Can Insurance Markets Handle Catastrophic Risks?

The Aftermath of Hurricane Andrew in 1992

The Congressional Research Service has a new report on insurance for catastrophic risk.  At least since Hurricane Katrina, there has been concern about whether the world’s insurance companies have the financial capacity to handle catastrophic risks.

As usual, the CRS report contains a lot of useful information on the subject.  It also highlights the need for giving much more attention to the problem.  There’s close to a one percent likelihood in any given future of losses too big for the system to handle.

The report also shows that catastrophic risks from natural events are highly concentrated geographically.  By far the greatest risk level is in Florida, which accounts for about 80% of total national risk, with California at about half this level and Texas half of that.   (Fortunately, there’s no correlation between earthquakes in California and hurricanes in Florida, or else the national risk exposure would be much higher.)  Most of the rest of the risk exposure is in the Gulf Coast.  Federal intervention, if any, should be to ensure that insurance is available, not to subsidize it.

There are a couple of other points worth discussing. First, there seems to be a growing reliance on sophisticated financial instruments to handle large risks.  For instance, investors purchase catastrophic risk bonds, which do not have to be repaid if the catastrophic loss occurs, thus providing the issuer with funds to cover the loss.  But there are also a variety of other instruments such as risk swaps, catastrophic equity puts,  and loss warrants.  Special purpose vehicles called “side cars” allow reinsurance companies to off-load risks to hedge funds and other investors.  This all seems a little too reminiscent of the brilliant risk management techniques that led to the financial crisis.  It would be nice to have some reassurance that all of the actors involved, including insurance regulators, really understand these instruments and the attendant risks.

Second, the report contains some worrisome but puzzling observations about the future trajectory of risks.  Some analysts have suggested that “catastrophic losses will likely double every 10 years and the present government insurance approach might not be adequate.” CRS cites a 2006 speech by an expert, who emphasized increased values of insured property at risk and higher replacement costs. That trend doesn’t leap out from the past record.  Putting aside Hurricane Katrina, insured property losses have hovered around $35 billion in peak years since 1990.  And I wonder whether the escalating rebuilding costs were due to the housing bubble rather than a long-term trend.  In any event, I’d like to see deeper analysis of trends.

This is obviously a very complicated, technical subject.  CRS does a good job, but its expertise and resources are limited.  It would be very useful if Congress would commission the National Research Council to do a thorough, in-depth analysis of the problem. We shouldn’t wait until after the fact to find out whether the insurance system can handle a mega-catastrophe.