The Financial Module

From Physical Damage to Insured Loss

The vulnerability module produces estimates of physical damage — damage ratios for buildings, contents, and time element losses. The financial module takes these estimates and converts them into insured losses by applying the contractual terms of insurance policies. This is where the cat model crosses from physical science into financial risk management.

Ground-Up Loss

The starting point is ground-up loss — the total physical loss to a structure before any insurance policy terms are applied. If a building worth $1 million suffers a 30% damage ratio, the ground-up loss is $300,000. This represents the total economic cost of physical damage, regardless of how it is financed.

Applying Policy Conditions

From ground-up loss, the financial module applies the specific terms of each insurance policy to calculate the insured loss. Key policy conditions include:

• Deductibles: The amount the policyholder pays before insurance responds. Deductibles reduce the insurer's exposure on smaller losses but have limited effect on very large events.
• Coverage limits: The maximum amount the insurer will pay. A policy with a $500,000 limit on a $1 million building will not pay more than $500,000 regardless of the damage.
• Sub-limits: Caps on specific types of damage within a policy — for example, a sub-limit on contents or on flood damage within a multi-peril policy.
• Coinsurance: A shared loss proportion, requiring the insured to maintain coverage equal to a specified percentage of replacement value.
• Attachment points and limits: For single or multiple location policies, the level at which coverage begins and the maximum covered.

Reinsurance Terms

For reinsurance analysis, the financial module applies an additional layer of financial terms — the reinsurance contract conditions that determine how losses are shared between the primary insurer (the cedant) and the reinsurer. These include the attachment point (where reinsurance cover begins) and the limit (where it exhausts). The output at this stage is the net loss to the cedant after reinsurance recoveries.

Validation Against Actual Loss Data

Estimates of insured loss are validated — and damage functions fine-tuned — using loss data from actual catastrophe events. This is particularly valuable for wind perils, where loss data are relatively plentiful. Actual event loss data typically consists of claims and paid losses by ZIP code and line of business, and sometimes by construction type and coverage. Such data allows modellers to check whether the model's financial module is producing results consistent with observed insured losses, and to identify systematic biases.

Loss Metrics Produced

From the full stochastic catalog of events and the financial module calculations, the model produces probability distributions of losses that can be summarised in several ways:

• The Exceedance Probability (EP) curve — showing the probability that losses exceed various thresholds in a given year
• The Annual Average Loss (AAL) — the expected loss averaged over the simulation period
• Return period losses (e.g. the 1-in-100, 1-in-200 year loss)
• Loss distributions by geography, line of business, construction class, and coverage

These metrics form the basis for all downstream business decisions — from premium setting to reinsurance purchasing to regulatory capital assessment.