Monday, October 11, 2010

Life Settlements Returning to Life?

The Wall Street Journal reported on October 9 that Leon Black’s Apollo Global Management is raising a $525 million fund to invest in life settlements, and the Financial Times notes that similar funds are being marketed in Europe. According to an SEC staff report, this rather obscure investment category has seen investor interest halve since its 2008 peak, when $15 billion in face value was transacted, but these high-profile fund raisings suggest that it may be reviving.

Investors in life settlements buy an actuarially diversified pool of life insurance policies from their owners. Payment for them is above surrender value but at a negotiated discount to face value, which is currently around 87%. Purchasers pay the policy premia (5 - 10% of face value per year) until the demise of the insured, at which point they claim the policies’ face value from the insurers that issued them.

Apart from the returns it can offer, the principal attraction of this type of investment is that it has little correlation to any other: it provides an income stream that is unrelated to security market trends, and since it is self-liquidating, exit involves uncertainty only as to timing and not as to the price that will be achieved. For example, a single policy purchased at 13% of face value with a 5% annual premium will return a compound annual rate of 17.5% if the insured expires after seven years. Obviously, the rate of return is strongly influenced by actual mortality outcomes: in this example, the return would be 23.4% if death occurs in six years and 13.3% if it occurs in eight. Thus the principal risk involved in life settlements relates to the soundness of the actuarial assumptions underlying the construction of the basket of policies purchased. If properly constructed, the same assumptions imply that later premia can be covered by earlier mortalities, making the position self-financing after a few years.

There are various reasons why life settlements have never attracted enormous investor interest and have recently fallen further out of favor. Not least are controversy and complex litigation surrounding policy acquisition practices, many of which have been questionable in the past. The industry has established codes of conduct, most states have legislated on these matters, and one of the recommendations of the SEC staff paper mentioned above was that life settlements should be brought within the definition of ‘security,’ which would allow federal regulation by the SEC. Thus uncertainty regarding investors’ title to policies should be in the process of resolution. However, this will not address issues regarding the return structure of these investments.

Consider the highly simplified caricature of a life settlement fund in the chart below. It consists of 500 policies with a face value of $1 million each, on the lives of average American males who simultaneously turned age 75 at the start of year one. Mortality assumptions are drawn from the Social Security Administration’s Period Life Table for 2006. The policies are purchased at an 87% discount to face value and each has a $50,000 annual premium. To keep matters simple, premium payments, receipt of death benefits accumulated over the course of the year and income distributions all occur on the last day of each year.

The compound annual return on this hypothetical investment is a very respectable 12.4% over thirty years. However, it is rather slow in coming: cumulative return is negative, dipping as low as -6.8% at the end of year two, and first turns positive at the end of year five. Premium payments due exceed death benefits received by $4.4 million in the first two years, so the first distribution is received at the end of year three. The investment does not return the initial $65 million investment until year eleven. In short, life settlements exhibit a fairly extreme J-curve, which is bound to limit investor interest in them to those with long investment horizons and the patience to wait for the first glimmer of a return. While this makes them sound ideal for many institutional purposes, it is an unusual (if exemplary) investment committee that can tolerate negative returns for five years and still have achieved only a 9.2% compound return after ten.

Of course, this example is contrived: real funds have much greater actuarial diversity. If the basket of policies includes some on older people or viatical settlements – that is, policies on people with a life expectancy of less than two years (the 75-year olds in the example have an average life expectancy of 10.5 years at time zero) – the investment’s unproductive period can be reduced. However, the J-curve cannot be completely eliminated. Regardless of the construction of the policy basket, life settlements will remain suitable only to patient investors.

So the audience for the funds being launched is rather limited, and it is unlikely that many such vehicles will be offered successfully. This is unfortunate, because many of the policies in the new crop of funds will be purchased from banks that need the liquidity. While there has been some discussion of securitization of life settlements (another reason for the SEC’s interest), most industry commentators seem to think that this is an unlikely development. Again, the return profile explains why. This also is unfortunate: despite the whiff of sulfur around an investment activity that appears to gamble on peoples’ deaths, and which has sometimes been characterized by dubious practices in the past, a more economic alternative to surrendering their policies is a very real benefit to some of the insured. While the life settlement market is unlikely ever to become big, it is not a bad thing that there is at least some life in it yet.