I dabble in financial history, not least because my experience in my own area - American foreign policy - has given me a healthy scepticism of both the consensus that often emerges among practitioners of a craft and the consensus held by their opponents, both of which are evident enough in the mass media. Finance, it seems to me, is clearly too important right now to be left to the financiers or the anti-financiers. If you lift up the rock on Wall Street you find things you could scarcely imagine underneath, and yet it all exists because it serves some sort of purpose for someone, somewhere, perhaps for a great many people; and surely one of the things we will have to do if we are to save our economies is to harness finance to the widest possible purpose, the general good health of the economy.
So I'm interested. And this interest led me to recently discover the existence of catastrophe bonds, which seem like they were designed specifically to invoke on their own the whole spectrum of mixed responses and moral ambiguities that modern finance engenders. The concept is simple, as it often is beneath all of the mathematics involved in complex financial deals.
Catastrophe bonds - or cat bonds for short - are a way for insurance companies to cope with extreme losses which threaten to ovewhelm their ability to pay out. Traditionally, this problem has been solved by reinsurance, which had a prosaic beginning in 1842 when Hamburg burnt to the ground and the German insurance industry was wiped out. Reinsurance deals involve one insurance company passing on some of the premiums it collects from customers to other insurance companies in return for those companies accepting the liabilities; if I crash my car, it might not actually be my insurance company which pays out, but another one who had been receiving my premiums all along. This allows risk to be spread around the industry and makes individual companies more resilient. It looks like a remarkably fraternal way for the industry to function, although of course it's not, because everyone's key motive is self-interest.
Reinsurance, though, isn't perfect. It generally keeps risk within the insurance industry, whose resources are finite. Some risks are so large that they can't be absorbed and diversified away within the insurance industry: try getting earthquake insurance in California. The insurance/reinsurance model doesn't work everywhere because the risks are just too large. To some financiers, this looks like a problem waiting to be solved: you have people who want insurance and can't get it, so is there a way to meet their need by matching them with investors willing to take outsized risks? Cat bonds are the answer.
Cat bonds work like this. An insurance company arranges to issue a bond to investors, paying a high rate of interest, but on the understanding that if some specific catastrophe occurs, the initial capital becomes forfeit. I might give California Insurance Co £5,000 in exchange for a cat bond which pays me 20% interest a year for two years, after which time I get my £5,000 back (they may have invested my money in the meantime, although if they made more than 20%, they're much smarter than me). But if an earthquake strikes California before the two years is up, I wave goodbye to my £5,000. Simple. But effective - the reason being that it encourages money to flow into the insurance industry from the broader financial sector, be it from banks or hedge funds or pension funds. If someone out there is willing to take the risks, cat bonds allow them to do so without starting their own insurance company.
This is how the financial system is supposed to work, from the point of view of the economy as the whole: efficiently allocating capital to where it is most needed. Cat bonds serve a social function by enabling the uninsured to get insurance, and ensure that the risk is allocated to those best able to bear it. Or, to put it another way, they help to spread the cost of coping with huge natural disasters across a broader part of the economy, meaning that the insurance industry won't be overwhelmed and bankrupted if an earthquake levels San Francisco. So that's good, right?
That remains to be seen. If you think you've heard this argument about spreading risk before, then you've obviously been paying attention during the credit crunch, because this was exactly the argument used to argue for the securitization of mortgages during the housing boom. Mortgages were packaged and resold to investors all around the world, supposedly spreading the risks associated with a huge foreclosure crisis to the point where they could be easily absorbed - unless, it turned out, the crisis was so huge that it actually brought down the western financial system as a whole, so thoroughly was the system now infested with this mortgage debt. The social benefits of mortgage securitization were analagous to cat bonds - get more capital into the mortgage industry and hence allow more people to get mortgages - but the risks turned out to be catastrophic in themselves.
But this analogy can only be strained so far, for two reasons. One is the fact that the amount of capital involved is of a completely different order of magnitude, and hence the threats to the financial system are different, and the other is the relative predictability of natural disasters compared to financial disasters. Before the credit crunch, repackaged mortgage debt was typically considered a very safe investment - AAA - because the unthinkable would have to happen for it to go sour; the problem with the unthinkable being, of course, that we haven't thought about it and hence haven't assessed the risk of it happening properly. Yet the safety of these mortgage investments was based on millions of little decisions made by homeowners all over the world, as well as millions of other fallible human factors which are beyond the capability of even our best computers to model.
Compared to human decisions and financial markets, natural disasters are grimly predictable. Seismologists know roughly how often an earthquake of a certain magnitude hits the San Andreas fault, and meteorologists know roughly how often a Katrina-sized hurricane strikes the Gulf coast. In fact, the predictability of the events is one reason the traditional insurance market breaks down in these cases. Cat bonds exist at the risky end of this market, and hence are typically graded as riskier investments, perhaps BB. And because they're based less on a complex understanding of economics and market dynamics, they're really much more of a plain old gamble than most traditional investment opportunities. Weather patterns are simpler than financial markets, but our understanding of them is still imperfect, although cat bonds have given rise to a burgeoning industry of companies modelling weather patterns so they can advise on cat bond investments - innovation which, no doubt, is also a social good.
The risk inherent in cat bonds should prevent them from becoming as mainstream as mortgage-backed securities, and from sucking so much capital into a black hole. But the market for them grows after every large catastrophe, and it looks set to grow again after Japan's tragic earthquake. And they remain, at base, a gamble. A gamble, in fact, on the possibility of human misery occurring, taken by those same old popular villains in the banks and hedge funds - perhaps for the young cat bond industry, whatever its social function, it's best that a rock obscure Wall Street's goings-on for some time yet.
Michael Lewis, "In Nature's Casino", The New York Times Magazine, available here, will probably satisfy the additional curiosity of most. Niall Ferguson's The Ascent of Money contains a good chapter on the history of the insurance industry as a whole.