Terror reinsurance markets have generally trended downward in activity and overall pricing since peaking during market conditions following the attacks of September 11, 2001 in 2002-2004. The dynamic being played out in the terrorism reinsurance market is not unlike other low-frequency, high-severity lines where a combination of the passage of time, the lack of further market defining loss events and constrained reinsurance budgets converge to allow the market to drift gently downward from supply/demand imbalances. On the surface, it would seem that lower pricing over time would increase activity levels, but interestingly we see simultaneous lower prices and lower activity levels.
While the trend is moderately down, differences in market conditions and activity levels certainly exist amongst global terror reinsurance markets. In the large US market, where the insured values and risk push the upper end of the spectrum and where interest in purchasing specific terrorism reinsurance peaked immediately after the attacks of September 11, 2001, overall activity levels are clearly down. In other major global markets, including some in Europe and Asia, activity levels are for the most part steady. That said, any of a number of factors can cause severity-driven markets to turn around very quickly, and elements are in place for a healthy global terror reinsurance market that will be prepared to respond as conditions evolve.
Wherever possible, insurers across the globe continue to seek to share with or transfer to the myriad of local governmental terror pools as much terror risk as possible. The government terrorism pools are where most insured terror risk currently resides (1). Accordingly, the reinsurances that protect these pools are by far the largest terror reinsurance transactions in the marketplace today. In the opinion of many reinsurers, they also happen to be the preferred vehicle for reinsurers to isolate and assume terror risk into their portfolios, as explained below.
Terror Reinsurance Pricing
Terror reinsurance pricing methodologies are comparatively less technical than methodologies used for pricing treaties exposed to natural perils only. One reason for this is that probabilistic commercial terror models lack sufficient credibility to strongly dictate a technical price for terror risk.
Nevertheless, reinsurers are using models to price terror more frequently now as they find their own comfort level with balancing technical and non-technical price drivers. Other factors that drive terror reinsurance pricing include:
- Relativities to other terror programs
- Perceived prevailing terror risk levels
- What the market will bear
One factor that could impact demand, and therefore pricing, is the more detailed questions asked by rating agencies regarding capital adequacy, taking fully into account various terrorism loss scenarios and their impact on capital after reinsurance.
In the United States, TRIPRA (2) is a commercial lines program, and personal lines policies of insurersare fully exposed to both TRIPRA-certified and noncertified acts of terrorism. In general, insurers have addressed these risks by having full terrorism - certified and noncertified - included in their property/catastrophe reinsurance programs for personal lines, typically excluding NBCR losses. This protection is frequently provided with no explicit cost breakout.
In September of 2009, Guy Carpenter administered an informal survey of treaty reinsurance underwriters to secure first hand observations on the broad terrorism reinsurance market. What we learned was largely consistent with our beliefs, yet also encouraging in terms of overall market outlook. The findings below elaborate on some key trends:
Key Survey Finding #1: Standalone terror deal flow has decreased over time.
- Buyer/seller disconnects on price
- Other priorities for reinsurance budget or target spend
- Passage of time since last major terror strike
- More terror coverage moved to core reinsurance programs
GC Perspective: Neither the overall conclusion nor the suggested potential drivers were a surprise as they generally align with our perspective. However, we believe this perspective emanates more from the US market than the global market.
Key Survey Finding #2: 83 percent of responding reinsurers are actively seeking new orexpanded terror reinsurance transactions.
GC Perspective: It is pleasing to see the strong market appetite for more terrorism business, but at the same time it emphasizes the current supply/demand imbalance. It is understandable that reinsurers would like to commit more capital to a diversified severity peril, but selling more standalone terrorism reinsurance still appears to require more attractive pricing for potential cedents.
Key Survey Finding #3: Reinsurers prefer geographically discrete opportunities.
GC Perspective: As with all catastrophic perils, we are not surprised at reinsurers’ stated preference to write a portfolio comprised of contracts from distinctly discrete geographies. Where we saw some discrepancy was around the definition of ‘discrete.’ For example, some reinsurers viewed country-specific governmental terror pools as an ideal example of a discrete capacity deployment and therefore a preferred risk, while others considered entire countries to be much too broad to be attractive, preferring instead to focus on insurers that only operate in one region of a given country.
Key Survey Finding #4: NBCR coverage is available through about two-thirds of markets and roughly one-third of reinsurers will consider pandemic covers.
GC Perspective: Both of these findings were somewhat surprising. In the robust terror market immediately following the attacks of September 11, 2001, it was uncommon for reinsurers to write NBCR coverage. Instead, many reinsurers preferred to take a clear underwriting position to avoid that sub-category of the terror peril. The fact that two-thirds of reinsurers will write the coverage today shows a true evolution in underwriting appetite, but ultimately the question of securing NBCR capacity hinges not just on the availability of capacity but also a question of price.
In terms of pandemic, the survey was administered in the midst of the H1N1 threat and we were surprised that roughly a third of reinsurers would provide pandemic coverage. Although the pandemic threat of H1N1 is not terrorism, per se, conceptually one form of terrorism could involve deadly disease spread on a mass basis through a biological attack.
As mentioned earlier, the largest terrorism reinsurance placements in the world are behind governmental pools in countries such as Australia, Belgium and France. These transactions are as large as USD2 billion of limit, which confirms that many reinsurers consider this to be the ideal platform to deploy capacity to terrorism. At USD2 billion, these terrorism placements are on par with the largest global reinsurance transactions for natural perils.
In countries like the United States, where companies individually manage their significant retentions under TRIA, we believe that the per program capacity for terrorism placements that include property is in the range of USD700 million, with that estimated figure fluctuating widely with the location and severity of the original insured policies. With other programs where the subject business is workers compensation, it may be feasible to secure more than USD1 billion of capacity for an individual program. Of course, as with all reinsurance transactions, the accessibility of capacity will expand or contract based on price, type and location of risk covered and overall market conditions, but at present it appears that supply is not an issue.
Many would argue that today we are at or near a peak in terms of hypothetical capacity availability due to a convergence of a variety of underwriting and economic factors in 2009. If that is so, it may also be true that we are just one market-changing event away from a major shift in the supply/demand balance.
Reinsurers typically allocate a fixed amount of capital to terrorism, and they tend not to leverage that capital as aggressively as they do with capital that is committed to natural perils. Because of this, a sudden shock event such as a major terror strike, a significant decrease in coverage from or outright repeal of TRIA, or even the mere threat that either of these events are imminent, could cause a scenario that rapidly depletes available terror capacity.
Should that occur, insurers that have already begun to explore terror risk transfer options with reinsurers may be better positioned to secure coverage in a fast-moving, tightening marketplace.
TRIA (TRIPRA) Update
In May of 2009, and again in February of 2010, President Obama proposed scaling back federal support for TRIA (TRIPRA) in the next fiscal budget process. The stated goal is to save more than USD250 million during the 2011-2014 timeframe to be accomplished via:
- Elimination of domestic terrorism coverage
- Increase in insurer deductible and co-insurance levels (amount to be determined)
- Increase in loss amount that triggers TRIA coverage (amount to be determined)
Commercial insurers remain strongly supportive of TRIPRA, as it provides them an ultimate safety net for their terrorism exposures. As expected, the cuts have been vigorously opposed by the insurance industry, and after passing the TRIA (TRIPRA) extension in 2007, Congress may be under-motivated to again take up the debate prior to the scheduled expiration date of 2014.
A significant reduction in TRIA without a commensurate change to the requirement to offer terrorism coverage could indeed be the type of event that shifts the global landscape toward a tighter terrorism market, but with the Obama Administration continuing to be focused on other agenda items we have our doubts as to whether the administration will be successful in cutting TRIA.
1 For more information on these governmental terror pools, please see Guy Carpenter’s Global Terror Update 2009 at http://www.guycarp.com/portal/extranet/insights/reportsPDF/2009/Global_terror_2009.pdf
2 TRIPRA was signed into law in 2007 and kept in place the basic framework of the Terrorism Risk Insurance Act (TRIA)