Deeper analysis reveals that, more than anything else, this pricing cycle has been capacity-driven.

Insurers' overall GAAP rate of return on average net worth for 2004 was 9.4 percent–5.5 percentage points less than the corresponding rate of return for the Fortune 500 and 0.5 percentage points less than the average rate of return for the insurance industry from 1971 to 2003. And despite the improvement in profitability since the industry's first-ever net loss after taxes in 2001, insurers' average rate of return fell from 11.8 percent during the decade ending 1984 to 10.1 percent during the decade ending 1994 to 6.9 percent during the decade ending 2004.

With insurers still earning subpar rates of return, reports attributing increases in competitive pressures to improvements in profitability must be discounted.

But changes in capacity explain both the hard market that started in mid-1999 and the soft market that is just beginning. Insurers' capacity depends on their surplus, or net worth.

In third-quarter 1999–the quarter when rates started heading upward–insurers' surplus, or net worth, dropped $17.7 billion, or 5.2 percent. And rate increases gained momentum through mid-2002 as insurers' surplus trended downward.

Finally, with industry surplus having risen to a record $393.5 billion at year-end 2004, ISO MarketWatch(R) data shows that changes in rates on commercial renewals dwindled to near zero last December, and surveys now indicate rates are declining.

So far, the declines in rates have been modest, leading some experts to conclude insurance markets have become more rational and this cycle won't be as severe as those of the past. That viewpoint is supported by changes in the investment environment. With stock markets being weak and interest rates still languishing near 40-year lows, combined ratios must now be better than they used to be for insurers to achieve the same overall rate of return.

For example, insurers' GAAP rate of return last surpassed 15 percent in 1987, when it was 17.3 percent and the combined ratio was about 105. To achieve that same rate of return with investment results, tax rates and financial leverage like those in 2004, ISO estimates the combined ratio would have to improve to about 88–roughly 10 percentage points better than the actual combined ratio for 2004.

Advances in technology and analytics may also limit the severity of the cycle. Recent advances have enabled individual insurers to price more accurately and more uniformly across their books of business. Recent advances also have enabled insurers to gauge the effects of their pricing decisions more quickly and, as a result, make quicker but more modest adjustments.

But as insurance markets cycle from hard to soft and then back again, there are inflection points, or brief interludes, when all appears to be in balance and markets appear to be operating in an orderly and rational fashion. We may be in such an interlude, with the current soft market poised to become every bit as severe as those that have preceded it.

Proponents of this view argue human nature and human behavior have not changed. These analysts recognize that technology and analytics have the potential to help temper cyclical swings in insurance prices, and they point out that pricing decisions are still made by people. Even when policies are rated by machine, people set the parameters.

Looking ahead, ISO projects written premium growth will slow from 4.7 percent in 2004 to 2.3 percent in 2005 but then accelerate to 3.3 percent in 2006 as declines in rates gradually taper off. ISO further projects that underwriting profitability will deteriorate as a consequence of recent and forthcoming developments in insurance markets, with the combined ratio rising to about 99 in 2005 and about 101 in 2006.

But a number of wild cards could change the outlook dramatically. For example, ISO's projections assume stock markets and interest rates will remain fairly stable. If stock markets or interest rates surge upward, insurance markets could soften more quickly and more severely, cutting premium growth and detracting from underwriting profitability. ISO's projections also assume normal catastrophe losses and modest loss reserve strengthening.

But the biggest wild card of all is the behavior of market participants themselves. Their actions will determine whether the pricing roller coaster glides to a soft landing or spirals downward.

Frank J. Coyne is the chairman, president and chief executive officer of ISO.

Quotebox (with mug):

“Changes in capacity explain both the hard market that started in mid-1999 and the soft market that is just beginning.”

Frank J. Coyne

Caption for roller coaster pix:

The biggest wild card of all is the behavior of market participants themselves. Their actions will determine whether the pricing roller coaster glides to a soft landing or spirals downward.