Property Per-Risk vs. Cat XL: Choosing Reinsurance Architecture
Property Per-Risk vs. Cat XL: Choosing the Right Reinsurance Architecture
By Hitul Mistry | Last reviewed: January 2026
Few decisions shape a property insurer's earnings volatility as much as how it structures its reinsurance. The same book can look stable or explosive depending on whether the program is built around per-risk protection, catastrophe cover, proportional treaties, or a blend of all three. The stakes have risen sharply: insured natural-catastrophe losses have exceeded USD 100 billion for several consecutive years, with 2024 again topping that mark (Swiss Re Sigma, 2025), while reinsurers pushed catastrophe attachment points materially higher through the recent hard market, leaving cedents to retain more frequency (Gallagher Re, 2025). In that environment, choosing between property per-risk excess of loss and catastrophe XL is no longer a technicality — it is a capital-efficiency decision that determines which losses a cedent keeps and which it passes on.
The confusion is understandable because the two structures look similar on paper: both are non-proportional, both attach above a retention, both use limits and reinstatements. But they solve different problems. Per-risk XL answers "what if one big risk fails?" while cat XL answers "what if one event hits many risks at once?" Getting the mix wrong leaves a gap that surfaces at the worst possible moment.
What is the core difference between per-risk XL and cat XL?
Per-risk excess of loss protects the severity of a single large loss on one risk, while catastrophe excess of loss protects the accumulation of many losses from a single event. The trigger and the loss aggregation logic are what separate them.
1. What triggers each structure
- Per-risk XL responds when a loss on one policy or risk exceeds the retention, regardless of cause.
- Cat XL responds only when a defined event, such as a windstorm or earthquake, causes losses across multiple risks above the retention.
2. How losses aggregate
- Per-risk sums the loss from a single insured location or policy.
- Cat XL sums losses from all affected risks within an event, usually bounded by an hours clause.
3. What each protects against
- Per-risk guards against a single major fire, explosion, or collapse at one site.
- Cat XL guards against correlated damage from a natural peril hitting a whole portfolio.
4. Where the gap appears
- A one-building total loss is too small for cat XL but can breach per-risk.
- A widespread hail event may leave each risk under the per-risk retention yet aggregate to a large cat loss.
When does each structure fit a property portfolio?
The right architecture depends on whether the book's tail is driven by individual large exposures or by geographic accumulation, and most real portfolios need elements of both. Diagnosing the dominant driver is the first step in program design.
1. Portfolios with large single risks
- Industrial, energy, and commercial property books with high per-location values need robust per-risk cover.
- The largest single insured value often sets the required per-risk limit.
2. Portfolios with geographic concentration
- Personal lines and SME books clustered in cat-exposed zones need cat XL as the priority.
- Aggregation from a single storm, not any one risk, drives the tail.
3. Mixed commercial books
- Most portfolios carry both, requiring layered per-risk plus a cat program on top.
- Per-risk recoveries can feed into cat event losses, so interaction must be modelled.
4. New or growing books
- Proportional treaties often lead while experience and data mature.
- Non-proportional cover is layered in as the cedent gains confidence in its retained profile.
How do proportional treaties compare to non-proportional cover?
Proportional treaties share premiums and losses by a defined proportion of each risk, whereas non-proportional cover responds only above a retention. Cedents often combine them, using proportional capacity to write business and excess of loss to cap volatility.
1. Quota share
- Cedes a fixed percentage of every risk and its premium.
- Simple, provides capacity and ceding commission, but shares even profitable small risks.
2. Surplus treaty
- Cedes only the portion of each risk above the cedent's retention line.
- Lets the cedent keep small risks net while ceding the top of larger ones, homogenising the retained book.
3. Non-proportional per-risk and cat XL
- Retention and limit replace proportional sharing.
- Cedent keeps all premium net of XL cost and pays only for the tail protection.
The table summarises how the main property structures behave.
| Structure | Responds to | Basis | Best suited to |
|---|---|---|---|
| Quota share | Every risk, proportionally | Proportional | Capacity, new books, smoothing results |
| Surplus | Portion of each risk above retention line | Proportional | Heterogeneous large-risk portfolios |
| Per-risk XL | Large loss on a single risk | Non-proportional | Severity from individual big exposures |
| Cat XL | Accumulated losses from one event | Non-proportional | Geographic catastrophe accumulation |
| Aggregate XL | Accumulated retained losses over the year | Non-proportional | Frequency of medium losses or events |
How are working layers, attachment, and limits designed?
Attachment points are set above routine large losses and limits are sized to the loss the cedent must survive, with working layers absorbing the more frequent hits in between. The design balances premium cost against retained volatility.
1. Working layers
- Lower per-risk layers expected to be hit regularly by moderate large losses.
- Priced on burning-cost or exposure-rated methods and often carrying paid reinstatements.
2. Higher and clash layers
- Upper layers respond to rarer, more severe single-risk losses.
- Some layers address clash, where one occurrence involves multiple policies or lines.
3. Attachment calibration
- Set above attritional and routine large losses using loss history and exposure curves.
- Informed by probable maximum loss (PML) and modelled return periods.
4. Limit sizing
- Per-risk limit benchmarked to the largest single insured value or a high-percentile single-risk loss.
- Cat limit sized to a target return-period event, such as a 1-in-200-year occurrence loss.
How do reinstatements and aggregate covers manage frequency?
Reinstatements restore an eroded layer for additional premium, while aggregate covers protect the accumulation of retained losses across a year. Together they address the risk of multiple events rather than a single large one.
1. Reinstatement provisions
- Determine how many times a layer can respond in the treaty year.
- Free, paid, or pro-rata reinstatements change both cost and the effective annual limit.
2. Occurrence versus aggregate limits
- Occurrence limits cap any single event or risk loss.
- Aggregate limits cap the total recoverable across the period, controlling frequency exposure.
3. Aggregate excess of loss
- Protects the cedent's accumulated net losses over an agreed threshold for the year.
- Backstops a season of several medium catastrophes that individually stay within retention.
4. Multi-year and structured features
- Some programs use multi-year terms or profit-sharing to stabilise pricing.
- Structured features can smooth results but must respect risk-transfer requirements.
How do analytics and AI shape modern program design?
Data quality and modelling now determine how well a program matches the underlying risk, and analytics let cedents test structures before committing capital. Better exposure insight directly improves capital efficiency.
1. Exposure and catastrophe modelling
- Vendor and internal cat models quantify PML, correlation, and return-period losses by peril and region.
- Model output drives cat XL attachment, limit, and pricing decisions.
2. Portfolio optimisation
- Scenario testing compares how alternative retentions and layers affect volatility and cost of capital.
- Cedents can visualise the trade-off between retained earnings volatility and reinsurance spend.
3. Data enrichment and exposure coding
- AI accelerates cleansing, geocoding, and occupancy classification across large property schedules.
- Cleaner exposure data reduces model uncertainty and negotiation friction at renewal.
4. Capital and solvency alignment
- Analytics link program structure to solvency capital relief and rating-agency expectations.
- Reinsurance is evaluated as a capital tool, not just loss protection.
How should cedents assemble a capital-efficient property program?
The most efficient programs layer proportional and non-proportional cover so each rand of premium buys the volatility reduction it targets, with no double-paying and no uncovered gap. Coordination between the layers matters as much as any single layer's terms.
1. Start from the retained risk profile
- Define the earnings volatility and single-loss shock the balance sheet can absorb.
- Let that appetite drive retentions rather than market convention.
2. Layer structures deliberately
- Use proportional cover for capacity and smoothing, per-risk for severity, cat XL for accumulation.
- Add aggregate cover where frequency, not severity, is the real threat.
3. Coordinate interactions
- Model how per-risk recoveries feed cat event losses to avoid gaps or overlaps.
- Align reinstatements and hours clauses across layers.
4. Review dynamically
- Re-test structure as exposure, inflation, and market pricing shift.
- Treat program design as an annual optimisation, not a static renewal.
Frequently Asked Questions
What is the difference between per-risk XL and cat XL?
Per-risk excess of loss responds to a large loss on a single risk or policy, while catastrophe excess of loss responds when one event causes losses across many risks at once. Per-risk protects severity on individual accounts; cat XL protects accumulation from a single occurrence.
When does a cedent need per-risk cover rather than cat XL?
When individual large exposures, such as a single high-value industrial site, can produce a loss big enough to threaten results on their own. Cat XL would not respond to a one-building fire, so per-risk fills that severity gap.
What is a working layer?
A working layer is a lower per-risk excess layer that is expected to be hit relatively frequently by moderate large losses. It is priced on a burning-cost or exposure basis and often carries limited paid reinstatements.
How do quota share and surplus proportional treaties differ?
Quota share cedes a fixed percentage of every risk, so the cedent and reinsurer share results proportionally on all business. Surplus cedes only the portion of each risk above the cedent's retention line, letting the cedent keep smaller risks net and homogenise the retained book.
How are attachment points and limits set?
Attachment is set above the level of expected attritional and routine large losses, informed by loss history, exposure curves, and probable maximum loss. Limits are sized to the largest single-risk or single-event loss the cedent needs to survive, often benchmarked to a modelled return period.
What are reinstatements and why do they matter?
Reinstatements restore the cover after a loss erodes the limit, usually for an additional premium. They determine how many times a layer can respond in a treaty year, which is critical for both frequent per-risk layers and multi-event catastrophe seasons.
How do aggregate covers fit alongside per-risk and cat XL?
Aggregate excess of loss protects the cedent's accumulated retained losses over the year rather than any single loss or event. It backstops frequency, such as a season of several medium catastrophes or numerous large risk losses that individually stay within retention.
How do analytics and AI improve property program design?
Exposure and catastrophe models quantify PML, correlation, and return-period losses, while portfolio analytics test how alternative structures affect volatility and capital. AI accelerates data cleansing, exposure coding, and scenario testing across large property schedules.
Editorial note: The statistics referenced here come from public industry research and are provided to illustrate market dynamics rather than to prescribe any specific program. Optimal structure depends on each cedent's exposure, appetite, and regulatory environment. InsurNest does not guarantee any particular underwriting, capital, or financial outcome.
Sources
- Swiss Re Sigma — Natural catastrophes and insured losses
- Gallagher Re — First View and Reinsurance Market Report
- Guy Carpenter — Property and catastrophe reinsurance insights
- Aon — Reinsurance Market Dynamics
- S&P Global Ratings — Global reinsurance sector outlook
- Verisk / AIR — Catastrophe modelling and exposure management
- Lloyd's — Property treaty and excess of loss market
The right property program is not the biggest one — it is the one where per-risk, cat XL, and proportional cover each earn their place, and InsurNest analytics prove it before renewal.
Visit InsurNest to learn more.
