Putting the Program Pieces Together
A large-deductible program may apply several terms in sequence. Each claim is first adjusted for a per-claim deductible and a per-claim insurer payment limit. The eligible amounts are then combined for the policy year, and an annual aggregate deductible determines how much of that total the insurer pays. The result is an allocation of covered losses between the insured and insurer; premium is a separate amount added to the insured’s cost.
The order matters. Applying the annual deductible separately to each claim, or treating a per-claim payment limit as a limit on gross claim size, can materially change the answer. The calculations below assume a single policy year, losses stated on the same basis, no annual maximum on insurer payments, and no discounting for payment timing. They are hypothetical examples for learning the method, aligned with the CAS Exam 8 (Advanced Ratemaking) syllabus.
Here \(L\) caps the insurer’s payment on an individual claim, not the claim’s gross amount. For example, with \(D=\$5{,}000\) and \(L=\$20{,}000\), a claim above \(\$25{,}000\) produces no more than \(\$20{,}000\) of eligible layer. The annual aggregate deductible is applied after those claim-level amounts have been determined and summed. The examples assume there is no separate annual aggregate payment limit; if a contract has one, that additional limit must also be applied.
Why the Calculation Is Ordered This Way
The per-claim terms define which dollars are eligible for insurer consideration. A claim below \(D\) has no eligible layer. A claim above \(D\) enters the layer, but the insurer’s claim-level contribution stops at \(L\). Only after this operation for every claim do we know the annual total to which \(A\) applies.
This follows directly from the contract structure: the insured absorbs the first \(D\) dollars of each claim, and the insurer’s eligible amount on that claim cannot exceed \(L\). The insured then absorbs the first \(A\) dollars of the combined eligible annual amount. If the combined amount does not exceed \(A\), the insurer pays zero; otherwise, the excess is paid. Thus \(I=(\sum_iR_i-A)^+\).
The retained-loss reconciliation follows from the definition of the insurer payment. The insured begins with gross losses \(G\), and the insurer pays \(I\) toward them, so the insured retains \(H=G-I\). This is a total-cost identity, not a claim-level allocation rule. When a claim-by-claim allocation is needed, one can apply the aggregate deductible in a stated order, while keeping the annual insurer total unchanged.
Worked Example: Reconcile Several Claims With Both Limits
A hypothetical policy has four covered claims of \(\$4{,}000\), \(\$14{,}000\), \(\$27{,}000\), and \(\$40{,}000\). The per-claim deductible is \(D=\$5{,}000\), the maximum insurer payment per claim is \(L=\$20{,}000\), and the annual aggregate deductible is \(A=\$12{,}000\). Find the eligible layer for each claim, the insurer’s annual payment, and the insured’s retained loss.
Assumptions. The listed claims are fixed hypothetical amounts; no probability distribution is used. There is no interest adjustment, mortality assumption, or annual insurer payment limit.
Step 1: Calculate the claim-level layers. For the \(\$4{,}000\) claim, the amount above the deductible is \((\$4{,}000-\$5{,}000)^+=\$0\), so \(R_1=\$0\). For the \(\$14{,}000\) claim, the amount above the deductible is \(\$14{,}000-\$5{,}000=\$9{,}000\), below the \(\$20{,}000\) cap, so \(R_2=\$9{,}000\). For the \(\$27{,}000\) claim, the amount above the deductible is \(\$22{,}000\), capped at \(\$20{,}000\), so \(R_3=\$20{,}000\). For the \(\$40{,}000\) claim, the amount above the deductible is \(\$35{,}000\), also capped at \(\$20{,}000\), so \(R_4=\$20{,}000\).
Step 2: Sum and apply the annual deductible. The eligible annual amount is \(\$0+\$9{,}000+\$20{,}000+\$20{,}000=\$49{,}000\). The insurer pays the excess over the \(\$12{,}000\) annual deductible: \(I=(\$49{,}000-\$12{,}000)^+=\$37{,}000\).
Step 3: Reconcile retained loss. Gross loss is \(\$4{,}000+\$14{,}000+\$27{,}000+\$40{,}000=\$85{,}000\). Therefore, \(H=\$85{,}000-\$37{,}000=\$48{,}000\). Check: \(\$37{,}000+\$48{,}000=\$85{,}000\), equal to total gross loss.
Method and trap avoided. The method is claim-level adjustment followed by one annual adjustment. Applying \(A\) separately to each claim would apply the annual deductible four times and would not follow the stated contract.
Claim-Level Limits and Aggregate Deductibles in Practice
An efficient calculation keeps three quantities distinct: gross loss, claim-level eligible layer, and insurer payment after the annual deductible. A ledger makes it harder to skip or repeat a contract term. In particular, a claim with a large gross amount may be capped at \(L\) before the annual calculation, while a small claim may contribute nothing to the eligible annual total.
Worked Example: Allocate the Annual Deductible Across Claims
Consider three hypothetical claims of \(\$5{,}000\), \(\$18{,}000\), and \(\$40{,}000\), with \(D=\$6{,}000\), \(L=\$22{,}000\), and \(A=\$15{,}000\). Determine the insurer payment and retained loss. For a claim-level illustration, apply the annual deductible to eligible layers in the order listed.
Assumptions. The claim amounts are fixed; no probability distribution is used. No discounting or mortality assumption applies, and no annual insurer payment limit applies.
Step 1: Calculate each eligible layer. The \(\$5{,}000\) claim is below \(D\), so its layer is \(\$0\). The second claim has \(\$18{,}000-\$6{,}000=\$12{,}000\) above the deductible, which is below \(L\), so its layer is \(\$12{,}000\). The third claim has \(\$40{,}000-\$6{,}000=\$34{,}000\) above the deductible, capped at \(L=\$22{,}000\). Its eligible layer is \(\$22{,}000\).
Step 2: Apply the annual deductible. The sum of eligible layers is \(\$0+\$12{,}000+\$22{,}000=\$34{,}000\). The annual insurer payment is \((\$34{,}000-\$15{,}000)^+=\$19{,}000\).
Step 3: Show the ordered allocation. The first claim contributes \(\$0\), leaving all \(\$15{,}000\) of the annual deductible. The second claim contributes \(\$12{,}000\), leaving \(\$15{,}000-\$12{,}000=\$3{,}000\) of the deductible. The third claim contributes \(\$22{,}000\), of which \(\$3{,}000\) completes the deductible and \(\$22{,}000-\$3{,}000=\$19{,}000\) is paid by the insurer.
Step 4: Reconcile. Gross loss is \(\$5{,}000+\$18{,}000+\$40{,}000=\$63{,}000\). Retained loss is \(\$63{,}000-\$19{,}000=\$44{,}000\). Check: \(\$19{,}000+\$44{,}000=\$63{,}000\).
Method and trap avoided. The aggregate deductible is applied to the sum of the eligible layers, not subtracted from each claim. In particular, the third claim’s excess above the per-claim deductible is \(\$34{,}000\), but its eligible layer is \(\$22{,}000\) because of the per-claim limit.
Premium and the Insured’s Total Cost
The insurer’s payment for a particular year is not the same as the expected insurer loss used to set a premium. Actual payment is calculated from the realized claims. A premium calculation may instead use a model of possible annual outcomes, an expense provision, and a loading. The pricing convention must be stated: there is no universal formula that turns a calculated expected loss into a quoted premium without specifying expenses and loads.
Once premium \(P\) is known, the insured’s total cost for a realized year is retained loss plus premium, \(H+P\). Do not add gross loss to insurer payment: the insured’s retained loss already equals gross loss less insurer payment. Nor should premium be mistaken for a reduction in the insurer’s claim payment.
Worked Example: Expected Insurer Loss, Premium, and Total Cost
A hypothetical program has \(D=\$4{,}000\), \(L=\$16{,}000\), and \(A=\$10{,}000\). Its annual claim outcomes are modeled as follows: with probability \(0.25\), one \(\$12{,}000\) claim occurs; with probability \(0.50\), claims of \(\$9{,}000\) and \(\$20{,}000\) occur; and with probability \(0.25\), claims of \(\$30{,}000\) and \(\$35{,}000\) occur. The quoted premium is defined as \(\$3{,}000\) plus \(1.10\) times expected annual insurer loss. If the realized year is the second outcome, calculate the expected insurer loss, premium, and insured total cost.
Assumptions. The stated three-outcome discrete distribution is the complete annual claim model, and its probabilities sum to \(0.25+0.50+0.25=1.00\). All amounts are in dollars. No interest discounting or mortality assumption applies. The premium formula is a stipulated hypothetical pricing convention, not a general market rule.
Step 1: Calculate insurer payments by outcome. In outcome 1, the eligible layer is \(\min(\$12{,}000-\$4{,}000,\$16{,}000)=\$8{,}000\); after the \(\$10{,}000\) annual deductible, insurer payment is \(\$0\). In outcome 2, the layers are \(\$9{,}000-\$4{,}000=\$5{,}000\) and \(\min(\$20{,}000-\$4{,}000,\$16{,}000)=\$16{,}000\). Their sum is \(\$21{,}000\), so payment is \(\$21{,}000-\$10{,}000=\$11{,}000\). In outcome 3, each claim has a layer capped at \(\$16{,}000\); the sum is \(\$32{,}000\), so payment is \(\$32{,}000-\$10{,}000=\$22{,}000\).
Step 2: Calculate expected annual insurer loss. Weight each outcome payment by its probability:
Step 3: Calculate premium. Under the stated formula, \(P=\$3{,}000+1.10(\$11{,}000)=\$3{,}000+\$12{,}100=\$15{,}100\).
Step 4: Calculate cost in the realized second outcome. The realized gross loss is \(\$9{,}000+\$20{,}000=\$29{,}000\), and the insurer payment in that outcome is \(\$11{,}000\). Thus retained loss is \(\$29{,}000-\$11{,}000=\$18{,}000\), and total insured cost is \(\$18{,}000+\$15{,}100=\$33{,}100\). Check: retained loss plus insurer payment equals \(\$18{,}000+\$11{,}000=\$29{,}000\).
Method and trap avoided. Expected insurer loss is calculated across all modeled outcomes, but the realized retained loss uses only the realized outcome. Using the realized \(\$11{,}000\) payment as though it were expected loss would happen to produce a different premium calculation; the correct expected loss is \(\$11{,}000\) here only after probability weighting.
Common Errors and a Reliable Workflow
Most errors in an integrated design arise from applying a contract term at the wrong stage or mixing quantities with different meanings. Use the following sequence and label each subtotal before moving on.
Identify the per-claim deductible, per-claim insurer payment limit, annual aggregate deductible, and any separate annual insurer payment limit. Confirm whether the limits apply to covered losses on the same basis.
For every claim, subtract the per-claim deductible, floor the result at zero, and cap it at the per-claim insurer limit.
Sum claim-level layers, subtract the annual aggregate deductible, and floor at zero. Apply any additional annual insurer limit if the contract specifies one.
Subtract insurer payment from gross loss to obtain retained loss. Add premium only when asked for the insured’s total cost.
A useful check is \(G=H+I\). If this identity fails, either gross loss, insurer payment, or retained loss has been calculated or interpreted inconsistently. A separate check for the claim-level work is \(0\le R_i\le L\) for every claim. For annual insurer payment, verify \(I\ge0\) and \(I\le\sum_iR_i\).
- Wrong: Subtract \(A\) from every \(R_i\), then add the results. Correction: Sum the eligible layers first and subtract \(A\) once.
- Wrong: Cap gross claim loss at \(L\). Correction: First apply the per-claim deductible, then cap the insurer’s eligible payment at \(L\).
- Wrong: Call the insurer’s payment the insured’s total cost. Correction: Total insured cost is retained loss plus premium.
- Wrong: Use a realized year’s insurer payment as expected insurer loss for pricing. Correction: Calculate the expected payment using the stated probability model, then apply the stated premium formula.
For exam calculations, explicitly writing \(R_i\), \(\sum_iR_i\), \(I\), \(G\), \(H\), and \(P\) helps preserve the distinction between the contract’s stages. This is especially important when an aggregate deductible is partly used by one claim and exhausted by a later claim: only the part of a claim’s eligible layer above the remaining deductible is paid.
Check Your Understanding
For each problem, calculate in the order described above. Amounts are hypothetical dollars; assume no discounting or separate annual insurer payment limit unless specified.
- A policy has \(D=\$3{,}000\), \(L=\$12{,}000\), and \(A=\$8{,}000\). Claims are \(\$2{,}000\), \(\$9{,}000\), and \(\$20{,}000\). Find insurer payment and retained loss. (Answer: insurer payment \(\$10{,}000\); retained loss \(\$21{,}000\))
- A policy has \(D=\$7{,}000\), \(L=\$18{,}000\), and \(A=\$20{,}000\). Claims are \(\$10{,}000\), \(\$30{,}000\), and \(\$40{,}000\). Find the eligible layer for each claim, insurer payment, and retained loss. (Answer: layers \(\$3{,}000,\ \$18{,}000,\ \$18{,}000\); insurer payment \(\$19{,}000\); retained loss \(\$61{,}000\))
- A policy has \(D=\$5{,}000\), \(L=\$15{,}000\), and \(A=\$10{,}000\). Claims are \(\$12{,}000\), \(\$28{,}000\), and \(\$40{,}000\). Find the annual insurer payment. (Answer: \(\$27{,}000\))
- A policy has \(D=\$4{,}000\), \(L=\$14{,}000\), and \(A=\$14{,}000\). In the realized year, claims are \(\$10{,}000\), \(\$24{,}000\), and \(\$35{,}000\). Expected annual insurer loss is \(\$24{,}000\), and premium is defined as \(1.15\) times expected insurer loss plus a fixed \(\$2{,}000\). Find the realized insurer payment, premium, and insured total cost. (Answer: insurer payment \(\$20{,}000\); premium \(\$29{,}600\); insured total cost \(\$78{,}600\))