ORSA as a Decision-Support Framework

ORSA as a Management Decision Tool

An insurer’s risk and solvency assessment is most useful when it changes how management evaluates choices. A plan to distribute capital, enter a new market, or retain more catastrophe risk can alter both the insurer’s risk profile and its ability to absorb adverse outcomes. An assessment that describes those effects but does not inform a decision has limited management value.

The earlier material on the insurer’s own view of risk, its risk profile, strategy, and capital management provides the inputs for this use. ORSA brings those inputs together to help decision-makers ask: What could change if we take this action, and would the resulting risk and solvency position remain acceptable? The answer depends on the insurer’s circumstances, its internal standards, and the assumptions used; there is no single ratio or scenario that automatically makes a decision appropriate.

Definition: ORSA as a decision-support framework is the insurer-owned process of using its understanding of material risks, forward-looking solvency considerations, and management objectives to inform business choices. It helps decision-makers compare the consequences of alternatives, identify constraints or safeguards, and determine what should be monitored after a decision.

Decision support does not mean that an ORSA calculation makes the decision by itself. A solvency measure can identify a constraint or reveal a trade-off, but management still needs to consider such matters as strategy, earnings, execution, uncertainty, and the quality of the evidence. The purpose is to make those considerations more disciplined and connected to the insurer’s risk profile.

From Risk Understanding to a Decision

A useful assessment starts with the proposed choice, not just with a list of risks. It identifies which material risks the choice may change, estimates the resulting resource and risk requirement effects over a relevant horizon, and considers whether the insurer remains within its own decision standards under appropriate conditions. A decision can be acceptable in a central projection but unsuitable under a severe, plausible scenario, or acceptable only if a mitigation or monitoring condition is in place.

State the decision and alternatives.
Specify what management is considering, the alternatives available, and the period over which consequences will be assessed.

Map the risk effects.
Identify which material risks, dependencies, and existing exposures could change under each alternative.

Assess solvency under relevant conditions.
Project resources and risk needs in a central view and in scenarios that test important vulnerabilities. State the assumptions and internal thresholds.

Compare and decide.
Explain which alternatives meet the stated constraints, what trade-offs remain, and whether safeguards or limits are needed.

Monitor the decision.
Identify indicators or changes that could invalidate the assessment or require management to reconsider the action.

For a simple illustration, let $R$ denote projected available financial resources and $K$ a projected internal capital requirement. The ratio $R/K$ can be used as one decision indicator when $K>0$. An insurer may set different minimum ratios for a central projection and a stress scenario, but these are internal decision standards in the example, not universal ORSA thresholds.

\text{Central condition: } \frac{R_{\mathrm{central}}}{K_{\mathrm{central}}}\geq c, \qquad \text{Stress condition: } \frac{R_{\mathrm{stress}}}{K_{\mathrm{stress}}}\geq s

The conditions are useful only if their components are defined consistently. Resources and requirements must use compatible bases, horizons, and assumptions. A ratio also summarizes only part of the decision: it does not by itself describe liquidity, uncertainty, risk concentration, or the operational feasibility of an action.

Deriving a Decision Constraint

A simple derivation shows how a solvency threshold can translate into a limit on a proposed action. Suppose a hypothetical distribution $D$ reduces projected resources dollar for dollar, while the requirement is unchanged. If resources before the distribution are $R_0$, the minimum acceptable ratio is $c$, and $K>0$ is the requirement, then the constraint is

\frac{R_0-D}{K}\geq c.

Multiplying both sides by the positive quantity $K$ preserves the inequality. Rearranging gives

R_0-D\geq cK \quad\Longrightarrow\quad D\leq R_0-cK.

The bound follows because each dollar distributed removes one dollar from resources under the stated assumption. This derivation does not apply unchanged if the action also alters the requirement, produces other resource effects, or changes the projection horizon. In those cases, the effects must be represented explicitly for each alternative and scenario.

Worked Example: Setting a Distribution Limit Across Two Conditions

A hypothetical insurer has projected resources of $320 million and an internal capital requirement of $250 million. It is considering a distribution of $D$ million dollars. Its central minimum ratio is 1.20. Under a specified stress, resources fall by $30 million and the requirement rises by $20 million; the stress minimum ratio is 1.05. Find the largest distribution that meets both conditions. Assume the distribution reduces resources dollar for dollar and does not otherwise change the requirement or stress effects.

Method. Derive a separate upper bound on $D$ for each condition, then apply the more restrictive bound.

Central condition. After the distribution, resources are $320-D$, and the requirement remains 250. Therefore,

\frac{320-D}{250}\geq 1.20 \quad\Longrightarrow\quad 320-D\geq 300 \quad\Longrightarrow\quad D\leq 20.

Stress condition. Stressed resources are $320-D-30=290-D$. The stressed requirement is $250+20=270$. Thus,

\frac{290-D}{270}\geq 1.05 \quad\Longrightarrow\quad 290-D\geq 283.5 \quad\Longrightarrow\quad D\leq 6.5.

Both conditions must hold, so the largest permitted distribution is $\min(20,6.5)=6.5$ million dollars. Check: at $D=6.5$, the central ratio is $313.5/250=1.254$, which is at least 1.20. The stress ratio is $283.5/270=1.05$, exactly the stress minimum.

Common mistake avoided: applying the stress requirement increase to resources, or calculating the stress bound without multiplying 1.05 by the full stressed requirement. The stress test requires $290-D\geq1.05(270)=283.5$, giving $D\leq6.5$, not 12.5.

Using ORSA to Compare Strategic Alternatives

For a strategic decision, the assessment should compare alternatives on the same basis. A plan that appears attractive because it increases expected resources may also increase the risk requirement or worsen a stressed outcome. Conversely, a less ambitious plan may preserve more flexibility. The decision-support value lies in making those effects visible together, rather than treating a projected capital figure as the whole answer.

Worked Example: Comparing Two Hypothetical Growth Plans

A hypothetical insurer projects $500 million of resources and a $400 million internal requirement at a one-year horizon. Its internal minimum ratio is 1.15 in the central projection and 1.05 under a specified stress. Management compares two plans. Plan A adds $15 million to projected resources and $50 million to the central requirement; in the stress, resources are $45 million below Plan A’s central amount and the requirement is $25 million above its central amount. Plan B adds $10 million to resources and $35 million to the requirement; in the stress, resources are $25 million below its central amount and the requirement is $15 million above its central amount. All figures are in millions. Which plans meet both stated ratio conditions?

Method. For each plan, calculate central resources and requirement, then calculate the stressed amounts. Compare each ratio with the corresponding internal minimum.

Plan A, central projection: resources are $500+15=515$, and the requirement is $400+50=450$. Its ratio is $515/450=1.1444$ to four decimal places, below 1.15. Plan A therefore fails the central condition.

Plan A, stress: resources are $515-45=470$, and the requirement is $450+25=475$. The ratio is $470/475=0.9895$ to four decimal places, below 1.05. Plan A also fails the stress condition.

Plan B, central projection: resources are $500+10=510$, and the requirement is $400+35=435$. The ratio is $510/435=1.1724$ to four decimal places, above 1.15.

Plan B, stress: resources are $510-25=485$, and the requirement is $435+15=450$. The ratio is $485/450=1.0778$ to four decimal places, above 1.05.

On these stated tests, Plan B meets both conditions and Plan A does not. This does not establish that Plan B is automatically preferable: management would still assess whether the projected risk effects, assumptions, execution risks, and strategic benefits are acceptable. The calculations are checked directly against both thresholds; Plan B’s lower growth in resources is not a reason to ignore its lower requirement increase.

Common mistake avoided: comparing only the change in resources or only the central ratio. The decision requires both resources and requirements to be assessed, and the stress condition can change the ranking or feasibility of alternatives.

Using Scenarios to Evaluate Risk Transfer

ORSA can also support decisions that change how an existing risk is retained or transferred. The relevant comparison is not simply the price of a risk-transfer arrangement. Management should consider its effect on resources, risk needs, stress outcomes, and any assumptions about the effectiveness and timing of recovery. A scenario result depends on what it includes; a modeled recovery should not be treated as certain if the assessment assumes no delay or counterparty impairment.

Worked Example: Assessing a Hypothetical Reinsurance Purchase

A hypothetical insurer has resources of $260 million and an internal requirement of $200 million. A specified catastrophe stress causes a $35 million reduction in resources and adds $10 million to the requirement. The insurer’s internal minimum stressed ratio is 1.10. A proposed reinsurance contract costs $4 million, reduces the stress loss by $18 million, and reduces the stressed requirement increase by $3 million. Assume the premium reduces resources immediately, the recovery is fully collectible within the assessment horizon, and the purchase has no other modeled effects. Does the contract meet the stressed minimum?

Without reinsurance: stressed resources are $260-35=225$ million and the stressed requirement is $200+10=210$ million. The ratio is $225/210=1.0714$ to four decimal places, below 1.10.

With reinsurance: resources after the premium, before the stress, are $260-4=256$ million. The remaining stress loss is $35-18=17$ million, so stressed resources are $256-17=239$ million. The stressed requirement increase is $10-3=7$ million, so the stressed requirement is $200+7=207$ million. The ratio is $239/207=1.1546$ to four decimal places, above 1.10.

Under the stated assumptions, the contract moves the insurer from below to above its stressed minimum. That result supports consideration of the purchase but does not settle the decision: management should also examine whether the recovery assumptions are credible, whether the premium is affordable, and what risks remain. If the recovery were delayed beyond the modeled horizon or not fully collectible, the calculation would need to reflect that.

Common mistake avoided: subtracting the contract’s recovery from the stress loss while forgetting the $4 million premium, or applying the $3 million reduction to the wrong quantity. The premium reduces resources; the specified $3 million effect reduces the stress-related requirement increase.

What Makes the Assessment Useful

A decision-support assessment is stronger when it tells a clear, reviewable story: which choice was considered, which risks could change, what assumptions were used, what the results imply, and what management will do if conditions change. A number without that context can be misleading. For example, a ratio above an internal minimum does not show whether the result depends heavily on one uncertain assumption, or whether a material risk is omitted from the scenario.

A practical decision record can include:

The decision and alternatives: what management considered and the relevant time horizon.
The risk connection: the material exposures that change and the reason they matter to the choice.
The evidence and assumptions: the basis for projected resources, requirements, stress effects, and any risk-transfer recoveries.
The decision implication: which conditions are met, which are not, and what trade-offs remain.
The follow-up: indicators, limits, or events that should trigger review or reconsideration.

The thresholds and scenarios should be relevant to the insurer’s own decision. A threshold is not a substitute for judgment, and a favorable result under one stress does not prove resilience to every adverse outcome. Equally, a result below a threshold is decision-relevant information: it may lead management to change the size or timing of an action, add a safeguard, or decline an alternative.

Common Decision-Support Traps

Using ORSA only to describe the current position. Connect the assessment to a choice and explain how the choice changes the risk and solvency outlook.
Testing only the central projection. A decision may meet a central condition but fail under a scenario that is important to the insurer’s risk profile.
Mixing assumptions or bases. Compare resources and requirements over compatible horizons and apply each stated stress effect to the correct quantity.
Treating an internal ratio as an automatic decision rule. The ratio informs judgment; it does not capture every strategic, operational, or uncertainty consideration.
Ignoring what happens after the decision. Identify conditions that could change the assessment and specify when the decision should be revisited.

Key takeaway: ORSA supports management when it connects a specific choice to the insurer’s material risks and forward-looking solvency considerations, compares the alternatives on stated assumptions, and identifies the follow-up needed to keep the decision appropriate.

Check Your Understanding

For each problem, use only the assumptions and internal thresholds stated. Amounts are in millions of dollars unless noted.

A hypothetical insurer has resources of 410 and a requirement of 320. A distribution $D$ reduces resources dollar for dollar. Its central minimum ratio is 1.15. Under stress, resources fall by 24 and the requirement increases by 16; the stress minimum is 1.05. Find the largest distribution meeting both conditions. (Answer: $D=33.2$; the central bound is 42 and the stress bound is 33.2.)
An insurer has projected resources of 600 and a requirement of 500. A proposed plan adds 20 to resources and 70 to the requirement. The central minimum ratio is 1.10. Under stress, resources fall by 30 from the plan’s central amount and the requirement rises by 25 from its central amount; the stress minimum is 1.00. Does the plan meet both conditions? (Answer: No; its central ratio is $620/570=1.0877$, below 1.10, and its stress ratio is $590/595=0.9916$, below 1.00.)
A hypothetical insurer has resources of 180 and a requirement of 150. Without reinsurance, a stress reduces resources by 18 and increases the requirement by 12. A proposed contract costs 3, reduces the stress loss by 12, and reduces the stress-related requirement increase by 4. Assume full and timely recovery. If the stressed minimum ratio is 1.05, does the contract meet it? (Answer: Yes; stressed resources are $180-3-(18-12)=171$, the stressed requirement is $150+(12-4)=158$, and the ratio is $171/158=1.0823$.)
A proposed action meets an insurer’s central ratio threshold but falls below its stress threshold. Name two decision-support responses management could consider, and one follow-up item that could be monitored if the action proceeds. (Answer: Possible responses include reducing or phasing the action, adding a suitable risk mitigation, or declining it; follow-up could monitor a specified risk exposure, solvency indicator, or stress-sensitive assumption.)

What You'll Learn