You can search for articles in back issues of Contingencies from July/August 2000 to March/April 2009 using the search box to the right. Simply type in subject words, author's name, or article title and click search. To search for articles from May/June 2009 to the present, go to the current digital issue of the magazine and use the search function on the left of the top navigation bar.
Making Good Sausage
By Michael G. McCarter
ACTUARIAL WORK PRODUCT IS INCREASINGLY VALUED by policy-makers in Washington and elsewhere. Academy volunteers have established a reputation for high-quality analysis that’s making the Academy increasingly likely to be called upon to shed light on important issues. Producing and communicating such analysis, however, continues to be a challenge for actuaries, because the deliberative, careful process actuaries are accustomed to using can be poorly suited to the realities of the political process. Legislators and their staffs often find themselves confronted with scores of complex issues in the waning days of a session, leading to a high-pressure whirlwind prone to lightning changes in the assumptions that the actuaries have relied on.
In August 2005, I was asked to chair the Academy’s subgroup that was just being formed to respond to congressional requests for actuarial analysis and modeling of proposals for extending or replacing the Terrorism Risk Insurance Act of 2002 (TRIA). I was fortunate to have a subgroup comprising very knowledgeable and capable Academy volunteers. Actually, I should say extremely fortunate, because we faced and overcame many hurdles to our work that I did not at all anticipate when I accepted the position of chair.
We had barely gotten started when Hurricane Katrina slammed into the Gulf Coast, which meant that several of our volunteers could devote much less time to our project than we had expected they could. The damage caused by Katrina and its successors continued to make volunteer participation difficult, but we also faced issues that had nothing to do with Mother Nature. Obstacles blown by political winds into the subgroup’s path actually caused us even more delays in completing our analysis.
When approaching a complicated insurance problem, actuaries are trained to consider it from the many perspectives that may be important to understanding both the problem itself and any constraints on its solutions. The subgroup’s desired first step was to understand clearly the legislative proposals being drafted so we could assure ourselves we were doing the right sort of analysis.
Laws are like sausages. It’s better not to see them being made.
— Otto von Bismarck, German Prussian politician (1815–1898)
Here is where Politics (notice the capital “P”) first reared its ugly head. Many of the key political players in Washington had already taken positions regarding terrorism insurance. No one developing a new proposal wanted to distribute a written draft that could fall into the hands of declared opponents who could then attack the draft before it was ready to be seriously discussed. This meant that the Academy’s TRIA subgroup had to winkle out the provisions of the proposals we were being asked to model, using only first, second, and even third-hand oral descriptions of those proposals.
We did our best to understand, and then we drafted descriptions of our understanding that our congressional clients could confirm or correct. However, Katrina and other issues were facing Congress, too, and the proposals themselves were evolving, so we never received final closure on the descriptions. Eventually, we had to proceed using what we had if we were to accomplish anything useful before the congressional session ended.
One of the proposals was an extension of the existing TRIA with some modifications, while another was a more complicated attempt to develop a market-based permanent solution to the problem of terrorism insurance. Clearly, a proposal for a new mechanism would generate far more questions and uncertainties than one that essentially extended current law, so the lack of access to a written draft was a far greater handicap for the second proposal than for the first.
The subgroup focused its modeling efforts on the first proposal and made what comments we could on the second. Our hope was that by providing Congress with a draft of our analysis for discussion, we could elicit more details on whatever proposals were then still alive politically.
The TRIA subgroup included volunteers from Insurance Services Office and AIR Worldwide, who had access to data and models based on industry-wide or countrywide exposures. One of the critical pieces of information arising from AIR Worldwide’s models was the potential insurance industry loss that could result from terrorist use of weapons of mass destruction. The TRIA subgroup prepared a discussion draft of its analysis that received Academy peer review and other reviews and then was presented to Congress in November.
Almost simultaneously with the delivery of our draft, however, both the House and Senate versions of the TRIA extension became publicly available. Neither was exactly like any proposal we had modeled, so that many of the details of the analysis we had worked so hard to complete didn’t apply to the further consideration of either version. At the same time, some of our analysis was extremely relevant to the question of the need for some sort of federal legislation in regard to terrorism insurance. Our analysis showed that some potential terrorism losses were large enough to disrupt the nation’s insurance reparations system.
Plan to Add Value
After discussion within the Academy, we decided that a public statement, excerpting from our analysis of the potential impact of very large terrorist events, would be a significant contribution to the debate in Washington. That public statement, based largely on relevant portions of the executive summary of our earlier report to Congress, was released on Dec. 2, 2005. The statement did not advocate a particular proposal among those being considered by Congress. It did express the basis for our conclusion that some sort of federal legislation was necessary if terrorism insurance were to be widely and readily available, and it did express our belief that there should be a deliberative process for developing recommendations for a permanent way of dealing with the risk of terrorism.
As I write this commentary, Congress has passed and President Bush has signed S. 467, the Terrorism Risk Insurance Extension Act of 2005 (TRIEA). I think that the work of the Academy’s TRIA subgroup helped to accomplish this result, but I wonder whether we can be more effective in helping Congress consider the more detailed aspects of any future terrorism insurance legislation. In 2006, we get a chance to find out.
One of the provisions of TRIEA is a requirement that the president’s Working Group on Financial Markets report to Congress by Sept. 30, 2006, on the long-term availability and affordability of terrorism insurance. The Academy’s TRIA subgroup plans to provide input to the president’s working group, as I’m sure many other groups plan to do as well. I hope we can add value by providing a professional actuarial perspective on the issues relating to providing a sustainable insurance market for the risk of terrorism. Based on what we’ve seen to date, there are likely to be a number of political arguments made that I hope actuarial analysis can shed some light on. Here are two of them:
While the competitive marketplace is a wonderful tool for encouraging improvements in product effectiveness and efficiency, it’s not a stand-alone entity independent of a society’s legal and economic structure.
A major purpose of government is to establish and maintain institutions that are preconditions for effective competitive markets. Examples are the court system and contract law. Enforcement of individual rights under these institutions allows for the reliable functioning of markets.
If terrorism coverage is to be offered to all comers, insurers need to be able to understand and price the risks they’re taking on, as well as to understand the accumulation of those exposures.
Unfortunately, there are several barriers to such understandings. One barrier is that certain insurance products, such as workers’ compensation, are defined so that insurers cannot control terrorism exposure except by entirely avoiding the underlying risk. No exclusions or deductibles related to terrorism losses can be used.
Another barrier is that terrorists respond to changes in the defenses against them, changing the likelihood that any given risk could suffer a terrorist attack. Still another barrier is that terrorists may have access to weapons of mass destruction and may deliberately unleash, in an unpredictable manner, devices that exceed natural disasters in their capacity to kill human beings or destroy the usefulness of property.
To summarize, our analysis has shown that providing terrorism coverage can expose the insurance industry to potentially massive losses for which we currently lack sufficient risk-assessment tools. No private insurance system can provide stable capacity when it can’t quantify the risks it faces with reasonable confidence. Under these conditions, the competitive marketplace, unaided by federal legislation, cannot be expected to make available widespread terrorism coverage.
First, many potential terrorism claims would fall entirely within the individual insurer deductibles under TRIEA, so that the taxpayer would pay nothing for such claims and the “subsidy” would be zero. Second, many other potential terrorism claims would fall within the levels where any payments must be recouped, making any conceivable “subsidy” limited to the time value of money, and subject to the third consideration below.
The third consideration is the required mandatory offer of terrorism coverage for the relevant lines. Insurers have no ability to tailor coverage to control their accumulations of terrorism exposure except by declining to write the underlying non-terrorism exposure. This mandated offer of coverage imposes capital costs to the extent uncompensated claims pose a risk to the insurer’s surplus.
To the extent the taxpayer is liable for some payments, those payments come from coverages the insurer had no option but to offer. It’s not correct to talk about a “subsidy” due to TRIEA without also talking about the costs imposed by that legislation.
There are a vast number of other issues that need to be considered in the development of a long-term solution to the issue of terrorism insurance. Here are a few of these issues:
Terrorism insurance is a complex and developing issue with many ramifications that actuaries can address. Unfortunately, in 2005, we saw evidence that some parties were content with simple slogans such as “subsidies” or “market competition,” without considering the issues underlying these words. Our challenge as actuaries is to find ways of dealing with the complexity of these issues and communicating our conclusions in ways that reduce the attractiveness of the simple but inadequate slogans.
Otto von Bismarck famously advised the squeamish or fastidious not to watch either sausage or legislation being made. Given the Iron Chancellor’s interest in insurance issues, it’s conceivable he had actuaries in mind when he made that remark. But making good legislation requires a good understanding of the issues, and actuaries have much to offer our nation’s policy-makers in that regard.
Our best hope for being effective is to help in establishing the terms of the debate. If we’re to help in making good sausage, we should provide our advice in time to avoid the meat grinder at the end of any legislative session.
MICHAEL G. McCARTER is vice president for industry and regulatory affairs at American International Group in New York City.
Contingencies (ISSN 1048-9851) is published by the American Academy of Actuaries, 1100 17th St. NW, 7th floor, Washington, DC 20036. The basic annual subscription rate is included in Academy dues. The nonmember rate is $24. Periodicals postage paid at Washington, DC, and at additional mailing offices. BPA circulation audited.
This article may not be reproduced in whole or in part without written permission of the publisher. Opinions expressed in signed articles are those of the author and do not necessarily reflect official policy of the American Academy of Actuaries.
Up To Code: