A number of factors have kept the development of insurance derivatives as a niche market, but some solutions are being found. Warren Cabral completes his discussion of ART begun in our December edition.

The ART question remains: if there are so many reasons for all the parties to do insurance derivative deals, when will the market grow beyond its current niche status? Insurers may not currently see a need for the extra capacity found in the capital markets, or they may simply fear undoing relationships with their reinsurers. They also regard derivatives deals as very labour and cost intensive compared to reinsurance.

Those who have done it say the effort is worth it for big deals. While small deals may not be cost effective using conventional structures, sooner or later issuers will turn to segregated account “protected cell” companies, as a way of commoditising the initial costs of the structure, thereby permitting “retail” derivatives.The cost of structuring bespoke securitisation deals arises from the necessity of establishing a new reinsurance company, known as a special purpose vehicle (SPV). The single purpose of this company is to reinsure the sponsoring insurer for a defined book of insurance business. The reinsurance structure itself is traditional, in that the treaty can be quota share, excess or catastrophe. The SPV itself then issues one of a possible number of capital market instruments, the proceeds of which are placed in a trust fund for the settlement of claims under the reinsurance contract, or, alternatively, for the repayment of the bonds if there are no losses.

The structure is broadly similar to a mortgage securitisation, in that the mortgage lender continues to service the mortgages which are the subject of the contract with SPV. In this way, the primary insurer (sponsor) remains liable to its insureds and handles claims but, as in a mortgage securitisation where the mortgagees receives cash, the SPV starts up on a fully-funded basis following the offering of capital instruments. In consequence of the trust arrangement, the underlying insurer (sponsor) enjoys a fully secured reinsurance treaty which takes its own capital off-risk and reduces exposure to its balance sheet. Thus, it can be seen that there is major deal documentation required, compared to a simple, old fashioned reinsurance slip and contract.

The instruments issued to fund the SPV are of several types. Early insurance derivative transactions were very conservative with only interest at risk. Protection for investors was achieved by matching funds or allowing the SPV a long time to repay. However, more recently the bond varieties have developed so that all principal may be lost up to coupon, but only if certain parameters are met or, more commonly, there are tranches where principal and/or interest can be lost. In all their variety, securitisations can take the following forms:

• CAT bond - an advance of funds is put in trust and the principal or interest is repaid depending on whether a specified CAT event has occurred. The trust funds earn LIBOR (thus no credit risk). The bonds may expose principal or interest or both. If there are no losses, the investors get LIBOR plus extra interest.• CAT excess of loss bond - same as above but with a defined attached point, and stipulated zone and time period.

• Property surplus note - investor returns are pro rata related to losses and gains on the reinsurance portfolio described in the bond.

• Portfolio linked swap - no principal is paid, but there is a debt note secured by a letter of credit, payable on the occurrence of a stipulated event. This leaves the investor free to invest principal as he wishes. During the term of exposure, the investor receives a premium from primary insurer for providing the debt note.• Surplus note - this is a direct equity (not debt) investment in an insurer which increases its statutory capital with no liability by the insurer to the investor. Thus, there is an increase in underwriting capacity. The notes pay interest.

• Contingent surplus note - this is an option on a surplus note; that is, the insurer pays an investor to commit to purchase a surplus note (as above) in a CAT event. Funds for the notes are placed in trust invested in liquid investments, for example T-bills. If there is a CAT event, the insurer can draw on the trust in exchange for issuing the surplus notes. The value of this structure is that there are pre-determined funds and a fixed interest rate in circumstances where raising funds and the interest payable might be much harder after a loss. Investors receive higher rates of interest than other liquid investments. The cost to the insurer is the difference between T-Bills and note interest rate.

• CAT lines of credit - similar to contingent surplus notes.

• CAT equity puts - insurers purchase the right to require an investor to purchase stock in the insurer at a set price after a CAT event.

Given the complexity of the deals, it is inevitable that technical problems should present themselves. The principal problem for investors, and for the bankers pricing the deal, is the difficulty in establishing the risk basis; that is to say, the basis on which loss projections are made. Notwithstanding the enormously complex modelling which goes into derivatives, where the main risks are short tail with low frequency but high severity, the advantage of the short-term exposure can evaporate if the losses are hugely higher than projected.

The SPV is usually capitalised so that its assets exactly match expected liabilities, but what if there are unexpected expenses or losses beyond even the most extreme projection? In these circumstances, thorough hedging against the bankruptcy of the SPV is needed, so for example, the reinsurance contract provides for the abatement of losses conterminously with the capital available or the SPV purchases stop loss cover, all together with conservative investment of assets of the SPV. This latter course, however, reduces the return to investors.

Aside from the question of economic return, there are also legal and practical problems for the investor. Depending on the domicile of the issuer and the nature of the instrument, it may be that an investor will itself be deemed to be an insurer. If so, the full force of licensing and capital requirements could come into play. There will also be compliance with state and federal securities and insurance laws for the issuer and holder. Likewise, there will be similar compliance requirements in each other relevant jurisdiction. In some cases, such as the United Kingdom, there are legal barriers for insurers to dealing in derivatives. The SPV itself must comply with local insurance laws.

At a practical level, the investing company will be risking capital without any particular underwriting expertise, as contrasted with a traditional reinsurance transaction where there is detailed knowledge of the risk and the method of underwriting it. Thus, an investor needs to understand the SPV's reinsurance contract, its scope and attachment point. The investor must also have a grasp on claims handling, as the net underwriting profit is the source of investor payments.

For the same reason, there is no off the shelf insurance derivative product without a rating. The market generally relies on the rating agencies to distinguish between the risks of different issuers and the underlying exposures, but in order to obtain a rating, there must be substantial disclosure, and disclosure in derivative deals poses a competitive problem. An issuing insurer will not want its rivals to know its market or its underwriting policies.

In response to this conundrum, new “warehouser” issuers have been established. The Goldman Sachs' Bermuda vehicle, Arrow Re, for example, does all the due diligence as a reinsurer of the issuer. The insurer-issuer does not mind such disclosure to its reinsurer.A further problem for the issuing insurer is that it is more difficult to account for the cover provided by the SPV as an asset for the insurer's own regulatory solvency position, because the SPV is not usually licensed in the insurer's own jurisdiction. Where the structure is a pure capital markets instrument, a different accounting problem arises, affecting the insurer's public reporting and, hence, stock price: reinsurance shows up in underwriting results and falls into the general offsets which produce the net results. Capital markets transactions, however, show up in operating results, which have a very different and less favourable accounting treatment.

Arrow Re and Lehman Re legislation
The pressure to find a solution to these problems has produced much innovation. In many respects, the Arrow Re and Lehman Re structures represent the cutting edge, in that both companies have sought to meet the difficulties of commoditisation of price, reduced client disclosure, ratings and legal compliance in one entity with unique corporate characteristics.

To do this they have taken advantage of Bermuda. The island offers access to its legislature for private clients. Accordingly, both Arrow Re and Lehman Re were able to petition for the enactment of private statutes granting them extraordinary powers to fire-wall deals one from another within one body corporate, and thereby spread the costs of an SPV over multiple transactions. The Arrow Re Act also allows the parties themselves to designate the legal nature of their contract, whether as insurance or as an investment, and accordingly direct the way in which both the issuer and the investor will be regulated and accounted for.

This starting point for the Arrow Re legislation was a Bermuda concept known as Segregated Accounts Acts, which allow the creation of multiple quasi-subsidiaries without the time, expense, drafting and rating exercises necessary if they really were separate bodies corporate. Bermuda has been at the forefront of such legislation, having introduced the first Segregated Account Act eight years ago. Other jurisdictions have subsequently introduced public legislation purporting to achieve the same effect, but liquidations experts and international public law practitioners have expressed doubt as to their effectiveness.The Bermuda government, drawing on its experience and having closely examined its competitors' attempts, will introduce public legislation in mid-1999 which goes way beyond the old standard. Similar in scope to the Arrow Re Act, the new public legislation will enable any industry, not just insurance, to avail of fast registration to open absolutely protected segregated accounts enforceable in any international jurisdiction.

Arrow Re's power to designate the legal nature of its contracts has already been addressed in Bermuda, when Parliament adopted the Insurance Amendment Act 1998. This act enables the registrar of companies to designate an insurance derivatives instrument as an investment contract only. The result is to lift the holder out of the application of the Bermuda Insurance Act. While this has only local effect, the declarative certificate may be persuasive in the jurisdictions of the holders.

The search for the appropriate structure to unlock the ART continues, whether in standardised instruments to be traded on the Chicago Board of Trade, the newly formed Bermuda Commodities Exchange or in the bespoke deals coming out of Bermuda or the Cayman Islands. Perhaps there will be no one final method, but what is sure is that the pressure to spread risk into the vast pool of market capital will continue and grow.

Warren Cabral is senior insurance partner with Appleby, Spurling & Kempe, Bermuda. Tel: 441 295 2244; fax: 441 292 8666; e-mail: wcabral@ask.bm