusion of Law No. 1.

C.

[7] The Commissioner also found in Finding of Fact 6 that national credibility factors were used to supplement the North Carolina credibility factors contrary to earlier rate-making procedures.

Kallop testified that in the former method of allocating losses among the individual classification groups the Rate Bureau assigned a percentage of credibility to actual North Carolina experience. The residual percentage of that class was assumed to parallel the loss changes in the broad industry group, “manufacturing”, “contracting”, or “all others” into which the class fell. Consequently, according to Kallop, the allocation of costs among the individual classifications was not accurate.

In the Filing sub judice, the Bureau assigned a percentage of credibility to North Carolina data following the established procedures. Then, half of the residual credibility was deemed to track the national rate of change for that classification. The other half of the residual credibility was treated as in previous filings in which the losses were assumed to track the rate of change in the broad industry group in North Carolina. This method, according to Kallop, does not produce any more or less premium. It is designed merely as a means of establishing the most accurate relationship of losses between the various classes of workers in North Carolina. There is no requirement that the Rate Bureau must always use the same rate-making formulas. See, State ex rel. Commissioner of Insurance v. Rate Office, 287 N.C. 192, 214 *101S.E. 2d 98 (1975). G.S. 58-124.19 provides that national data may be utilized where credible North Carolina experience- is unavailable. Here, the countrywide data is used only to supplement the North Carolina evidence, so it is clear that the Filing is not in violation of G.S. 58-124.19. In addition, Kallop testified that the new procedure did not affect overall rates, and that under the new formula the class relativities were reflected more accurately. There is no evidence whatsoever that credible North Carolina evidence was ignored, or given less weight than before, or that the new formula caused excessive or discriminatory rates in violation of Article 12B, Chapter 58 of the General Statutes. The Commissioner’s Finding of Fact 6 is unsupported by substantial evidence and cannot support his Conclusion of Law 1.

Therefore, there remains no finding of fact, based on substantial evidence, to support the Commissioner’s Conclusion of Law 1.

V. Expenses of Companies

[8] In Finding of Fact 7, the Commissioner found that the proposed rates were excessive because the expense allowance was based solely upon the expenses of stock companies, without considering the expenses of mutual companies, and that stock companies generally have higher expenses than mutual companies.

The Rate Bureau contends that the Commissioner’s finding was not based upon substantial and material evidence.

The testimony of Kallop tended to show that the insurance companies which write workers’ compensation insurance are either mutual companies or stock companies. Stock companies, owned by stockholders, generally market insurance through commissioned agents. Mutual companies, on the other hand, utilize no agents since they are owned and operated by the policyholders. Mutual companies sell insurance through company offices thereby eliminating the additional costs of agents and commissions. The mutual companies then return the excess portion of the premiums to the policyholders by way of dividends. Kallop testified that if the expense costs are averaged, as urged by the Department of Insurance, the stock companies are unable to obtain a fair rate of return, and the mutual companies are unable to pay sufficient dividends to compete with the stock companies. Averaging the expenses of mutual and stock companies would not provide a fair rate of return for either type of company.

*102Bryon Tatum, witness for the Insurance Department, testified that calculating the expenses based solely upon the expenses of stock companies would provide excessive profits for mutual companies because of the higher expenses of stock companies. Tatum also testified that expenses are averaged in calculating expenses for automobile insurance.

G.S. 58-124.19 provides that the Commissioner may disapprove a rate filing if the proposed rates are “excessive, inadequate or unreasonable.” According to Kallop’s testimony, stock companies and mutual companies, because of their different structures, by definition have different expenses. The statutory scheme, however, does not allow deviations and therefore the Rate Bureau must generate a unified rate providing a fair rate of return for both types of companies. Kallop testified that the averaging method proposed by the Department of Insurance would create a rate that is inadequate to cover the expenses of stock companies and which did not accurately reflect the expenses of mutual companies. The proposed rate clearly provides an accurate expense allowance for stock companies, and also provides a fair rate of return for mutual companies since the excessive profits are returned to the policyholders in dividends.

The Commissioner’s disapproval must be based on an affirmative showing that the proposed Filing fails to comply with statutory standards. State ex rel. Commissioner of Insurance v. Rating Bureau, 30 N.C. App. 487, 228 S.E. 2d 261 (1976), aff’d 292 N.C. 70, 231 S.E. 2d 882 (1977). The Commissioner may not reject as untrustworthy uncontradicted testimony or data. State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 471, 234 S.E. 2d 720 (1977).

There is substantial and uncontradicted evidence presented by the Department of Insurance that the Filing considered only the expenses of stock companies and that stock companies have greater expenses than mutual companies. This is conceded by the Rate Bureau. Therefore, the Department of Insurance contends, the rates are excessive for mutual companies. This contention is refuted by Kallop’s testimony that the rates for mutual companies are not excessive because dividends paid by the companies to the policyholders compensate for any extra charges. The Department of Insurance presented no evidence contradicting or impeaching *103Kallop’s testimony. That part of the Commissioner’s Finding of Fact 7 which finds that the proposed rates are excessive is not based on material and substantial evidence and thus there is no support for the Commissioner’s Conclusion of Law No. 2.

VI. Breakdown of Loss Reserves

[9] In Findings of Fact 9 through 19, the Commissioner stated that the Filing did not contain a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported losses, and that an overstatement of such reserves would result in excessive rates. The Commissioner also found that the necessary information on the breakdown of reserves was available to the Bureau. The Commissioner concluded as a matter of law that the Rate Bureau had “not carried the burden of proving that it has not overstated the indicated rate level change by its failure to provide a breakdown of incurred losses into paid losses, reserves, bulk reserves and incurred but not reported (IBNR) reserves which was available from its member companies and which is necessary in order to properly analyze and audit incurred losses for an indication of overstatement of incurred losses.”

The Notice of Public Hearing requested that the Rate Bureau submit a breakdown of incurred losses to the Commissioner. On 4 November 1977, the Bureau notified the Commissioner that such a breakdown was unavailable, and indicated that the annual change in loss reserves was reported on page 14 of Best’s Executive Data Service. Kallop testified that the reserve figure in Best’s data was a reflection of calendar year changes. In other words, that figure represented paid losses and incurred losses for the calendar year minus the initial loss reserve. The losses included all losses reported and incurred during that particular calendar year, regardless of when the policy covering the loss was written. Kallop also testified that, assuming that the reserves were excessive, the loss development factor which tracked losses over a period of years would correct for any underreserving or overreserving. The change in reserve would be reflected in the calendar year data, and that would feed into the next year’s calendar year data and there would consequently be a correction for excess reserves. Therefore, the rates would not be excessive regardless of any overstatement of reserves. In addition, Kallop *104testified that it is impossible to determine whether or not the reserves are excessive by a breakdown of reserves into the various components because the determination of liability must await the final resolution of each claim. The setting of reserves is essentially an estimate of liability. If the estimate should prove too high or too low, this would be corrected in the loss development factor.

Tatum testified that in the medical malpractice insurance companies, the case reserves were overstated and that the insurance companies could not justify the method of calculating incurred but not reported loss reserves. There is no uniform method currently utilized by the companies for computing reserves. The reports filed with the National Association of Insurance Commissioners lists the states and has columns for earned premiums, incurred losses and the loss ratio. In order to report such statistics, the workers’ compensation insurance companies first had to determine the case reserves and incurred but not reported loss reserves. Therefore, the companies reported incurred losses, which is by definition the sum of paid losses, reserves and IBNR.

The Commissioner’s findings as to the availability of the reserve statistics were based on substantial evidence. Tatum testified that such a breakdown of loss reserves was available. The Commissioner has the authority pursuant to G.S. 58424.18(d) to compel the production of data necessary to compile statistics, and the Rate Bureau is required to maintain reasonable records of the experience of its member companies and the data used in rate-making. G.S. 58424.20(c).

The purpose of the hearing before the Commissioner is to determine whether the proposed rates are “unreasonable, excessive or discriminatory.” Although the breakdown of loss reserves was not provided by the Rate Bureau, there was no evidence that the proposed rates were excessive, or that the data requested would establish that the rates were excessive. On the contrary, there was substantial evidence that the loss development factors would compensate for any overstating of reserves. There was no evidence presented by the Insurance Department which tended to show that the loss development factors utilized by the Rate Bureau were not sufficient to compensate for poten*105tial excessive reserves. The information requested by the Commissioner was therefore immaterial to a determination of whether the rates were excessive due to overstated reserves. The Commissioner may not reject as untrustworthy uncontradicted testimony or data. State ex rel. Commissioner of Insurance v. Rating Bureau, 292 N.C. 471, 234 S.E. 2d 720 (1977). There was therefore no substantial evidence upon which the Commissioner could base his Findings of Fact, and thus the Commissioner’s Conclusions of Law 3 and 4 are unsupported by adequate findings of fact.

In the Commissioner’s Finding of Fact 20, the Commissioner stated that the witness Tatum testified that the Rate Bureau was dilatory in failing to report a breakdown of loss reserves. We note that the Commissioner did not make a finding of fact that the Rate Bureau was dilatory, but only a finding that Tatum so testified. The Commissioner’s so-called Finding of Fact 20 lends no support to Conclusion of Law 4.

VII. Investment Income

[10] In Findings of Fact 21 and 22 the Commissioner found that Tatum testified that the Filing contained no calculation of anticipated income and that investment income should be calculated at a risk-free rate of return. Again we find mere assertions of a witness’ testimony, which are not findings of fact and cannot support the Commissioner’s Conclusion of Law 5. In addition, Finding of Fact 23 states that an understatement of anticipated earnings would cause an overstatement of the indicated rate level change. The Commissioner’s “Finding,” however, is merely an observation that overstated anticipated earnings cause an overstatement of indicated rates and not a finding that the earnings are understated in the proposed rates. Therefore, the Commissioner’s Conclusion of Law 5 is not supported by any competent findings of fact.

Nevertheless, even if we assume that the purported Findings of Fact 21-23 are competent, there was insufficient evidence to support them.

The Commissioner “found” that the Rate Bureau failed to include in the Filing the anticipated income on investments of unearned premium reserves and loss reserves, that investment in*106come of 6% at a risk-free rate of return should be anticipated as future earnings and that an understatement of investment income would cause an overstatement in the indicated rate level change.

Kallop testified that income from investments was not considered in the rate increase but that the investment income was reflected in the 2.5% underwriting profit. Kallop testified that the companies earned a 4.3% rate of return on unearned premium and loss reserves.

Tatum testified that in order to determine whether or not the 2.5% return on underwriting was reasonable, the rate of return on investment income must be considered as well.

The Rate Bureau contends that it is not required that investment income be considered in workers’ compensation insurance. In State ex rel. Commissioner of Insurance v. Attorney General, 19 N.C. App. 263, 198 S.E. 2d 575, cert. denied, 284 N.C. 252, 200 S.E. 2d 652 (1973), this Court held that the Commissioner was not required to consider the investment income earned by the companies in reviewing rates. In State ex rel. Commissioner of Insurance v. Attorney General, 16 N.C. App. 724, 193 S.E. 2d 432 (1972), the Attorney General contended that the Commissioner must consider investment income. In construing the statute, this court noted that there was no specific statutory requirement that investment income be considered in the pertinent statute. Therefore, the Commissioner only needed to consider the underwriting profit. The Rate Bureau contends that those cases cited above are dispositive here since G.S. 58-124.19(2) provides that “Due consideration shall be given within this State ... to a reasonable margin for underwriting profit . . . .” (Emphasis added). Since the new statutory provisions refer specifically to “underwriting profit,” and not to “investment income,” the Commissioner need not consider investment income. See, State ex rel. Commissioner of Insurance v. Attorney General, 16 N.C. App. 724, 193 S.E. 2d 432 (1972).

The cases cited above, however, merely hold that the Commissioner is not required to consider profits from investment income in every rate filing. The cases are not dispositive of the issue of whether the Commissioner may consider such evidence. In State ex rel. Commissioner of Insurance v. Attorney General, 16 N.C. App. at 729, 193 S.E. 2d at 435, the court noted that *107“[w]hether, under the statutory provisions governing this proceeding, the Commissioner would have committed error had he required evidence on [investment income] is not before us.”

We must therefore consider whether the Commissioner may consider investment income in reviewing rates. In State ex rel. Commissioner of Insurance v. Attorney General, 19 N.C. App. at 271, 198 S.E. 2d at 581, the court noted that the evidence indicated that “investment income from the unearned premium and loss reserves was taken into consideration in the establishment of the 2.5% allowance for profit and contingencies.” Kallop’s testimony in the case sub judice also indicated that investment income was considered in arriving at the 2.5% margin of profit for underwriting. Therefore, it is clear that the Rate Bureau and the Commissioner have implicitly considered the rate of return on investments in arriving at a reasonable rate of return for underwriting. It was not error for the Commissioner to request information on investment income or to consider investment income in determining whether or not the 2.5% underwriting profit was reasonable.

Kallop testified that the total return from investment income on both unearned premium and loss reserves, expressed as a percentage of premiums before taxes, was 4.3%. When combined with the 2.5% allowance for underwriting profit, the total return from both sources of income equalled 6.8%, before taxes. The after tax income was approximately 5.01%.

The Department of Insurance, however, contended that the rate of return on investment income should be calculated at a risk-free rate of return which amounted to 6.0%. The Insurance Department witness Tatum testified that this risk-free rate of return is the appropriate rate of return to consider. In support of this contention, the Department of Insurance cites Attorney General v. Commissioner of Insurance, 353 N.E. 2d 745 (Mass. Sup. Ct. 1976), in which the court held that investment income should be computed at a risk-free rate of return rather than the actual rate of return experienced by the companies.

G.S. 58-79.1 sets forth guidelines for investment by insurance companies. There is no requirement in the statute that insurance companies invest in risk-free ventures; rather, the statute pro*108vides that insurance companies may engage in a variety of investments.

To require the investment income to be calculated at a risk-free rate of return would limit the range of investments by insurance companies, contrary to the intent of the legislature, evidenced by the provisions of G.S. 58-79.1.

Although the Commissioner is empowered to consider investment income in determining whether the 2.5 °/o margin for underwriting was reasonable, the Commissioner erred in requiring the investment income to be considered at a risk-free rate of return rather than the rate of return actually experienced by the companies.

There was considerable evidence by Kallop that the 2.5 °/o allowance for underwriting profit was reasonable and that the 2.5% margin of profit for underwriting reflected the profits from investment income, amounting to a total of 5.01% profit allowance after taxes.

There was no evidence whatsoever that the 2.5% margin was unreasonable or excessive when the profits from investment income equaled 4.3% of the premiums. The Commissioner’s Findings of Fact 21 to 23 were not based on substantial and material evidence and therefore cannot support the Commissioner’s Conclusion of Law 5.

The Commissioner’s order disapproving the Filing by the Rate Bureau is vacated, and the rates proposed in the 9 September 1977 Filing by the Rate Bureau are deemed approved, and remain in effect until changed as by law provided, and the escrowed funds placed by the member insurance companies in the escrow account pursuant to G.S. 58424.22(b) shall be remitted to the member insurers by the escrow agents.

Vacated.

Judges Vaughn and Hedrick concur.