Kenneth A. Froot spends more time thinking about natural disasters than the average business school professor. In addition to the rise and fall of the Dow and the long-term implications of the financial crisis in Greece, he has natural perils—hurricanes, earthquakes, floods, and unusually severe European winters—on the brain.
And he thinks you should, too.
Froot, the André R. Jakurski Professor of Business Administration at Harvard Business School, has spent more than 15 years researching how the reinsurance industry—which provides insurance for insurers—manages catastrophic risk. And while he's seen improvements in the way reinsurers distribute risk over the last two decades, he still questions whether reinsurers in their current form can survive a major catastrophic event.
"Reinsurers assume the risk of, and hold the capital for, those catastrophic events for which insurers are poorly suited," explains Froot.
Why can't they produce enough of this coverage at a lower price?
A local property-casualty insurance agency can easily hold enough capital to cover losses from house fires or burst pipes; such losses are relatively infrequent and their numbers easy to predict. Natural disasters are a different story: though they occur much less frequently, when they do happen it's akin to thousands or even tens of thousands of pipes bursting and homes burning all at once. Hurricane Irene, for instance, inflicted an estimated $6.6 billion in insured losses. And they are much harder to predict. Because of this, it becomes very expensive for insurance firms to hold the capital necessary to cover catastrophic losses.
So they turn to reinsurance firms, paying them a fee to assume that risk. Ideally, reinsurers are able to absorb the risk of, say, a hurricane blowing through Vermont, because they have diversified their own risk across the possibility of hurricanes in Vermont, floods in England, and earthquakes in California.
"One relies on the insurance sector to promote risk sharing," says Froot. "One of the features my research raises is that the level of risk sharing in many instances is very incomplete, very poor."
Froot examines where insurance companies purchase protection asking, "Do they buy reinsurance contracts for small losses, for in-between losses, for big losses, for gargantuan losses?" Logic dictates that insurers will choose to cover gargantuan losses "because those will wipe you out," says Froot.
But that's not where they spend their money.
"For medium and even smaller large losses, there's some risk sharing—maybe 30, 40, up to 60 percent of the losses are covered," says Froot. "That's not great protection, but at least it's some coverage. But for really devastating kinds of large losses, the coverage levels actually fall from there. They should continue to increase, but don't—and even fall for the most devastating events."
Prices Too High
Part of the problem comes down to behavioral issues. Say you work at an insurance agency. Your job is to devise protection, and you've discovered, as many insurers do, that catastrophic reinsurance is expensive. Meanwhile, your boss has been hounding you for months to keep costs down, and it's almost time for your annual review. Purchasing catastrophic reinsurance coverage protects your job—but only if a hurricane hits. Spending all that cash will raise your boss's eyebrows because it raises the pressure on him. "People are shortsighted," explains Froot.
Froot found that the major issues arise at a more macro level; the fault lies less with the buyer that's unwilling to purchase catastrophic reinsurance and more with reasons that make reinsurance so expensive in the first place.
"The beauty of property-casualty insurance is we actually know something about what it should cost to produce it," he says. And if you know what it costs to produce something, you know what constitutes a fair price. "If the annual chance of loss is 1 in 100 for a potential $100,000 insurance claim, then the expected loss is about $1,000 a year." But Froot found that reinsurers often charge 10 times that amount. "Why is it so costly to produce? It shouldn't be. Where's the inefficiency in our production by reinsurance companies? Why can't they produce enough of this coverage at a lower price? A lot of the research goes into trying to understand why it is costly."
One reason catastrophic reinsurance is so expensive is because the cost of capital is unfairly high. When reinsurers sell securities to raise money, buyers expect high returns, just like from any other equity investment, Froot says.
If you look at the risks that compose most reinsurers' portfolios, they're very concentrated in relatively few events.
The problem is, reinsurance isn't like any other equity investment: it isn't really exposed to market risk, so expecting high returns is unrealistic. A financial adviser might be willing to accept lower returns because she is investing in reinsurance as a diversifier, with little exposure to today's volatile equity markets. "In fact," he continues, "reinsurance stocks are plummeting also. They get contaminated by the equity market even though their underwriting returns continue apace."
That the equity market for reinsurance isn't auction efficient is another problem. Prices are set by a few big reinsurers, and they have every incentive to set those prices relatively high. "There's not enough price competition," says Froot.
Unfortunately, prices are one of the few things the companies can control. Reinsurers are, oddly and importantly, at the mercy of Floridian gales and Californian fault lines. "If you look at the risks that compose most reinsurers' portfolios, they're very concentrated in relatively few events—wind blowing in Florida, a quake happening in California," says Froot. "Now that wasn't the idea. The idea was to get widespread diversification.
"Reinsurance holds out the prospect that diversification will cheapen the cost of reinsurance and make it attractive to share risk. But if [reinsurers] don't diversify, they can't offer that. … The losses in a single down day on the world equity markets are 10 times what big event losses would be. And that's just another day on the stock market. We should be able to bear these big event losses relatively easily, and yet because they're so concentrated these losses could wipe out reinsurers, which would be devastating."
Although there hasn't been a catastrophe big enough to wipe out large numbers of reinsurers simultaneously—and it may end up being several catastrophic events in close succession that does the job, says Froot—individual companies did fail, and the reinsurance industry took a long time to recover from events like the Northridge earthquake and Hurricanes Andrew and Katrina. In certain cases catastrophe insurance costs more than doubled and took eight years to work their way back down. This year,the earthquakes in Japan and New Zealand, and the devastating floods caused by Hurricane Irene, may trigger similar price increases.
Diversification The Wrong Way
Reinsurance firms are pushed to diversify, diversify, diversify, but they need to do it well, which Froot hasn't found to be the case. Reinsurers often diversify in order to garner good scores from ratings agencies—and those agencies don't take profitability into account when they're assigning scores.
"To get a really high rating, a reinsurer has to say, 'I can't devote more than 25 percent of my risk exposure to California, so I'm going to devote some to European freeze,' " he explains. "European freeze is not a big risk. The profitability of underwriting risk in European freeze is poor. You shouldn't do diversification that, after risk adjusting, is guaranteed to lose money."
Appropriate diversification is just one item on the list of things reinsurers should be doing to ensure their survival. (And it's important to note that while reinsurers may one day cease to exist in their current form, reinsurance will go on.) They should also act more like risk-taking investors and less like risk-averse corporations, their capital fluctuating "with people's decisions about whether they thought reinsurance was an attractive bet at the time, as opposed to the stock market," says Froot. A reinsurer that acts more like an investor than a corporation is also going to be more transparent about how it does business—another plus.
Froot likens this new form of reinsurance to a floodgate. "Rather than keep capital captive in reinsurers as we do now, it's much better to have the gates be open like a conduit that would allow water levels to fluctuate according to the tide. If prices are too low, capital comes in; if prices are too high and returns going forward are poor, capital leaves."
In the big picture, says Froot, "all this is interesting and useful because it's a microcosm for the larger financial sector, made up of banks and shadow banks. We have a huge institutional setup for bearing these sorts of 'out-of-the-money' risks—very unusual, low frequency, high severity events. That's much like the chance of a Greek or Italian default today or the risks in the mortgage market in the United States viewed from the perspective of 2006."
A Peek To The Future
What fate befalls policyholders should reinsurers default after the next big calamaties? In the United States, insurance is regulated on a state-by-state basis, and many have funds or other programs in place that could help policyholders get some form of reimbursement. "But these account for only a small part of the risk; the rest is kind of gray," says Froot. "We don't know what would happen." The federal government could also be expected to provide emergency relief, but in many cases might not fully compensate claimants.
Unfortunately, the world appears to becoming more susceptible to catastrophic events, so this question of who pays becomes ever more pressing.
"As we move forward into a world of global warming, rapid melting of polar ice, and an increasingly fragile ecosystem where infectious diseases of all kinds are more likely to be successful, we run greater risks of all sorts of perils," says Froot. "In some cases—such as the 2001 terrorist attacks or the nuclear damage in Japan—insurance or reinsurance contracts do not cover the losses. In other cases the language will be ruled that coverage is implied. Either way there are potential large damages, and the question will become who will pay and what will it do to them going forward."