Retention Limit: Definition and How It Works

A retention limit is the maximum amount of loss that a business, self-insured entity, or insurer agrees to absorb before coverage from an excess insurer, reinsurer, or stop-loss carrier begins to apply. Below the retention, the risk-bearing party pays losses directly from its own funds; above the retention, the excess layer pays. The term appears in several distinct but related contexts: self-insured retentions (SIRs) in commercial liability policies, retained risk layers in captive insurance programs, and retention limits in reinsurance treaties. In all contexts, the core concept is the same — it defines the boundary between self-funded and insured risk. See also the companion concept Aggregate Limit, which caps total payments across all occurrences, while the retention limit determines who pays first for each occurrence.

Self-Insured Retention (SIR) vs. Deductible

The two most common mechanisms for retaining risk in a commercial policy are the deductible and the SIR, and they are meaningfully different:

With a deductible, the insurer defends and pays the claim from dollar one, then bills the insured for the deductible amount — up to a per-claim or per-occurrence deductible limit, and sometimes an annual aggregate deductible cap. The insured is in a creditor relationship with the insurer.

With an SIR, the insured handles and pays claims within the retention itself. The insurer's policy is excess — it does not attach until the SIR is satisfied. This gives the insured more control over defense strategy and settlement decisions below the SIR, but also exposes it to the full cost of each claim up to the SIR amount. SIRs are common in professional liability (E&O) and directors and officers (D&O) policies for large organizations.

Retention in Captive Insurance

In a captive insurance structure, the retention limit defines how much risk the captive retains before ceding to a reinsurer. A single-parent captive might retain the first $500,000 of each occurrence, with a $1,000,000 per-occurrence cap and $5,000,000 aggregate, and purchase excess reinsurance above those retentions. The retention level is a core actuarial decision — too high and the captive's capital is at risk; too low and reinsurance premiums eliminate the cost savings that motivated the captive.

The retention limit in a captive context often has two dimensions:

• Per-occurrence (per-risk) retention: The maximum retained for any single event
• Aggregate retention: A stop-loss aggregate that caps the total retained losses across all occurrences in the captive year

See Captive Insurance Strategy for the full framework of how retentions are structured in captive programs.

Retention in Reinsurance

For insurance carriers, the retention limit (or "retention") in a reinsurance treaty is the amount the ceding insurer keeps for its own account before the reinsurer's treaty attaches. In a quota share treaty, the retention is expressed as a percentage — the ceding insurer keeps 20% of every loss, and the reinsurer takes 80%. In an excess of loss treaty, the retention is a dollar threshold — the ceding insurer pays losses up to $500,000 per occurrence, and the reinsurer pays the excess above that per-occurrence retention.

Stop-Loss Retention in Self-Funded Health Plans

In self-funded employee health plans, the retention limit concept appears as the stop-loss attachment point — the dollar amount of claims the employer absorbs per covered employee (specific stop-loss) or across the entire plan (aggregate stop-loss) before the stop-loss carrier reimburses. See Stop-Loss Insurance for Self-Funded Plans for how specific and aggregate stop-loss attachment points interact to cap employer health plan exposure.

Determining the Right Retention Level

The optimal retention level balances three factors:

1. Financial capacity: The retained layer should not exceed the organization's ability to fund losses from operating cash flow without material disruption. A common rule of thumb is that the per-occurrence retention should not exceed 5% of annual revenue for well-capitalized businesses.

2. Premium savings: Higher retentions reduce the insured portion of the risk and typically reduce premiums proportionally. The expected value of the premium savings should exceed the expected value of additional retained losses — if not, the higher retention is not economically justified.

3. Administrative cost: Higher retentions require claims management infrastructure. If the organization lacks claims handling capability, a high SIR may produce poor loss outcomes that offset the premium savings.

Related Terms

• Aggregate Limit — the total payout cap for the policy year; the retention limit determines who pays below the policy, while the aggregate caps what the policy pays above the retention
• Self-Insured Retention (SIR) — the specific mechanism for per-claim retention in commercial liability programs; covers defense cost treatment, collateral requirements, and large deductible program comparisons in depth
• Insurance Deductible — a related but distinct mechanism for cost-sharing within an insurance policy; the deductible is applied after the insurer pays, while the SIR is applied before the insurer's obligation attaches
• Captive Insurance Strategy — the formal structure for retaining risk through a licensed insurance subsidiary; retention limits are a core design element
• Stop-Loss Insurance for Self-Funded Plans — the health plan analogue to retention limits, using specific and aggregate attachment points
• Umbrella vs. Excess Liability — umbrella and excess policies sit above the primary policy's per-occurrence limit; when the primary has an SIR, the umbrella's attachment is measured differently depending on the policy language

How Insurance Brokers Use Retention Limits in Practice

Retention level discussions are most productive during the annual renewal strategy meeting, not as a reactive response to a premium increase. A broker who proactively models three or four retention scenarios — showing the premium savings and worst-case retained exposure at each level — gives the client a genuine decision rather than just accepting the insurer's default structure.

For clients with stable, predictable claims histories (e.g., a large contractor with five years of consistent loss data), raising the per-occurrence SIR from $25,000 to $100,000 may produce significant premium savings with manageable incremental retained exposure. For clients with volatile or low-frequency, high-severity exposures, higher retentions shift unquantifiable tail risk onto the insured — a different risk proposition than the actuarial average suggests.

When structuring an SIR, confirm that the primary insurer's policy language clearly defines when the insurer's duty to defend attaches relative to the SIR exhaustion, and that the client has the collateral required by the insurer (usually a letter of credit or trust fund) to avoid policy cancellation.