Over the past several weeks, several industry observers and analysts have tried to put a number on the insurance industry’s aggregate subprime-related loss exposure. At one end, Bear Stearns on January 24, 2008 estimated the industry’s exposure at $8-9 billion (refer here). By contrast, on February 8, 2008, Lehman Brothers estimated (here) that the insurance industry’s losses might range up to $3 billion, and on February 6, 2008, Advisen announced (here) that it will be releasing a report estimating that the industry’s ultimate losses at $3.6 billion.

I don’t envy these experts whose job it is to try to quantify something as big, amorphous and evolving as the subprime-related litigation wave. Nor do I profess to have any particular insight into whose estimate is more accurate or what the ultimate number will be. I do have some observations about some considerations that are or should be being taken into account in making these kinds of estimates, in light of the circumstances surrounding the evolving subprime meltdown. (I should add that in making these observations, I have not had the benefit of reading the entire Advisen report, which as of this writing is not yet available; I have only had an opportunity to review the press release summary.)

In general, I think the various estimates have correctly noted that a potentially large portion of the amounts to be paid in settlements or judgments in the subprime litigation may not represent insured loss. In particular, the observers have correctly noted that many of the largest commercial and investment banks that are involved in the subprime-related litigation carry very large self-insured retentions and also often carry only Side A insurance programs (covering only nonindemnifiable loss, unlikely to occur here for these entities) or in some cases no insurance at all for certain exposures. These various observers have made a number of other valid observations concerning other factors that could restrict the impact of subprime losses for D & O insurers.

But at the same time, it seems to me that there are a number of other considerations that these observers have undervalued or even overlooked in assessing the possible impact of the subprime meltdown on insurers.

First and foremost, I think it is important to stress that we are only at the very earliest stages of the emergence of the subprime-related litigation. To be sure, there are (as documented here) already 43 subprime-related class action lawsuits, as well as nine subprime-related ERISA lawsuits, but before all is said and done, there are going to be many, many more of these and other kinds of lawsuits. We have not even completed the first round of subprime loss truth-telling (refer here), and it is probable that there will be even further deterioration in the mortgages underlying the subprime-backed assets as homeowners find it easier to walk away that to continue to pay down debt on a house that is declining in value (about which refer here).

As Couglin Stoia attorney Sam Rudman observed at last week’s PLUS D & O Symposium, there are likely to be more securities class action lawsuits in 2008 than any year since the passage of the PSLRA (Rudman is himself already involved as plaintiffs’ counsel on many of the subprime-related lawsuits).The subprime-related litigation wave is likely to continue to emerge well into 2009 and possibly beyond (just as the options backdating litigation wave continues to emerge). The possible extent of this future litigation threat may be discerned from the recent litigation commenced against the Cadawalader firm (about which refer here), in which the allegations relate to commercial mortgage securitization documents the firm prepared in 1997. In other words, any dollar estimate of the possible subprime-related insurance losses should be accompanied by a healthy appreciation of how little of the ultimate amount of subprime-related litigation we can currently even see. Since we still don’t know how big of an event this ultimately will be, and because it is likely to be years before we have clear idea, any attempt at quantification should carry some very substantial caveats.

Second, many of these estimates seem to presume that the insurance industry’s subprime-related losses will be limited to the financial institutions sector. I do not think this is a conservative assumption. To the contrary, I think it should be assumed that the subprime-related litigation wave will both spread beyond subprime and beyond the financial sector (as I discuss at greater length here). The recent securities class action lawsuits against student loan company SLM Corporation (about which refer here) and Levitt Homes (about which refer here) underscore that the claims have already spread. Bristol Myers Squibb’s recent $275 million write-down for subprime-related investment losses (refer here) further highlights that the credit crisis is no longer just about the financial sector. The possibility of further credit-related losses in many sectors outside the financial sector, and for ensuing claims, at this point seems likely — or at least that would appear to be the conservative assumption.

Third, much of the analysis of the insurance industry’s exposure has been concentrated largely (although, it must be recognized, not exclusively) on potential losses for D & O insurers. To be sure, the growing number of subprime-related securities class action lawsuits represents a very substantial threat to the D & O insurance industry. But the potential for insured losses in coverage lines outside of D & O could also be very substantial. By way of illustration, State Street’s recent $618 million charge for anticipated subprime-related litigation expenses was in connection with lawsuits that do not (as discussed in my recent post, here) appear to implicate D & O coverage, but that could present significant fiduciary liability or even investment management E & O losses.

By the same token, the recently revised complaint in the subprime-related securities litigation involving Countrywide (about which refer here) added accountant liability claims, as well as claims against Countrywide’s offering underwriters. Other professionals undoubtedly will find themselves caught up in subprime related litigation, including, for example, lawyers; hedge fund and pension fund managers; mortgage brokers; appraisers and surveyors; real estate brokers; and insurance agents, among many others. The cumulative losses from claims against other professionals could be very substantial, and at this early stage particularly difficult to prognosticate.

Even with respect to the analysts’ breakdown of the likely D & O losses, the breadth of the current and likely future claims may or may not be being fully taken into account. That is, while it is true that some of the lawsuits against the largest financial institutions may not, because of the way that these entities structure their insurance, involve the prospect for insured losses, most of the current and likely future subprime litigation defendants do not have these types of insurance arrangements. As the claims spread to secondary players and targets in the hinterlands (about which refer here), the claims are hitting defendants that have more traditional insurance structures. Those (far more numerous) claims may involved a greater percentage of insured losses than (the relatively few claims, as a percentage matter) against the largest banks and financial institutions.

Fourth, I am well aware that one of the issues with which these analysts have had to grapple is the need to try and put the subprime meltdown into context. The challenge is not just to say how it compares, for example, to the S & L crisis or the bursting of the dotcom bubble, but also to come up with a figure for those prior events in order to compare the current subprime crisis. I don’t have data for those prior events, but I do know that the still unfolding options backdating scandal may present a useful comparison. As I have detailed in another post today, the options backdating losses, on the few cases that have been resolved so far, already represent in aggregate some very impressive numbers. There are many more options backdating cases yet to be resolved. The total options backdating related losses are likely to by very substantial. Given that just about everyone assumes that the subprime related crisis represents an even greater threat to insurers than the options backdating scandal, the implication is that the subprime related losses could be very significant indeed.

Fifth, whatever else might be said, nothing meaningful about the extent of the subprime threat can be derived from the D & O insurers’ current marketplace behavior. My comment here relates specifically to the comment in the Lehman Brothers report linked above that "if insurers were concerned about suffering multi-billion dollar subprime D & O losses that could spread outside financial institutions sector, the market would tighten significantly." If the D & O industry had a long track record of skillfully adjusting its prices to changing exposures, this remark might have greater validity. Unfortunately, the industry’s consistent history suggests that the industry is only capable of disciplining itself when losses become so painful that it is forced to change its ways. The current D & O pricing environment is a reflection only of the amount of available capacity, not of any calibration to emerging exposures. The marketplace will remain competitive until cumulating losses force the changes of necessity, and then any changes would be abrupt and disruptive — as they have always been in the past.

Sixth, as most of the analysts have noted, the defense expense associated with the subprime cases in and of itself could be staggering. As an example of how expensive these cases can be, Apollo Group recently reported (here) that it had spent $25 million dollars taking the securities lawsuit pending against the company through trial. Because of the legal and factual complexity surrounding the subprime cases, they could be extremely costly to defend. Much of the associated defense expense, other than for the large investment bank defendants, is likely to be covered loss. For each of the securities cases, the defense expenses are likely to be many millions of dollars, and, for the cases in the aggregate (including those already filed and those yet to be filed), to be many hundreds (and possibly thousands) of millions of dollars. To these costs must be added the costs of defending the claims raised against other professionals.

Finally, it would be unfortunate if the subprime hype were to obscure the fact that the subprime-related litigation is only one of several very important current developments affecting D & O insurers’ exposure. As I have noted elsewhere (refer here) securities litigation levels would be elevated compared to the prior two years’ activity levels even without the subprime-related litigation. The Securities Litigation Watch blog recently noted (here) that January 2008 securities activity remained at elevated levels, only in part because of the subprime related litigation. None of this could be discerned from D & O insurers’ current conduct. It has been ever thus.

Blog Warning: This week I hope to be making some long needed adjustments to The D & O Diary. While these changes are taking place, I will not be adding any new blog posts (although the current posts will remain available). These adjustments should result in several improvments to The D & O Diary. I will report further on the adjustments once they have been completed.