Why Traceability Matters
An ORSA capital conclusion is more useful when a reader can follow how it was reached: which material risk judgment informed a decision, who approved that decision, and how the decision affected the stated capital position. A conclusion such as “capital remains adequate” is difficult to evaluate on its own. The supporting record should make clear what was assessed, what changed, why the change was accepted, and how the capital result follows.
Traceability is not simply having a risk report, meeting minutes, and capital calculations somewhere in the same file. Those documents must connect. For example, if management accepts a higher concentration of a particular exposure, the record should identify the judgment about that concentration, document the decision and its conditions, and show how the capital assessment reflects the resulting exposure. If any link is absent, a reviewer may be unable to tell whether the capital conclusion reflects the approved decision or an outdated assumption.
The chain does not require every judgment to be expressed as a precise formula. It does require enough information to distinguish a judgment from a fact, a proposed action from an approved action, and an expected capital effect from one that has actually been incorporated in the assessment. This tutorial is aligned with the CAS Exam 9 (Risk Management for Actuaries) syllabus.
The Evidence Chain
A practical trace can be tested in sequence. The aim is not to produce more documentation for its own sake; it is to determine whether the conclusion is supported by a clear account of the risk and the decisions that shaped its assessment.
State the relevant view of the exposure, its drivers, or a management action assumed in the assessment. Record the date and the basis for the judgment.
Show which assessment or calculation uses the judgment, what inputs it changes, and what basis is being compared. An estimate of an effect is not proof that the effect was incorporated.
Record whether the relevant decision-maker approved, rejected, deferred, or conditionally approved the action. State any limits, conditions, or follow-up requirements.
Connect the approved decision to the capital resources, capital need, or other stated measure. Explain material differences between the analysis and the reported conclusion.
Verify that the reported result uses the approved decision, the correct version of the analysis, and a consistent measurement basis. State any remaining limitation or condition.
The same chain can be summarized as a question: What judgment led to what decision, on what evidence, and how is the decision reflected in the capital conclusion? A “yes” at each step is not enough if the documents contradict one another. The dates, amounts, conditions, and assumptions must also agree.
Relating Capital Changes to a Conclusion
For a simple internal comparison, let \(R\) denote capital resources and \(K\) denote assessed capital need, both measured on the same stated basis. The surplus is \(H=R-K\). If \(K>0\), a coverage ratio can be defined for the comparison as \(C=R/K\). These are illustrative internal measures; they are not universal definitions of regulatory capital adequacy. The measure and its basis must be identified before a result is interpreted.
These measures provide a simple arithmetic check on a documented capital bridge. Starting from \(C=R/K\), subtract one from both sides:
Thus a stated surplus and a stated coverage ratio must be mutually consistent if they use the same \(R\) and \(K\). A positive surplus alone does not establish that a chosen coverage threshold is met; nor can a ratio be interpreted correctly if the resources and need use different dates or bases. A change in capital need also does not automatically change resources. The record should show which quantity changed and why.
Following a Decision Through the Numbers
Worked Example: A Concentration Decision Changes the Capital Comparison
A hypothetical insurer is considering a reduction in a commercial property concentration. On a stated internal assessment basis, capital resources are $150 million. Assessed capital need is $126 million before the proposed reduction and is estimated at $118 million afterward. The board approves the reduction, and the assessment records that resources are unchanged. The board uses a coverage threshold of 1.20 for this comparison. Determine whether the documented capital result supports the conclusion that the threshold is met after the decision.
Method. Trace the decision to the stated before-and-after capital need, confirm the resource assumption and threshold, then calculate the surplus and coverage on each side. All amounts are in millions of dollars, and the example assumes the same internal assessment basis applies before and after the decision.
Before the decision: \(R=150\) and \(K=126\), so the surplus is \(H=150-126=24\). The coverage ratio is \(C=150/126=1.190476\ldots\), or 1.1905 to four decimal places. This is below 1.20.
After the approved decision: the recorded estimate reduces need by \(126-118=8\), while resources remain $150 million. The surplus is \(H=150-118=32\). The coverage ratio is \(C=150/118=1.271186\ldots\), or 1.2712 to four decimal places. This is above 1.20.
Traceability check. The conclusion is supported only if the $118 million need is the estimate associated with the approved reduction, the assessment uses the same basis in both cases, and the unchanged-resource assumption is justified. The numbers show the threshold is met on those stated assumptions.
Common mistake avoided: using the old $126 million need after approval and then claiming the decision improved capital adequacy. The old-basis ratio is \(150/126=1.1905\), which does not meet the threshold. The correction is to use the post-decision need of $118 million, while documenting why that estimate is applicable.
Worked Example: An Unreconciled Reinsurance Benefit
A hypothetical insurer has $90 million of capital resources and an assessed capital need of $80 million before a proposed reinsurance purchase. The documented analysis estimates that the purchase reduces need by $12 million, while a counterparty adjustment adds $3 million. The committee approves the purchase. A summary report states that post-purchase need is $68 million and that coverage is 1.32. Identify whether the stated conclusion follows from the documented bridge.
Method. Reconstruct the post-decision need from each stated component, calculate the resulting coverage using unchanged resources, and compare those results with the summary. Amounts are in millions of dollars; the example assumes all figures are on the same internal assessment basis and that resources do not change.
The net change in need is a reduction of $12 million plus an addition of $3 million, for a net reduction of \(12-3=9\) million. Therefore post-purchase need is \(80-12+3=71\) million. The post-purchase surplus is \(90-71=19\) million, and coverage is \(90/71=1.267605\ldots\), or 1.2676 to four decimal places.
The summary’s $68 million need does not reconcile to the stated components: it is $3 million below the calculated $71 million. Its stated ratio also does not match the reconstructed result. Using the summary need gives \(90/68=1.323529\ldots\), or 1.3235 to four decimal places, which rounds to 1.32 to two decimal places. That ratio is arithmetically consistent with $68 million, but the $68 million input is not supported by the documented bridge.
Conclusion. The committee’s approval is documented, but the summary does not establish that its capital conclusion reflects the approved analysis. The record needs to identify and support the additional $3 million reduction, correct the reported need, or explain another change in inputs. Until then, the conclusion based on $68 million is not traceable to the stated analysis.
Common mistake avoided: accepting a plausible-looking ratio as evidence that the analysis reconciles. The correction is to test the ratio and the underlying need separately: arithmetic agreement between $90 million and $68 million does not explain why need is $68 million.
Worked Example: A Conditional Decision and a Threshold
A hypothetical insurer currently reports $120 million of resources and $100 million of capital need on one internal basis. A proposed product expansion is expected to increase resources by $6 million and capital need by $14 million. The committee approves the expansion only if a specified exposure cap is implemented. The assessment reports post-expansion resources of $126 million and need of $114 million, and the committee’s comparison threshold is 1.10. Determine the result if the cap is implemented, and explain what must be checked if the report does not establish whether it was implemented.
Method. First calculate the post-expansion amounts from the stated changes. Then calculate coverage and compare it with the threshold. Treat the committee’s condition as part of the decision, not as a footnote.
Before expansion, the surplus is \(120-100=20\) million and coverage is \(120/100=1.20\). After the stated changes, resources are \(120+6=126\) million and need is \(100+14=114\) million. The new surplus is \(126-114=12\) million. Coverage is \(126/114=1.105263\ldots\), or 1.1053 to four decimal places. On the stated figures, the result exceeds 1.10.
That numerical comparison supports a conclusion only for the conditional decision as described. The record must show that the exposure cap was implemented and that the $126 million and $114 million figures reflect the capped expansion. If implementation is not established, the calculation alone cannot show that the committee’s condition was met; the capital conclusion is therefore conditional, not an unconditional approval of the expansion.
Common mistake avoided: recording “approved” while omitting the condition and treating approval as unconditional. The correction is to carry the condition into the evidence chain and verify that the figures supporting the conclusion correspond to the condition being satisfied.
Common Breaks in the Chain
When reviewing a capital conclusion, focus on breaks that prevent a reader from connecting the evidence to the decision and result. A missing link does not necessarily mean the decision was unsound. It does mean the record may not demonstrate why the conclusion is supportable.
| Break in the record | Why it matters | What makes the link clearer |
|---|---|---|
| A judgment is stated without its basis | The reader cannot distinguish an evidence-based assessment from an unsupported assumption. | Record the evidence, principal uncertainty, and date of the judgment. |
| A recommendation is presented as an approved action | The capital assessment may assume a decision that governance has not made. | Identify decision status, decision-maker, date, and any conditions. |
| A decision is recorded without its effect on the assessment | It is unclear whether the reported result uses the approved action or an earlier assumption. | Show which input or amount changed and connect it to the relevant assessment. |
| The reported result does not reconcile | A reader cannot tell whether the difference is an error, an omitted adjustment, or a separate assumption. | Provide a bridge that accounts for each material change and resolves discrepancies. |
A common exam trap is to equate a complete set of documents with a complete decision trail. Minutes can establish that a committee discussed an issue; they do not necessarily show that the capital calculation used the resulting decision. Likewise, a calculation can be correct for its inputs yet fail to support the conclusion if those inputs do not reflect the approved action. The useful test is not merely whether each document exists, but whether their contents agree.
Check Your Understanding
For each hypothetical case, trace the stated decision to its capital conclusion. Amounts are in millions of dollars unless stated otherwise.
- A committee approves a risk reduction. Resources are $84 before and after; need falls from $72 to $66. Calculate the post-decision surplus and coverage to four decimal places. (Answer: Surplus = $18 million; coverage = \(84/66=1.2727\).)
- A report begins with need of $55, subtracts a $7 benefit, and adds a $2 adjustment. Resources are $64 and unchanged. A summary states need is $49. What need and coverage follow from the stated bridge, and what discrepancy should be raised? (Answer: Need = \(55-7+2=$50\) million; coverage = \(64/50=1.2800\); the summary understates need by $1 million.)
- Resources are $105 and need is $90. An approved action increases resources by $3 and need by $9. Calculate post-action surplus and coverage; does coverage exceed a 1.08 threshold? (Answer: Resources = $108 million, need = $99 million, surplus = $9 million, coverage = \(108/99=1.0909\); yes, it exceeds 1.08.)
- A committee approves a purchase only if a $4 million exposure limit is put in place. The capital calculation assumes the limit, but the decision record does not say whether it was implemented. What is the traceability issue, even if the calculation is internally correct? (Answer: The calculation is conditional on implementation, but the evidence does not establish that the condition was met; the result cannot support an unconditional conclusion.)
- A report states resources of $96 million, need of $80 million, and surplus of $20 million, all on the same basis. Which figure fails the arithmetic check, and what is the corrected surplus? (Answer: The stated surplus fails; \(96-80=$16\) million, so the corrected surplus is $16 million.)