A few years ago, I posed the question of whether the nonadmitted market had reached a “new normal” in terms of market share, or whether it was destined to retreat to its historic norms. To date, there has been little sign of retreat. According to A.M. Best’s annual survey of the market, surplus lines market share edged up to 7.1% in 2014. 1
With the previous hard market peak a decade past, the market’s resilience suggests that maybe, instead of fretting about retreat, we should be considering a different question: Could the surplus lines market be heading even higher?
In this post, I’ll cover three ongoing trends that could propel the market beyond a 10% share: 1) declining interest rates, 2) big data, and 3) catastrophe risk.
In 2004, I took out a 30-year purchase mortgage with a fixed rate of 5.375%, and I crowed to my wife about how low the rate was, and how we couldn’t possibly expect to see a rate this low ever again. In 2009, we again took out a 30-year purchase mortgage, this time with a fixed rate of 4.375%. I again crowed to my wife about how low the rate was, this time opining that it was a “once-in-a-lifetime opportunity” created by the recent financial crisis, and that we couldn’t possibly see rates this low again. Nevertheless, in 2012, we refinanced, with no closing costs and points, to a 30-year mortgage at 3.50%. This time, I held my tongue. With plenty of egg on my face already, I decided that I should get out of the business of forecasting the bottom of the interest rate cycle.
Only time will tell whether my capitulation was a case of “better late than never” or a sure sign that bottom was in the offing. What is indisputable, however, is that intermediate and long-term interest rates have been in secular decline for over 30 years:
Moreover, strange as it seems, yields have significant room to fall if we continue along the path of other rich countries. As of the time of this writing in February of 2016, Japan’s 10-year government bond yield had hit zero, while Germany’s stood at 0.2%. The 10-year US Treasury yield? A comparatively lofty 1.6%.
As I’ve written before, low investment yield environments wreak havoc with casualty insurance pricing benchmarks. The standard profitability metrics—such as loss ratios and combined ratios—are not well-suited for measurement in long-tailed lines because they have no built-in adjustments for interest rates. As a result, the headline figures can be misleading.
To take an example, current Workers Compensation results seem quite good in the context of historical combined ratios; unfortunately, the figure needs to be interpreted in the context of the interest rate environment. For example, the calendar year 2013 combined ratio of 101 matched the industry’s combined ratio for the 1997 calendar year. The 101 in 1997, however, occurred when the 5-year Treasury yield was north of 6%, while the 5-year Treasury yield started 2013 south of 1%. Insurance companies were earning far more on their “float” in 1997, so the 1997 underwriting result was a cause for celebration. The 2013 result? Maybe a shoulder shrug.
It is not easy to communicate this to regulators. Falling investment yields mean that insurers, in typical cases where the impact is not offset fully by a lower cost of capital, need higher underwriting profit and contingencies (UPC) factors in their rates for long-tailed business. But requests for higher “profit” provisions are coming at a time of seemingly high underwriting “profits,” which makes for a challenging sales pitch in a rate filing.
Therein lies the potential boon for the surplus lines market. The divergence between economic realities and traditional metrics will expand if investment yields continue to fall. To the extent that regulators reject arguments for increased UPC provisions, rate suppression will become more acute in long-tailed lines, leading more business to surplus lines and to residual markets.
After confessing to my ignorance about the course of interest rates, there’s not much point to throwing my hat in the three-ring circus of “Big Data” prognostication. But, just for sport, let’s take the pundits at their word. Suppose that Big Data is on the verge of transforming insurance. Whenever an individual or business submits an application for insurance, an army of cyber-bots will be unleashed, scouring cyberspace for every scrap of recorded digital activity relevant to assessing the application. Every like, dislike, friend, defriend, and hashtag from social media will be assimilated and dissected, along with a torrent of data from every phone application to which the user unwittingly granted access by tapping “Accept” without reading the disclosure.
However, if this scenario in fact materializes, my guess is that only a fraction of the information utilized will be revealed in filed risk classification schemes. This is partly due to the nature of regulation: It is hard to imagine regulators balking at the use of credit scoring in insurance rating (several large states currently ban the practice) but embracing cyber-stalking. Moreover, given the multitude of possible pricing factors, companies may be reluctant to tip their hands to their competitors and possibly also to their policyholders. Thus, even as Big Data becomes “big” in insurance, it may remain in the shadows and out of the spotlight of the standard market.
Nevertheless, the Big Data genie has escaped the bottle, and companies are going to use it in some way. If insurers don’t have pricing and contracting flexibility corresponding to their data-enhanced understanding of the risk, it is hard to avoid the conclusion that a growing number of accounts scoring poorly on cyber-metrics will be redirected to the surplus lines market due to “underwriting decisions” in the standard market.
Property catastrophe risk has been a huge growth area for surplus lines underwriters over the past two decades. Yet, in some ways, we are seeing just the tip of the iceberg.
Much catastrophe risk is uninsured, in part because lenders often don’t compel borrowers to carry earthquake and flood insurance. For example, the Report from the California Department of Insurance reveals that the only about 10% of both residential and commercial policies carry earthquake insurance. In flood insurance, FEMA reports that only 53% of homes in Special Flood Hazard Areas (SFHAs—where federally backed mortgages are required to carry insurance) have coverage; outside of SFHAs, very few homes have coverage. There is evidently much room for the private market to grow.
Demand has not been helped by the absence of major disasters in recent years. The last significant earthquake for the US market was Northridge in 1994, and, though the wind has blown at times in the East, nothing has approached the high-water mark set by Katrina in 2005. This relatively tranquility also has effects on the supply side. Standard carriers have ventured back into the market in certain areas, with the result being an ongoing depopulation of various government-sponsored catastrophe facilities, such as Citizens Property Insurance Corporation (CPIC) in Florida.
Mother Nature will strike again at some point, shocking the current insurance system. How our society will respond is hard to predict, and societal responses are not always favorable for private markets: indeed, various forms of nationalized catastrophe insurance exist in developed countries. However, it is worth noting that the recent history of federal catastrophe insurance market policy in the terrorism insurance market (through the Terrorism Risk Insurance Program) was conducive to private underwriting, and the renewal of both the Terrorism Risk Insurance Program and the National Flood Insurance Program featured reduction of risk subsidies and of federal exposure. Should the federal response to the next event follow in this line, huge opportunities could emerge for the surplus lines market.
The changes discussed above are unlikely to happen overnight. Indeed, in some cases, the changes may not happen at all: Interest rates are not guaranteed to fall, nor is demand for private catastrophe insurance guaranteed to materialize. That said, these are potentially powerful potential catalysts for growth in a surplus lines market that possesses the expertise and flexibility to manage risks in a changing world. The last 15 years were truly spectacular for surplus lines growth. Could the next 15 offer even more?
- A.M. Best’s market share figures are higher than those I have reported in the past due to their inclusion of Lloyd’s and other alien underwriters, as well as some methodological differences. ↩