Setting Materiality Thresholds for ORSA

Why ORSA Uses Materiality Thresholds

An ORSA risk inventory can contain many exposures, but not every exposure needs the same depth of analysis. Materiality thresholds help an insurer decide which risks warrant closer assessment and which can be monitored through routine processes. A threshold is a prioritization tool: it guides attention, not whether a risk exists or whether it should be governed.

A useful framework combines quantitative screens, such as a possible financial impact compared with available resources, with qualitative criteria, such as an exposure that could disrupt a critical service or threaten the insurer’s ability to meet obligations. Quantitative screening can make prioritization consistent, but a single number cannot capture every strategically or operationally important concern.

Definition: A materiality threshold is a documented criterion used to identify an exposure, change, or combination of exposures for enhanced ORSA assessment or escalation. A quantitative threshold compares a defined measure of impact with a stated limit. A qualitative threshold identifies circumstances that warrant attention even when a numeric limit is not reached.

Thresholds are specific to the purpose for which they are set. A screen for a solvency assessment should use a measure relevant to solvency, whereas a screen for operational continuity may use service interruption, recovery time, or dependency on a critical provider. The threshold, the unit of measurement, and the decision it triggers should all be clear.

Designing a Quantitative Screen

A quantitative screen is meaningful only when its numerator and comparison basis are defined. For example, an insurer might compare a one-year adverse financial impact with available financial resources at the beginning of that period. The impact and the resource measure must use consistent units and a compatible basis. A dollar impact over one year should not be compared with an annual percentage threshold without stating what the percentage is applied to.

One simple design expresses the threshold as a percentage of a stated resource base. Let $R$ be the selected resource measure and $p$ the threshold percentage, written as a decimal. The dollar threshold $T$ is:

$$ T=pR $$

This formula follows directly from the definition of a percentage: an impact equals $p$ percent of the resource base when it is $pR$. An impact $I$ reaches the screen when $I \geq T$, provided that $I$ and $R$ are measured on the stated, consistent basis. This comparison does not itself establish that the insurer is solvent or insolvent; it identifies an exposure for further assessment under the organization’s process.

A policy may also specify a fixed-dollar screen, or use more than one trigger. For example, an insurer could require attention if an impact exceeds either a percentage-based threshold or a fixed amount. The rule must say whether crossing either limit or both limits triggers escalation. If a policy uses the lower of two limits, the screen is $T=\min(pR,C)$, where $C$ is the fixed-dollar cap. Such choices are governance decisions, not universal actuarial constants.

State the decision the screen supports.
Specify whether crossing the screen prompts analysis, management escalation, or another defined action. A threshold should not be left as an unexplained number.

Define the impact measure and basis.
Identify what is being measured, the time horizon, the units, and whether the estimate is gross or net of specified recoveries or mitigations.

Choose the comparison basis and limit.
State the resource measure, percentage or fixed amount, and whether multiple limits operate as an “either/or” or “both” test.

Assess aggregation and qualitative triggers.
Check whether exposures share or overlap impacts, then apply relevant qualitative criteria independently of the numeric screen.

Record the result and rationale.
Document the inputs, assumptions, calculation, judgment, and any reason for escalation or continued monitoring.

Worked Example: Applying a Percentage Screen to Several Risks

A hypothetical insurer sets a one-year quantitative screen at 3% of available financial resources. Its available resources are $240 million. Three estimated adverse impacts are $4.6 million, $3.4 million, and $1.1 million. For this example, the impacts are explicitly assumed to be non-overlapping and additive, and all three use the same one-year basis. Determine which individual impacts meet the screen and whether the combined impact does.

Method. Calculate the dollar threshold, compare each impact with it, and then add the impacts under the stated non-overlap assumption.

Step 1: Calculate the threshold. Here $p=0.03$ and $R=\$240$ million, so:

T=0.03(\$240\text{ million})=\$7.2\text{ million}

Step 2: Compare each impact. The three impacts are $4.6 million, $3.4 million, and $1.1 million. Each is below $7.2 million, so none individually reaches the quantitative screen.

Step 3: Aggregate on the stated basis. Because the problem states that the impacts are non-overlapping and additive:

I_{\text{combined}}=\$4.6+\$3.4+\$1.1=\$9.1\text{ million}

The combined impact exceeds the threshold by $\$9.1-\$7.2=\$1.9$ million. As a cross-check, $\$9.1/\$240=0.0379167$, or approximately 3.79% of resources, which is greater than 3%.

Conclusion and trap avoided. The combined exposure reaches the screen even though none does individually. The addition is justified here only by the explicit assumption that the impacts are non-overlapping and additive. Calling exposures “distinct” would not, by itself, establish that their impacts do not overlap.

Aggregation Without Double-Counting

Aggregation matters because several individually small exposures can create a material combined impact. But adding separate estimates can overstate the total if they include the same underlying loss. The analyst must establish whether the estimates are incremental, mutually exclusive, or partially overlapping before using a sum.

For two estimates that each include the same shared impact, the non-duplicated total is the sum less the shared amount. This is the two-set inclusion-exclusion rule: the shared portion appears once in each estimate, so subtract it once to count it only once in the total.

I_{\text{combined}}=I_1+I_2-I_{\text{overlap}}

This expression applies when the stated overlap is the only duplicated component and the estimates otherwise use a consistent basis. If the overlap is unknown, or several interactions exist, the combined impact cannot be established by simple addition alone. The analyst should obtain a justified combined estimate, describe the uncertainty, or escalate the unresolved aggregation issue rather than assert a precise total.

The same care applies across business units and risk categories. A single interruption, claim event, or market movement may affect several reported exposures. Conversely, some exposures can amplify one another rather than merely overlap. The aggregation approach should reflect the scenario and the purpose of the screen; a mechanical sum is not a substitute for understanding the relationship between impacts.

Worked Example: Adjusting for a Shared Impact

A hypothetical insurer uses 4% of $200 million of available resources as its one-year screen. Two separately prepared impact estimates are $5.2 million and $4.1 million. Both estimates include the same $1.5 million cost of a particular service interruption; all other components are non-overlapping. The interruption also lasted 36 hours, and the insurer’s documented qualitative criterion calls for escalation when a critical service is interrupted for more than 24 hours. Determine the quantitative and qualitative results.

Method. First calculate the dollar screen. Then remove the duplicated impact once from the combined estimate and assess the independent qualitative criterion.

Step 1: Calculate the threshold.

T=0.04(\$200\text{ million})=\$8.0\text{ million}

Step 2: Check the individual estimates. The first estimate, $5.2 million, is below $8.0 million. The second estimate, $4.1 million, is also below $8.0 million.

Step 3: Calculate the non-duplicated combined impact. The shared $1.5 million appears in both estimates, so count it only once:

I_{\text{combined}}=\$5.2+\$4.1-\$1.5=\$7.8\text{ million}

The combined amount is $0.2 million below the quantitative screen. The ratio check is $\$7.8/\$200=0.039$, or 3.90%, also below 4%.

Step 4: Apply the qualitative criterion. The 36-hour interruption is longer than the stated 24-hour criterion. Therefore, the exposure is escalated on qualitative grounds even though the non-duplicated financial estimate does not reach the numeric screen.

Conclusion and trap avoided. The quantitative result is below the screen, but the overall assessment still calls for escalation. Adding $5.2 million and $4.1 million without subtracting the shared impact would produce $9.3 million and count the same $1.5 million twice.

Qualitative Criteria and Combined Materiality

Some risks are hard to summarize with one financial estimate. A low-frequency event, a new dependency, or a weakness affecting a critical process may be important even when the available impact estimate is below a quantitative threshold. Qualitative criteria make these concerns visible without pretending that they can always be reduced to a precise dollar amount.

Criteria should be sufficiently specific to support consistent decisions. Depending on an insurer’s risk profile, documented triggers could include a threat to the ability to meet obligations, a material change in the risk profile, a critical service interruption beyond a stated duration, or a serious control concern. These are examples of possible criteria, not universal thresholds. The insurer must define which circumstances apply and who reviews them.

Individual and combined assessments answer different questions. An individual screen asks whether one exposure reaches a criterion on its own. A combined screen asks whether related or independent exposures together reach a criterion, after overlaps and interactions have been considered. The combined screen should not be restricted to risks that are already individually material: that would miss precisely the case where several small exposures add up.

Worked Example: Choosing a Relevant Screening Basis

A hypothetical insurer has $120 million of available financial resources and $450 million of annual premium. For the purpose of its one-year solvency-prioritization screen, its documented policy calls for enhanced review when an adverse impact reaches the lower of 4% of available resources or $6 million. A proposed one-year exposure has an estimated impact of $6.2 million. Determine the applicable threshold and whether the exposure reaches it. Explain why comparing the impact only with premium could mislead.

Method. Calculate both policy limits, apply the specified lower-of rule, and compare the impact with that limit.

Step 1: Calculate the percentage-based limit.

0.04(\$120\text{ million})=\$4.8\text{ million}

Step 2: Apply the lower-of rule. The two limits are $4.8 million and $6.0 million. Therefore:

T=\min(\$4.8\text{ million},\$6.0\text{ million})=\$4.8\text{ million}

Step 3: Compare the estimated impact. Since $6.2 million is greater than $4.8 million, the exposure reaches the policy’s quantitative screen. It exceeds the threshold by $\$6.2-\$4.8=\$1.4$ million.

Step 4: Check the alternative denominator. The impact is $\$6.2/\$450\approx0.01378$, or about 1.38% of annual premium. That ratio does not change the result: the policy specifies available resources and a fixed-dollar cap for this solvency screen, not premium. Premium could be useful for a different purpose, but substituting it without justification would answer a different question.

Conclusion and trap avoided. The exposure reaches the screen at $4.8 million. The relevant denominator and the “lower of” instruction both matter; selecting a more convenient denominator or treating the two limits as an “and” test would contradict the stated policy.

Applying and Documenting Thresholds

A threshold system supports sound judgment only if its application is traceable. For each assessment, record the risk or combination considered, the impact measure and time horizon, the resource base and threshold, the aggregation assumptions, the qualitative criteria checked, and the resulting decision. If an estimate is uncertain, record the uncertainty rather than presenting the threshold comparison as more precise than the inputs allow.

Review thresholds when the risk profile, business model, resource base, or decision purpose changes. A fixed dollar limit may become less informative as the insurer grows or contracts. A percentage limit can also change in practical significance if its denominator changes. Reviews should preserve an explanation of any change so that a different result is not mistaken for a change in the underlying exposure when it actually reflects a revised basis.

A common exam trap is to treat “below the threshold” as synonymous with “immaterial.” The correct conclusion is narrower: the exposure did not meet that particular numeric screen on the stated basis. It may still warrant attention because of qualitative criteria, uncertainty, aggregation with other exposures, or a governance decision to escalate. Similarly, crossing a screen identifies a need for attention; it is not, by itself, a conclusion about solvency.

Key takeaway: Set thresholds for a defined decision, compare consistent measures on a stated basis, aggregate only after checking overlap, and apply qualitative criteria independently. A sound ORSA process uses thresholds to prioritize assessment—not to make risks disappear.

Check Your Understanding

For each problem, state whether the quantitative screen is reached and note any specified qualitative result.

A hypothetical insurer has $320 million of available resources and a screen of 2.5% of resources. Three one-year impacts are $3.1 million, $4.2 million, and $1.0 million. They are explicitly non-overlapping and additive. Do any impacts reach the screen individually, and does their combined amount reach it? (Answer: The threshold is $8.0 million; none reaches it individually, while the $8.3 million combined impact exceeds it.)
A hypothetical insurer has $90 million of available resources and a 6% screen. Two impact estimates are $3.8 million and $2.5 million, with a shared $1.1 million component and no other overlap. Is the non-duplicated combined impact above the screen? (Answer: The threshold is $5.4 million; the combined impact is $3.8 million + $2.5 million − $1.1 million = $5.2 million, so it is below the screen.)
A hypothetical policy triggers review at the lower of 4% of $175 million of available resources or a fixed $9 million. An exposure is estimated at $7.2 million. What is the threshold, and is it reached? (Answer: The two limits are $7.0 million and $9.0 million, so the threshold is $7.0 million and the $7.2 million impact reaches it.)
A hypothetical insurer has $100 million of available resources and a 5% screen. Two estimates are $2.6 million and $3.0 million, and their shared component is $0.8 million; all remaining impacts are non-overlapping. The documented qualitative criterion also requires escalation for a critical-service interruption longer than 24 hours. The service was interrupted for 30 hours. What are the quantitative and qualitative conclusions? (Answer: The threshold is $5.0 million; the non-duplicated combined impact is $2.6 million + $3.0 million − $0.8 million = $4.8 million, below the screen, but the 30-hour interruption meets the qualitative escalation criterion.)
A hypothetical insurer has $140 million of available resources. Its documented rule triggers enhanced review if either the impact exceeds 3% of resources or it exceeds a fixed $5 million. A one-year impact is $4.7 million. Does the rule trigger review? (Answer: The percentage limit is $4.2 million; the impact exceeds that limit even though it is below $5 million, so the “either” rule triggers review.)

What You'll Learn