Why Case Reserves and Settlement Patterns Matter
Reported claim results reflect both the cost of claims and the way those claims are handled. A claim may be paid and closed quickly, or remain open with a case reserve for an extended period. Even if two portfolios ultimately pay the same amount, their paid losses and outstanding balances can look quite different at a valuation date. Case reserve adequacy matters as well: reserves that are too high or too low affect incurred losses before the claim is finally settled.
The key interpretive question is whether a change in reported results signals a change in expected ultimate claim costs, or instead reflects how much has been paid, which claims remain open, and how those open claims have been reserved. This distinction is aligned with the CAS Exam 5 (Basic Ratemaking and Estimated Claim Liabilities) syllabus.
For a claim cohort, let \(P_t\) denote cumulative paid losses at valuation \(t\), \(C_t\) denote case reserves at that valuation, and \(I_t\) denote incurred losses on a paid-plus-case basis. Then \(I_t=P_t+C_t\). These quantities describe a reported view of the cohort at a particular date; the ultimate estimate can differ from incurred losses if case reserves are inadequate or if other unpaid amounts are not represented in the case reserves.
Separating Settlement Activity from Reserve Adequacy
A settlement changes the form in which a claim appears in the accounts. When an open claim is paid and closed, paid losses rise and the case reserve for that claim is removed. If the payment equals the case reserve, incurred losses do not change solely because of the settlement. If the payment exceeds the case reserve, incurred losses rise; if it is less, incurred losses fall. The comparison is between the payment and the case reserve immediately before settlement, not between the payment and an unrelated portfolio average.
The second equation follows by subtracting \(I_0=P_0+C_0\) from \(I_1=P_1+C_1\). It shows why incurred development cannot be interpreted from payments alone: the change in case reserves also contributes. In a settlement period, \(P_1-P_0\) is typically positive while \(C_1-C_0\) is typically negative for claims that close. Their net effect on incurred depends on how payments compare with the reserves released.
Worked Example: An Overstated Case Reserve on Open Claims
Consider a hypothetical cohort of 100 similar claims. At the valuation date, 40 claims have been settled for a total of $400 thousand. The other 60 remain open. Their case reserves are $9 thousand each, while the current estimate of their average future payment is $8.5 thousand each. Assume no other unpaid amounts, no discounting, and no change in the estimate of the future payments. What are the reported incurred losses and estimated ultimate, and what happens to incurred losses if 30 open claims settle for $8.5 thousand each?
Method and assumptions. Reconcile paid losses and case reserves, then compare the reported amount with the estimated ultimate. All amounts below are in thousands of dollars. The expected ultimate is the $400 thousand already paid plus the estimated future payments on the 60 open claims.
1. Calculate the remaining case reserves. There are 60 open claims, each with a case reserve of $9 thousand:
2. Calculate incurred losses. Add the $400 thousand paid to the $540 thousand of case reserves:
3. Calculate estimated ultimate losses. Expected future payments on the 60 open claims are \(60(8.5)=510\) thousand. Therefore:
The reported incurred amount exceeds this estimated ultimate by \(940-910=30\) thousand. Equivalently, the case reserves exceed expected future payments by \(9-8.5=0.5\) thousand per open claim, or $500 per claim. Across 60 open claims, that difference is \(60(0.5)=30\) thousand.
4. Reconcile the next settlement activity. Suppose 30 of the open claims settle for $8.5 thousand each. Payments increase by \(30(8.5)=255\) thousand. The case reserves removed are \(30(9)=270\) thousand. The change in incurred losses is therefore:
After these settlements, paid losses are \(400+255=655\) thousand and case reserves for the remaining 30 claims are \(30(9)=270\) thousand. Incurred losses are \(655+270=925\) thousand. Estimated ultimate remains \(400+510=910\) thousand: the total expected payments have not changed simply because 30 claims settled. The remaining reserve excess is \(30(0.5)=15\) thousand, which agrees with \(925-910=15\).
Check. Incurred decreased by $15 thousand because the settled claims were paid for $8.5 thousand each but had $9 thousand of case reserve each. The result is consistent with both the paid-plus-case reconciliation and the remaining reserve excess. The common mistake is to call the reserve excess $1 thousand per claim; it is $0.5 thousand per claim.
How Settlement Speed Changes Reported Patterns
Settlement speed changes how much of a cohort’s cost appears as paid losses and how much remains in case reserves at a given date. Faster settlement generally produces more paid losses and fewer open claims, while slower settlement generally produces fewer paid losses and more case reserves. These are tendencies, not guarantees: claim mix, payment size, and case reserve adequacy can also differ.
If case reserves accurately reflect expected future payments, a change in settlement speed by itself need not change the estimated ultimate. It does, however, affect paid development and the composition of incurred losses. Therefore, paid results at one maturity are not directly comparable without considering how many claims have settled and what remains open. Incurred results also require care because case reserve levels can change as claims are reviewed or settled.
Worked Example: Faster Settlement with the Same Estimated Ultimate
Two hypothetical portfolios each contain 100 similar claims, with an estimated ultimate cost of $10 thousand per claim. At the same valuation date, Portfolio S has settled 30 claims and Portfolio F has settled 70 claims. Assume every settled claim has been paid $10 thousand, and every open claim has a case reserve of $10 thousand that equals its expected future payment. Compare their paid losses, case reserves, incurred losses, and estimated ultimates.
Method and assumptions. Calculate paid and outstanding amounts separately, then add them to obtain incurred losses. All dollar amounts are in thousands. There is no discounting or change in the estimated ultimate per claim.
Portfolio S, with slower settlement. It has 30 settled claims, so paid losses are \(30(10)=300\). The remaining \(100-30=70\) claims each carry a $10 thousand case reserve, so case reserves are \(70(10)=700\). Thus:
Its estimated ultimate is \(100(10)=1{,}000\) thousand: $300 thousand already paid plus $700 thousand expected on open claims.
Portfolio F, with faster settlement. It has 70 settled claims, so paid losses are \(70(10)=700\). Its 30 open claims have case reserves of \(30(10)=300\). Thus:
Its estimated ultimate is also \(700+300=1{,}000\) thousand. The two portfolios have equal incurred losses and estimated ultimates, but Portfolio F has $400 thousand more paid losses and $400 thousand less in case reserves at this valuation date.
Settlement-period check. If 20 open claims in each portfolio next settle for $10 thousand each, paid losses rise by \(20(10)=200\) thousand and case reserves fall by the same $200 thousand. In each portfolio, incurred losses remain $1,000 thousand. The settlement changes the paid-versus-outstanding split, not the expected total cost under the stated assumptions.
Common mistake. The higher paid amount in Portfolio F does not establish that its claims cost more. The portfolios have the same number of claims and the same assumed cost per claim; their settlement maturity differs. Comparing paid totals without that context confuses timing with cost.
When Reported Development Reflects Case Reserve Adequacy
A case reserve is an estimate, not a final payment. When an open claim settles, the difference between its payment and its pre-settlement case reserve flows into incurred development, all else equal. For a group of settled claims, that difference can be summarized as total payments minus the case reserves released. A negative result is favorable incurred development for those claims; a positive result is adverse development.
This relationship explains why apparent improvement can be misleading. If case reserves were set too high, settlements below those reserves can generate favorable development even when the ultimate claim costs have not fallen. If case reserves were too low, settlement payments above those reserves can generate adverse development even when the expected ultimate for the overall cohort has not changed. A change in reported incurred losses is evidence about the estimates and emergence being reported; it is not, by itself, proof of a change in underlying ultimate costs.
Worked Example: Faster Settlement Reveals Under-Reserved Claims
Two hypothetical groups, A and B, each contain 100 claims with the same expected ultimate cost of $10 thousand per claim. At valuation, Group A has paid 40 claims at $10 thousand each and has 60 open claims with case reserves of $9 thousand each. Group B has paid 70 claims at $10 thousand each and has 30 open claims with case reserves of $10 thousand each. In the next period, 30 claims in A and 20 claims in B settle, each for $10 thousand. Calculate the incurred development for each group and explain whether the results demonstrate different ultimate costs.
Method and assumptions. Apply the paid-plus-case reconciliation separately to each group and period. All values are in thousands. Assume claim counts are as stated, payments are $10 thousand per settlement, there are no other payments or reserve changes, and the expected ultimate remains $1,000 thousand for each group.
1. Initial results for Group A. Paid losses are \(40(10)=400\). Case reserves are \(60(9)=540\). Therefore:
2. Initial results for Group B. Paid losses are \(70(10)=700\), and case reserves are \(30(10)=300\). Therefore:
Both groups have the same expected ultimate, \(100(10)=1{,}000\) thousand, although Group A’s initial incurred amount is $60 thousand below that estimate.
3. Group A’s next-period development. Thirty claims settle for \(30(10)=300\) thousand. The case reserves released are \(30(9)=270\) thousand. Thus:
Group A records $30 thousand of adverse incurred development. After settlement, paid losses are \(400+300=700\), and case reserves for 30 remaining claims are \(30(9)=270\). Incurred losses are \(700+270=970\) thousand.
4. Group B’s next-period development. Twenty claims settle for \(20(10)=200\) thousand, and the case reserves released are \(20(10)=200\) thousand. Thus:
Group B’s incurred losses remain $1,000 thousand: paid losses are \(700+200=900\), and the remaining 10 claims have case reserves of \(10(10)=100\).
Interpretation and check. The different incurred development is consistent with different case reserve adequacy: Group A’s initial case reserves were $1 thousand below the assumed $10 thousand future payment per open claim, while Group B’s matched it. At the end of the period, Group A’s incurred amount is still $30 thousand below its $1,000 thousand expected ultimate, matching the shortfall on its 30 remaining claims. The example does not show that Group A’s underlying claims cost less; under the stated assumptions, both groups have the same expected ultimate.
Common mistake. It would be wrong to interpret Group A’s adverse development as proof that its ultimate costs increased during the period. The stated assumptions hold ultimate costs constant; the difference arises because the initial case reserves for its open claims were below their expected future payments.
A Practical Interpretation Sequence
When reported results change, separate the accounting movements before drawing a conclusion about ultimate cost. The following sequence keeps settlement, case reserve changes, and the ultimate estimate distinct.
Confirm that the valuation dates and claim cohorts being compared are consistent. A different maturity can change the paid-versus-open split even when ultimate cost is unchanged.
For each date, calculate incurred losses as paid losses plus case reserves. Then explain the change using the change in payments and the change in case reserves.
For settled claims, compare payments with the case reserves removed. Payments above those reserves produce adverse incurred emergence; payments below them produce favorable emergence, subject to any other movements.
Ask whether the expected total cost changed, or whether the reported pattern is explained by settlement timing or changes in case reserve adequacy. Do not infer a lower ultimate from favorable development alone.
This sequence complements the earlier material on reserve development and reserve releases without treating a reported movement as self-explanatory. The interpretation should identify what changed and why. For example, a lower outstanding balance may result from faster claim closure, a lower selected reserve per open claim, or both. Those explanations have different implications for the estimate of ultimate losses.
Common Interpretive Traps
- Equating paid growth with higher claim costs. Paid losses can rise because claims settle sooner. Compare the paid increase with the associated reduction in case reserves.
- Treating incurred losses as automatically equal to ultimate losses. Incurred losses equal ultimate only if the recorded amounts adequately represent all expected costs for the cohort.
- Ignoring the amount of case reserve released. A settlement payment alone does not determine incurred development. The pre-settlement case reserve matters too.
- Calling favorable settlement emergence a lower ultimate without evidence. Payments below case reserves may indicate prior over-reserving, not a reduction in the underlying claim costs.
- Comparing groups with different settlement maturity as if they were equally developed. State how many claims are paid and how many remain open before interpreting paid or outstanding differences.
Check Your Understanding
For each hypothetical problem, distinguish the reported movement from the implication for estimated ultimate losses.
- A cohort has paid losses of $520 thousand and case reserves of $180 thousand at one valuation. At the next valuation, payments increase by $70 thousand and case reserves decrease by $65 thousand. What is the change in incurred losses? (Answer: $5 thousand increase.)
- There are 40 open claims with a case reserve of $12 thousand each. The current estimate of future payments is $11.5 thousand per claim. Calculate the total case reserve excess over expected future payments. (Answer: $20 thousand.)
- A hypothetical portfolio has 80 claims with an expected ultimate of $9 thousand per claim. At valuation, 50 claims have settled for $9 thousand each; the other 30 have case reserves of $9 thousand each. If 10 open claims next settle for $8.5 thousand each, with their full $9 thousand case reserves removed, what is the incurred development on those settlements? (Answer: $5 thousand favorable.)
- Two groups each have 60 claims with an expected ultimate of $7 thousand per claim. Group X has paid 20 claims and has case reserves of $7 thousand for each of its 40 open claims. Group Y has paid 45 claims and has the same $7 thousand reserve per open claim. Assuming settled claims were each paid $7 thousand, calculate each group’s incurred losses and explain the difference in paid losses. (Answer: Each has incurred losses of $420 thousand; Group Y has $175 thousand more paid losses because more claims have settled.)
- A cohort has 25 open claims, each reserved at $8 thousand. During the period, 15 settle for $8.4 thousand each, and the other 10 case reserves remain unchanged. There are no other movements. Calculate the incurred development and state whether it alone proves the cohort’s ultimate increased. (Answer: $6 thousand adverse development; no, it does not alone prove the ultimate increased.)