Tutorials › CAS Exam 8 - Advanced Ratemaking › Incentive Effects of Experience Rating Design

CAS Exam 8 · Experience Rating Plan Design · Tutorial 510 of 989

Incentive Effects of Experience Rating Design

Learn to measure the financial incentives created by experience rating and identify how plan features can encourage prevention or distort reporting decisions.

Intermediate 12 min read

What You'll Learn

  • Separate the effect of a loss reduction on retained losses from its effect on future premiums.
  • Calculate the present value of experience-rated premium changes with renewal timing and uncertainty.
  • Compare reporting a claim with self-funding under stated contractual assumptions.
  • Analyze why caps and thresholds can make incentives nonlinear.
  • Identify design features that may weaken prevention incentives or distort claim reporting.
  • Distinguish a financial incentive calculation from a recommendation to violate policy or reporting requirements.

Why Experience Rating Creates Incentives

An experience-rated premium can respond to an insured’s past losses. That response does more than allocate premium: it can change the insured’s financial payoff from preventing a loss, reporting a claim, or retaining a loss when permitted. The size and timing of the premium response matter. A premium adjustment that is small, delayed, uncertain, or capped may create a different incentive from an immediate, uncapped adjustment.

The analysis is a comparison of hypothetical financial outcomes, not a claim that every insured can control every loss or choose whether to report it. Policy terms, applicable law, and reporting duties govern what must be reported. Any comparison involving self-funding below assumes that choice is permitted and does not remove a contractual reporting obligation.

Definition: An experience-rating incentive is the change in an insured’s expected financial cost caused by a change in losses or reported claims when the experience-rated premium responds to that experience. The incentive includes the insured’s change in retained loss cost and the present value of the resulting premium change.

A Basic Financial Framework

Let \(L\) denote the loss amount entering the experience-rating calculation, and let \(P(L)\) denote the corresponding premium, holding other plan inputs fixed. If an action reduces covered loss by \(\Delta L>0\), assume the insured retains a proportion \(r\) of each loss dollar. The retained-loss saving is \(r\Delta L\). If the premium reduction caused by the action is \(\Delta P=P(L)-P(L-\Delta L)\), and it is payable \(t\) years later if renewal occurs, its present value at effective annual interest rate \(i\) is \(q\Delta P/(1+i)^t\), where \(q\) is the probability of renewal.

This comparison assumes the loss reduction does not change exposure, coverage, or other rating inputs; the retained proportion is constant over the losses considered; and the premium change occurs at the stated time. If there is more than one future premium change, value each change at its own payment date and include its probability. If the plan has a cap, floor, threshold, or other nonlinear feature, calculate the premium at each loss level directly rather than assuming one marginal response applies throughout.

$$ \text{Present value of benefit} = r\Delta L+\frac{q\Delta P}{(1+i)^t} $$

The retained-loss term is measured at the loss date in this simplified framework. The premium term is discounted because it is paid later. If the loss payment itself occurs later, its timing must also be reflected. A premium increase from reporting can be analyzed in the same way, with the sign reversed: it is a cost to the insured, not a benefit.

Why the Formula Holds

Suppose the action reduces loss by \(\Delta L\). Without the action, the insured pays \(rL\) in retained loss; with the action, the insured pays \(r(L-\Delta L)\). The difference is

$$ rL-r(L-\Delta L)=r\Delta L. $$

The action also lowers the future premium by \(\Delta P\) if renewal occurs. The expected future saving is \(q\Delta P\), and its present value is \(q\Delta P/(1+i)^t\). Adding the two distinct savings gives the stated formula. The terms should not be combined before their timing and meaning are clear: one is a retained-loss saving, while the other is a discounted premium saving.

When \(P(L)\) is differentiable near the current loss level, a small loss reduction has approximately \(\Delta P\approx P'(L)\Delta L\). Thus the local benefit per dollar of loss reduction is approximately \(r+qP'(L)/(1+i)^t\). This local approximation is not valid across a cap, floor, or threshold if the marginal premium response changes there.

Worked Example: Valuing a Prevention Action

A hypothetical insured retains 20% of losses. A prevention action is expected to reduce loss by $2,500. Near the current loss level, the plan reduces the insured’s premium by $0.10 for each dollar of loss reduction. Renewal is certain, the premium change applies one year later, and the effective annual interest rate is 4%. Find the present value of the benefit before subtracting the action’s cost.

Method. Add the retained-loss saving to the discounted premium saving. The stated local premium response is assumed to apply across the $2,500 reduction, with no cap or threshold intervening.

Retained-loss saving. The insured retains 20% of the loss reduction:

$$ 0.20(2{,}500)=\$500. $$

Premium saving. The loss reduction lowers the future premium by:

$$ 0.10(2{,}500)=\$250. $$

Discount that one-year premium saving at 4%:

$$ \frac{\$250}{1.04}=\$240.3846\ldots $$

The total present value before the action’s cost is \(500+240.3846\ldots=\$740.3846\ldots\), or $740.38 to the nearest cent. This is the break-even amount for the action’s cost under the stated assumptions. The common mistake is to add the $250 future premium saving without discounting it; that would overstate the present value.

Claim Reporting and Retention Choices

A reported claim may produce an insurer payment while also increasing a future experience-rated premium. A useful simplified comparison separates the insured’s deductible or other direct retained amount from the future premium effect. It also makes explicit when the insurer payment and premium change occur. This comparison does not determine whether a claim must be reported; that is governed by the policy and applicable requirements.

Worked Example: Reporting Versus Self-Funding

A hypothetical covered loss is $1,200, and the policy deductible is $300. If reported, the insurer pays the covered amount above the deductible at the time of loss. Reporting causes a $250 premium increase at a certain renewal one year later. The effective annual interest rate is 5%. Compare the present-value cost of reporting with the cost of self-funding, assuming self-funding is permitted and does not violate a reporting duty.

Method. Calculate the insurer payment and the insured’s direct deductible, then discount the future premium increase. For this example, the full $1,200 is the insured’s direct cost under self-funding, with no separate claim expense.

Insurer payment if reported.

$$ 1{,}200-300=\$900. $$

Present value of the premium increase.

$$ \frac{\$250}{1.05}=\$238.0952\ldots $$

The present-value cost of reporting is the $300 deductible plus the discounted premium increase:

$$ 300+238.0952\ldots=\$538.0952\ldots\approx \$538.10. $$

Self-funding costs $1,200 under the stated assumption. Reporting therefore has the lower present-value cost by \(1{,}200-538.0952\ldots=\$661.9048\ldots\), or $661.90. Equivalently, the insurer’s $900 payment less the present value of the premium increase is $661.90. The common trap is to compare the insurer payment with the premium increase but forget that the insured also pays the deductible when reporting.

Renewal Uncertainty and Nonlinear Plan Features

A premium response matters only when it is actually paid. If a renewal is uncertain, a simplified expected-present-value calculation multiplies the future premium change by the probability of renewal, provided that probability is appropriate for the scenario. A prevention action can still be worthwhile even when renewal is uncertain because it may reduce retained losses immediately.

Worked Example: Prevention with Renewal Uncertainty

A hypothetical insured retains 25% of losses. A project costing $500 is expected to reduce loss by $1,600. It also reduces a one-time renewal surcharge by $180. Renewal is 75% likely and occurs in one year; the effective annual interest rate is 5%. Find the expected net benefit after project cost.

Method. Add the retained-loss saving to the expected present value of the surcharge reduction, then subtract the project cost.

The retained-loss saving is:

$$ 0.25(1{,}600)=\$400. $$

The expected present value of the surcharge reduction is:

$$ \frac{0.75(\$180)}{1.05} = \frac{\$135}{1.05} = \$128.5714\ldots $$

Therefore the expected net benefit after the project cost is:

$$ 400+128.5714\ldots-500 = \$28.5714\ldots $$

The project has a positive expected net benefit of $28.57 under the stated assumptions. The common mistake is to count the full $180 surcharge reduction as certain; the renewal probability must be applied before discounting.

Worked Example: A Cap Changes the Marginal Incentive

A hypothetical plan adds a surcharge equal to 10% of losses above a $10,000 benchmark, subject to a $1,000 cap. An insured’s experience loss falls from $22,000 to $18,000. By how much does the surcharge fall?

Method. Apply the excess calculation and cap separately at each loss level, then take the difference.

At $22,000, the uncapped surcharge is:

$$ 0.10(22{,}000-10{,}000)=\$1{,}200. $$

The $1,000 cap applies, so the actual surcharge is $1,000. At $18,000, the uncapped surcharge is:

$$ 0.10(18{,}000-10{,}000)=\$800. $$

This is below the cap, so the actual surcharge is $800. The surcharge falls by \(1{,}000-800=\$200\). The common mistake is to apply the uncapped 10% rate to the entire $4,000 loss reduction and conclude that the surcharge falls by $400. That calculation ignores that the original surcharge was capped; the actual premium response is only $200.

Design Choices That Shape Incentives

Experience-rating design determines how strongly loss experience changes the premium and whether that response is predictable. The relevant question is not simply whether a plan responds to losses, but how the response behaves over the range of losses an insured could influence. Several design features can strengthen, weaken, or distort that response.

  • Premium responsiveness. A larger marginal premium response can increase the expected financial return to prevention. It can also make the premium more sensitive to random loss variation, so the plan’s responsiveness should be interpreted alongside its stability provisions.
  • Caps and floors. A cap can make additional loss reduction produce little or no premium saving once the premium adjustment reaches its limit. A floor can similarly limit the premium benefit of improved experience. Calculate the actual premium at both endpoints before inferring an incentive.
  • Thresholds. A threshold can create a discontinuity or a sharp change in the marginal premium response. A small loss change that crosses the threshold may have a different premium effect from an equally sized change that remains on one side.
  • Timing and experience windows. A delayed response has a smaller present value than an otherwise identical immediate response. If experience affects more than one future renewal, value each affected premium change at its payment date and reflect whether the policy is expected to renew.
  • Claim reporting effects. If reported losses affect experience but unreported losses do not enter the rating calculation, the premium incentive may differ between reporting and not reporting. Such an incentive does not change contractual duties, coverage terms, or legal requirements.

An observed premium response is not the same as the total social value of a prevention action. A plan calculation measures the insured’s financial incentives under specified assumptions; it does not by itself measure safety benefits, administrative costs, effects on other policyholders, or changes in the underlying frequency of loss. Keep the conclusion limited to the comparison actually calculated.

Common Exam Traps

  • Mixing amounts from different dates. Discount future premium changes to the comparison date, and apply the stated timing to claim payments when relevant.
  • Ignoring renewal probability. A future premium effect that occurs only upon renewal should not be treated as certain when renewal is uncertain.
  • Using a marginal rate beyond its range. A local premium response may not describe a change that crosses a threshold or encounters a cap or floor.
  • Confusing an insurer payment with the insured’s net benefit. Account for the deductible and any other stated retained costs before comparing reporting with self-funding.
  • Treating a financial incentive as permission. A favorable financial comparison does not override policy terms or reporting requirements.
Key takeaway: To analyze an experience-rating incentive, identify the loss change that enters the plan, calculate the insured’s retained-loss change, calculate the actual premium change under the plan’s boundaries, and align each amount to its timing and uncertainty. State the assumptions that make the comparison valid.

Check Your Understanding

For each hypothetical situation, use the stated assumptions and distinguish retained-loss effects from future premium effects.

  1. An insured retains 30% of losses. A prevention measure reduces loss by $4,000 and lowers a future premium by $0.06 per dollar of loss reduction. The premium change occurs in two years if renewal occurs; renewal is certain and the effective annual interest rate is 5%. Find the present value of the benefit before the measure’s cost. (Answer: $1,417.69)
  2. A covered loss is $1,800 with a $500 deductible. If reported, the insurer pays the covered amount at the time of loss and the premium increases by $360 at a certain renewal two years later. The effective annual interest rate is 4%. Assuming self-funding is permitted, how much lower is the present-value cost of reporting than self-funding the full loss? (Answer: $967.16)
  3. A project costing $900 reduces loss by $2,500 for an insured retaining 40% of losses. It also reduces a one-time surcharge by $220 at a renewal one year later. Renewal is 60% likely, and the effective annual interest rate is 10%. Find the expected net benefit after project cost. (Answer: $220.00)
  4. A plan surcharge is 15% of losses above an $8,000 benchmark, capped at $900. Experience loss falls from $16,000 to $12,000. Calculate the reduction in the actual surcharge, applying the cap separately at each loss amount. (Answer: $300.00)
Clinical Biostats