In recent years, a surge in agency captives, in their various forms, has contributed substantially to the growth of captives. One of the most nagging problems for an agency captive owner is “gap”, being the difference between the loss fund (which is gross written premium less expenses) and the aggregate stop loss reinsurance attachment point. Finding ways to mitigate the impact of this gap is not simple, but can only benefit the captive owner.

Agency captives generally fall into two camps, sponsored and unsponsored. Sponsored refers to agency captives set up by writing companies such as Crum & Forster, Atlantic Mutual and Chubb to name a few, by which those companies' preferred sources of business share in the underwriting profits and investment income with the carrier. It is often an effective way to further reward business source partners and to engender loyalty to the sponsoring carrier. It is also a means to give incentive to renew good business and introduce profitable new business to the sponsor.

Unsponsored agency captives are set up either as homogeneous, niche vehicles or for heterogeneous books of business. Often, the homogeneous book will be generated from an existing managing general agent (MGA) that specialises in underwriting a specific class of business. The MGA might have been underwriting for fees and/or commissions for risk taking carriers for years. With time, credibility and confidence, there is a natural allure to taking on risk and sharing in the positive results that the MGA has generated for its risk-taking partner.

Unsponsored, heterogeneous captives are often set up for the same reasons as for a homogeneous book- the interest in sharing in the financial rewards (again, underwriting profit and investment income) that historically had been ceded from the agent to the insurance carrier. Generally, the starting point for such a captive is a controlled book, where the agent/owner has historic loss information on a diverse book of business that can be analysed, and to an extent, relied upon to produce profits.

Which is more suited to a captive? Both have benefits and disadvantages. A homogeneous book of business can be more readily underwritten with an eye towards the specific risk issues surrounding that book. Unfortunately, there is strong correlation risk inherent in these books. For example, if the legal environment surrounding the niche business class deteriorates, it is likely to deteriorate across all insureds. In a heterogeneous book, there is a natural hedge. The downside of the heterogeneous book is that the risk profile of the book can change over time, and the results of the past may or may not be indicative of the future. For example, if an agent reviews its book historically and projects a ultimate loss ratio of 65%, that is only true to the extent that both coverage and risk profile remain static.

For both types of book, a very attractive aspect of a captive is the ability to bear risk in less risky, more predictable positions, and then grow the level of risks as the MGA's capital base and confidence in results improve. The retention level of a captive can increase or decrease over time as the MGA's financial situation and/or market conditions improve or deteriorate. For example, the net retention may start with the captive reinsuring 50% of the first $100,000 per loss, but increase to 100% of the first $500,000 over time.

The reinsurer's perspective

Whether looking at single parent, group/association, or agency captive, a reinsurer (especially one that has an intimate understanding of alternative risk and specifically captives) is likely to look as or more favourably upon a captive's risk than risk from a professional cedant, all things being the same. The rationale is that there is the added element of alignment of financial interests. Imagine as an underwriter hearing that the producer of the business (who traditionally is motivated by volume more than profit) is so confident about its book of business that it will take risk in the venture. This is very compelling, in that the producer is taking on downside risk and cares about the underwriting viability of the book.

What kind of risk do reinsurers prefer? Obviously, it is predicted on the particular reinsurer's set of experiences, expertise and financial structure. In general terms, though, there are some basic preferences, most of which are self-evident:

  • An existing book is preferable over a start-up;

  • A larger book is preferable to smaller, since size generally reduces volatility;

  • A tested underwriting template is preferable over a programme that is being remediated due to poor performance.

    Reinsurance issues

    In a word, one of the most nagging problems for an agency captive owner is “gap”. Gap is the difference between the loss fund (which is gross written premium less expenses) and the aggregate stop loss reinsurance attachment point. As a numerical example, consider the following structure, albeit an over-simplified one:

    With almost any agency captive writing business via a US admitted carrier, the gap must be fully collateralised, most often by a letter of credit (LOC). This LOC frequently must be backed by cash deposits. In our scenario, that means that the captive owner needs to tie up $1,250,000 in cash to support the captive's net position - a tidy sum to become illiquid. This illiquidity dilemma is compounded by the fact that this sum is only for the first year of ownership. This amount “snowballs” in the coming years, especially for long tail business.

    Take the example further by assuming a five-year, straight-line payout for the book of workers' compensation. The following is the aggregate collateral requirement per year.

    Obviously, the real pattern of paid loss and remaining outstanding reserve and IBNR is much more complicated and subject to differing interpretations between captive owner and the fronting company, but the snowballing effect and its impact on cash resources and liquidity are evident. The cash demands keep growing until the first year's underwriting results level out. In a growing book, and with even longer tailed business, this effect is even more dramatic. Even after the captive ceases writing new business, the collateral obligation continues until the payout patterns terminate.

    So, in the face of such financial difficulty, what can be done? The easiest answers are to change the simple maths of the equation. To reduce the amount of gap, either the size of the loss fund as a percentage of written premium must increase, or the aggregate stop loss attachment point must decrease. To increase the size of the loss fund, acquisition costs must come down, either in the form of reduced commissions, fronting fees, administrative fees or unallocated claims costs, etc. Negotiations and leverage must be used to accomplish this and, often, the best way of securing favourable rates are size, longevity and solid business relationships.

    The other side of the equation is the stop loss reinsurance attachment point. This is somewhat a function of the marketplace. If you are (re)insuring a class of business that is out of favour at that moment (currently, examples might include trucking liability or contractors workers' compensation), negotiating a lower attachment point could be tricky. Here are some possible tools.

    Quota share levels: The most straightforward way to reduce gap is to reduce the net position. If the collateral obligation for a net retention of 100% of the first $250,000 per loss is $1,000,000, a reduction of the net to 50% of the first $250,000 will be $500,000. The ratio of retained premium to gap remains the same, and the ability to share in underwriting profits is cut in half, but the dollar value of the collateral obligation is easier to absorb.

    Multi-line approach: All things being the same, the more lines of business, especially non-correlating lines, that can be batched together, the more stable the results will be. An example might be the combination of a captive's workers compensation and errors and omission lines. Losses from these two lines are unlikely to correlate with each other. A reinsurer will generally assume that there will be a “netting out” effect that makes the overall bet better the same exposures on a monoline basis. Batching correlating lines is less attractive, since auto liability, general liability and workers compensation might all suffer in a bad winter. In short, a reinsurer may need a smaller buffer between expected loss and the attachment point in a blended programme than when looking at lines separately.

    Multi-year approach: Along the same lines as the multi-line approach, it is generally less volatile than single year programs. Again, a netting out effect could occur, and a stop loss attachment point that is calculated from two years of experience should be more predictable, thus the overall percentage gap between the loss fund and the attachment points should reduce.

    Other gap solutions

    Alternative financing: Instead of securing an LOC, there are companies, such as premium finance companies, that might be willing to lend the cash deposits or provide a guarantee for the gap position. The collateral requirements from the premium finance company may take the form of a lien against future commission income, personal assets or other sources. While the obligation and risk still exist for the captive owner, the stress of having to make illiquid millions of dollars is reduced.

    Negotiating with the fronting carrier: With a predictable book of business and solid relationship, it could be possible to convince the front either to relax its full collateralisation requirement, credit expected investment income against the full requirement or maybe even to “invest” in the gap with the captive. It is by no means an easy task, but a captive with confidence in its book, and with the data to back it up, may be successful going this route.

    The last tool to consider is loss portfolio transfers (LPTs). For more mature policy years, it is possible to engage an LPT reinsurer to evaluate the case reserve and IBNR levels, and to offer terms that trade the existing collateral obligation for a lump sum premium payment. In essence, an LPT reinsurer will provide a limit equal to the difference between paid losses and the aggregate stop loss attachment points for the years being transferred.

    The premium charged will be the net present value of the expected ultimate net losses, plus some additional charges for risk, administration and profit. The premise is that there is an arbitrage between the fronting carrier's opinion of the ultimate losses and the LPT reinsurer's opinion. The LPT route generally makes more sense on larger books with predictable results. Trying to purchase an LPT for a smaller, more volatile book is difficult.

    In the end, despite the existence of the problem of gap, agency captives continue to proliferate. The underlying profit potential and control that a captive offers are quite compelling, and the gap, if nothing else, serves as a means for a captive owner to demonstrate its confidence in its own abilities. Nonetheless, finding ways to mitigate the impact of the financial obligations due to the unfunded portion of the net position can only benefit the captive owner. Releasing liquidity allows the captive owner more manoeuvrability and a greater ability to reinvest in its business instead of tying it up behind an LOC.

  • R. Lincoln Trimble is a vice president of Chubb Atlantic Reinsurance Specialists Ltd. (CARS) in Bermuda, which is the reinsurance marketing arm of Chubb Atlantic Indemnity Ltd., a subsidiary of the US based Chubb Corporation. Chubb Atlantic specialises in reinsuring alternative risk vehicles, such as captives, risk retention groups and other pooled arrangements.