Diarmuid Brennan asks whether it is possible to realise the benefits of a commutation without sacrificing the benefits of the associated reinsurance?

Commutation is possibly the single most important tool available to run-off managers but it is not without its problems; and reinsurance associated with the commuted policies is the thorniest of them. The big issue is, of course, that commutation equals lost reinsurance; and for a market that is, historically, as heavily 'geared' as London, this is a big thorn. The problem applies whether the policies being commuted are direct or reinsurance.

The London Market

The London Market can be divided into many parts - but the most obvious division separates pre and post 1992 business.

Business written in and before 1992 is characterised principally by the fact that so many of its participants have ceased writing new business.

Most of those that continue have seen to it that discontinued lines or businesses have been isolated from the rest. In London terms it is a run-off market.

One consequence of this transformation for the 1992 and prior Market is that its former 'clubby' nature has disappeared. Now each company looks out for itself and will pay claims only if it must. At one time 'unwritten rules' protected cedants and were respected by the Market, even though on a strict legal analysis claims might be refused. These have largely disappeared.

The transition from 'club' to run-off was probably most obviously manifested when, almost overnight, companies began using Time Bar1 as a defence.

This happened in the course of a few months in around 1997 and without any obvious trigger. Behaviour that was formerly considered to be unacceptable from Market participants became standard practice very quickly. The torturous legal wrangling of the Exxon and Kuwait Airways claims would have been far less involved had the Market's participants had an interest in continuing to write new reinsurance business.

Continuing market

For some time the continuing market carried on treating claims in the former, more forgiving manner - that is, after all, what insurance and reinsurance companies do - ie, pay claims. Clearly, when companies have continuing relationships they take a less legalistic approach to claims.

But this seems also to have altered. The effects of the 9/11 losses have lead to a dramatic change in the claims process. One of the principal effects is that the continuing market has adopted some of the rigour of the run-off market when adjusting claims - particularly claims from cedants that are no longer clients.

The claims issue that, above all others, has caused a major headache for the run-off market, and now appears to be a real problem for continuing businesses, is commutation.


Commutation is one of those subjects that is not mentioned in polite society because, like religion and politics, it causes arguments. Principally the arguments take place between the cedant and its reinsurer over the collection of the reinsurance associated with the commuted 'inwards' policies.

Which side of the argument you are on generally depends on whether you wish to collect reinsurance or avoid paying it. And because there is so little guidance in terms of court decisions, personal preference has much more room for expression.

The absence of what lawyers call authority means that to attempt to understand the issue it is necessary to go back to basics - and reinsurance law is very basic.

Reinsurance and settlement

Reinsurance is a contract. More specifically it is a form of insurance - a contract of indemnity. It is a simple promise by the reinsurer to make good a certain type or types of loss, as defined in the contract.

Typically this loss is defined by certain of its details:

- Date. The loss must occur or be notified within a given time - the policy period.

- Cause. The loss must be caused by a certain type of peril. For reinsurance this generally boils down to loss caused by the perils identified in the original insurance contract.

- The reinsured must have become bound to pay either by reason of a judgment, arbitration award or simple agreement of the claim.

This does not obviously include an obligation to provide an indemnity for payments by the cedant of an estimate of claims that it thinks it is likely will be made but of which it has not yet been notified (IBNR).

Less obviously, it does not include an obligation to indemnity payment for a group of claims that have been made but which have not been adjusted or proven (case reserves).

This is, in essence, what a commutation does. The reinsured does not look at each claim and take a view as to its liability but, in general, takes a view as to the pattern of prior payments, of claims notifications and its exposure. When the cedant pays what it considers to be its likely future claims obligations it is not paying claims but paying to terminate the insurance contract. Normally, when a claim is made the reinsurer is entitled, at the least, to see a claims file showing that the cedant has exercised reasonable skill and care in agreeing to pay. When the cedant commutes it is unable to deliver a claims file containing the information that would justify the payment of any individual claim.

It is easy to understand why reinsurers don't consider commutations to be, automatically, covered by reinsurance2. They have no control over the deal but are expected to pay their part; the deal goes contrary to the way reinsurers intend to make money since they are asked to pay when the reinsured wants to get paid; and, if the reinsured wanted the reinsurer to pay for commutations it could have negotiated this as one of the reinsurance contract's terms. Clearly, it is much harder for a reinsurer to do business if its cedants can arbitrarily accelerate the payment of claims and can impose estimates of likely claims on the reinsurer.

In legal terms a commutation is simply a payment in return for termination of the contract. Of course the price of the deal is dependent on what the parties consider to be the likely cost of claims.

Follow the Settlements

This is where the Follow the Settlements clause kicks in, at least for proportional reinsurance. The broader Follow the Settlements terms require the reinsurer to follow settlements or compromises (usually of claims but some refer to loss3) made by the reinsured in good faith.

Case reserves are clearly claims - though they vary greatly - from simple notification of circumstances likely to give rise to a loss, to a fully documented claim. Settlement of case reserves is arguably a claims settlement, though the cedant will be unable to provide evidence, for any particular claim, as to why it considered that claim to be payable. For IBNR claims, the cedant's argument is even more tenuous because there simply is no claim to settle - though there is, probably, a loss. The cedant will not, however, be able to provide more about the loss than statistical probability arguments as to the likelihood of losses of that type occurring and resulting in future claims.

In non-proportional reinsurance, Follow the Settlements clauses are typically more restrictive and generally contain a requirement that the claim fall 'within the terms and conditions of the original contract and this reinsurance contract'. This has been taken to mean4 that the cedant must prove that it had a legal liability to pay the claim. For IBNR claims this is clearly an impossible requirement to meet in the strict sense because no claim has been made. The most the cedant can prove is that on the basis of prior valid claims, future claims are likely to be made to the estimated value.

What this all appears to assume is that, by making a payment to the insured, or agreeing to do so, the cedant is able to transform mere estimates of likely future claims into real claims. Once so transformed, these estimates are immediately binding on the reinsurer. But what gives the cedant the right to impose this estimate on its reinsurer when the only certain thing about the estimate is that it is wrong? Could you imagine the cedant requiring that the reinsurance contract include a term that the cedant have the right, when it chooses, to determine the likely losses under the contract and the timing of their payment, and require that the reinsurer pay for its share of the estimate on demand? It seems unlikely.


One of the ways that cedants have attempted to avoid the commutation problem is by creating a 'capping' mechanism. In this mechanism the deal is not treated simply as payment for termination of the contract. Instead the insurer agrees to pay the insured for the latter's agreement to place a limit or cap on the claims that can be made under the contract.

The quid pro quo for the cap is that the insurer must agree to pay an amount equal to the capped value. Here is the snag because, if the insurer pays for these claims reserves and the cedant agrees not to make any claims in excess of the agreed figure, this looks just like a commutation. What commonly happens instead is that the insurer calls its payment a 'pre-payment' of claims. The cedant remains obliged to provide claims to the insurer which will adjust the claims and pass them to its own reinsurers.

The problem with this mechanism becomes obvious as it is described. Calling the payment by the insurer a pre-payment is misleading. The pre-payment cannot be clawed back by the insurer should the insured fail to provide valid claims. In practice the insured could submit blank claims forms and ask the insurer to fill them in itself. Payment has been made before the claim has arrived and the outcome of the claims adjustment is immaterial to the payment the insured receives. Reinsurance is a contract of indemnity - the claim submitted by the insured gives rise to no payment by the cedant and there is, therefore, no payment for the reinsurer to indemnify.


Are there any solutions? The main tactic at present appears to be to keep quiet and simply submit claims hoping that reinsurers do not pick up the point. Others hope to avoid the problem by commuting the reinsurance first. For companies that have no option but to commute inwards contracts the risk of losing reinsurance is a big threat and in many cases prevents otherwise good deals from being done.

There are at least two structures that can be used to achieve the benefits of a commutation without losing the benefit of the associated reinsurance.

Both require continued processing - at least until deals are done with the reinsurers. In both cases the insurance contract remains in existence and claims must continue to be made and adjusted.

One structure involves capping the claims from the insured. The insured receives a pre-payment of claims but is obliged to repay should actual claims fall short of the estimate.

The second structure wraps a new finite reinsurance product around the existing insurance and reinsurance contracts.

In both cases the insured gets its payment, the cedant fixes its liability and reinsurers are bound to pay the claims that flow through as a consequence.

1 Limitation of actions. The statute of limitations for contract claims bars enforcement six years from the accrual of the contractual right.

Rather oddly, since reinsurance is a contract of indemnity and the right to indemnity is a claim to damages - this right accrues on the happening of the insured or reinsured event - not the refusal or failure of the reinsurer to pay.

2 "IBNR ... is not an indemnifiable loss but an estimate of future losses and therefore not covered" Nicholas Roenneberg - Munich Re (run off business magazine - number ten - Autumn issue 2004).

3 This point could well be critical in respect of claims to indemnify commutation of IBNR which is not a claim but is certainly a loss to the insured if not to the cedant.

4 Commercial Union Assurance Co et al v NRG Victory Reinsurance Ltd.

Diarmuid Brennan, Director and Solicitor, James, Brennan & Associates.