Why Catastrophe Loss Costs Belong in Property Pricing
A property rate indication must fund more than ordinary expected claims. It must also recognize the expected cost of catastrophe claims, the cost of transferring some of that risk through reinsurance, expenses, and the insurer’s required return on capital. A catastrophe loading that omits one of these items can understate the premium needed; one that counts the same cost twice can overstate it.
The calculations below use a hypothetical annual policy-year and express costs in dollars per policy unless stated otherwise. A policy-year is one unit of exposure. The examples are pricing illustrations, not real company or industry data. They assume the catastrophe loss costs supplied are suitable expected annual values for the rating basis being used. The earlier material on catastrophe results and gross-to-net catastrophe loss views provides context for distinguishing gross losses from reinsured outcomes.
The distinction between loss cost and premium loading matters. Expected catastrophe losses are costs per exposure, while variable expenses and a target premium margin are often expressed as proportions of premium. A dollar of expected loss therefore does not necessarily translate into exactly one dollar of additional indicated premium.
Build the Indication From Consistent Cost Components
For a simple annual indication, let \(A\) be attritional expected loss cost per exposure, \(C\) be gross expected catastrophe loss cost per exposure, \(H\) be reinsurance premium cost per exposure, and \(Q\) be expected reinsurance recoveries per exposure. The net reinsurance expense is \(H-Q\). Let \(e\) be fixed expense per exposure and \(K\) be the capital charge per exposure. Finally, let \(v\) be variable expenses as a proportion of premium and \(m\) be the required underwriting margin as a proportion of premium.
This form assumes that \(v\) and \(m\) are proportions of the indicated premium, that the numerator contains dollar costs per exposure, and that \(1-v-m>0\). It also assumes the stated cost components are on a consistent annual and exposure basis. If an expense or capital provision is already included in another component, it must not be added again.
Sum the attritional and gross catastrophe loss costs per exposure. Keep the peril components visible when the rating plan needs peril-specific indications.
Add the reinsurance premium and subtract expected recoveries, or equivalently add the net reinsurance expense \(H-Q\) to gross expected claims.
Include fixed expenses and an explicit capital charge, provided neither is already reflected elsewhere.
Divide the dollar costs by \(1-v-m\) to allow for variable expenses and target margin that are percentages of premium.
The capital charge \(K\) can be represented, for example, as a cost-of-capital rate \(r\) multiplied by allocated capital \(B\) per exposure: \(K=rB\). This is a pricing assumption, not an identity that determines the correct capital amount or rate. The choice of capital basis and cost-of-capital rate must be made consistently with the insurer’s pricing framework.
Why the Premium Formula Holds
The indication can be derived by equating premium to the costs and provisions it must fund. Expected claims, reinsurance expense, fixed expenses, and capital charge are dollar amounts per exposure. Variable expenses consume \(vP\), and the required underwriting margin is \(mP\). Thus the balance equation is:
Move the premium-proportional terms to the left and factor out \(P\):
When \(1-v-m>0\), division gives the indicated premium formula. This derivation also explains why simply adding \(v\) or \(m\) dollars to the loss cost is incorrect: those amounts depend on the premium being solved for.
Peril costs can be added because expected value is linear. If each peril’s expected annual loss cost is measured per the same exposure unit, the expected total catastrophe loss cost is the sum of those peril-level expectations. This does not require the peril losses to be independent. It does require the components to be defined consistently, with no duplicated losses.
Worked Example: Indicating a Property Rate With Two Perils
A hypothetical property program has annual attritional expected loss cost of $420 per policy. Its expected catastrophe loss costs are $96 for wind and $54 for hail. Reinsurance premium is $48 per policy and expected recoveries are $18 per policy. Fixed expense is $45 per policy. Allocated capital is $600 per policy and the assumed annual capital cost rate is 4%. Variable expenses are 20% of premium and the target underwriting margin is 5% of premium. Calculate the indicated premium and the catastrophe-related increment in premium.
Method. First calculate the gross catastrophe pure premium, then the net reinsurance expense and capital charge. Apply the premium gross-up only after all dollar costs have been combined.
The catastrophe pure premium is:
The net reinsurance expense is:
The capital charge is:
The dollar cost numerator is \(420+150+30+45+24=669\). The premium proportion remaining after variable expenses and margin is \(1-0.20-0.05=0.75\). Therefore:
To isolate the catastrophe-related increment under these same assumptions, use the catastrophe loss cost plus net reinsurance expense: \(150+30=180\). Its premium increment is \(180/0.75=240.00\). The remaining indicated premium is \(892.00-240.00=652.00\), which also equals \((420+45+24)/0.75=652.00\). The results reconcile. The method avoids treating the $150 catastrophe loss cost as if it were the full premium increment, while also avoiding counting expected recoveries twice.
Peril and Territory Loadings
A rate indication may need to vary by peril, territory, or both. The basic calculation is to determine each cell’s expected loss cost per exposure and then apply the relevant expense, reinsurance, and capital assumptions. A peril-specific loading should be labeled clearly: it may mean a pure premium amount, a factor relative to a base, or a contribution to the final indicated premium. Those quantities are not interchangeable.
Where costs are allocated across territories, the allocation basis should be stated. For example, a portfolio reinsurance expense may be allocated in proportion to gross catastrophe expected loss cost. That basis makes the allocated amounts sum to the portfolio total, but it is only one possible allocation convention. An allocation changes how costs are assigned to rating cells; it does not change the total portfolio cost.
Worked Example: Allocating Catastrophe and Reinsurance Costs by Territory
A hypothetical portfolio has 500 policy-years in each of Territory North and Territory South. Attritional expected loss cost is $360 per policy in both territories. Catastrophe pure premium is $120 in North and $240 in South. Total reinsurance expense for the portfolio is $90,000, allocated in proportion to territory catastrophe expected loss costs. Fixed expense is $30 per policy in each territory. Capital charges are $20 per policy in North and $40 in South. Variable expenses are 20% of premium and the target margin is 5%. Find the territory indications and check the portfolio average.
Method. Allocate the portfolio reinsurance expense using each territory’s share of total expected catastrophe losses, then calculate each territory’s premium using the common gross-up factor.
North catastrophe expected loss cost for the portfolio is \(500(120)=60{,}000\). South catastrophe expected loss cost is \(500(240)=120{,}000\). The total is \(180{,}000\), so the allocation shares are \(60{,}000/180{,}000=1/3\) for North and \(120{,}000/180{,}000=2/3\) for South. The allocated reinsurance expenses are therefore \(90{,}000(1/3)=30{,}000\) and \(90{,}000(2/3)=60{,}000\). Dividing each by 500 policies gives $60 per North policy and $120 per South policy.
The denominator for each territory is \(1-0.20-0.05=0.75\). North’s numerator is \(360+120+60+30+20=590\), so its indication is:
South’s numerator is \(360+240+120+30+40=790\), so:
The portfolio average is \((500(786.67)+500(1{,}053.33))/1{,}000=920.00\), using the unrounded territory results for the calculation. As a direct check, average the dollar numerators: \((590+790)/2=690\), then divide by 0.75 to obtain \(920.00\). The agreement confirms that the allocation preserves the portfolio total. Final indications are rounded to the nearest cent.
Worked Example: Deriving a Catastrophe Loading From Event Scenarios
A hypothetical portfolio has 2,000 policy-years. For a simplified annual model, there are three mutually exclusive outcomes: no modeled wind event, a wind event with probability 0.02 and gross portfolio loss of $1,500,000, or a second wind event outcome with probability 0.05 and gross portfolio loss of $400,000. The remaining probability is no event. Expected recoveries under the existing reinsurance arrangement are $600,000 in the first event and $100,000 in the second. The annual reinsurance premium is $25,000. Attritional loss cost is $300 per policy, fixed expense is $20 per policy, and capital is $100 per policy with an assumed capital cost rate of 6%. Variable expenses are 25% of premium and target margin is 5%. Calculate the catastrophe pure premium, net reinsurance expense, and indicated premium.
Method. Calculate expected gross loss and expected recovery by probability-weighting each event outcome. Convert portfolio expectations to per-policy values before using the indication formula.
Expected gross catastrophe loss for the portfolio is:
The catastrophe pure premium is \(50{,}000/2{,}000=\$25\) per policy. Expected recovery is \(0.02(600{,}000)+0.05(100{,}000)=12{,}000+5{,}000=\$17{,}000\), or \(17{,}000/2{,}000=\$8.50\) per policy. The reinsurance premium is \(25{,}000/2{,}000=\$12.50\) per policy, so net reinsurance expense is \(12.50-8.50=\$4.00\) per policy.
The capital charge is \(0.06(100)=\$6\) per policy. The dollar numerator is \(300+25+4+20+6=\$355\). The denominator is \(1-0.25-0.05=0.70\). Thus:
The indicated premium is $507.14 per policy, rounded to the nearest cent. The catastrophe loss cost and reinsurance expense together contribute \((25+4)/0.70=\$41.43\) to the indicated premium. The check is that the other components contribute \((300+20+6)/0.70=\$465.71\); $41.43 plus $465.71 is $507.14 after rounding.
Common Pricing Traps
The most frequent errors arise from mixing gross and net loss views or from treating premium-based provisions as dollar costs. In the event-scenario example, one incorrect calculation would add the $25 catastrophe pure premium and the $12.50 reinsurance premium, producing a $37.50 catastrophe-related cost per policy before gross-up. That omits the $8.50 expected recovery. The correction is \(25+(12.50-8.50)=29\) dollars per policy before gross-up, giving a $41.43 premium increment after division by 0.70.
Another incorrect approach is to subtract expected recoveries from gross claims and then add net reinsurance expense. That counts the expected recovery twice. The consistent alternatives are:
- Use gross expected claims and add reinsurance premium less expected recoveries; or
- Use net expected claims after recoveries and add the full reinsurance premium.
Both alternatives produce the same total before other costs. Do not combine one part of the first method with one part of the second. Also, do not divide one component by a premium gross-up factor while leaving the other components ungrossed if the goal is a total indicated premium on one consistent basis.
Check Your Understanding
Use the stated assumptions in each hypothetical problem. Give dollar rates to the nearest cent.
- A policy has expected annual catastrophe loss costs of $62 for wildfire and $38 for convective storm. What is its catastrophe pure premium per policy? (Answer: $100.00 per policy)
- A portfolio has 1,200 policy-years, expected gross catastrophe losses of $72,000, expected recoveries of $18,000, and reinsurance premium of $30,000. Find gross catastrophe pure premium and net reinsurance expense per policy. (Answer: $60.00 catastrophe pure premium and $10.00 net reinsurance expense per policy)
- For a territory, attritional expected losses are $500 per policy, catastrophe expected losses are $80, net reinsurance expense is $20, fixed expense is $40, and capital charge is $30. Variable expenses are 22% of premium and target margin is 3%. Find the indicated premium. (Answer: $893.33)
- Two territories each have 100 policies. Their catastrophe pure premiums are $50 and $150 per policy. A $12,000 portfolio reinsurance expense is allocated in proportion to catastrophe expected loss cost. Find the reinsurance expense allocated per policy to each territory. (Answer: $30.00 for the first territory and $90.00 for the second)
- A hypothetical portfolio has 1,000 policy-years. Two mutually exclusive catastrophe outcomes have probabilities 0.03 and 0.04 and gross losses of $900,000 and $250,000, respectively. Recoveries are $300,000 and $50,000 in those outcomes. Reinsurance premium is $15,000. Find the catastrophe pure premium and net reinsurance expense per policy. (Answer: $37.00 catastrophe pure premium and $4.00 net reinsurance expense per policy)