As if facing a worldwide economic crisis weren’t enough, life settlement investors have found out in recent months that insureds are going to live longer than was previously thought. At least 2 major medical underwriters, the companies who evaluate insureds current medical status and project how long they’ll live, have announced changes to their projections in recent months. The changes lengthen their life expectancy estimates. While insureds may welcome this new, if only statistical, lease on life, for life settlement investors, it means that they’ll pay more premiums and see less profit from the transactions.
Minneapolis-based 21st Services has announced changes to its projections that could be more than 20% longer in some cases. Kennesaw, Ga.- based AVS Underwriting, LLC, followed with an announcement of a roughly 10% adjustment. Part of the changes stem from the fact that “the life settlement business is still very young in terms of data that is available,” according to 21st Century’s Chief Actuary Vincent Granieri. As an industry, medical underwriters have probably evaluated less than half a million lives since they began providing life expectancy reports, Granieri estimates. “The good news,” he says, “is that all of them are life settlement lives” and are therefore relevant in terms of data. “The bad news is that it’s not enough to be credible” in terms of basing a model solely on those lives. In general, he says, actuaries “are used to having millions” of data points to work with. “Over time that data will emerge,” he says, although he estimates that it may take a decade for the data to reach 1 million lives, “let alone 2 or 3 million.” Compounding the problem, he says, is the nature of much of the data that is out there. This data comes from life insurance carriers, he points out, and while there are ample individuals to use for basing expectancy evaluations, he says the data tables were created for individuals seeking to obtain life insurance rather than those looking to sell an existing policy. Effectively, Granieri says, as life insurance actuaries are building their data tables, they are going to factor in a higher mortality to rate as a conservative stance. By comparison, for life settlement companies, a conservative stance would be the opposite–a low mortality rate that assumes individuals will live longer. “Conservative for them (life carrier actuaries) is quite aggressive for us,” he says. As a result, he says, 21st Services has worked to combine the data of the 2008 Valuation Basic Table put out by the Society of Actuaries with the company’s own data. It has also begun efforts to test models by comparing them with data from actual lives in the Medicare database during the 1990s. For now, the question is how much damage has been done already to the settlement market due to the changes. Michael Fasano, president of Fasano Associates, a Washington D.C.-based underwriting firm, notes that “clearly it’s had a chilling effect” on the market, and is one of the reasons for the turmoil being seen in the life settlement industry. Michael Coben, senior vice president of national distribution at Coventry, Fort Washington, Pa., says the changes could have a “significant impact.” But he adds that the effect would be “specific to each provider” based on which medical evaluator the provider uses. At the moment, he says, “firms are re-pricing, and a lot of capital sources are trying to figure out what they want to do.” Coventry has used its own underwriting department to evaluate life expectancy, and has not been affected by the changes, Coben adds. More specifically, Fasano says he expects one form of transaction–those involving financed policies sold not long after the 2-year contestability period has ended–will fade away entirely. “The premium financed part of the market may never recover,” he says, adding that such deals are premised on the idea that the policy could be sold after 2 years, and priced as such. “The only way those numbers work is if your life expectancy is too short,” he says. With longer and more accurate life expectancy medical underwriting, Fasano adds, “there’s really no way you could make a profit.” With a shorter life expectancy report, policies attracted significantly higher prices from funders. But when all the involved players get their piece of the funding pie, and the life insurance company gets its premiums, “the pie is not going to be big enough,” he says. “That may well kill the premium financed part of the life settlement market.” Fasano notes that his company has routinely provided longer life expectancies, and has at times lost business as a result. One of the contributing factors to the problems of inaccurately short life expectancy estimates, he says, has been “an unhealthy misalignment of interests” in some parts of the business, in which brokers and providers put a policy’s potential price and the ability to complete more transactions as their top priorities. “There has been some shopping for short life expectancy estimates,” he said. As a result, significant discrepancies had existed among medical underwriters, or as much as 25% to 30%. This, Fasano notes, “scares off a lot of capital.” Fasano also says he found it difficult to understand the reasoning behind the variance. “We’re all looking at the same data,” he says. While determining a life expectancy is not as simple as some have argued it is, “it’s not rocket science either and I would not expect the differences to be that significant,” he maintains. Now that the announced changes have taken effect, Fasano says he expects it will take about 6 months for the market to work out its uneasiness. After that, he predicts, funders will regain their confidence, or new capital will begin to enter the market. Perhaps most importantly, “I would not imagine, going forward, that you would have changes of that order of magnitude that we have seen in the last few months,” he says. As with his own models, he expects individual companies to fine tune and refine their life expectancy models, but those changes “will be much less extensive,” he believes. Similarly, there will still be differences due to the specifics of their individual models, he says, but “those differences will be significantly less.”