The cork is still in the bottle of Project Corkscrew, but this first effort to unwind the legacy of the LMX spiral will not be the last.

If Ipe Jacob, architect of the ambitious Project Corkscrew, is disappointed by the failure of th first attempt to get his spiral-busting commutation project off the ground, he doesn't show it. Corkscrew, devised and promoted by the Insurance section of the Financial Markets Group at consultancy Grant Thornton, has at least provided good public relations for his firm, he told Global Reinsurance.

And the possibility remains that a less ambitious version of the commutation process, designed to unwind the London market excess of loss spiral arising from a selection of coded catastrophe losses between 1983 and 1992, may still proceed. “I'm neutral now, rather than optimistic,” Mr Jacob says. “Certainly people are interested, but whether they are interested enough, and whether they are the people who can commit to paying, I don't know.”

Corkscrew was launched on 1 June this year, after Grant Thornton invited 47 active and ceased London market insurers to the London Underwriting Centre to hear Mr Jacob's proposal to unwind the spiral. In the words of the project brochure, the scheme is about “reducing the wasteful costs of processing claims through several layers of the London market spiral, and bringing certainty and finality to balance sheets in respect of spiral losses.”

Execution requires the crunching of participants' data regarding thousands of outwards reinsurance contracts related to specific LMX spiral losses through a computer-driven ‘commutation matrix' that simulates claims processing and yields the balances between participants at a chosen future date. Participating companies settle up immediately, avoiding another 25 or more years of handling. The commutation would put an end to any and all claims between participants related to the named catastrophes.

Spiral Expense
Grant Thornton believes the spiral costs the market about $500m annually. Mr Jacob proposed to stop the bloodletting for a cost of 5% of the relative outwards reinsurance balance due to or owing by each participant, with a fixed entry fee of £1.75m in total. “The music should stop,” the brochure declares. “All participants' incurred data should be frozen, at, for instance, 31 December 2000, and that incurred position should be treated as if it were a paid loss, and the cover under which each specific loss is being recovered be deemed to be utilised.”

It's all a game, Mr Jacob says, complete with winners and losers. But of course everyone's a winner, since the longer-term savings will be significant, providing enough risk carriers play the game. Part of the challenge was making the game attractive enough to encourage people to play, win or lose. “There is a running theme of equity, because everybody plays by the same rules. Inequity is likewise spread around the entire group.”

Under Grant Thornton's order of play, activation of Corkscrew required a quick cash commitment from the invited players. Mr Jacob allowed four weeks following the launch meeting for the insurers to stump up £1.75m between them, at a minimum of £50,000 each, to give the mass commutation the green light. “We were not funded up to the £1.75m, but I won't say who paid how much,” he says, adding that the money that was paid in has been returned.

Project Corkscrew didn't work. “I think the principal reason we did not reach the funding limit was that four weeks was not enough time,” Mr Jacob says. “We had one invited company say they could not participate, as it was not in their business interests, because they have a preference for retaining risk. Three foreign companies said they did not want to participate in the first round, but wanted to present a commercial case that they should be able to come in for less than £50,000. Finally, there was a whole raft of commentary about technical issues, but not a single company said it will not or cannot work, or that there is anything wrong with the technology behind the project.”

Other companies had other objections, not least of which was the decision to base the finality amounts on a total including outstandings. “A great bugbear is that some people will not respond to commutations involving outstandings. Some significant players in the market are taking that stance, and there are two arbitrations in train on that issue.” One option is to allot a zero value to outstandings so that the technical issue drops away, but that raises problems for many companies on a cash flow basis, Mr Jacob says.

Market attitudes
Individual company concerns aside, Mr Jacob believes that wider market attitudes also led to the first-round failure of Corkscrew. “Another issue is whether they really have an appetite to do something collectively to solve the problem. I am afraid the track record is appalling,” he admits. “I am quite grateful that we have banked our marketing gain already, because whether or not they will actually pay anything to anybody, they are genuinely interested in talking and debating as long as the problem might exist. But whether they have the appetite to actually solve the issue, I genuinely don't know.”

Equitas, the run-off reinsurer with responsibility for all Lloyd's syndicates' 1992 and prior losses, and thus for all Lloyd's Corkscrew claims, is less specific but more clear-cut in its assessment of the project. “We are not participating,” says James Burcke, head of communications at the reinsurer, which has been the major driver of commutation in the London market heretofore. “We don't think it is going to achieve the needs of the market.” Mr Burcke was unable to provide more details of the reasons behind Equitas' refusal to back the project, or specify just what needs it would fall short of meeting, but according to another reinsurer, one who requested anonymity, the project is doomed without Equitasian support. “The key is Equitas. They must be 40% or 50% of the market. Without them, can it really work?” the reinsurer asks.

At least two players have said it will only work if everybody plays, Mr Jacob admits, but he believes the game has merit as long as at least three companies want in. It will, of course, get smaller with a smaller number of players, but there is still value to be realised, he insists. “The ideal would involve all of the people playing, for their own benefit, and of course it works best if more, rather than less people play. But the commutation matrix will work with as many or as few people as want to be in the game.” The difference is the potential gain, he says. “It is only worthwhile if you get a benefit, so if it is only two reinsurers, they don't need me. They can do a handshake deal ... So we have designed it to make it unattractive to not play, and very attractive to play.”

Mr Jacob intentionally peppers his talk about Project Corkscrew with words such as winners, players and rules. Seeing the London market in part as a continuous, symbiotic game of pass-the-parcel, he has incorporated so-called ‘game theory' throughout the scheme. “A technical solution of itself would not work. People had lots of different agendas, and different perspectives of where they themselves and others stood,” Mr Jacob says. “We viewed the marketplace, in some senses, as a cooperative game, with winners arising from cooperative behaviour. But there is a kind of uncooperative game equilibrium, where the players are not doing anything, and are content with their own position solely because they might benefit competitors by moving from it. We had to design a game that would incentivise them to play.”

By setting some new rules and clearly identifying the players, Project Corkscrew could move the goal posts, he believes. “If we made it very complex it wouldn't work, but simplicity had to be set off against equity, making sure that everybody gained. Game theory applies all through the structure of entry and exit points. The rules are open for everyone to see,” the consultant explains. “We are not in the role of advising how to steal a march on another in the market. It all has to be open all at once, and we have to structure a series of incentives to get [reinsurers] to join, and consequences if they do not join, assuming the game runs.” That approach explains many of the fundamental aspects of the project. Anonymity until the final endgame, for example, and the fact that the players would set many of the rules, including the up-front ante for non-joiners who opt to be dealt in whilst the game is underway.

When it came time to play, however, too few players showed up for the match. The first round has been called off. Ipe Jacob is disappointed, more in the market's collective lack of will to attempt to claw its way out from the burden of the LMX spiral than from their partial rejection of his commutation matrix. However, he hopes it has a future, and is rewriting the rules before trying again. He could opt to amend the confidentiality rule, for example.

“We may suggest a change in the rules so that we flag who has joined, then ask Company A if they will play, and post their answer on our notice board.” Another option is to alter the 5% fee on balances – both positive and negative – that Grant Thornton intends to levy for its services. Some companies are apparently worried that five points could equal a very large number. But it is not just the rules but the game itself which must be amended, Mr Jacob realises. “There are a lot of emotional agendas around the marketplace that have nothing to do with commercial aspects of the business, and a lot of history that is very real. Next time we play the game we will have some behavioural economists involved, because the market's behaviour is not always rational.”

Necessary changes
He is attempting to get to grips with that irrationality by moderating an ongoing, online discussion and debate over the changes that need to be made to get Corkscrew off the ground. “All the major players, bar one, are engaged with us, and we are working through it,” he reports. He says another next step may be to run a small-scale trial. “We might have to run a trial. We would pick Hurricane Alicia, probably, and run the maths through the commutation matrix so people can see how it runs for them.” Meanwhile, he is aware that reinsurers have other things on their minds at the moment. “We don't anticipate that anything will take off this year, and I suspect, because the first half is busy, it will take some considerable time.”

The consultation may be the answer. Grant Thornton's commutation matrix was built based on data provided by half a dozen ‘big players' in the LMX spiral (data which has since been destroyed), and it seems obvious that those companies are keen on Corkscrew. For others, it is Ipe Jacob's continuing goal to bring them into the game. “Some have said, ‘We know companies X, Y and Z are out of retro cover, or almost, and they are going to go bust. We are quite prepared to bear the pain until we see them drop out of the game.'” Such a strategy is dangerous, he warns. “Each company that drops out of the spiral could be as much a reinsurer as a cedant, so what might look good on one side of the balance sheet could come back to bite you on the other.”

Mr Jacob hopes to issue a modified proposal before year-end. “This time I will probably have to require pre-commitment, on a financial basis, by the companies. Last time we were very keen that our offer should be to the entire marketplace, but we are now narrowing down the circle.” So how will Project Corkscrew, The Sequel look? “I expect it to be rather smaller,” Mr Jacob says, “but I cannot see that it makes any sense to curtail an offering of the solution to a small group of people. It needs a core group who will say it is justifiable for us to do even if it is only for us, then, hopefully, we would agree to invite everybody else again.”

Philip Marcel, a long-standing London market figure and one of Grant Thornton's consultants on Project Corkscrew, sums up both sides of the challenge. “As a solution, it is the best one that I have seen, and it is the only one that has a real chance of running,” he declares, noting his bias. “But it is always difficult to get any group of people to subscribe to a collective scheme. There is always the inevitable question: am I doing better than my neighbour?” It is the wrong question, he contends. Spiral players should ask: “Is this better than doing nothing?” So far, it seems, many are not yet ready to leave old practice behind and take that bold step.