It has been suggested that the idea of a property/casualty pricing cycle is now outmoded. James V. Davis believes a cycle still exists and explains what this means to risk managers.
Two of my Willis Corroon colleagues, both fellows of the Casualty Actuarial Society, recently prepared an analysis of likely future trends in the pricing of property/casualty insurance products.1 While the full scope of their analysis cannot be replicated here, it is important for the remainder of this paper to understand their central thesis.
Prices of US property/casualty products have been in a decade-long decline. It has become accepted wisdom that, in the absence of a major catastrophe, prices will continue to soften indefinitely. (Even the size of the catastrophe required to change the direction of the market has continued to escalate. It now is hypothesized that it would take a series of losses of the magnitude of the $15-$16 billion insured loss associated with Hurricane Andrew to turn the market. Alternatively, one insured loss in the $50 billion range might be enough.) Some pundits have suggested that the concept of a pricing cycle is itself an outmoded concept.
In the face of selected industry fundamentals, that point of view is understandable. One of the industry's measures of financial leverage, the premium-to-surplus ratio, was at 1.1:1 at the end of 1996. That was the lowest point for the ratio in 50 years. The ratio of loss and loss-adjustment expense reserves to surplus at 1.4:1 was the lowest in 25 years. With such fundamentals in place, the stage would appear to be set for a long-run continuation of the present fierce level of price competition.
There is, however, reason to believe that by the end of 1999, empirical evidence will show that, yes, there still is an insurance cycle, and that the current one will have reached its bottom. Price decreases spanning all lines of coverage are likely to be in the process of ending, and there may even be modest upward movement in the pricing of some lines.
The change in direction will result from accident year losses growing more rapidly than reflected in calendar year loss reserves. Calendar year reserves are the loss reserves posted for financial reporting purposes. At this stage in the previous pricing cycle, insurers were optimistic in setting initial workers' compensation reserves. It is believed that this situation is being repeated again in this cycle for workers' compensation and/or other long tail lines.
The importance of this change, which Messrs Alff and Davenport believe began in 1997, is that accident year loss reserves will need to be strengthened going forward. This will erode the ability of property/casualty insurance companies to produce the quarter-over-quarter earnings improvement that their shareholders and security analysts are demanding. Pricing their products more adequately in relationship to the losses that are flowing from current policies will be needed to regenerate upward earnings momentum.
(In addition to their analysis of industry data, Messrs Alff and Davenport also conducted interviews with senior actuaries of seven major property/casualty companies. These individuals confirmed the pressures through June of 1997 were to increase writings in the key lines, products, and states, but to work hard to maintain pricing integrity. Messrs Alff and Davenport believe that as this pricing cycle draws to a close late in 1999, the developing loss experience for accident years 1998 and 1999 will increase the pressure to improve calendar year experience through increased product pricing.)
If this forecast is accurate, it means that risk managers and brokers will need to step back from the techniques of "hard market" marketing. A more disciplined market will give professionals of all backgrounds a chance to better display the depth of some aspects of that professionalism.
The selling of a risk to underwriters will take on added importance. (Even in the current market, those skills are needed for that relatively small subset of risks with difficult to place exposures.) An important part of the process will be the creation of in-depth underwriting submissions. For complex risks, these submissions are likely to be computer-generated multimedia presentations which will combine video, graphs, text, and numerical output. Submitted to underwriters on CDs, they will provide the former with a better understanding of the risks being written.
(Such submissions, of course, already are being used. One firm with a problematic products exposure partially credits such a submission with leading to a multi-million dollar reduction in its premiums. Another insured so liked its presentation that it now uses it with security analysts to help the latter gain a better understanding of the firm.)
In a tightening market, completeness of loss data and its development to ultimate incurred will once again be central to the underwriting process. The use of insured-specific loss development factors will be needed to counterbalance the use of insurer or insurance industry loss development factors by others. Use of other than the insured's own factors may lead to an erroneous perception of the effectiveness of loss prevention and safety programs in the reduction of losses.
Accurate description of the loss prevention and safety programs will also help favorably influence pricing in an environment in which each insured's loss ratio is important. A major objective of the underwriting submission will be to remove as much uncertainty as possible from the underwriter's mind.
It may be necessary to more carefully choose the underwriters to which a risk is submitted. This is not likely, however, to be as important as it was in the mid 1980s when insurers were under pressure to allocate limited capital to their most desirable risks. (Some insurers then were writing at 5:1 or worse premium-to-surplus ratios as opposed to the approximately 1:1 ratio that exists now.) This time the emphasis will be on matching a risk to an underwriter's particular area of expertise.
In a tighter pricing environment, more care will need to be taken in the analysis of an insurer's charges. Internal charges such as profit and administration and claims supervision should be scrutinized. In addition, selected pass through charges (for example, state premium taxes, residual market loads) should be reviewed.
There will be pressure to restrict the scope of coverage and to increase retentions. (These moves actually are likely to precede any upward movement in prices.) It then will be interesting to observe how underwriters deal with the wide variety of risk financing options that have been utilized in recent years. Historically in a tightening market, underwriters have retrenched and become less flexible in the program designs they will entertain. Will risk managers still be able to buy guaranteed cost (NPV), all-line aggregate, integrated risk, multi-year single limit, and embedded derivative programs?
One suspects that they will be able to do so. The hit to underwriter credibility would be substantial if there was widespread abandonment of the programs now being aggressively sold. That aside, underwriters have substantially improved their data bases and can selectively target those lines of coverage or program designs that are producing their losses. Underwriters are also very sensitive to the alternative risk transfer market and the business that will be lost to it if there are significant increases in premiums or reduction in an insured's ability to approximate, at least in some instances, earnings per share coverage. In addition to the restraining influence of the ART market, primary underwriters will be concerned about those reinsurers who will make their capacity available behind captive insurance companies.
The impact of these influences does mean that the purveyors of capital market alternatives will probably find that their products will still be too expensive to enjoy wide-spread acceptance. Most of the capital market transactions done so far have been substitutes for traditional reinsurance where a number of the buyers have acknowledged that they were paying a premium to gain experience with the process.
This year or next may be the time for risk managers to avail themselves of the guaranteed rate, non-cancelable, multi-year contracts that underwriters are selling. If entered into this year, the potential also exists to cancel and rewrite them in 1999 if both parties are agreeable.
(A fringe benefit of a multi-year contract may be that it will force an underwriter to reveal its year 2000 intentions, ie, the introduction of year 2000 exclusions for specified coverages. Based on sketchy data, programs extending into the year 2000 and beyond have been written with and without such exclusions.)
In sum, the bottom of the current pricing cycle may be reached late in 1999. If this indeed happens, risk managers will need to represent their risks to underwriters even more effectively than they do now. An in-depth underwriting submission will be part of that effort. Availability of expansive coverage and program designs may become more limited. The potential exists to lock in these kinds of program features now through utilization of multi-year programs.
James V. Davis, is chairman and ceo of Advanced Risk Management Services, Willis Corroon Corporation.
1. Gregory Alff and Edgar Davenport, "An Actuarial Perspective: Insurance Industry Pricing and Reserving Through 2000", Risk Management (VL: January 1998), 19-26.