William M. Wilt describes Moody's analysis of the Unicover situation, and offers a view on how the situation is most likely to be resolved, as well as on the exposure characteristics of companies most likely to seek early resolutions to this growing industry dilemma.
Moody's believes that the combination of substantial potential losses and the unsavoury tinge of Unicover involvement will drive most Unicover participants toward a resolution that will 1) limit the total business booked under Unicover, 2) mitigate the gains to the primary ceding companies, and 3) moderate the losses for Unicover's many layers of reinsurers.
Some participants in the Unicover facilities - mainly cedants, their customers and the brokers - have the potential to derive substantial economic gain from the favourable reinsurance terms. Other participants - including some fronting carriers, reinsurers and retrocessionaires - are exposed to sizable potential losses. However, despite the lopsided economics of the arrangements we see no clear winners.
Since the initial disclosure of problems within the workers' compensation reinsurance facilities managed or brokered by Unicover Managers, Inc., Moody's has taken a number of rating actions on companies that were direct or indirect participants in that series of business arrangements. These include changes in rating outlooks and reviews for possible downgrade.
In this article we will describe our analysis of the Unicover situation in greater detail. The principal analytic considerations in Moody's ongoing assessments of Unicover participants are 1) the range of their potential gross and net exposures to the pool, and 2) the implications of their involvement in Unicover for the quality - of design and execution - of risk control processes in effect at those insurers and reinsurers. In addition, we will offer Moody's views on how this situation is most likely to be resolved, as well as on the exposure characteristics of companies most likely to seek early resolutions to this growing industry dilemma.
Unicover - an unfolding story
There are those who would argue that the Unicover situation will take years to unwind. Others would suggest that business and market forces would compel the parties involved to find a more immediate solution to their problems. There are a number of compelling arguments to support either position. For example, those who suggest that the Unicover debacle will unfold slowly could argue:
1. There is a lengthy reporting delay for premiums, and an even longer reporting delay for losses. The reinsurance pipeline through which these dollars are (or, will be) flowing is long and convoluted. Despite the publicity, there are quite possibly retrocessionaires several levels down who are not aware that they are at risk on any Unicover business.
2. The Unicover programmes could be largely cash flow positive until sometime in mid-or-late 2000. Under most programmes, premiums are paid from the primary companies to the brokers (and on down the chain) net of loss payments made on behalf of the reinsurers. With the reporting lag in premiums and the slow-to-evolve nature of workers' compensation losses, premiums are expected to exceed loss payments for the next several quarters. The real pressure to find a solution of some sort will come when the primary companies come knocking on the reinsurers' doors - hat in hand.
3. Workers' compensation is a “long-tailed” line of business and loss payments stretch out for many years in the future. This argument is, in Moody's opinion, of limited merit. Roughly 75% of workers' compensation loss payments are made within the first four years of the inception of a policy and this business is, after all, low-layer reinsurance.
Alternatively, the forces pushing these parties toward some type of settlement are strong, and may be gaining momentum. Consider the following:1. All companies involved in these reinsurance programmes have “better ways to spend their time”. Certain companies, in Moody's opinion, have other pressing business matters that could be impacted by the dark cloud of Unicover. For example, both Sun Life and John Hancock are in the process of demutualising. Reliance is facing a number of other business challenges, including the maturing of over one-half of its debt in 2000. These companies have many incentives to “get this behind them”. Reliance, for example, has disclosed that their financial role in ongoing settlement discussions “(if consummated) would result in an after-tax charge to the company, which would not be expected to exceed 10% of consolidated shareholders' equity.” Such a meaningful cost for a company that took virtually no net underwriting risk suggests that the opportunity costs stemming from the Unicover cloud of uncertainty are, indeed, very real.
2. Unicover has been a major distraction for the management teams at these companies in 1999. With the rapid pace of evolution in the financial services industry, management can ill afford to spend too much time fixing past problems instead of looking ahead to future challenges.
3. The primary companies who generated this business may also have incentives to seek reasonable solutions, and many may be willing to settle for something less than 100 cents on the dollar. For example, if some reinsurers are slow to pay claims, the primary companies will have their own working capital tied up until the reinsurers reimburse them. Given the difficult workers' compensation market conditions over the past few years, and particularly the market in California, few of the smaller, specialty companies that took part in these buy-down programmes have much capital to spare.
4. Lastly, one could reasonably expect the brokers involved in placing this business to be agreeable to a settlement. This is probably true on the primary side, but particularly true for the brokers who placed the hugely unprofitable business for the multiple levels of retrocessionaires. Setting aside any questions of legal, fiduciary, or professional responsibilities for placing fairly priced business, the reputational risk at stake is sizable. Although much damage has undoubtedly been done already, much of it could be mitigated if reasonable settlements are sorted out.
Moody's exposure analysis
As part of its analysis, Moody's independently evaluated the potential gross and net exposure of the various participants in the Unicover programme (the Unicover programme described here covers all Unicover facilities and the known cedants, reinsurers and retrocessionaires - many of whom are cited above). The steps involved in our quantitative modelling are described briefly below.
1. Ceded premiums were estimated. Moody's designed a screen, or filter, that captured all workers' comp insurance companies that exhibited key characteristics one would associate with involvement in potentially problematic reinsurance programmes. The filter looked for, among other items, rapid growth in workers' comp premiums, and sizable differences between projected ceded and net ultimate loss ratios on workers' comp business. Once these companies were identified, their ceded premiums were aggregated. To project ultimate ceded premiums it was necessary to estimate the “pipeline” premiums using the ceded premiums at year-end 1998 and our knowledge of the Unicover contracts.
2. Ceded loss ratios were estimated. The process described above allowed us to capture the primary companies' ceded loss ratios on this business. These loss ratios were applied to the ceded premiums to generate aggregate ceded losses.
3. Losses were allocated to layer of risk. In what is one of the most challenging, and subjective, aspects of the analysis, it was necessary to take the total losses and spread them to layer of reinsurance. To accomplish this we used industry loss distribution curves for workers' comp losses. We understand that each ceding company may actually retain different amounts of risk (e.g., the first 20K of risk, 50K, etc.). As a result, we constructed a model that allowed for varying assumptions about the amount and layers of risk ceded to the reinsurance facilities.
4. The reinsurance structure was overlaid on the allocated losses. Through a variety of sources, Moody's has been able to reasonably approximate the Unicover reinsurance structure. We should note that these reinsurance arrangements are complex and we are not able to take the reinsurance scheme beyond a certain level of retrocessions. As a result, our gross loss estimates (i.e., estimates before giving credit for ceded reinsurance) are more reliable than our net loss estimates. Moreover, it is recognised that Unicover losses will likely touch many carriers beyond those listed in this report.
5. Low and high ranges were put around each company's gross and net exposure. Recognising the dynamic nature of the many variables that can impact the exposure for each company (such as premium volume, loss ratio, and underlying retentions by ceding company) we reviewed a multitude of scenarios to simulate the reasonable low and high exposure estimates for each company. To determine net exposure we assumed that all reinsurance would be collectible.We have been able to test the reasonableness of our exposure estimates through a variety of sources. Most importantly, a number of companies have “gone public” with their Unicover exposure.
• Cologne Life Re took a $275 pretax million charge to earnings for its involvement in Unicover.
• Sun Life Assurance Company has cited gross exposure of between $700 million and $910 million for its involvement in Unicover. In its press release, the company went on to say that, along with Cologne and Phoenix, the three companies could face claims of up to $1.6 billion based on its current understanding of the contracts.
• Reliance disclosed in its 3Q99 10-Q that it has entered into settlement discussions and it “...believes that there is a reasonable likelihood that these discussions will result in contingent settlement agreements...” The company has estimated that the cost of consummating these discussions “...would not be expected to exceed 10% (after-tax) of the consolidated shareholders' equity.” Based on this measure, the after-tax charge to Reliance would not be expected to exceed roughly $98 million.
Exposure speaks to risk management
Numbers only tell part of the story. A critical consideration for virtually all of the companies involved in the Unicover facilities will be the adequacy of their risk management processes. Moody's will be seeking to answer the question of whether the controls in place at each company are consistent with the current insurance financial strength ratings.
Simply put, the insurance financial strength rating speaks to the ability of each company to repay, punctually, senior policyholder claims and obligations. In addition, each rating category carries with it an expectation for the likelihood of adverse financial events that might impact the company's ability to honor its obligations. For example, for a Aaa-rated company the events that can be visualized are most unlikely to impair the company's fundamentally strong position. For a Aa-rated company the long-term risks appear somewhat larger but are, nonetheless, very remote1.
With regard to Unicover, in light of the magnitude of some companies' gross exposures, one must ask whether the risk controls that are currently in place (such as internal audit procedures or underwriting authorities granted) are able to effectively guard against such event risks consistent with the ratings assigned. In some cases, the fact that the insurer put itself at significant risk from a single exposure might alone justify a rating action even if the insurer manages to avoid all but the most modest realised losses from Unicover business.
Settlements - on a limited scale - seem likely
Moody's does not, of course, have a crystal ball that can be used to forecast the resolution of this debacle. However, we are inclined to believe that the forces pushing toward settlements, at least on a limited scale, will prevail. Recall that the parties involved do not have to seek one grand ‘industry solution'. Rather, negotiations and settlements can take place with individual cedants. Moreover, we would anticipate that by closing one or two major settlements, the likelihood of more following would increase significantly.
Predicting settlements is much easier than predicting where the losses will fall. It is logical to think that companies will consider both their gross and net exposure when deciding on their bargaining position. For example, those with a great disparity between their gross and net exposure (as in the case of companies that ceded the majority of their business) would likely be more willing to pay a premium over their net exposure so as to avoid facing the realisation of their gross losses.
Conversely, companies with high net exposures would seemingly have the incentive to argue against the validity of the arrangements or use the implied threat of non-payment to chip away at their exposure. Not surprisingly, most of the parties at the top of this reinsurance drain are in the former position (that is, high gross exposure and low net exposure). Thus, it would not be surprising if those companies fought hard for settlements and in the end paid a premium (above their lower net exposures) to avoid the spectre of uncollectible reinsurance.
William Wilt is a vice president and senior analyst for Moody's Investors Service in New York and a Fellow of the Casualty Actuarial Society. He is also the author of the December, 1999 Special Comment from which this article was drawn: “The Unicover Facilities - Can the Principal Parties Find a Way to Turn a Mountain back into a Molehill?”
1 Taken from Moody's Investors Service rating definitions - Insurance Financial Strength Ratings