A poll released last week by the Risk and Insurance Management Society — the trade association for risk managers of corporations, non-profits and local government agencies, who together are the major purchasers of commercial insurance — found the group’s membership deeply troubled by the scheduled expiration of the Terrorism Risk Insurance Act at the end of 2014.
Roughly half the group’s membership expect the availability of affordable terrorism insurance to shrink should Congress fail to extend the $100 billion federal reinsurance backstop for terrorism risks. About a quarter of RIMS members, the survey found, expect the market to disappear altogether.
You hear similar things from all three of the trade groups that represent U.S. property/casualty insurers: the American Insurance Association, the Property Casualty Insurers Association of America and the National Association of Mutual Insurance Companies. All have talked of TRIA expiration in nearly apocalyptic terms, warning that expiration would leave the government as effectively the only payer for terrorism coverage.
We take such concerns seriously. While a lot of our compatriots at other free-market oriented think tanks believe it’s obvious that TRIA should be abolished, we find the question to be thornier than that. There really are difficulties in modeling terrorism risks, particularly for their frequency, because human behavior doesn’t lend itself to simple models. Some big risks, like the infamous “nuclear, biological, chemical and radiological events,” really are probably uninsurable. State-level price controls makes it difficult for insurers to charge appropriate rates in some segments of the market and that’s especially true for workers’ compensation insurance, the one line of business that is required to cover all losses, regardless of cause.
All that said, we also see plenty of opportunities to improve the TRIA program. The industry’s co-payments and deductibles should probably be adjusted upward — especially for commercial property — to reflect the fact that industry appetite for writing terrorism has grown since the last extension. The backstop probably shouldn’t exist at all for commercial liability insurance. And we’d like to see the program charge an upfront premium, because there is no way to know how much capacity private reinsurers might theoretically extend in a competitive market so long as the government is giving it away for free.
That’s why it is heartening to see recent comments from Ed Noonan, chairman and CEO of Bermuda reinsurer Validus Re, during the company’s third quarter conference call that affirm our contention the industry is being somewhat disingenuous about whether it’s capable of underwriting conventional (that is, non-NBCR) risks:
We think the industry doesn’t service itself well by claiming that terrorism risk can’t be priced and modeled effectively, with the exception of NBCR. We’re one of the world’s largest terrorism insurers and reinsurers, and we’ve spent a tremendous amount of time creating robust risk modeling and management tools that do, in fact, enable pricing of conventional terrorism risk.
The question isn’t whether it can be priced, but rather, the precision of the parameters in a pricing model. There’s very good data on damageability from various blast sizes with secondary effects. And while frequency doesn’t have a rich data set, there are many other classes of risk that suffer the same shortcoming, and the industry uses simulation tools to help guide our judgment. In fact, the reinsurance market does price conventional terrorism today. It’s a $40 billion market in terms of limit purchase, and pricing is implicitly based on frequency assumptions.
The commercial terrorism models have only limited value, not least because of their limited coverage areas. To counter this, we and many other companies use spider mapping to track exposures. And we’ve gone further, and we’ve built a database that divides the world into 4.4 million individual grids for aggregate and total insured value tracking.
The argument that we can’t underwrite conventional terrorism was a classic example of driving business out of the market and into governmental solutions, think about the National Flood Insurance Program. With better discipline around data collection, data scrubbing and exposure tracking, the industry can and should take on the risk of conventional terrorism.
The industry can’t address NBCR today as the breadth of the potential events are either unknowable or could potentially bankrupt the market. I would include cyber terrorism in this category, different from cyber liability currently written in the market. The scope, duration, potential damage and economic loss from cyber terrorism is currently unknowable and therefore, uninsurable. So we’ll watch Washington closely.
Later in the call, during the question and answer period with equity analysts, Noonan expounded on his position, saying he doesn’t “see any reason why the U.S. government needs to cover conventional terrorism anymore.” Moreover, he projects that the catastrophe bond market could soon be taking on terrorism risk, as well.
Perhaps not on day one, but it won’t take too long before you can start packaging it up into the (insurance-linked securities) market. We do know investors in the ILS market today who would like to take that risk. And so it won’t take very long, I think, for that to start to provide the high-layer type of capital required.
So we don’t know what Congress will do, but I think it’s almost a certainty that there’s more opportunity for us because the two likely options, I think, are renewing the TRIA with higher-scope participations and industry retention, and that means that people will look to buy more reinsurance to protect their risk, or the approach that we think makes sense and that is, get the government out of conventional terrorism and let the government just deal with nuclear biological, chemical and cyber terrorism, which creates a lot of demand for reinsurance.
Responding to a question about whether, even if the private market could provide terror coverage, the government would nonetheless probably provide it cheaper, Noonan acknowledged that the federal government would provide coverage cheaper — noting “it’s free now; it’s hard to compete with that” — but also explained the unique position he was in as both the head of a Bermuda reinsurer and an American taxpayer:
When I come to work in Bermuda, I get paid for taking terrorism risk. When I go home at night, I’m actually providing terrorism reinsurance to the world for nothing. I like coming to work for that reason. So it may be cynical, and I think there’s a cynical aspect of the industry position and say, “No, we can’t deal with this risk.” When, come on, in fact, we can, with the conventional terrorism risk.