Assess Incentives Created By Per-Claim Caps

Why a Per-Claim Cap Changes the Incentive

In a retrospective rating plan, the premium can respond to losses after the policy period. When the plan limits the loss amount used for each claim, an additional dollar of loss does not always produce the same premium change. Below the cap, the plan responds to eligible loss increases through the loss conversion factor. Once a claim is at or above the cap, further loss on that claim does not increase the plan’s loss-based premium, assuming no other plan feature changes.

This claim-level response matters when interpreting incentives. A policyholder may have a premium reason to reduce a claim while it is below the cap, but little or no additional premium reason to reduce the part of a claim that remains above the cap. The cap changes the premium impact; it does not change the underlying claim amount or determine the policyholder’s total economic cost.

Definition: Let $X_j\geq 0$ be the eligible loss for claim $j$, $C>0$ the per-claim cap, $L>0$ the loss conversion factor (LCF), and $B$ the portion of premium held fixed in the comparison. With a per-claim cap applied before conversion, and no aggregate cap or other changing premium component, the adjusted premium is $$ P=B+L\sum_{j=1}^{n}\min(X_j,C). $$ The limited loss for claim $j$ is $\min(X_j,C)$. The formula assumes the number of claims, $B$, $L$, and $C$ remain fixed when comparing the loss scenarios.

The order in the formula is important: limit each claim’s eligible loss first, then multiply the total limited loss by $L$. A cap is applied to each claim separately, not once to the portfolio total. The $B$ term does not affect the premium change when it stays fixed.

Marginal Premium Impact at the Claim Level

For one claim with eligible loss $x$, its contribution to premium is $L\min(x,C)$, apart from the fixed portion $B$ and the contributions of other claims. The marginal change depends on the claim’s position relative to $C$.

\frac{d}{dx}\left[L\min(x,C)\right] = \begin{cases} L, & 0\leq x<C,\\ 0, & x>C. \end{cases}

At exactly $x=C$, the function has a kink, so there is no single two-sided derivative. A small increase from the cap has no premium impact; a small decrease from the cap reduces the capped loss and therefore reduces premium at rate $L$. This one-sided distinction is important when a question says a claim is “at” the cap.

Result: For an increase $\Delta>0$ in a claim initially at $x$, the premium increase is $$ \Delta P=L\left[\min(x+\Delta,C)-\min(x,C)\right]. $$ If $x<C$, only the part of the increase up to $C$ is reflected in premium. If $x\geq C$, the increase has no premium impact. The result concerns the loss-based premium component under the stated assumptions, not any separate expenses or plan provisions.

Why the Marginal Formula Holds

The capped amount for one claim is the smaller of its eligible loss and the cap. Below the cap, a small increase in loss increases that amount dollar for dollar. Above the cap, the smaller amount remains $C$, regardless of a further increase in loss. Multiplying that capped amount by the fixed LCF scales its premium contribution.

More formally, define $g(x)=L\min(x,C)$. For $x<C$, $g(x)=Lx$, so the change from $x$ to $x+h$, while both values remain below $C$, is $L(x+h)-Lx=Lh$. For $x>C$, $g(x)=LC$, so the same change is $LC-LC=0$. If an increase begins below and ends above the cap, split it at $C$: the portion from $x$ to $C$ changes premium by $L(C-x)$, and the portion above $C$ changes it by zero. Therefore the whole increase changes premium by $L(C-x)$, not by $L$ times the full loss increase.

Worked Example: An Increase That Crosses the Cap

A hypothetical plan has $B=\$92{,}000$, $L=1.10$, and a per-claim cap $C=\$18{,}000$. Its four eligible claims are $\$10{,}000$, $\$16{,}000$, $\$22{,}000$, and $\$18{,}000$. The second claim increases from $\$16{,}000$ to $\$21{,}000$. Find the premium before and after, and the premium increase. Assume all other plan terms remain fixed.

Method. Apply the cap separately to each claim, sum the limited amounts, and then multiply by $L$. This avoids treating the entire increase as eligible when it crosses the cap.

Before the change. The limited amounts are $\$10{,}000$, $\$16{,}000$, $\$18{,}000$, and $\$18{,}000$. Their sum is $\$10{,}000+\$16{,}000+\$18{,}000+\$18{,}000=\$62{,}000$. The premium is $\$92{,}000+1.10(\$62{,}000)=\$92{,}000+\$68{,}200=\$160{,}200$.

After the change. The second claim is now limited to $\min(\$21{,}000,\$18{,}000)=\$18{,}000$. The limited total is $\$10{,}000+\$18{,}000+\$18{,}000+\$18{,}000=\$64{,}000$. The premium is $\$92{,}000+1.10(\$64{,}000)=\$92{,}000+\$70{,}400=\$162{,}400$.

Change and check. The premium increases by $\$162{,}400-\$160{,}200=\$2{,}200$. Equivalently, the claim’s limited amount rises only from $\$16{,}000$ to $\$18{,}000$, a $\$2{,}000$ increase, and $1.10(\$2{,}000)=\$2{,}200$. The remaining $\$3{,}000$ of claim growth is above the cap and has no premium effect.

Common mistake avoided: Multiplying the full $\$5{,}000$ loss increase by 1.10 gives $\$5{,}500$, which is wrong because it ignores the cap. Only the $\$2{,}000$ increase in limited loss is converted.

Combining Claim-Level Changes

When several claims change, calculate the change in limited loss for each claim before adding the results. One claim’s increase may be partly capped, while another claim’s increase or decrease may be fully reflected. The net premium change is $L$ times the net change in the sum of limited losses. A claim already above the cap contributes no additional limited loss when it increases, but a decrease that brings it down below the cap can reduce the limited amount.

Worked Example: Offsetting Changes Across Four Claims

A hypothetical plan has $B=\$140{,}000$, $L=0.92$, and $C=\$25{,}000$. Four claims change as follows: Claim A rises from $\$21{,}000$ to $\$28{,}000$; Claim B rises from $\$31{,}000$ to $\$35{,}000$; Claim C falls from $\$24{,}000$ to $\$20{,}000$; and Claim D rises from $\$5{,}000$ to $\$9{,}000$. Calculate the premium change and identify each claim’s effect.

Method. Compare each claim’s limited amount before and after, then add the claim-level changes. This makes increases and decreases visible instead of applying the LCF to the raw portfolio change.

Before the changes. The limited amounts are $\$21{,}000$, $\$25{,}000$, $\$24{,}000$, and $\$5{,}000$. Their total is $\$75{,}000$, so the premium is $\$140{,}000+0.92(\$75{,}000)=\$140{,}000+\$69{,}000=\$209{,}000$.

Claim-level limited-loss changes. Claim A changes from $\$21{,}000$ to $\$25{,}000$ limited, an increase of $\$4{,}000$. Claim B stays limited at $\$25{,}000$, so its change is zero. Claim C decreases by $\$4{,}000$, from $\$24{,}000$ to $\$20{,}000$, both below the cap. Claim D increases by $\$4{,}000$, from $\$5{,}000$ to $\$9{,}000$. The net limited-loss change is $\$4{,}000+\$0-\$4{,}000+\$4{,}000=\$4{,}000$.

After the changes and check. The new limited total is $\$25{,}000+\$25{,}000+\$20{,}000+\$9{,}000=\$79{,}000$. The new premium is $\$140{,}000+0.92(\$79{,}000)=\$140{,}000+\$72{,}680=\$212{,}680$. Thus the increase is $\$212{,}680-\$209{,}000=\$3{,}680$, which agrees with $0.92(\$4{,}000)=\$3{,}680$.

Common mistake avoided: Claim B’s $\$4{,}000$ raw increase does not increase limited loss because the claim was already above the cap. It must not be included as an increase in the LCF base.

Interpreting the Incentive

The cap creates a piecewise claim-level premium response. For an increase that stays below $C$, the premium changes by $L$ times that increase. For an increase that begins below and crosses $C$, the premium responds only up to the cap. For an increase starting at or above $C$, the premium does not change through this capped-loss component. Decreases work in the reverse direction: a reduction entirely above the cap may have no premium impact, while a further reduction that takes the claim below the cap begins reducing premium.

This is an incentive in the narrow sense of the premium response to eligible claim loss under the assumed plan formula. It should not be confused with the policyholder’s total financial incentive to prevent or reduce a loss. Coverage terms, deductibles, claim handling, and other costs can matter to that broader decision. The per-claim cap alone supports the conclusion about the premium formula.

Worked Example: Comparing Reductions on Capped and Uncapped Claims

A hypothetical plan has $B=\$80{,}000$, $L=1.15$, and $C=\$12{,}000$. Two claims are $\$15{,}000$ and $\$9{,}000$. Consider a reduction of the first claim to $\$10{,}000$ and a reduction of the second claim to $\$4{,}000$. Find the premium reduction from each change and the combined reduction.

Method. For each claim, compare the capped amount before and after the reduction. The premium reduction equals $L$ times the decrease in that capped amount.

Initial premium. The limited amounts are $\min(\$15{,}000,\$12{,}000)=\$12{,}000$ and $\min(\$9{,}000,\$12{,}000)=\$9{,}000$, totaling $\$21{,}000$. Thus $P=\$80{,}000+1.15(\$21{,}000)=\$80{,}000+\$24{,}150=\$104{,}150$.

First claim reduction. Its limited amount falls from $\$12{,}000$ to $\$10{,}000$, a $\$2{,}000$ decrease. The premium reduction is $1.15(\$2{,}000)=\$2{,}300$. The claim itself fell by $\$5{,}000$, but the first $\$3{,}000$ of reduction moved the claim from $\$15{,}000$ down to the cap without changing its limited amount.

Second claim reduction. Its limited amount falls from $\$9{,}000$ to $\$4{,}000$, a $\$5{,}000$ decrease. The premium reduction is $1.15(\$5{,}000)=\$5{,}750$.

Combined result and check. The total premium reduction is $\$2{,}300+\$5{,}750=\$8{,}050$. After both changes, limited losses total $\$10{,}000+\$4{,}000=\$14{,}000$, giving a premium of $\$80{,}000+1.15(\$14{,}000)=\$96{,}100$. The direct difference, $\$104{,}150-\$96{,}100$, is also $\$8{,}050$.

Common mistake avoided: Applying the LCF to the full $\$5{,}000$ reduction in the first claim would overstate its premium impact. Only the $\$2{,}000$ reduction in limited loss changes premium.

Common Exam Traps

Using raw losses instead of limited losses. Compute $\min(X_j,C)$ for every claim before applying $L$.
Applying one cap to the portfolio total. A per-claim cap means $\sum_j\min(X_j,C)$, not $\min(\sum_jX_j,C)$.
Applying the cap after conversion. Under the stated formula, the eligible loss is limited before multiplication by the LCF. Capping $LX_j$ at $C$ would describe a different formula.
Assuming a change is fully reflected because it is below the cap at the start. If an increase crosses the cap, only the part up to $C$ affects limited loss.
Ignoring the direction of change at the cap. An increase from exactly $C$ has no effect; a decrease from $C$ reduces the limited amount.

Key takeaway: Under a fixed-$B$, fixed-$L$, per-claim capped-loss formula, the premium impact of changing a claim is the LCF multiplied by that claim’s change in limited loss. Below the cap, marginal premium impact is $L$ per dollar; above the cap, it is zero.

Check Your Understanding

For each problem, use a per-claim cap applied before the LCF. Assume other plan terms remain fixed.

A plan has $L=1.05$ and $C=\$8{,}000$. One claim increases from $\$6{,}000$ to $\$9{,}000$. By how much does the loss-based premium change? (Answer: increase of $\$2{,}100$)
A plan has $L=0.90$ and $C=\$20{,}000$. One claim rises from $\$18{,}000$ to $\$23{,}000$, while another rises from $\$25{,}000$ to $\$29{,}000$. What is the net premium change? (Answer: increase of $\$1{,}800$)
At $L=1.20$, a claim is exactly at a cap of $\$14{,}000$, then decreases by $\$3{,}000$. What is the premium reduction? (Answer: $\$3{,}600$)
A plan has $B=\$100{,}000$, $L=1.10$, and $C=\$15{,}000$. Claims are $\$8{,}000$, $\$17{,}000$, and $\$12{,}000$; the third increases to $\$19{,}000$. Find the initial premium and the premium increase. (Answer: initial premium $\$138{,}500$; increase $\$3{,}300$)
A claim of $\$22{,}000$ is reduced to $\$14{,}000$ under $C=\$17{,}000$ and $L=0.95$. What is the premium reduction? (Answer: $\$2{,}850$)

What You'll Learn