Market Forecast Cloudy For Workers' Comp Insurers

("Premium leakage" is a catch-all category for other factors that would impact premium. Examples include movement to or from alternative markets, and changes in premium collectibility–such as due to policyholder bankruptcies.)

NCCI has leveraged reported data from workers' comp policies to further analyze price changes during 2000. Significant differences were observed, for example, on new business as compared to renewals (that is, those policies remaining with the same insurer). Little or no change was observed in the bottom-line price being paid by renewed policyholders, while rates increased for business on the move.

Despite the increase in overall premium, however, many carriers did not make money in workers' comp during 2000. NCCI has preliminarily projected that the calendar year combined ratio for 2000 was about 121 for the industry–the fifth year in a row that there was deterioration from the prior year.

For accident-year numbers, which are typically more reflective of profitability on more recent policies, NCCI has projected a combined ratio of 136 for 2000 based on preliminary information. (These combined ratios and some of the other numbers discussed in this article will be updated shortly based on more complete information and they will be posted at www.ncci.com.)

The numbers on a combined ratio basis do not, by definition, reflect investment income–investment combined with underwriting results yields operating results. In the middle-to-late-1990s and into 2000, the industry experienced fairly stable and favorable investment returns. For 2000, NCCI expects to find that operating results continued to decline, reflecting a very slight operating loss on an industrywide basis for workers' comp.

Reserve deficiencies have also grown larger in recent years. At this time, NCCI is projecting about a $20 billion workers' comp reserve deficiency at the end of 2000. For accident years 1998 and 1999, the industrys reserves reflected some strengthening during 2000. However, the accident years prior to 1998 continued to experience a bit more in the way of reserve releases.

Reserve deficiency can be anticipated at any given point in time due to discounting of reserves. However, our analysis has revealed that the overall impact of reserve discounting has remained fairly stable at the same time that our estimates of reserve deficiency have grown over the past few year-ends.

Interestingly, loss costs filed by workers' comp rating bureaus havent changed much between 1999 and 2001. While the last two years have seen an overall average increase in loss costs (after six years of decreases), this is almost entirely driven by increases in California loss costs. Focusing on states where NCCI files loss costs, we find negligible changes in the overall average loss cost level.

How can loss costs be so stable with combined ratios in the 130s in recent accident years? Taking a deeper look at the numbers and making a couple of adjustments–namely, taking out California results and reflecting premiums at bureau loss cost levels (prior to any individual carrier pricing actions)–the combined ratio would be about 100. And a combined ratio of 100, even in todays lower leverage environment, would likely yield a reasonable return on capital for the industry on average.

This analysis leads to the conclusion that the primary driver of workers' comp underwriting results today is pricing decisions made by insurers, not inadequate loss costs.

On the claims side, a mixed bag of higher costs and lower frequency is also sending mixed messages to insurers.

Statistics clearly show that claims frequency has been persistently declining since the early 1990s. NCCI recently published a study showing that the decline in frequency of workers' comp claims spanned almost all occupations, and thus that the overall shift in occupational mix had no impact on the decline in frequency.

This decline in frequency is a dramatic story–a great development for American workers and employers. The improvements in frequency, along with much greater workplace safety, are obviously a win/win for everyone involved in the system.

As frequency has declined, however, indemnity and medical costs per workers' comp claim have been on the upswing since 1994, with average annual increases of 5 percent for indemnity and 7 percent for medical. Anecdotal evidence suggests that medical inflation will continue to accelerate, although that has not emerged in the reported numbers. An important area to observe is whether frequency will continue to improve and help keep overall workers' comp claim costs in check.

Looking at the economics of property-casualty industry performance, we find that the health of workers' comp is inextricably linked to the health of the broader p-c business. Unfortunately, for the past few years, p-c economics has continued to be characterized by rising surplus and falling returns.

As reported by Value Line, the return on equity for p-c insurance averaged just an estimated 10 percent in 2000. Thats not only worse than many other industries (drug manufacturers averaged over 33 percent return on equity, for example), it is well below even the cost of capital–which we estimate to be in the 13-to-14 percent range.

At the same time, the surplus available to the p-c industry has grown tremendously for a long time. In fact, over the last 25 years, theres been about a 20-fold increase–or about 13 percent per year–in surplus. Since premium over the same period increased at a much slower pace, industrywide leverage (as measured by the ratio of premium to surplus) has decreased significantly.

It can be instructive to examine return on surplus as opposed to return on net earned premium. Based on a 10-year rolling average, return on premium has been remarkably consistent, averaging between 6 and 9 percent. During the same period, return on surplus ranged from more than 15 percent in 1979 to less than 9 percent over the past two years. In fact, return on surplus has now fallen below return on premium for the first time in more than two decades.

What does that mean? Ten years ago, the industry had a 13 percent return on surplus. A breakdown of that return on surplus shows that 7 percentage points of it came from earnings on surplus, and 6 percentage points of it came from earnings on reserves and profits in the insurance operation.

But in 2000, while earnings on surplus were still 7 percent, the reserves and profits from insurance operations contributed just 1 percentage point of the 8 percent reported return on surplus. In hindsight, many companies would have been better off ceasing insurance operations altogether and instead investing their surplus with three or four financial managers. Thats the effect this negative leverage has–it dramatically dilutes the contribution of earnings from insurance operations.

Meanwhile, there can be other downsides to this growth in surplus.

For example, deploying surplus can be a challenge because insurance regulators are unlikely to allow insurers to take capital out of the business until the last claim is paid–particularly for workers' comp. So as surplus rises, rating agencies and regulators start to raise the bar, requiring a higher surplus amount to maintain an "A" rating.

The result is that capital gains dont produce capital that can be leveraged further. Instead, they create more surplus despite premium growth.

In short, excess surplus and declining earnings are expected to affect p-c insurance operations for the foreseeable future. Surplus for the p-c industry declined by approximately 5 percent during 2000. It remains to be seen what might happen to surplus in the future.

Robert Blanco is head actuary for the National Council on Compensation Insurance, Inc. in Boca Raton, Fla.

Reproduced from National Underwriter Property & Casualty/Risk & Benefits Management Edition, August 27, 2001. Copyright 2001 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.

Contact Webmaster