Why Compliance-Only ORSA Is a Diagnostic Problem
An Own Risk and Solvency Assessment (ORSA) can be completed on schedule, approved, and filed while having little influence on how an insurer is managed. The central diagnostic question is therefore not simply whether an ORSA document exists. It is whether the assessment is part of governance: whether accountable leaders engage with it, business areas own their contributions, and conclusions inform decisions or deliberate decisions not to act.
A compliance-only pattern is not established by one imperfect meeting minute or one decision that did not change. It is a conclusion drawn from a body of evidence. The examiner or reviewer should distinguish a missing record from evidence that no consideration occurred, and should not treat a title, assigned owner, or monitoring task as proof of ORSA use by itself. This tutorial is aligned with the CAS Exam 9 (Risk Management for Actuaries) syllabus.
A Practical Diagnostic Checklist
Use a checklist to organize evidence, not as a substitute for judgment. The relevant question for each area is whether practice matches the stated governance arrangement. Policies and committee terms of reference describe intended responsibilities; minutes, decision records, interviews, and follow-up evidence help establish how those responsibilities operate.
| Diagnostic area | Evidence consistent with management use | Warning sign of compliance-only practice |
|---|---|---|
| Board engagement | Records show informed challenge of material conclusions, discussion of implications, and follow-up where appropriate. | Approval is routine, with little evidence that directors considered the assessment or its implications. |
| Executive accountability | Executives can explain their responsibilities and how ORSA conclusions affect management oversight. | Responsibility is assigned on paper, but executives treat ORSA as a risk-function deliverable. |
| Business ownership | Business leaders provide, explain, and take responsibility for relevant inputs and responses. | Inputs are assembled centrally and business leaders cannot explain or validate them. |
| Decision use | Records connect an ORSA conclusion to an action or a reasoned decision not to act. | Actions are listed without a demonstrated connection to ORSA, or conclusions receive no recorded consideration. |
| Follow-through | Assigned actions have appropriate oversight, and completion or changed circumstances are considered. | Actions appear in a report but have no meaningful follow-up or connection to the original conclusion. |
These signs require context. For example, a board may appropriately approve an assessment without changing strategy if its conclusions support the existing plan. The diagnostic issue is whether the board understood and considered the relevant conclusions, not whether the outcome was a change. Similarly, a risk function may coordinate the process without owning the business judgments made by executives and business units.
Specify what the organization says ORSA does, who is accountable, and which governance or management decisions it is intended to inform.
Review decision records, meeting materials, action tracking, and explanations from the people assigned responsibility. Do not rely on a policy statement alone.
For a claimed use, ask whether the record links a conclusion to an action, a reasoned non-action, or a monitoring response. If the link is missing, classify the evidence as unresolved rather than assuming it.
Consider whether gaps recur across governance, ownership, and decisions. Separate isolated documentation weaknesses from evidence that ORSA is routinely treated as a filing task.
Classifying Evidence of Decision Use
A consistent classification prevents favorable interpretations from outrunning the evidence. A demonstrated use has a traceable connection between an ORSA conclusion and an action. A demonstrated reasoned non-action records that the conclusion was considered and explains why no change was warranted. No evidence means the available record contains no indication of consideration. Unresolved means the facts suggest possible use, but the connection or rationale has not been established.
An assigned monitoring response needs particular care. Monitoring can be an appropriate response to an ORSA conclusion, but the existence of a task and a named owner does not show why it was created. Count it as demonstrated use only when evidence establishes that the monitoring response was prompted by, or otherwise linked to, an ORSA conclusion. If that link is not established, classify it as unresolved.
Worked Example: Separating Use from Unresolved Evidence
A hypothetical insurer reviews five strategic decisions from its latest planning cycle. Decision A changed a proposed retention after the minutes cited an ORSA stress conclusion. Decision B kept the existing plan after executives documented why the ORSA finding did not warrant a change. Decisions C and D have no evidence of ORSA consideration. Decision E has an assigned monitoring task and named owner, but the record does not say whether the task arose from an ORSA conclusion. Classify the decisions and calculate the proportion with established evidence of use.
Method. Classify each decision based on the connection actually documented; do not infer a connection from timing or task ownership.
Step 1: Classify A. The record links the retention change to an ORSA stress conclusion. A is demonstrated use.
Step 2: Classify B. The executives considered the ORSA finding and documented a rationale for retaining the plan. B is demonstrated reasoned non-action, which is evidence of use even though no change occurred.
Step 3: Classify C and D. Neither has evidence of consideration. Both are classified as no evidence; the available facts do not establish ORSA use.
Step 4: Classify E. A monitoring task and owner are present, but no fact connects the task to an ORSA conclusion. E is unresolved, not demonstrated use.
Step 5: Calculate the established proportion. Two of the five decisions have established use: A and B. Thus the descriptive proportion is \(2/5=0.40\), or 40.0%. The full classification is two demonstrated, two with no evidence, and one unresolved.
Check. The categories account for all five decisions: \(2+2+1=5\). If later evidence establishes that E's monitoring response was prompted by an ORSA conclusion, the count would become three of five, or \(3/5=60.0\%\). Without that evidence, reporting three would overstate what is known.
Governance and Ownership: Test Practice, Not Labels
Formal roles are necessary, but they are not sufficient to show that ORSA is embedded. A useful review compares assigned responsibility with observable behavior. For board oversight, ask whether directors receive information that enables challenge and whether the record shows consideration of important conclusions. For executive accountability, ask whether executives can explain their role and how conclusions enter management oversight. For business ownership, ask whether the relevant leaders can stand behind their inputs and explain material judgments.
The diagnostic should not demand that every business leader reproduce technical analysis. Instead, test whether the people responsible for a business area understand the assumptions and conclusions relevant to that area, provide appropriate input, and take part in responses. Central coordination by a risk function can be consistent with sound governance. A concern arises when coordination becomes substitution: the risk function prepares the entire assessment while executives and business units neither own the judgments nor use the conclusions.
Worked Example: Distinguishing Formal Accountability from Active Ownership
A hypothetical insurer's policy names the chief executive as accountable for ORSA, the risk function as coordinator, and product leaders as owners of their business inputs. In practice, the risk function drafts the assessment and requests product data by email. Product leaders return spreadsheets but cannot explain a material assumption used for their products. The chief executive signs the final report, while the committee record contains no discussion of the conclusions. Diagnose the evidence without assuming that the written role assignments prove active use.
Step 1: Compare intended and observed roles. The policy assigns accountability and business ownership. Observed practice establishes that the risk function coordinates drafting and product leaders provide data, but it does not establish that they validate or own the relevant judgments.
Step 2: Assess executive evidence. The chief executive's signature shows approval or acknowledgment, but the stated facts do not show substantive engagement with conclusions. A signature alone is insufficient evidence of active accountability.
Step 3: Assess business ownership evidence. Returning spreadsheets is evidence of input provision. Inability to explain a material assumption is a warning sign that the business leaders may not own the judgment behind the input. It is not, by itself, proof that they had no involvement; follow-up evidence could clarify their role.
Step 4: Assess committee evidence. The absence of discussion in the available record is a governance warning. It establishes a documentation gap, but without further evidence it should not automatically be described as proof that no discussion occurred.
Conclusion. The evidence shows formal responsibilities and process completion, alongside concerns about executive engagement, business ownership, and recorded consideration. The appropriate diagnosis is a potential compliance-only pattern requiring corroboration, not a definitive claim that no one used ORSA. The common mistake would be to treat the chief executive's signature and completed spreadsheets as proof that governance and ownership were effective.
Using Counts Without Turning Them into a Score
A small review of decisions can help make a diagnosis concrete. Define the sample before calculating, classify every item, and report unresolved evidence separately. A fraction such as demonstrated uses divided by sampled decisions is a descriptive indicator for that sample. It is not a universal ORSA effectiveness score, a regulatory threshold, or proof of the quality of the entire program. The representativeness of the sample and the quality of the records matter.
Worked Example: Reporting a Decision Sample Carefully
A reviewer examines eight hypothetical management decisions. Three have actions explicitly linked to ORSA conclusions. Two have documented, reasoned decisions not to act after considering ORSA conclusions. One has a monitoring response explicitly prompted by an ORSA conclusion. One has an assigned monitoring response but no evidence linking it to ORSA. One has no evidence of ORSA consideration. Calculate the established-use proportion and state the unresolved count.
Step 1: Count established uses. The three linked actions, two reasoned non-actions, and one explicitly linked monitoring response show established use. The numerator is \(3+2+1=6\).
Step 2: Classify the remaining decisions. The monitoring response without an established ORSA link is unresolved. The final decision has no evidence. These account for the remaining two decisions.
Step 3: Calculate using all sampled decisions as the denominator. The established-use proportion is \(6/8=0.75\), or 75.0%. Report alongside it that one decision is unresolved and one has no evidence.
Check. The classification totals \(6+1+1=8\), matching the sample. The unlinked monitoring response is not included in the numerator. Counting it would incorrectly produce \(7/8=87.5\%\). If new evidence explicitly links that response to an ORSA conclusion, the established count would become seven of eight, or \(87.5\%\).
Common Traps and a Sound Conclusion
- Equating completion with use. A timely report, required approval, or completed template establishes process activity, not necessarily management influence.
- Equating no change with no use. A documented, reasoned decision to retain the existing plan can demonstrate consideration and use.
- Inferring a link from an assigned task. A named owner and monitoring action do not establish ORSA use unless the task is connected to an ORSA conclusion.
- Overstating what missing minutes prove. Missing documentation is a concern and may itself weaken governance evidence, but it does not always prove that discussion did not occur.
- Treating a sample fraction as a rating. A descriptive proportion depends on the selected decisions and available records. It should support, not replace, a broader evidence-based assessment.
A balanced conclusion identifies the evidence, the gaps, and what remains uncertain. For example, a reviewer might conclude that reporting requirements appear to be met, while sampled records show limited documented decision use and several ownership concerns. That conclusion is stronger than either an unsupported declaration that the program is effective or an overconfident claim that it is purely administrative. The diagnostic purpose is to establish whether ORSA is functioning as a management tool and where the evidence does or does not support that view.
Check Your Understanding
Classify the evidence before calculating or drawing a conclusion. Treat an unestablished connection as unresolved, not as demonstrated use.
- A hypothetical insurer samples four decisions. One action cites an ORSA conclusion; one decision not to act includes a rationale tied to ORSA; one has an assigned monitoring task with no stated ORSA link; and one has no evidence of consideration. How many show established use, and what is the proportion of all four? (Answer: Two; \(2/4=50.0\%\). The monitoring task is unresolved and one decision has no evidence.)
- A board approves an ORSA report. The minutes show no discussion, but directors later state that they considered its conclusions. What does the available record establish, and what evidence would help resolve the issue? (Answer: Approval is established, but the record alone does not establish substantive challenge or consideration. Contemporaneous materials, recorded questions, or other corroborating evidence could clarify the engagement.)
- In a sample of six decisions, two actions are explicitly linked to ORSA conclusions, one reasoned non-action is documented, two have no evidence, and one monitoring task is assigned without a stated ORSA link. How many have established use and what is the proportion? (Answer: Three; \(3/6=50.0\%\). The monitoring task is unresolved, not established use.)
- Business leaders supply requested data, but cannot explain a material assumption in their area. Is business ownership established? Give the diagnostic implication. (Answer: Not fully established. Data submission is evidence of input provision, but inability to explain the material assumption is a warning sign requiring follow-up about validation and ownership.)
- A sample of eight decisions contains four actions explicitly linked to ORSA, one documented reasoned non-action, one monitoring response explicitly linked to an ORSA conclusion, one unlinked monitoring response, and one decision with no evidence. Calculate the established-use proportion and report the other categories. (Answer: Six of eight, or \(75.0\%\), show established use; one is unresolved and one has no evidence.)