Optimal Reinsurance Retention: A Balance Sheet Guide

How to Determine the Optimal Reinsurance Retention for Your Balance Sheet

Last reviewed: November 2026

The optimal reinsurance retention is the retention that minimises the total economic cost of risk, which is the reinsurance margin you pay plus the cost of the capital you hold, while keeping net losses within your earnings, capital and liquidity limits. It is a balance sheet decision, not a purchasing one.

Too low a retention pays reinsurers to carry losses the balance sheet could absorb cheaply. Too high a retention saves premium but ties up capital and exposes earnings to shocks the board never agreed to.

This guide is for CFOs, CROs, chief actuaries and heads of reinsurance at cedents and reinsurers buying retro. It sets out a practical method, a worked comparison and the mistakes that distort the answer.

Key Takeaways

• The optimal reinsurance retention balances the ceded margin paid to reinsurers against the cost of capital needed to retain risk.
• Pure economics gives a starting point, but earnings, rating and liquidity constraints often set the final retention.
• Lower layers carry more expected loss and less capital relief, so they are often the least efficient to buy.
• The cost of capital assumption can move the optimum materially, so test it rather than assume it.
• Counterparty risk and diversification change how much capital a layer really releases.
• The board should set the risk tolerance; the reinsurance team should show its price.

What Is the Optimal Reinsurance Retention?

The optimal reinsurance retention is the point at which keeping one more unit of risk costs more in capital than it saves in reinsurance margin. Below that point, retaining more is cheaper; above it, buying more is cheaper.

The answer must then pass a set of constraints. A retention that is economically efficient but could wipe out half a year's profit in a moderate event may still be unacceptable.

Retention applies to every structure. It is the per-risk or per-event retention on excess of loss, the retained share on quota share, and the attachment on aggregate covers.

What Drives the Right Retention?

Six factors drive the right retention. The first three are about the balance sheet, and the last three are about the reinsurance market and the portfolio.

Capital Strength

A larger capital base can absorb more retained loss without breaching solvency or rating thresholds. As capital grows, the optimal retention usually rises.

Earnings Tolerance

Investors and boards care about the volatility of annual results, not just solvency. The retention should keep a plausible bad year within an agreed tolerance.

Liquidity

Large retained losses must be paid in cash before recoveries arrive. Retention should reflect how quickly the balance sheet can fund claims.

Reinsurance Pricing

When reinsurance margins are high, retaining more becomes attractive. When they soften, buying lower can become efficient again.

Diversification

A retained catastrophe loss may need less marginal capital if it diversifies against other risks. The diversified capital effect, not standalone capital, is what matters.

Counterparty Quality

Capital relief is reduced by the credit charge on recoverables. A cheap panel with weaker security releases less capital than the gross figures suggest.

How Do You Measure the Economic Cost of Reinsurance?

The economic cost of a reinsurance layer is its ceded margin, which is premium minus expected recoveries. The economic benefit is the capital it releases multiplied by the firm's cost of capital.

A layer creates value when its capital benefit exceeds its ceded margin. Comparing these two figures across retention options gives a clear, defensible starting point.

Two refinements matter in practice:

1. Use diversified marginal capital. Measure how much the total capital requirement changes, not the layer's standalone capital.
2. Include all costs. Add reinstatement premiums, brokerage, counterparty capital and the administrative cost of the treaty.

Worked Example: Comparing Retentions on a Cat Programme

The example below is illustrative and simplified, using USD millions. An insurer is choosing the retention for its catastrophe excess of loss programme, with a fixed exit point of 550.

Assumptions:

• Gross 1-in-200 event loss: 600, so 50 is retained above the exit point
• Net 1-in-200 loss = retention + 50
• Marginal diversified capital = 60% of the net 1-in-200 loss
• Cost of capital: 10% a year
• One event a year, with reinstatements ignored for simplicity

Step 1: Layer pricing

Step 2: Total economic cost by retention

On economics alone, a retention of 100 is optimal at 19.0. Raising it to 150 saves 2.5 of margin but adds 3.0 of capital cost.

Step 3: Apply the earnings constraint

The board will accept a net loss of no more than 75 from a 1-in-10 event, which is 25% of planned pre-tax profit of 300. The gross 1-in-10 event loss is 120, so the net 1-in-10 loss is the lower of 120 and the retention.

The constrained optimum is a retention of 75. The earnings constraint costs 19.5 − 19.0 = 0.5 a year compared with the unconstrained optimum.

Step 4: Test the cost of capital

At an 8% cost of capital, the capital costs become 3.6, 4.8, 6.0, 7.2 and 9.6. Total costs become 20.4, 19.2, 18.0, 17.2 and 17.1, so the unconstrained optimum moves up to 150.

The lesson for boards and reinsurance buyers: the economic optimum depends heavily on the cost of capital and diversification assumptions. The earnings constraint then decides the final answer, and its price, here 0.5 a year, is what the board should consciously accept or reject.

Which Constraints Should Override Pure Economics?

Four constraints commonly override the economic optimum. Each should be explicit, approved by the board and tested every year.

If no constraint binds, the economic optimum stands. If one binds, the constraint, not the economics, sets the retention.

When Should You Raise or Lower Retention?

Retention should move when the balance sheet, the portfolio or the reinsurance market moves. The table maps common triggers to the usual direction.

Movements should be tested, not assumed. A scenario view of exposure, claims and capital side by side, as described in treaty renewal scenario rooms, makes the trade-offs visible.

Which Mistakes Lead to the Wrong Retention?

Six mistakes repeatedly push retentions away from the optimum. Most come from looking at reinsurance price without capital, or capital without constraints.

How Should Boards Run a Retention Review?

A retention review should give the board a small number of options, each with its economic cost and its effect on constraints. The reinsurance team prices the options; the board chooses the tolerance.

Steps for an Annual Retention Review

1. Confirm risk appetite. Agree the earnings, solvency, rating and liquidity tolerances before any pricing.
2. Model gross losses. Build per-event and annual aggregate distributions from current exposure.
3. Price options. Obtain indicative pricing for several retentions, not just the expiring one.
4. Measure capital. Calculate diversified marginal capital for each option, including counterparty risk.
5. Compare total cost. Combine ceded margin and capital cost for each option.
6. Apply constraints. Remove options that breach tolerances and show the price of each binding constraint.
7. Decide and record. Document the rationale so the next review starts from evidence.

What Data Does the Review Need?

Retention is one part of a wider buying strategy linked to capital objectives. That wider view is covered in why CFOs and CROs need one view of reinsurance buying.

How Do Capital Frameworks Affect the Retention?

Each capital framework measures net losses differently, so the same retention can look different under each. The board should see all relevant views together.

Solvency II and Solvency UK

Under Solvency II, the Solvency Capital Requirement is calibrated to a 99.5% value at risk over one year. A higher retention increases the net loss in that scenario and therefore the capital requirement, while reducing counterparty default risk.

The UK applies its own version, Solvency UK, overseen by the PRA. Internal model firms see the retention effect directly in their own loss distributions.

Rating Agencies

Rating agency models stress net losses at high return periods and charge for the credit risk of recoverables. A retention that suits regulatory capital can still strain rating capital, an issue covered in rating-agency capital surprises.

IFRS 17

Under IFRS 17, reinsurance contracts held are measured separately from the underlying contracts. Retention choices therefore change both reported volatility and the presentation of reinsurance results.

Counterparty Concentration

Lower retentions increase recoverables and therefore exposure to reinsurers. Shared ownership and asset managers across the panel can concentrate that risk, as discussed in reinsurance counterparty contagion.

What Is the Market Context for Retention Decisions?

Since the 2023 market reset, many cedents have carried materially higher catastrophe retentions. With capacity returning and pricing softening through 2025 and 2026, some are now testing whether lower retentions have become economic again.

The right response is not to follow the market. It is to rerun the same economic and constraint tests with current pricing, capital and exposure.

Checklist for CFOs, CROs and Heads of Reinsurance

Use this checklist before you set or renew a retention.

• The board has approved explicit earnings, solvency, rating and liquidity tolerances
• Gross per-event and aggregate loss distributions reflect current exposure
• Indicative pricing covers several retention options, not just the expiring one
• Capital relief is measured on a diversified, marginal basis
• Counterparty risk and collateral are included in capital relief
• The cost of capital assumption has been tested for sensitivity
• The price of each binding constraint is shown to the board
• Treaty terms and historic recoveries come from a controlled system rather than spreadsheets

For the last point, a reinsurance treaty management platform keeps treaty terms, premium and settlement calculations consistent across renewal years.

Key Terms

Retention: The amount of loss the cedent keeps before reinsurance responds, per risk, per event or in aggregate.

Optimal retention: The retention that minimises total economic cost while staying within the board's risk tolerances.

Ceded margin: Reinsurance premium minus expected recoveries, the economic price paid for reinsurance.

Cost of capital: The annual return the firm requires on capital held to support retained risk.

Marginal diversified capital: The change in the firm's total capital requirement caused by retaining or ceding a layer, after diversification.

Exit point: The top of the reinsurance programme, above which losses are retained again.

Return period: The average interval between losses of a given size, such as 1-in-10 or 1-in-200.

Earnings tolerance: The maximum net loss from a defined bad year or event that the board is prepared to accept.

Sources

• Directive 2009/138/EC (Solvency II)
• Bank of England: Solvency II and Solvency UK
• IFRS 17 Insurance Contracts, IFRS Foundation
• EIOPA: Solvency II resources