Event Scenario Design for Catastrophe Management

Why Design an Event Scenario?

Catastrophe decisions often require a clear account of what event is being considered and which parts of the portfolio it affects. A scenario makes those assumptions explicit. It can reveal a geographic concentration, show how a proposed policy changes loss under a specified event, or help compare mitigation options.

An event scenario is not, by itself, a prediction that an event will occur, nor does a single scenario establish its annual probability. It is a structured description of a plausible event and its consequences under stated assumptions. Scenario losses are useful for decision-making precisely because the event, affected exposures, and loss calculations can be inspected and challenged.

Definition: An event scenario is a specified hypothetical catastrophe event, described by its peril (the event type), footprint (the area affected), intensity (the event’s severity across that area), and affected exposure (the insured property or other portfolio value exposed to the event). A scenario loss is the loss calculated for the affected exposures under those specified event conditions and the stated loss assumptions.

A well-constructed scenario answers four questions. The peril might be earthquake, flood, or windstorm. The footprint identifies the locations reached by the event. Intensity describes how severe the event is at each location or zone. Affected exposure identifies the values in the portfolio at those locations. A loss calculation then connects the intensity to estimated damage and, if relevant, to contract terms.

For a simple property scenario, suppose each location $j$ has insured value $V_j$, and the event produces a damage ratio $r_j$ at that location. The ratio is the proportion of insured value expected to be lost under the scenario. If there are no deductibles, limits, business-interruption losses, or other adjustments, the scenario loss is:

L_{\text{scenario}}=\sum_j V_j r_j

The damage ratio is an assumed or estimated translation from event intensity to loss; it is not the intensity itself. For example, wind speed is not a damage ratio. The method requires consistent exposure values and ratios, and its result applies only to the loss components included in the assumptions.

From Event Description to Portfolio Loss

Scenario design should be concrete enough that another reviewer can reproduce the calculation. A useful sequence is to define the event, map its footprint, assign intensity, identify exposed values, and calculate loss using assumptions that match the decision being considered.

Specify the peril and event.
State what happens, such as a hypothetical earthquake of a specified severity or a hypothetical flood affecting a defined watershed. Avoid vague labels such as “severe catastrophe” without a physical event description.

Define the footprint.
Identify which locations or zones are affected and which are outside the event footprint. The footprint should align with the geographic units used to summarize the portfolio.

Assign intensity and loss response.
Describe event severity by location, then state the assumed damage ratio or another explicit loss function for each affected area. Do not apply one ratio everywhere unless that is the stated assumption.

Identify affected exposure and calculate.
Use the exposure actually located within each intensity area. Multiply each exposure by its corresponding damage ratio, apply any included policy terms, and add the resulting losses.

This is a scenario calculation, not a replacement for a full catastrophe model. Its usefulness depends on whether its assumptions are suitable for the decision. A simple calculation may omit vulnerability differences, policy wording, demand surge, or secondary losses. Those omissions should be stated rather than silently treated as zero.

Worked Example: Measuring a Hypothetical Earthquake Accumulation

An insurer is reviewing a hypothetical earthquake affecting three zones. Zone A has $80 million of insured property value and a scenario damage ratio of 20%; Zone B has $50 million and a ratio of 10%; Zone C has $30 million and is outside the footprint. Assume the ratios apply to insured property values, and ignore deductibles, limits, business interruption, and all other loss components. Calculate the scenario loss and the effect of adding $20 million of insured value in Zone B.

Method. Match each zone’s exposure to the ratio for the event intensity in that zone. An out-of-footprint zone has a zero ratio under this scenario.

Zone A loss is $80 \times 0.20 = 16$ million dollars. Zone B loss is $50 \times 0.10 = 5$ million dollars. Zone C loss is $30 \times 0 = 0$ million dollars. Therefore:

L_{\text{scenario}}=16+5+0=21\text{ million dollars}

Total insured value is $80+50+30=160$ million dollars. The scenario loss as a proportion of total insured value is $21/160=0.13125$, or 13.125%.

Now add $20 million of insured value in Zone B. Its incremental scenario loss is $20 \times 0.10=2$ million dollars. The revised scenario loss is $21+2=23$ million dollars, and revised total insured value is $160+20=180$ million dollars. The revised scenario-loss proportion is $23/180=0.127777\ldots$, or 12.7778% to four decimal places.

Check. The loss increases by $2 million, as expected for new exposure in a zone with a 10% ratio. The loss proportion falls because the new exposure is subject to a lower ratio than the portfolio’s original average. A common mistake is to apply Zone A’s 20% ratio to all $180 million; that would give $36 million, incorrectly treating every location as having Zone A’s intensity.

Using Scenarios for Underwriting Decisions

A scenario can provide a consistent measure of how much a proposed risk adds to a particular event loss. It does not, on its own, determine whether a policy should be written: that decision can also depend on price, contractual terms, portfolio objectives, and other constraints. The scenario contribution is most informative when the proposed exposure is mapped to the same footprint and intensity assumptions as the existing portfolio.

Worked Example: Setting a Scenario-Based Line Size

A hypothetical coastal portfolio has $40 million of insured value in a hurricane footprint, subject to a 30% scenario damage ratio, and $25 million inland, subject to an 8% ratio. Another $35 million is outside the footprint. Assume no deductibles, limits, or other loss components. The underwriting team will not allow this scenario loss to exceed $15.5 million. A proposed coastal risk has an insured value of $10 million and would have the same 30% ratio. Determine whether to accept the full risk, and calculate the largest line that meets the scenario limit.

Method. Calculate the existing portfolio’s scenario loss first, then calculate the incremental loss from the proposed risk. For a maximum permitted loss $M$, existing loss $L_0$, and proposed exposure $V$ with ratio $r$, the maximum exposure satisfying the limit is $(M-L_0)/r$, provided $r>0$ and $M\geq L_0$.

The existing coastal loss is $40 \times 0.30=12$ million dollars. The inland loss is $25 \times 0.08=2$ million dollars, and the outside-footprint loss is zero. Thus, $L_0=12+2=14$ million dollars.

The full proposed risk adds $10 \times 0.30=3$ million dollars. Accepting it in full would produce a scenario loss of $14+3=17$ million dollars, which exceeds the $15.5 million limit by $17-15.5=1.5$ million dollars. The full line does not meet the stated constraint.

The available loss capacity is $15.5-14=1.5$ million dollars. At a 30% damage ratio, the largest additional insured value is $1.5/0.30=5$ million dollars. The scenario loss at that line is $14+(5 \times 0.30)=15.5$ million dollars, exactly the limit.

Check. The line is half the proposed $10 million value, so it adds half the full proposed loss, or $1.5 million. A common mistake is to treat the $1.5 million of remaining loss capacity as $1.5 million of acceptable insured value. The limit is stated in loss, so the capacity must be divided by the 30% damage ratio to obtain insured value.

Using Scenarios to Compare Mitigation

Mitigation changes the event’s estimated effect on exposure; it need not change the footprint or the amount insured. A scenario comparison should hold the event and exposure basis constant, then change only the assumptions affected by the mitigation measure. This isolates the scenario loss reduction attributable to that measure.

The difference between the pre-mitigation and post-mitigation scenario losses is a conditional reduction for the specified event. It is not an expected annual benefit unless event probabilities and the relevant range of outcomes are also considered. Similarly, a reduction in scenario loss does not by itself prove that an investment is economically worthwhile.

Worked Example: Comparing Two Hypothetical Flood Mitigations

A hypothetical portfolio has three flood-exposed properties. Property A has $12 million of insured value and a 50% scenario damage ratio; Property B has $8 million and a 25% ratio; Property C has $10 million and a 10% ratio. Assume these ratios already reflect the event intensity and ignore policy terms and other losses. Elevating Property A would reduce its ratio from 50% to 20% and cost $0.9 million. Installing barriers at Property B would reduce its ratio from 25% to 15% and cost $0.3 million. A $1.0 million budget allows only one measure. Calculate the scenario loss reduction for each measure and choose the larger reduction within the budget.

Method. Calculate the baseline loss for each property. Then recalculate only the property affected by a selected mitigation. The reduction is baseline scenario loss minus mitigated scenario loss.

Before mitigation, Property A’s loss is $12 \times 0.50=6$ million dollars, Property B’s is $8 \times 0.25=2$ million dollars, and Property C’s is $10 \times 0.10=1$ million dollars. The baseline scenario loss is $6+2+1=9$ million dollars.

With elevation at A, its loss becomes $12 \times 0.20=2.4$ million dollars. The new portfolio scenario loss is $2.4+2+1=5.4$ million dollars. The reduction is $9-5.4=3.6$ million dollars.

With barriers at B, its loss becomes $8 \times 0.15=1.2$ million dollars. The new portfolio scenario loss is $6+1.2+1=8.2$ million dollars. The reduction is $9-8.2=0.8$ million dollars.

Both measures individually fit the $1.0 million budget. Elevation at A produces the greater reduction under this scenario: $3.6 million rather than $0.8 million. The respective scenario loss reductions per dollar of cost are $3.6/0.9=4$ and $0.8/0.3=2.6667$, in millions of scenario-loss reduction per million dollars of cost.

Check. The portfolio’s unaffected losses remain in each mitigated total: changing Property A does not remove losses at B or C. The answer is a ranking for this specified event and the stated assumptions, not an estimate of annual savings. A common mistake is to subtract the mitigation cost from the scenario loss and call the result the mitigated loss; the cost is an investment, not an insured event loss.

Justification and Interpretation

The simple scenario-loss formula follows directly from the definition of a damage ratio. If $r_j$ is the fraction of value $V_j$ lost at location $j$, then the loss at that location is $V_jr_j$. The total loss across the portfolio is the sum of the location losses:

L_{\text{scenario}}=\sum_j V_jr_j

This addition is appropriate when the location losses are expressed in the same monetary units and the scenario calculation includes each loss once. It also explains why exposure location matters: moving the same insured value from a zone with ratio $r_1$ to one with ratio $r_2$ changes scenario loss by $V(r_2-r_1)$. The formula does not capture interactions or additional loss components unless they are included in the location-level loss estimates.

For a proposed exposure of value $V_{\text{new}}$ in a zone with ratio $r_{\text{new}}$, the incremental loss under the same assumptions is $V_{\text{new}}r_{\text{new}}$. Thus, before policy terms or other adjustments, the portfolio’s scenario loss changes from $L_0$ to $L_0+V_{\text{new}}r_{\text{new}}$. This is a transparent way to examine accumulation consequences, but only for the specified event and portfolio data.

Common Scenario Design Traps

Using a peril label without a footprint. “Hurricane” does not identify which locations are affected. Specify the geographic reach and the exposure included.
Applying one intensity or ratio to every location. If event severity varies by zone, use the corresponding zone-level assumptions. An out-of-footprint exposure is not automatically affected.
Confusing insured value with loss. Exposure is the value at risk; scenario loss also depends on the intensity-specific damage ratio and any applicable policy terms.
Mixing scenario loss with probability or expected loss. A scenario amount is conditional on the stated event. Without probability information, it is not an annual expected loss or a return-period estimate.
Changing multiple assumptions in a comparison. To attribute a change to underwriting or mitigation, hold the event and other assumptions constant unless the purpose is explicitly to compare different scenarios.
Leaving out important scope limits. State whether the calculation includes deductibles, limits, business interruption, or other components. An omitted item should not be mistaken for a modeled zero.

Key takeaway: A decision-useful catastrophe scenario links a defined peril and footprint to location-specific intensity, affected exposure, and explicit loss assumptions. Its calculated loss can support accumulation, underwriting, and mitigation decisions, but its conclusions remain conditional on the event and assumptions specified.

Check Your Understanding

For each problem, use the stated scenario assumptions only; do not infer an event probability or add loss components that are not specified.

A hypothetical windstorm affects Zone North, with $24 million of insured value and a 25% damage ratio. Zone South has $16 million of value and is outside the footprint. Ignoring policy terms, what is the scenario loss? (Answer: $6 million.)
A flood scenario produces a 12% damage ratio on $30 million of exposed property. A proposed location adds $5 million in that same area. What is the incremental loss and the revised scenario loss, assuming the existing scenario loss is $4 million? (Answer: $0.6 million incremental; $4.6 million revised.)
An earthquake scenario loss is $11 million before a proposed policy. The portfolio limit is $12.2 million, and the proposed location’s damage ratio is 20%. What is the maximum additional insured value that meets the limit? (Answer: $6 million.)
A hypothetical hail scenario has $10 million of exposure at a 40% damage ratio and $15 million at a 20% ratio. A retrofit reduces the first ratio to 25%, costs $0.5 million, and does not change the second exposure. What is the baseline loss, the mitigated loss, and the scenario loss reduction? (Answer: $7 million baseline; $5.5 million mitigated; $1.5 million reduction.)
A wildfire scenario loss is $8 million. A proposed risk adds $4 million of insured value at a 25% damage ratio. The underwriting limit for this scenario is $9 million. What is the revised loss, and does the full proposed risk meet the limit? (Answer: $9 million; yes, it meets the limit exactly.)

What You'll Learn