Item E-1409—Enhancement to NCCI’s Experience Rating Plan Methodology
PURPOSE
This item revises certain underlying components provided in the methodology used in NCCI’s Experience Rating Plan (Plan). The underlying Plan components impacted are:
- Primary/excess loss split point (split point)
- State per claim accident limitation (SAL) and United States Longshore and Harbor Workers’ Compensation (USL&HW) per claim accident limitation
- G value
- Credibility parameters underlying calculation of the weight (W) and ballast (B) values
- Discount ratios (D-ratios)
The formulas and general structure of the Plan remain unchanged.
BACKGROUND
The experience rating modification is intended to predict an employer’s future loss experience using its historical loss experience.1 For example, an experience rating modification of 1.20 is a prediction that the employer’s future loss experience will be 20% worse than the average employer in the same classification. The experience rating modification is intended to reflect the employer’s past loss experience only to the extent that it is considered predictive of future loss experience. The weight given to actual losses in the experience rating modification calculation is called credibility; as the size of the employer increases, credibility also increases.
NCCI periodically evaluates Plan methodology and performance. During the latest review, NCCI identified some opportunities to improve Plan performance, with revisions that will result in:
- A more accurate and predictive experience rating modification
- Experience rating modifications that reflect a more equitable determination of primary and excess losses across states with varying cost levels
- More comparable Plan performance in states with claim costs that vary significantly from the countrywide average
- Experience rating modifications less sensitive to large outlier claims without sacrificing predictive accuracy
- More consistent calculation of each employer’s expected claim count, which is expected to result in more appropriate credibility being assigned to each employer’s loss experience
- Recalibrated credibility parameters underlying the weight and ballast values to increase equity across employers
- The elimination of complex calculations where no value is added
PROPOSAL
This item proposes the following changes to the Plan:
1. Implement state-specific split points reflecting an average D-ratio of approximately 40% by state.
The primary/excess loss split point divides the losses from each historical claim into two layers: primary losses (those beneath the split point) and excess losses (those above the split point). For example, if the split point is $15,000, a claim totaling $50,000 would contribute $15,000 to the primary layer and $35,000 to the excess layer. Primary losses receive a greater weight than excess losses in the experience rating modification formula. Because of this, primary losses have a greater impact on the experience rating modification.
The D-ratio is the expected percentage of losses that fall below the split point. Currently, the split point is set at a common level across states where NCCI provides ratemaking services. Because the average D-ratio in a state depends not only on the split point, but also on that state’s average claim costs, having a common split point results in an average D-ratio that varies widely across states.
The proposed methodology level-sets the average D-ratio across states to approximately 40% by introducing a state-specific split point, which allows the split point to better reflect each state’s average claim costs. As a result, experience rating modifications will reflect a more equitable determination of primary and excess losses across states with varying cost levels.
For example, instead of the current split point value of $18,500 applying to all states, under the proposed Plan a state with higher-than-average claim severity may have a split point value of $25,000, while a state with lower-than-average claim severity may have a split point value of $15,000. The use of state-specific split point values that reflect individual state cost differences is intended to better align across states the weight given to actual employer loss experience in the experience rating modification calculation. In turn, this is expected to produce improved and more comparable Plan performance in states with claim costs that vary significantly from the countrywide average.
Average claim costs vary significantly across NCCI states. The experience rating modification applies to premium that is based on manual loss costs/rates. Because loss costs/rates vary by state in accordance with cost differences, it is paramount that the experience rating modification similarly reflects state cost differences to achieve comparable performance across states. Tailoring the split point to reflect these cost differences is a big step towards aligning performance across states and results in a more accurate and predictive experience rating modification compared to applying a countrywide split point uniformly across states.
To keep up with changes in claim costs and preserve alignment with other experience rating parameters, it is anticipated that the split point value will be indexed concurrent with each state’s annual loss cost/rate filing based on an estimate of annual severity changes between the average loss date for experience rating modifications in the initial implementation year and the effective year.
2. Revise the calculation of accident limitations to reflect the 95th percentile of lost-time claims.
The state per claim accident limitation (SAL) is used to curtail the impact of large claims on the experience rating modification, because large outlier claims are generally not expected to be predictive of future loss experience. The use of a state-level 95th percentile results in an SAL that is expected to impact the largest 5% of lost-time claims. A similar change is proposed for the USL&HW per claim accident limitation to reflect the 95th percentile of lost-time claims reported under USL&HW Act benefits for F-classifications. Note that this change renders the state reference point (SRP) obsolete (the SAL was previously calculated as 10% of SRP).
The new definition of the SAL and USL&HW per claim accident limitation results in lower limits in every state, making experience rating modifications less sensitive to large outlier claims without sacrificing predictive accuracy.
The state per claim accident limitation and USL&HW per claim accident limitation will be calculated under the new methodology in each state’s loss cost/rate filing.
3. Revise the calculation of G to reflect accident limitations and the reduction of medical-only losses, where applicable.
G represents state average claim severity (in thousands of dollars). While G currently reflects the unlimited average severity in a state, this item proposes to update the methodology under which G is calculated to reflect accident limitations and the 70% reduction of medical-only losses (per the experience rating adjustment [ERA]), where applicable.
In the calculation of an employer’s expected claim count, the employer’s expected losses are divided by the G value (average claim severity). Because expected losses already reflect ERA and accident limitations, this change in how G is calculated makes for a more consistent calculation of each employer’s expected claim count. In turn, this is expected to result in more appropriate credibility being assigned to each employer’s loss experience.
The G value will be calculated under the new methodology in the state’s loss cost/rate filing.
4. Update the credibility parameters underlying the weight and ballast values.
The proposed credibility parameters underlying the weight and ballast values have been recalibrated to increase equity across employers. All other things being equal, the new credibility parameters would increase credibility for larger risks and decrease credibility for smaller risks. However, because of the interaction of the split point change with the credibility change, the combined impact will vary across states and across employers of difference sizes.
5. Simplify the D-ratio calculation by removing the classification-level weighting adjustment.
NCCI currently produces D-ratios that vary by classification code. D-ratios are first calculated by hazard group (HG) and then adjusted to reflect a classification’s share of indemnity and medical pure premium. While the classification-level adjustment adds complexity to the calculation, NCCI’s research found that this adjustment does not add value. As such, this item proposes to eliminate the classification-level adjustment, meaning that the D-ratio will vary only by HG and not by classification code.
IMPACT
No statewide premium impact is anticipated from the changes proposed in this item. The overall average experience rating modification in each state is not expected to be impacted by these changes.
Impacts to experience rating modifications at the individual employer level will vary, and may be offset by changes in loss experience and routine updates to rating values. Experience rating modifications are expected to change by less than +/- 5% for most employers.
Overall, the proposed changes to the experience rating modification calculation are expected to produce Plan performance that is both improved and more comparable across states.
IMPLEMENTATION
This item will become effective for experience rating modifications with rating effective dates on and after each state’s anticipated loss cost/rate filing effective on and after November 1, 2023.
For example, this item will become effective for experience rating modifications with rating effective dates on and after January 1, 2024, for states with loss cost/rate filings that have an anticipated January 1, 2024 effective date. Similarly, this item will become effective for experience rating modifications with rating effective dates on and after July 1, 2024, for states with loss cost/rate filings that have an anticipated July 1, 2024 effective date.
[top]
