ORDERS:
FINAL ORDER AND DECISION
STATEMENT OF CASE
On November 30, 1998, the National Council on Compensation Insurance, Inc., (NCCI) filed
two applications with the Director of the South Carolina Department of Insurance (Department)
relating to workers' compensation insurance loss costs and rates to be utilized by insurers transacting
business in South Carolina. One application, the loss costs filing, requested approval of an overall
average increase of 10.0% for workers' compensation insurance voluntary market loss costs
applicable to "F" classifications and an overall average decrease of 10.9% for all other workers'
compensation insurance voluntary market loss costs. The other application, the assigned risk filing,
requested approval of an overall average increase of 14.9% for workers' compensation assigned risk
rates applicable to "F" classifications, and an overall average decrease of 6.8% for all other workers'
compensation insurance assigned risk rates. These changes were measured against the current
approved loss costs and rates. The Petitioner sought an April 1, 1999 effective date in both
filings. (1)
The Department requested a contested case hearing on the loss costs filing, which was
assigned to me and designated as Docket No. 98-ALJ-09-0682-CC (Case 682). Afterwards, the
Consumer Advocate for the State of South Carolina (Consumer Advocate) moved to intervene in
Case 682. The Department also requested a contested case hearing on the assigned risk filing, which
was assigned to the Honorable Alison Renee Lee and designated as Docket No. 99-ALJ-09-0010-CC
(Case 010). The Department named the Consumer Advocate as a Proposed Intervenor in Case 010.
By motion dated January 28, 1999, NCCI moved for an Order, pursuant to Rule 19(D) of the
Rules of Procedure of the Administrative Law Judge Division, consolidating Case 682 and Case 010.
Finding that both cases involved the same parties, substantially similar evidence, and substantially
similar issues, these cases were consolidated for hearing purposes. Additionally, the Court took
notice of the fact that the parties may not have received thirty days notice of the consolidated
hearing, as required by the Administrative Procedures Act. All parties waived their right to such
notice. Also, with the consent of the other parties, the Court granted the Consumer Advocate's
Motion for Leave to Intervene in Case 682. The consolidated hearing on this matter was held before
the Administrative Law Judge Division on February 24, 1999.
FINDINGS
Having carefully considered all testimony, evidence and arguments presented at the hearing
in this matter, I make the following Findings of Fact by a preponderance of evidence as to the
requested revision in workers' compensation insurance voluntary loss costs and assigned risk rates:
General Findings
1. The Department published notice of hearing referencing Petitioner's filing to change
workers' compensation insurance rates in newspapers of general, statewide circulation at least thirty
days in advance of this hearing. Though the published notices of this hearing only referenced Case
682, the notice did not identify the hearing as addressing or being limited to only the loss costs filing
or the assigned risk filing. I therefore find that proper notice of a hearing concerning the revision
in the voluntary loss costs and assigned risk rates was given to the Petitioner, Respondents, and the
Public.
2. No member of the public appeared to present a protest or other position with respect
to either filing, and no member of the public contacted the Court about this hearing.
3. The Petitioner sought an overall average increase in the voluntary market loss costs
of 10.0%, applicable to "F" classifications policies written on or after April 1, 1999, and an overall
average decrease in the loss costs of 10.9%, applicable to all other voluntary market policies written
on or after April 1, 1999.
4. The Petitioner sought an overall average increase of 14.9% in assigned risk rates,
applicable to "F" classifications for policies written on or after April 1, 1999, and an overall average
decrease of 6.8% for rates applicable to all other assigned risk policies written on or after April 1,
1999.
5. The Department concluded that the loss costs filing complies with all statutory
requirements and will not produce loss costs that are excessive, inadequate, or unfairly
discriminatory. Furthermore, the Department concluded the assigned risk filing complies with all
statutory requirements; that it will not produce rates that are excessive, inadequate, or unfairly
discriminatory; and that it will produce an assigned risk plan that is self-sustaining. Therefore, the
Department did not oppose either the loss costs filing or the assigned risk filing.
6. The Consumer Advocate agreed with the Petitioner and the Department that the same
calculations and methodologies were and should be used in both filings. Furthermore, the
Consumer Advocate presented no testimony regarding loss costs or rates applicable to "F"
classifications. However, the Consumer Advocate opposed the loss development methodology and
calculation used by NCCI and the trend factors selected by NCCI in these filings.
Loss Development
7. Loss development is the actuarial process used to estimate the ultimate amount of
losses in a particular time period. This requires the estimation of the loss ratio (i.e., the ratio of paid
losses to earned premiums) for that time period. Loss development is particularly appropriate in
workers' compensation insurance because indemnity payments may extend many years into the
future and, in some circumstances, the conditions giving rise to losses may take many years to even
manifest themselves.
8. NCCI examined the results of loss development projections based on: (a) paid losses;
(b) paid losses plus case reserves (i.e., paid losses plus reserves on reported claims); and (c) total
incurred losses (i.e., paid losses plus case reserves plus losses that have been incurred but not
reported). NCCI discarded the last methodology because of a significant possibility that it produced
distorted results in this instance. The rate indications based on the other two methodologies differed
by less than one percentage point each, giving credence to both methodologies. Consistent with prior
workers' compensation filings in South Carolina and to promote stability in the market, NCCI
utilized paid losses to calculate the loss development factors in these two filings.
9. Having selected its loss development methodology, NCCI reviewed the number of
years of loss development data that could be used to calculate the loss development factors.
Consistent with at least the last ten years of NCCI's filings in South Carolina, NCCI utilized a two-year average to balance stability and responsiveness.
10. The Consumer Advocate objected to NCCI's "loss development calculation
methodologies" based on the possibility that NCCI's use of two years of data may produce distorted
results that could have been avoided if time periods longer than two years had been used. The
Consumer Advocate utilized NCCI's methodology and calculation of the loss development factors
over a five-year period, and then applied these factors in his trend calculation. However, the
Consumer Advocate did not establish that such distortions actually occurred or that the use of five
years of data, as the Consumer Advocate proposed, would produce results that were actuarially more
credible than those produced by NCCI's methodology.
11. The Department concurred with the loss development methodology and the loss
development calculation utilized by NCCI.
12. I find that the loss development projections utilized by NCCI is appropriate.
Trend
13. The trend component of an insurance rate formula is used actuarially to estimate
changes in the historical loss ratios from the time period of the historical data underlying the filing
to the midpoint of the time period that the proposed rates will be in effect. Whereas loss
development estimates the ultimate loss ratio, trend estimates the change in that loss ratio from the
historical period to a future period. Trend accounts for the impact of inflation on both losses and
premiums over this time period. The trend factors are calculated separately for indemnity and
medical loss ratios.
14. NCCI used the same approach to calculate the medical and the indemnity trend factors
but calculated them separately. NCCI reviewed the current approved trends and also calculated trend
factors under each of the following methodologies: eight-point policy year exponential trend model;
seventeen-point policy year exponential trend model; trends resulting from frequency/severity
analysis; and country-wide trends. NCCI synthesized the information developed by that process as
a result of those calculations and selected an annual medical trend of 1.010 and an annual indemnity
trend of 0.980 as most appropriate trends. These trends did not equal the exact result of any of the
foregoing calculations but, instead, represented NCCI's best judgment of the most appropriate trends
in light of the results of those calculations.
15. The Consumer Advocate opposed the trend factors selected by NCCI on the ground
that NCCI's trend was not the result of the mechanical application of an established trend
methodology. The Consumer Advocate calculated trend factors by averaging the eight-point policy
year exponential trend factors and the trend factors resulting from the frequency/severity analysis.
He argued that an application of equal weighting of the eight-point policy year exponential trend
factors and the frequency/severity analysis trend factors would produce loss costs and rates that are
adequate and not excessive.
16. The Department concurred with NCCI's methodology and calculation of trend and
determined that the use of NCCI's trends would produce loss costs and assigned risk rates that are
not inadequate, excessive, or unfairly discriminatory. However, the Department also agreed that the
Consumer Advocate's calculation was reasonable but did not accede that his calculation would
produce loss costs and assigned risk rates that are not excessive, inadequate, or unfairly
discriminatory.
17. The conflict in the testimony regarding the trend results from a disagreement among
the actuaries as to whether the improvements in the market have "bottomed out." This disagreement
is exacerbated by the fact that the latest data underlying these filings is from 1996 and the actuaries
are using that data to predict results in 1999-2000. Furthermore, the trends would generate
substantial decreases in South Carolina's workers' compensation insurance premiums. In the interest
of maintaining a stable workers' compensation insurance market, I find that these substantial
changes in loss costs and assigned risk rates should be implemented cautiously. Therefore, I find
that the most appropriate trend factors in this case are neither the ones presented by NCCI nor those
presented by the Consumer Advocate. Instead, the appropriate trend factors should represent the
arithmetic average of the results of the eight-point policy year exponential trend model, the
seventeen-point policy year exponential trend model, the trend factors resulting from the
frequency/severity analysis, and NCCI's country-wide trends.
When the trend factors used in the NCCI filings are replaced with the trend factors calculated
as described above, the resulting average decrease in overall voluntary market loss costs (exclusive
of "F" classifications) is 11.7% and the resulting average decrease in overall assigned risk rates
(exclusive of "F" classifications) is 7.6%. The use of these trends will produce voluntary market loss
costs and assigned risk rates that are not excessive.
CONCLUSIONS
1. The South Carolina Administrative Law Judge Division is empowered to hear this
case pursuant to S.C. Code Ann. § 38-73-1320 (Supp. 1997) and Chapter 23 of Title I of the South
Carolina Code of Laws, as amended.
2. Generally, a request for insurance loss cost and assigned risk rates revisions are
governed by S.C. Code Ann. §§ 38-73-10 et seq. (Supp. 1998).
3. The Administrative Procedures Act requires that a notice of the hearing be published
at least thirty days in advance. S. C. Code Ann. 1-23-320(a) (Supp. 1998). The notice of this hearing
was published more than thirty days prior to this hearing. The notice set forth that the contested
case hearing would determine whether the Petitioner's filing to change workers' compensation
insurance rates was lawful. In light of those facts and having consolidated Case 682 with Case 010
for hearing purposes, I conclude that adequate and sufficient public notice for this hearing was
provided.
4. S. C. Code Ann. §§38-73-10(a)(1) and 38-73-430(4) (Supp. 1998) require that
insurance rates not be excessive, inadequate, or unfairly discriminatory. Furthermore, the rates
shall include a reasonable margin for underwriting profit. S.C. Code Ann. § 38-73-430(1) (Supp.
1998). The rates for the assigned risk market must also be self-sustaining. S. C. Code Ann. §38-73-540(c) (Supp. 1998). A public hearing is required as a condition of approval of an increase in
workers' compensation insurance rates. S. C. Code Ann. §38-73-910 (Supp. 1998). Also, the
findings of fact in a contested case must be based upon the evidence and matters officially noted
during the course of the hearing. S.C. Code Ann. §1-23-320(I) (Supp. 1998).
5. By Order dated March 1, 1990, the Chief Insurance Commissioner of the Department,
acting with authority granted by S.C. Code Ann. § 38-73-1430 (Supp. 1998), directed that workers'
compensation rate filings be treated differently for the voluntary and the assigned risk markets.
Assigned risk market filings should propose complete rates, and voluntary market filings should
propose only the projected losses and the projected loss adjustment expenses (i.e., the loss costs).
NCCI's loss cost filing and the assigned risk filing comply with this Order.
6. NCCI is the nonpartisan rating bureau required by statute for the workers'
compensation insurance. S. C. Code Ann. §38-73-510 (Supp. 1998). The subject applications are
made by NCCI on behalf of all licensed workers' compensation insurance carriers doing business in
the State of South Carolina. S.C. Code Ann. § 38-73-1210 (Supp. 1998).
7. I find that an overall average decrease of 11.7% in voluntary market loss costs
(exclusive of "F" classifications) would produce loss costs that are not inadequate, excessive, or
unfairly discriminatory. Additionally, an overall average decrease of 7.6% in assigned risk rates
(exclusive of "F" classifications) would produce rates that are not inadequate, excessive, or unfairly
discriminatory, and will produce an assigned risk market that is self-sustaining. Furthermore,
NCCI's applications for a 10.0% overall average increase in workers' compensation insurance
voluntary market loss costs for "F" classifications and a 14.9% overall average increase in workers'
compensation insurance assigned risk rates for "F" classifications would produce loss costs and rates
which would not be excessive, inadequate or unfairly discriminatory, and would be in compliance
with S. C. Code Ann. §38-73-430 (Supp. 1998).
ORDER
IT IS THEREFORE ORDERED that the current overall workers' compensation insurance
voluntary market loss cost level, on average, (other than for "F" classifications) be reduced 11.7%
and the overall assigned risk rate level, on average, (other than for "F" classifications) be reduced
7.6%.
IT IS FURTHER ORDERED that the proposed changes in the workers' compensation
insurance voluntary market loss costs and assigned risk rates for "F" classifications be as sought in
the loss costs filing and in the assigned risk filing, respectively.
IT IS FURTHER ORDERED that all changes ordered above shall be effective for new and
renewal policies May 1, 1999.
IT IS SO ORDERED
Ralph King Anderson, III
Administrative Law Judge
March 30, 1999
Columbia, South Carolina
1. After the hearing, the Petitioner and the Department conducted discussions as to the amount of time that
would be reasonably required to give carriers, agents, and other relevant parties appropriate notice of the changes, and
both agreed to amend the filings to provide that these changes be effective May 1, 1999, rather than April 1, 1999. |