SJC 13807

Mass.

Jul 6, 2026

THE WORKERS’ COMPENSATION RATING AND INSPECTION BUREAU OF MASSACHUSETTS vs. COMMISSIONER OF INSURANCE.

SJC-13807

Supreme Judicial Court of Massachusetts

March 4, 2026. - July 6, 2026.

Suffolk. Present: Budd, C.J., Gaziano, Kafker, Wendlandt, Georges, Dewar, & Wolohojian, JJ.

NOTICE: All slip opinions and orders are subject to formal revision and are superseded by the advance sheets and bound volumes of the Official Reports. If you find a typographical error or other formal error, please notify the Reporter of Decisions, Supreme Judicial Court, John Adams Courthouse, 1 Pemberton Square, Suite 2500, Boston, MA, 02108-1750; (617) 557-1030; SJCReporter@sjc.state.ma.us

Commissioner of Insurance. Insurance, Rate setting, Workers’ compensation insurance, Commissioner of Insurance. Practice, Civil, Review of decision of Commissioner of Insurance. Administrative Law, Evidence, Judicial review, Rate setting, Findings.

Civil action commenced in the Supreme Judicial Court for the county of Suffolk on July 22, 2024.

The case was reported by Kafker, J.

Civil action commenced in the Supreme Judicial Court for the county of Suffolk on June 13, 2025.

The case was reported by Gaziano, J.

Melissa C. Allison (Mina S. Makarious also present) for the petitioner.

John R. Hitt, Assistant Attorney General (Coleman G. Smith also present) for the respondent.

KAFKER, J. The Workers’ Compensation Rating and Inspection Bureau of Massachusetts (WCRIB) seeks review of two decisions by the Commissioner of Insurance (commissioner) that disapproved WCRIB‘s filings for proposed workers’ compensation rate revisions. In the first filing, made in December 2023 for new rates to be effective on July 1, 2024, WCRIB proposed a 7.6 percent rate decrease. The commissioner disapproved the filing and ordered a 14.6 percent decrease. In the second filing, made in November 2024 for new rates to be effective on July 1, 2025, WCRIB proposed a 7.1 percent rate increase. The commissioner disapproved the filing and left the then-existing rate intact (i.e., the rate that went into effect on July 1, 2024, that reflected a 14.6 percent decrease from the prior rate). Although the commissioner properly exercised his authority to disapprove WCRIB‘s proposed rates, for the reasons discussed below, he failed to explain, as to the 2024 decision, the basis for his decision to order a 14.6 percent rate decrease. We remand the matter to the commissioner for further consideration.

1. Background. a. Statutory scheme.

WCRIB is a licensed rating organization pursuant to G. L. c. 152, § 52C (a). As such, it is required to file “classifications of risks and premiums . . . at least every two years and on any additional date that the commissioner . . . may designate.” G. L. c. 152, § 53A (2). The commissioner, in turn, is required to thereafter conduct a hearing “to determine whether the classifications and rates are not excessive, inadequate or unfairly discriminatory for the risks to which they respectively apply and that they fall within a range of reasonableness.” Id. If the commissioner “disapproves proposed premiums and classifications, stating his reasons for disapproval, any . . . rating organization may file new proposed classifications and premiums.” G. L. c. 152, § 53A (7). Additionally, if the commissioner “determines that any already effective premium is excessive, he shall order a specific decrease in that premium to be effective six months from the date of the . . . rating organization filing under consideration.” G. L. c. 152, § 53A (8).

WCRIB is the only licensed rating organization for workers’ compensation insurance in Massachusetts, and all companies licensed to sell workers’ compensation in Massachusetts are members of WCRIB. Additionally, the commissioner has designated WCRIB the “advisory organization” for self-insurance groups pursuant to G. L. c. 152, § 25O (1), which provides that “[e]very workers’ compensation self-insurance group shall adhere to the uniform classification system, uniform experience rating plan, and manual rules filed with the commissioner of insurance by an advisory organization designated by said commissioner.”¹

b. 2024 decision.

On December 22, 2023, as directed by the commissioner, WCRIB submitted its “General Revision of Workers’ Compensation Insurance Rates and Rating Values,” proposing rates to be effective July 1, 2024.² WCRIB initially proposed a Statewide decrease in average rates of 8.3 percent, which it later changed to 7.6 percent. A public comment hearing took place on February 7, 2024, at which representatives spoke for WCRIB, the State rating bureau (SRB),³ and the Massachusetts affiliate of the National Association of Housing and Redevelopment Officials (NAHRO), a Massachusetts self-insurance group (SIG).

The February 7 hearing was followed by several administrative hearings in March, April, and May 2024. The Attorney General and the SRB each submitted advisory filings and participated fully in the administrative process.⁴ The Attorney General challenged several aspects of WCRIB‘s filing and recommended a rate decrease of 17.5 percent. The SRB also challenged several aspects of WCRIB‘s filing but did not recommend a specific rate decrease. The commissioner ultimately disapproved WCRIB‘s filing for several reasons, including WCRIB‘s use of a two-year look-back period for indemnity paid losses, rather than a five-year look-back period, and WCRIB‘s use of a new methodology for calculating the underwriting profit provision. The commissioner concluded that the then-existing rates were excessive and ordered a Statewide decrease in average workers’ compensation insurances rates of 14.6 percent (2024 decision). No explanation was provided for how the commissioner reached that number. The commissioner also directed WCRIB to submit a filing in December 2024, for rates to take effect on July 1, 2025.

c. 2025 decision.

On November 15, 2024, WCRIB submitted its “General Revision of Workers’ Compensation Insurance Rates and Rating Values,” to be effective July 1, 2025, proposing a Statewide increase in average rates of 7.1 percent.⁵ A public comment hearing took place on January 14, 2025, at which representatives spoke for WCRIB, SRB, and NAHRO. That hearing was followed by several administrative hearings in February and March 2025.

The Attorney General and the SRB each submitted advisory filings and participated fully in the administrative process. The Attorney General challenged several aspects of WCRIB‘s filing and recommended a rate decrease of 6.2 percent. The SRB also challenged several aspects of WCRIB‘s filing and recommended a rate decrease. The commissioner again disapproved WCRIB‘s filing for at least some of the same reasons that he had disapproved the previous filing -- WCRIB‘s use of only two years of data for purposes of assessing indemnity paid losses and the methodology used in connection with the underwriting profit provision. The commissioner concluded that he could not determine that WCRIB‘s proposed rate was not excessive, inadequate, or discriminatory and fell within a range of reasonableness. He also concluded that no party had presented convincing evidence that the current rate was excessive and therefore declined to order any specific decrease, leaving the then-existing rate, i.e., the rate that had gone into effect on July 1, 2024, intact (2025 decision).

WCRIB sought review of both decisions by filing petitions in the county court pursuant to G. L. c. 152, § 53A (10), which provides that any party aggrieved by the commissioner‘s decision may petition this court for review. In August 2025, a single justice reserved and reported the first case, involving the 2024 decision, to the full court, and in September 2025, a different single justice reserved and reported the second case, involving the 2025 decision, to the full court. The cases were thereafter consolidated.⁶

2. Discussion. a. Standard of review.

The commissioner “may disapprove rates or withdraw his approval only if rates are inadequate, excessive or unfairly discriminatory.” Workers’ Compensation Rating & Inspection Bur. of Mass. v. Commissioner of Ins., 391 Mass. 238, 245 (1984) (WCRIBMA), quoting Liberty Mut. Ins. Co. v. Commissioner of Ins., 366 Mass. 35, 42 (1974). “He must determine whether the rates are inadequate, excessive or unfairly discriminatory, based upon their falling within a ‘range of reasonableness.‘” WCRIBMA, supra. Although the “court does not require the commissioner to make findings on every controverted issue of law or fact,” we do “require findings . . . that indicate the over-all basis of [the commissioner‘s] decision and that permit effective appellate review.” Massachusetts Auto. Rating & Acc. Prevention Bur. v. Commissioner of Ins., 401 Mass. 282, 292 (1987), and cases cited. Indeed, if the commissioner disapproves a proposed rate, he must, pursuant to G. L. c. 152, § 53A (7), “stat[e] his reasons for disapproval.”⁷

“Our review of the commissioner‘s rulings . . . is limited to whether the evidence reasonably supports his findings that the rates either did or did not comply with the standards set out in G. L. c. 152, § [53A].” WCRIBMA, 391 Mass. at 245. Here, the commissioner‘s findings, in his written decisions, adequately indicate the bases for his assessment of various controverted issues and his decision to disapprove WCRIB‘s proposed rates as inadequate, excessive, or unfairly discriminatory. The commissioner‘s findings, in other words, had reasonable support in the evidence. See id. at 244. What is missing, however, is any indication -- any reasoned explanation -- as to how, after making his various findings, the commissioner determined in the 2024 decision that a 14.6 percent decrease in rates was warranted. We address these points in turn.

b. Excessive rates.

In seeking review of the commissioner‘s decisions, WCRIB argues, among other things, that the commissioner improperly imposed a ratemaking methodology that deviated from WCRIB‘s long-established methodology for forecasting future losses.⁸ Essentially, WCRIB uses historical data to estimate or predict the ultimate value of workers’ compensation claims after the claims are fully settled or paid. Because, in the workers’ compensation context, some benefits are paid out over long periods of time, losses do not necessarily reach their final value for many years. As explained by the commissioner in his 2024 decision, “[u]nderlying the methodology for calculating [loss development factors] is the assumption that losses develop in a consistent pattern over time and that historical experience has predictive value for future development.”

As WCRIB accurately notes, there is no dispute here regarding the general model of loss development as a way to set rates. What is in dispute is the number of years of historical data that should underlie the “indemnity paid loss” estimate.⁹ In its rate filing, WCRIB used the two most recent years of data, as it had done for more than twenty years. Here, for purposes of the 2024 decision, that entailed using data from policy years 2021 and 2022. The commissioner, however, concluded that five years of data, from 2018 through 2022, should be used to estimate indemnity paid losses. In so doing, the commissioner noted, and no one disputes, that, due to the COVID-19 pandemic, conditions and the relevant data were different in 2020. The data indicate that losses were historically lower in 2020 and historically higher in 2021 and

2022.¹⁰ A five-year loss development period accounts for two pre-COVID-19 years (2018 and 2019), the peak COVID-19 year of 2020, and two post-COVID-19 years (2021 and 2022). Essentially, the five years of data account for the highs and lows stemming from the COVID-19 pandemic and, as is relevant here, their effect on the workplace and labor force; considering only the two most recent years of data, 2021 and 2022, does not.¹¹

As a part of its argument that the commissioner erred in his decision necessitating consideration of five years of data rather than two, WCRIB argues that, where historically only two years of data have been relied on, the commissioner‘s decision lacks consistency. The last time that the commissioner considered how many years of data should be used was in 2003,

when, in considering WCRIB‘s rate filing for rates to be effective on September 1, 2003, the commissioner declined to deviate from the two-year data period. At the time, the commissioner stated that there was no evidence that the established methodology produced unreasonable results. The commissioner also noted in that decision that the “use of a predictable, consistent methodology promotes stability in ratemaking.” Here, WCRIB argues that, in shifting from the use of two years of data to five years of data, the commissioner is being inconsistent.

It is certainly true that “[a] party to a proceeding before a regulatory agency . . . has a right to expect and obtain reasoned consistency in the agency‘s decisions.” Boston Gas Co. v. Department of Pub. Utils., 367 Mass. 92, 104 (1975). This does not mean, however, that an agency “may never deviate from its original position, only that any such change must be explained.” MCI WorldCom Communications, Inc. v. Department of Telecomm. & Energy, 442 Mass. 103, 116 (2004), citing Robinson v. Department of Pub. Utils., 416 Mass. 668, 673 (1993). Here, the commissioner has provided just such an explanation. WCRIB may not agree with the decision, but there is no question that the commissioner provided a reasoned explanation. The changing assessment of the number of years of data to be relied upon in

assessing the proposed rates stems directly, and understandably, from the COVID-19 pandemic, which resulted in anomalous data.

Although the commissioner based his decision that he could not approve WCRIB‘s proposed rates as “not excessive, inadequate . . . and . . . within a range of reasonableness” on a number of factors, including the loss development data issue and an issue related to the underwriting profit provision, the loss development data issue is the only rate-premium issue that WCRIB contests on appeal.¹² And, on that point, the evidence reasonably supports the commissioner‘s findings that WCRIB‘s proposed rates -- its proposed decrease -- did not comply with the parameters of G. L. c. 152, § 53A. See WCRIBMA, 391 Mass. at 245.

In opposing the commissioner‘s disapproval of its proposed rate decrease, WCRIB does not suggest that the difference between using five years of data and two years of data for estimating indemnity paid losses and a different methodology for determining the underwriting profit provision will result in a

significantly different percentage reduction. Instead, WCRIB argues that the commissioner must determine a specific quantifiable range of reasonableness and then conclude that such a range has been violated, rather than proceed as the commissioner did here to identify unreasonable or excessive aspects of WCRIB‘s request.

The commissioner‘s approach was reasonable. He did not need to set out a specific quantitative range but rather could identify unreasonable or excessive aspects of WCRIB‘s methodology for calculating its rate request and then conclude that he could not find that such request was not excessive and was within a range of reasonableness.¹³ See Massachusetts Ass‘n of Older Ams. v. Commissioner of Ins., 393 Mass. 404, 408 n.7 (1984), citing WCRIBMA, 391 Mass. at 245 (declining to limit “range of reasonableness” to specific percentage in context of health insurance rate setting). We have recognized no particular threshold for determining that a proposed rate is excessive. See WCRIBMA, supra at 264 (predecessor statute to G. L. c. 152, § 53A, did “not require the commissioner to

approve elements of filings which would lead to rates falling within a range of excess, no matter how small“).¹⁴

c. The 14.6 percent decrease.

We turn, next, to the commissioner‘s conclusion, in the 2024 decision, that a 14.6 percent decrease in over-all rates was warranted. Disapproving the proposed rate decrease and specifying the amount of the decrease present two different issues. As we have explained, the commissioner‘s determination that five years of data rather than two should be used to calculate the “indemnity paid loss” estimate is supported by the record, and his determination that a new methodology should not be used to calculate the underwriting profit provision has not been contested on appeal. For at least these two reasons, the commissioner reasonably rejected WCRIB‘s proposed rate decrease of 7.6 percent on the basis that it would result in rates that were inadequate, excessive, or unfairly discriminatory.

But then, without further explanation, the commissioner stated that he was “exercise[ing his] authority” under G. L. c. 152, § 53A (8), “to order a statewide decrease in average workers compensation insurance rates of 14.6 [percent].” WCRIB does not contest the commissioner‘s authority to order a specific rate decrease. Indeed, doing so is expressly provided

by G. L. c. 152, § 53A (8), which states that if the commissioner “determines that any already effective premium is excessive, he shall order a specific decrease.” What WCRIB does contest, and reasonably so, is the lack of any explanation for the specific percentage decrease. That WCRIB, the Attorney General, and the SRB all advocated for a rate decrease is certainly relevant to the commissioner‘s decision to order a decrease, but it does not address how the commissioner reached the 14.6 percent figure. Additionally, that the figure lies between WCRIB‘s proposed rate decrease of 7.6 percent and the Attorney General‘s proposed rate decrease of 17.5 percent may also be relevant but, again, does not explain the end result. As our high school mathematics teachers reminded us: you need to show your work. That is true for the commissioner‘s calculations as well. See Massachusetts Auto. Rating & Acc. Prevention Bur., 401 Mass. at 292 (“[T]he amount of the savings is only conjecture when the method of computation is not disclosed. At this point, we can only speculate on the effects of the tax because the commissioner has not revealed his methodology“).

Without that analysis, we have no way of knowing how the commissioner arrived at the 14.6 percent figure. We do not require, as noted above, that the commissioner “make findings on every controverted issue of law or fact.” Massachusetts Auto.

Rating & Acc. Prevention Bur., 401 Mass. at 292. We do, however, “require findings . . . that indicate the over-all basis of his decision and that permit effective appellate review.” Id. We do not have that here in connection with the 14.6 percent decrease and thus “have no way of knowing how [that figure was reached,] and, hence, effective review is frustrated.” Id.

The commissioner must provide a reasoned explanation as to how he determined, as he necessarily did here, that the 14.6 percent decrease in rates falls within a range of reasonableness. He cannot simply order a percentage decrease without more. It is not the court‘s role to “supply a reasoned basis for the [commissioner‘s] action that the [commissioner himself] has not given.” Attorney Gen. v. Commissioner of Ins., 442 Mass. 793, 807 (2004), quoting Costello v. Department of Pub. Utils., 391 Mass. 527, 536 (1984). That the commissioner must do himself.

d. Class code 9033.

We turn, finally, and briefly, to WCRIB‘s argument that the commissioner exceeded his authority when, in the 2024 decision, he ordered WCRIB to change its methodology for calculating rates in a particular business classification, class code 9033.¹⁵ Class code 9033 involves

primarily public housing authority employees, and a single SIG, NAHRO, conducts approximately eighty-five percent of the business within the class code. In determining rates and rate modifications for the class code, Massachusetts data -- from WCRIB -- are “complemented” for credibility with data from an additional source.¹⁶ Typically, WCRIB and other rate makers across the country use “countrywide data,” which consists of data from dozens of States, as a complement to credibility.

In 2023, however, the commissioner ordered WCRIB to address the use of NAHRO‘s SIG data, rather than countrywide data, for determining rates in class code 9033. The commissioner thereafter determined in the 2024 decision that using the countrywide data as a complement to credibility was inappropriate and ordered that all future rate filings should be consistent with this determination. Relying on the SIG data rather than the countrywide data will, it appears, effectively lead to an increase in rates within the class code. In WCRIB‘s view, the commissioner is, in essence, requiring WCRIB‘s members, and any SIG who competes with NAHRO, to charge higher rates so that NAHRO can better compete.

In contesting the order to use SIG data rather than countrywide data, WCRIB argues, among other things, that the commissioner‘s decision broke with a prior order issued in 2000 without reasoned explanation. Deviations from a prior position sometimes reasonably will occur, as we noted above in addressing the use of the five-year data for determining certain losses. It does not necessarily follow that such deviations are problematic or demonstrate an inconsistency, so long as they are reasonably explained. See MCI WorldCom Communications, Inc., 442 Mass. at 116.

That the commissioner might have legitimate concerns with the use of the countrywide data as the complement to credibility in this context does not, however, establish the reliability of, or bear on the propriety of using, NAHRO‘s SIG data for that purpose. As WCRIB explains, SIGs are not regulated like insurers. Their data are not subject to the same cost control measures, and to the extent that NAHRO‘s advocating for the use of its data is related to certain of its financial difficulties, WCRIB asserts that NAHRO‘s financial difficulties and higher expenses may be the product of “inadequate cost controls, deficient premium collection, poor accounting, or organizational problems.”

WCRIB also argues that because the Division of Insurance‘s examinations of NAHRO‘s finances are kept confidential, WRCIB cannot fairly evaluate and contest these issues. In his 2024 decision, the commissioner questioned WCRIB‘s claims regarding its lack of access to certain of NAHRO‘s data, but he provides no response to WCRIB‘s current arguments on this point on appeal. WCRIB, in short, has raised legitimate questions regarding the use of SIG costs to establish the rates of regulated insureds for which the commissioner has not provided an adequate response or explanation.

Furthermore, the commissioner appears to have singled out the 9033 class code for a different methodology and, again, has not provided a reasoned explanation for so doing. For all these reasons, as with the commissioner‘s declaration of the 14.6 percent decrease in the 2024 decision, his decision regarding class code 9033 requires further explanation.

3. Conclusion.

For the reasons set forth herein, judgment shall enter in the county court affirming so much of the 2024 decision as determined that the then-existing workers’ compensation insurance rates were excessive. Because the commissioner must provide a specific, reasoned explanation for the order in the 2024 decision decreasing rates by 14.6 percent, the single justice is directed to remand the matter to the commissioner for further consideration. In addition to providing a reasoned explanation for the 14.6 percent figure in the 2024 decision, the commissioner must also address the issues discussed in this opinion related to class code 9033 in connection with both the 2024 and 2025 decisions.¹⁷

So ordered.

Notes

This provision, and the role of WCRIB as the “advisory organization,” is relevant to the commissioner‘s treatment of rate calculation issues, and WCRIB‘s arguments in opposition thereto, in a particular category of insurance, class code 9033, discussed in part 2.d, infra.

In April 2023, the commissioner had approved a stipulation on rates that would become effective on July 1, 2023. The commissioner also directed WCRIB to submit a new filing in December 2023 for the following year, i.e., for rates that would be effective on July 1, 2024, essentially on the basis of the COVID-19 pandemic‘s effect on the economy and the structure and operations of the workforce, and the fact that relevant data for certain years, particularly 2020, were anomalous.

Pursuant to G. L. c. 26, § 8E, the SRB, which describes itself as the “technical arm” of the Division of Insurance, advises the commissioner on various matters, including determining appropriate workers’ compensation insurance rates.

The Attorney General participated in the administrative process as a statutory intervener pursuant to 211 Code Mass. Regs. § 110.02 (1996). Both the Attorney General, as a statutory intervener, and the SRB submitted advisory filings pursuant to 211 Code Mass. Reg. § 110.05 (1996).

WCRIB‘s proposed rate increase was an increase from the rates that had gone into effect on July 1, 2024, i.e., the rates that reflected the 14.6 percent decrease ordered by the commissioner in his 2024 decision.

The petitions filed in the county court named the commissioner, the Attorney General, and the SRB as respondents. After the cases were entered in the full court, the parties stipulated to the voluntary dismissal of the Attorney General and the SRB.

The scope of the commissioner‘s authority under G. L. c. 152, § 53A, differs from that of his authority under some other statutes. See, e.g., . Although he has the authority under certain statutes to set rates, he has no such authority under G. L. c. 152, § 53A, except to determine a specific reduction, as explained infra. See (noting difference between G. L. c. 175, § 113B, pursuant to which commissioner may set rates related to car insurance, and G. L. c. 152, § 52, repealed by St. 1987, c. 691, § 13 [predecessor to § 53A], where he did not have that authority).

WCRIB raised the ratemaking methodology issue in connection with both the commissioner‘s 2024 decision and his 2025 decision. We address the issue in the context of the 2024 decision, but the analysis holds true for the 2025 decision as well.

As a part of its process for developing rates, WCRIB analyzes historical patterns of loss data in four categories: indemnity paid losses; indemnity paid losses plus case reserves; medical paid losses; and medical paid losses plus case reserves. The only category that is at issue here is the indemnity paid loss category.

The calculation of data also involves so-called “link values” or “link ratios.” WCRIB annually calculates link ratios between claims paid on policies that are in effect during a calendar year but were written in a prior year. Here, the commissioner noted that “the 2021 and 2022 link values are the highest of any within the past five years and the highest and fifth highest shown over the past twenty years.”

WCRIB does not propose using two years of data for all four categories. See note 9, supra. Rather, for paid losses plus case reserves, WCRIB agrees that a five-year data period is appropriate. “Paid losses plus case reserves” includes amounts set aside by the insurer to address case loss development, which is the anticipated future costs of claims, usually involving serious or particularly expensive injuries. As explained by the commissioner, WCRIB asserts “that the difference arises from the premise that paid [losses] plus case [loss] development, because of the presence or absence of large losses in a single year, is more volatile and that averaging the past five years of data mitigates that volatility.”

The commissioner also based his disapproval of WCRIB‘s rate filing on WCRIB‘s underwriting profit provision methodology. Although WCRIB contested the commissioner‘s decision on this issue in its initial petition for review in the county court, it has not raised the issue in this court, and thus it is waived. This uncontested issue provides further support for the commissioner‘s disapproval of the rates. See (addressing commissioner‘s authority and discretion in considering underwriting profit provisions).

As noted above, the commissioner found that the use of two rather than five years of data for estimating indemnity paid losses was unreasonable and that the methodology for calculating the underwriting profit provision provided an excessive rate.

The same rationale applies with equal force to both the 2024 decision and the 2025 decision.

The commissioner argues that this issue is not ripe for review because his order, in the 2024 decision, that WCRIB

The use of additional data as a “complement to credibility” is necessary when there is not enough data from WCRIB members to be statistically reliable.

The commissioner ordered the change in the methodology in the 2024 decision, and WCRIB implemented the change in its rate proposal submitted in November 2024 for rates that became effective on July 1, 2025. Although WCRIB did not raise the 9033 class code issues in its appeal from the 2025 decision, it had already raised the issues in its appeal from the 2024 decision, which was pending at the time WCRIB had to submit its next rate filing. The issues therefore apply to both the 2024 and 2025 decisions.

Read the detailed case summary