Why Finite Risk Contracts Require a Trade-Off Analysis
A finite risk contract can provide insurance protection while spreading funding across multiple years or linking future payments to the experience of the contract. That structure can reduce volatility in an insurer’s retained losses and make cash flows more predictable. It can also cost more than a simpler indemnity arrangement, provide only limited protection against severe losses, expose the insurer to counterparty default, or require repayments when experience is favorable.
The key is to distinguish the contract’s stated payment from its net economic effect. A promised recovery is not necessarily the amount ultimately collected; an experience repayment can offset earlier protection; and a fixed premium is a financing cost even if claims are low. The calculations below treat each arrangement as hypothetical and specify its terms and assumptions.
For an insurer, it is useful to track two quantities separately. Retained loss measures the loss remaining after amounts actually collected and any repayments. Net contract cost measures the present value of premiums and repayments less recoveries. The first concerns loss exposure; the second concerns the financial price and timing of the arrangement. Neither alone captures every relevant consideration.
Worked Example: Protection, Repayment, and Net Cost
Worked Example: Protection, Repayment, and Net Cost
A hypothetical one-year contract has a premium of $55 payable now. At year-end, it pays the lesser of the insurer’s loss and $100. If the loss is zero, the insurer must also repay $20 at year-end under an experience provision. Loss \(X\) is $0 or $200, each with probability 0.50. Assume an effective annual interest rate \(i=5\%\), full collection of contractual payments, and no other cash flows. Find the expected retained loss before premium, its variance, and the expected present value of net contract cost.
Method. First calculate the payment and repayment in each loss state. Then compare the retained-loss distribution with the gross-loss distribution and discount the contract cash flows to time zero. This avoids treating an experience repayment as if it were an additional recovery.
When \(X=0\), the contractual payment is \(\min(0,100)=0\), and the repayment is $20. Retained loss before premium is therefore \(0-0+20=\$20\).
When \(X=200\), the contractual payment is \(\min(200,100)=\$100\), and there is no repayment. Retained loss before premium is \(200-100+0=\$100\).
The expected retained loss is \(0.50(20)+0.50(100)=\$60\). Its variance is \[ 0.50(20-60)^2+0.50(100-60)^2 =0.50(1{,}600)+0.50(1{,}600)=1{,}600\text{ dollars}^2. \] The gross loss has mean \(0.50(0)+0.50(200)=\$100\) and variance \(0.50(0-100)^2+0.50(200-100)^2=10{,}000\text{ dollars}^2\). Thus the contract reduces the variance of loss before premium in this model, but retained loss is still positive in both states.
Let \(Y\) be the contractual payment and \(R\) the repayment. The net contract cost at time zero is the premium less the present value of \(Y\), plus the present value of \(R\). Here \(E[Y]=0.50(0)+0.50(100)=\$50\) and \(E[R]=0.50(20)+0.50(0)=\$10\). Therefore,
The expected net present cost is $16.90, rounded to the nearest cent. The calculation uses unrounded present values until the final step. This is a cost measure for the stated contract cash flows, not a conclusion that the contract is or is not worthwhile: the value of reduced volatility and other objectives must also be considered.
Common mistake. Counting the $20 repayment as a recovery would produce the wrong sign. It is money paid by the insurer, so it increases net cost and retained loss. The correct retained-loss calculation is \(X-Y+R\), not \(X-Y-R\).
How Experience Repayment Changes Protection
The example also illustrates why a finite contract’s protection should be evaluated across its full experience period. A year with no claims may not mean that the insurer keeps all of the premium or has received a lasting economic benefit: the contract may return some of that value through an experience account, corridor, or repayment provision. Conversely, a repayment that depends on favorable experience can leave the insurer with some risk-sharing benefit while still making the eventual net cost sensitive to loss outcomes.
A short justification helps organize the expected-cost calculation. Let \(P_0\) be a premium paid at time zero, and let \(Y_t\) and \(R_t\) be recoveries collected and repayments made at the end of year \(t\). If the effective annual rate is \(i\), then one dollar at the end of year \(t\) has time-zero present value \(v^t\), where \(v=1/(1+i)\). Present value is linear, so the expected net present cost is
This holds when the listed cash flows are the only contract cash flows, their timing is specified, and the discount rate is appropriate for the stated valuation. It follows by writing the realized cost as \(C_0=P_0-\sum_{t=1}^{n}v^tY_t+\sum_{t=1}^{n}v^tR_t\) and taking expectations term by term. If a recovery is uncertain because of default, use the expected amount collected rather than the contractual amount. If a repayment is conditional, include its expected amount under the specified loss model. The formula does not itself account for the value of risk reduction, taxes, accounting recognition, or changes in the discount rate.
Worked Example: A Limit and Counterparty Default
Worked Example: A Limit and Counterparty Default
A hypothetical contract covers the layer above an attachment of $100, subject to a $250 limit. The insurer’s loss \(X\) is $100, $400, or $800 with probabilities 0.50, 0.30, and 0.20, respectively. Independently of loss, the counterparty defaults with probability 0.10; on default, the insurer collects 40% of the contractual payment, and otherwise collects the full amount. Find the expected gross loss, expected contractual payment, expected collectible recovery, expected residual loss, and residual loss at the $800 loss in both default states.
Method. Apply the attachment and limit to each loss first. Then apply the collection fraction to the resulting payment. This separates the contractual promise from the amount actually collected.
For a loss \(X\), the contractual payment is \(Y=\min((X-100)^+,250)\). At \(X=100\), \(Y=0\). At \(X=400\), \(Y=\min(300,250)=250\). At \(X=800\), \(Y=\min(700,250)=250\). The expected contractual payment is
The expected gross loss is
The expected collection fraction is \(0.90(1)+0.10(0.40)=0.94\). Because default is independent of loss, expected collectible recovery is \(125(0.94)=\$117.50\). Expected residual loss is gross loss less collectible recovery: \(330-117.50=\$212.50\).
At \(X=800\), the contractual payment is capped at $250. In default, collection is \(0.40(250)=\$100\), leaving residual loss \(800-100=\$700\). Without default, the full $250 is collected, leaving \(800-250=\$550\). The maximum contractual payment remains $250; the maximum amount collected in default under this assumption is $100.
Common mistake. Subtracting the contractual payment of $250 in the default state understates residual loss. The insurer collects only $100 in that state, so the correct residual loss is \(800-100=\$700\), not $550.
Multi-Year Protection Has a Financing Price
A multi-year structure can make protection available across several periods and can reduce the need to renegotiate or fund a new arrangement each year. But a multi-year promise does not automatically provide unlimited protection or a low-cost source of funds. The relevant comparison includes the premium, the timing and amount of recoveries, any experience repayments, and the insurer’s retained losses in each period. Discounting is essential when cash flows occur at different times.
The following example isolates the price of a repayment provision. It assumes no default and compares the finite arrangement with an otherwise identical hypothetical indemnity cover. This is a cash-flow comparison, not a claim that the two contracts have identical legal or risk-transfer characteristics.
Worked Example: Comparing Two-Year Contract Cash Flows
An insurer faces two annual losses, \(X_1\) and \(X_2\). In each year, loss is $0 or $100 with probability 0.50; the two years are independent. A finite contract charges a $70 premium now, pays $60 at the end of each year in which loss is $100, and requires a $20 repayment at the end of year 2 for each loss-free year. Assume \(i=4\%\), full collection, and no other cash flows. An otherwise identical hypothetical two-year indemnity alternative charges $68 now, pays the same $60 per loss year, and has no repayment. Compare expected net present costs.
Method. Calculate expected cash flows by year, not by treating the two-year contract as a one-year arrangement. Independence is not needed for the expected value of the sum, but the stated joint model confirms the possible outcomes.
There are four equally likely outcomes for \((X_1,X_2)\): \((0,0)\), \((0,100)\), \((100,0)\), and \((100,100)\), each with probability \(0.25\). In each year, the expected recovery is \(0.50(0)+0.50(60)=\$30\). The expected number of loss-free years across the two years is \(0.50+0.50=1\), so the expected year-2 repayment is \(1(20)=\$20\).
For the finite contract, the expected net present cost is
Rounded to the nearest cent, the expected net present cost is $31.91. For the indemnity alternative, it is
The alternative’s expected net present cost is $11.42. The finite arrangement costs $20.49 more in present-value terms: it has a $2 higher initial premium and an expected repayment with present value \(20/1.0816=\$18.49\). Both contracts provide the same stated recoveries in this simplified comparison. The finite contract’s cash-flow terms therefore need a separate justification, such as a valued multi-year feature or other contractual benefit; the protection alone does not make it cheaper.
The experience repayment also affects the timing of cash flows. The insurer may receive protection in a loss year but owe money in year 2 if either year is loss-free. A review limited to year-1 claims or to the premium would miss that obligation. Do not round intermediate present values before summing; retain precision and round only the final reported amounts.
Practical Trade-Offs and Exam Traps
A useful assessment keeps distinct dimensions visible rather than compressing them prematurely into a single expected-cost figure.
- Earnings volatility: A contractual recovery can reduce the variation in retained losses, but an experience repayment can return some of that benefit in favorable outcomes. Measure the resulting net cash flows under the contract terms.
- Financing cost: Compare premiums, recoveries, and repayments at consistent dates using the stated interest assumption. A lower current cash outlay does not necessarily imply a lower present-value cost.
- Peak-loss protection: An attachment and limit cap contractual payments. A severe loss above the covered layer remains substantially with the insurer, even before considering default.
- Counterparty exposure: Expected contractual recovery and expected collectible recovery differ when collection is uncertain. State the default and recovery assumptions, and apply them to the covered payment rather than to the gross loss.
- Future repayments: Identify when repayments are due and what experience triggers them. Treat them as outflows, and include them in both cost and retained-loss calculations when appropriate.
A recurring trap is to interpret a contract’s maximum limit as the amount of protection in every scenario. The limit is a ceiling, not a guaranteed payment: the loss must first exceed the attachment, and default may reduce the amount collected. Another trap is to describe a lower expected retained loss as proof that all relevant risk has been transferred. Expected loss, volatility, peak exposure, counterparty risk, and the contract’s substantive risk-transfer assessment answer different questions.
Check Your Understanding
For each hypothetical arrangement, use the stated assumptions and keep intermediate values unrounded.
- A loss is $80 with probability 0.25 and $240 with probability 0.75. A contract pays the loss above a $40 attachment, subject to a $120 limit; assume full collection and no repayment. Find the expected gross loss, expected payment, and expected residual loss. (Answer: expected gross loss $200; expected payment $100; expected residual loss $100.)
- A contract charges $120 now, pays expected recoveries of $50 at the end of year 1 and $40 at the end of year 2, and requires an expected repayment of $15 at the end of year 2. If \(i=5\%\), find expected net present cost. (Answer: $49.71.)
- Losses of $100, $300, and $600 have probabilities 0.20, 0.50, and 0.30. A contract covers the layer above $150 with a $200 limit. Default is independent, has probability 0.08, and yields 25% collection of the contractual payment; otherwise collection is full. Find expected contractual payment, expected collectible recovery, and expected residual loss. (Answer: expected contractual payment $135; expected collectible recovery $126.90; expected residual loss $223.10.)
- A loss is $0 or $200 with equal probability. A contract pays the lesser of the loss and $100; if the loss is zero, the insurer repays $30 at year-end. Assume full collection and ignore premium. Find the mean and variance of retained loss \(X-Y+R\), and compare the variance with gross loss. (Answer: retained-loss mean $65; retained-loss variance $1,225 dollars squared; gross-loss variance $10,000 dollars squared.)