I wasn’t planning to write about this weekend’s New York Times article about the securitization of life settlements after reading Felix Salmon’s post saying there was no new news there. But I was thinking about it some more and thought it was an interesting concept, whether or not it gets off the ground.
Life settlements already exist. The idea is that someone has a whole life insurance policy with a death benefit of, say, $1 million. The insured bought it when he was 35 and had two kids; now he’s 70, the kids are working on Wall Street and don’t need the death benefit, but they’ve cut him off and he needs some cash to fill the prescription drug donut hole and pay his Medicare co-pays. The insurance company will give him a cash settlement value of, say, $100,000. I don’t know what this actual number is, but the key point is that it is less than $1 million at the insured’s expected date of death, discounted back to the present (let’s call that the current actuarial value of the policy). In a life settlement, an investor pays the insured a lump sum that is greater than $100,000 – say, $200,000 – and makes the premium payments (if any are left to be made) on his behalf; in return, the investor becomes the beneficiary on the policy. Again, this already happens, although there are concerns about churning, misrepresentation, the whole deal.
In a securitization, an investment bank would buy a whole lot of life insurance policies, pool them, and issue bonds in tranches (just like a CDO) to fund the purchase of the policies. The idea is that investors would get an asset that is uncorrelated or only loosely correlated with other assets, while insureds would get higher prices for their policies because there would be greater demand for them. This is not happening, although the Times articles makes it seem like it is going to start.
I think this is conceptually interesting because basically it is just securitizing an arbitrage trade. The obvious thing to compare it to is residential mortgages. When you securitize residential mortgages, you are expanding the pool of people willing to invest in buying houses, which spills over quite significantly into construction of new houses or improvement of existing houses. That is, you are moving capital to places where it is (theoretically) being productively invested. I’ve written several times about how this went too far, but the basic point is that securitization is promoting value-generating investment.
A life settlement, by contrast, is pure arbitrage: the ultimate thing you are investing in is just a financial product. Today, insurers make money because the cash settlement value is less than the current actuarial value; some people alternatively let their premiums lapse, and then they lose their death benefit, which is even better for the insurer. Expanding the market for life settlements would help those insureds because they would get the current actuarial value (less fees) instead of the cash settlement value. So the first order effect would be a transfer of money from insurers to insureds. But life insurance is a reasonably competitive industry; insurers will predictably raise premiums on everyone since they can’t count on people taking the paltry cash settlement values or letting their policies lapse. So the second order effect will be a transfer of money from insureds who keep their policies until death to insureds who sell them prior to death, with the insurers just as well off as before.*
So at the end of the day, all we’ve done is pushed cash around – except for those fees! So the people involved in originating, packaging, and selling the securitized life settlements take home their 4%, and the rest is a zero-sum game. I’m not sure any value has actually been destroyed, since investment bankers are people, too. It’s just that there used to be $100 shared among insureds and their beneficiaries, and now there are only $96 to share between insureds, their beneficiaries, and investors, and $4 to share among the brokers and the bankers.
Now arguably the way the $96 is being shared is more efficient than the way the $100 was being shared, since you no longer have the people who take cash settlements subsidizing the people who collect death benefits like in the current system. But I don’t think that we’ve gotten any aggregate social welfare in exchange for those $4 in fees. I suppose there might be third-order benefits from efficiency here (might the insurance market function better in some sense? would this increase demand for life insurance? and is that a good thing?), but I’m skeptical.
* I got in an argument about this yesterday with someone who thought premiums could go down, but I can’t tell you the details because it was off the record.
By James Kwak