Imágenes de páginas
PDF
EPUB

Risk-Linked Securities Do
Not Have Broad Investor
Participation

Catastrophe bonds have not attracted a wide range of investors beyond institutional investors. Investor participation in risk-linked securities is limited in part because the risks of these securities are difficult to assess. Investment bank representatives and investment advisors we interviewed noted that catastrophe bonds have thus far been issued only to sophisticated institutional investors and a small number of large investment fund managers for inclusion in bond portfolios that include noninvestment-grade bonds. Most catastrophe bonds carry noninvestmentgrade bond ratings from the rating agencies, but a low rating by itself has not been a barrier to active investor interest in other types of bonds, such as corporate bonds. The investment fund managers told us that catastrophe bonds comprise 3 percent or less of the portfolios in which they are included. On the one hand, the managers like the diversification aspects of catastrophe bonds because the risks are generally uncorrelated with the credit risks of other parts of the bond portfolio. On the other hand, managers stated that they have concerns about the limited liquidity and track record of catastrophe bonds as well as the lack of in-house expertise to understand the perils, indexes, and other features of the bonds."

As requested, we explored the potential for individual investors to purchase shares in mutual funds that purchase catastrophe bonds for inclusion with other securities in a mixed asset fund. We analyzed the SEC rules governing catastrophe bond issuance and mutual fund composition and confirmed with SEC that these rules and regulations do not preclude mutual funds from purchasing catastrophe bonds. One of the investment advisors we interviewed told us that his firm included a small amount of catastrophe bonds in mutual funds sold to the public. However, a mutual fund industry association official told us that the mutual fund companies that the association surveyed-including three of the largest-have not included catastrophe bonds in funds available to individual investors because the companies lack the capacity to evaluate the risks. The mutual fund industry official also raised the issue of whether the risk associated with risk-linked securities would be appropriate or suitable for investments by a broad range of investors, including moderate-income investors.

The September 9, 2002, comment letter from RAA notes that no catastrophe bond
contracts have been triggered by catastrophic events.

Agency Comments and
Our Evaluation

We received written comments on a draft of this report from NAIC, RAA,
and BMA. We also obtained technical comments from Treasury, SEC,
CFTC, NAIC, RAA, and BMA that have been incorporated where
appropriate. NAIC commented that it supports developing alternative
sources of reinsurance capacity, the securitizing of catastrophic risk within
the United States, and subjecting SPRVS to U.S. insurance regulation. As
stated in our report, a group of insurance industry representatives
interacting with NAIC's working group on securitization is considering how
to structure a legislative proposal to make the onshore SPRV a tax-exempt
entity. Our report also indicates that such legislation also could result in tax
revenue losses and other potential costs. NAIC stated that SPRVs, however,
would be subject to onshore supervision by U.S. regulators, but it is not
clear to us how risk-linked securities would actually be regulated once
brought onshore.42

RAA commented that our report provides an excellent summary on the use of risk-linked securities in providing coverage for catastrophes. However, RAA took exception to (1) our characterization of reinsurance industry capacity and (2) our description of risk-linked securities as an alternative to reinsurance. RAA noted that in recent occurrences of major catastrophic events in the United States, insurers and reinsurers had sufficient capital to meet their obligations and added that most of the California and Florida market was underwritten by insurers that relied very little, if at all, on reinsurance capacity. First, we note that while the reinsurance industry has been able to meet its obligations from recent events with existing capacity, the industry's capacity must be considered along with issues related to (1) the price and availability of catastrophic reinsurance in high-risk areas and (2) its ability to handle multiple, sequential catastrophes. Some insurers who self-reinsure might do so partially because they believe that the price of reinsurance to cover their exposure to catastrophic events is not attractive. Second, RAA asked that we characterize risk-linked securities as a supplement to reinsurance rather than as an alternative because of the relatively small amount of reinsurance coverage currently provided through risk-linked securities. We agree, and our report states that risklinked securities add to or supplement reinsurance capacity, but we also

*In one case, companies experienced an estimated $1 to $2 billion in losses in reinsuring the occupational accident portion of workers' compensation insurance policies. See GAO-01977T.

84-418 D-9

note that sponsors of catastrophe bonds view these securities as alternatives to traditional reinsurance when they are more cost-effective.

BMA stated that our report was accurate and well-researched and commented on several policy issues raised in the report. Their letter raised several concerns with our discussion of tax treatment, accounting treatment, and investor interest in risk-linked securities. First, BMA disagreed with concerns cited in our report that pass-through tax treatment for risk-linked securities could result in (1) tax revenue losses and unfair tax and (2) regulatory and tax advantages that are not afforded to existing U.S.-licensed and taxed reinsurance companies. BMA commented that because a large percentage of entities that provide reinsurance coverage is based outside of the United States, including all reinsurance companies established since September 11, 2001, the tax impact would not be dramatic. In addition, BMA noted that any potential loss of U.S. tax revenue must be weighed against the policy benefits associated with creating additional private-sector capacity to absorb and distribute insurance risk. We agree that many reinsurance entities are not U.S.-based, but the potential tax revenue losses would depend on a number of factors, including business lost by taxable entities and the regulatory requirements used to implement such legislation. We also agree that many considerations must be weighed in the policy decision to grant special tax treatment for onshore SPRVs, including potential tax revenue losses and the extent to which an uneven playing field is created for domestic reinsurance companies.

Second, BMA commented that our description of FASB's SPE consolidation proposal was not based on the final exposure draft and that they interpret the proposal to allow SPRVS to apply only a variable interests approach and not satisfy a particular outside equity threshold. Our draft report discussion of the FASB proposal was based on the final exposure draft. While we did not evaluate BMA's interpretation of the FASB proposal, we included their position in our report. Finally, BMA commented that our discussion of reasons for the lack of broader investor participation in risk-linked securities was incomplete and somewhat inaccurate. They noted that several mutual funds have purchased risk-linked securities as part of their overall portfolios, that mutual fund managers are well-equipped to evaluate the risk associated with these securities, and that lack of broader investor participation may be due to limited issuance. We agree that some mutual funds have purchased risk-linked securities and that lack of broader participation may be attributed to some degree to limited issuance of risklinked securities. However, information we obtained indicates that some of

the largest mutual fund companies did not include risk-linked securities in their mutual fund portfolios mainly because of their unusual and unfamiliar risk characteristics.

Unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies of this report to the Ranking Minority Member of the House Committee on Financial Services and the Chairmen and Ranking Minority Members of the Senate Committee on Banking, Housing and Urban Affairs; and the House Committee on Ways and Means. We also will make copies available to others upon request. In addition, this report will be available for no charge on GAO's Internet home page at http://www.gao.gov.

Please contact Bill Shear, Assistant Director, or me at (202) 512-8678 if you or your staff have any questions concerning this report. Key contributors to this work were Rachel DeMarcus, Lynda Downing, Patrick Dynes, Christine Kuduk, and Barbara Roesmann.

Sincerely yours,

Davi M. D'Agostino

Director, Financial Markets and
Community Investment

Appendix I

Scope and Methodology

You asked us to report on the potential for risk-linked securities to cover
catastrophic risks arising from natural events. As agreed with your office,
our objectives were to (1) describe catastrophe risk and how insurance and
capital markets provide for insurance against such risks; (2) describe how
risk-linked securities, particularly catastrophe bonds, are structured; and
(3) analyze how key regulatory, accounting, tax, and investor issues might
affect the use of risk-linked securities.

Even though we did not have audit or access-to-records authority with the private-sector entities, we obtained extensive documentary and testimonial evidence from a large number of entities, including insurance and reinsurance companies, investment banks, institutional investors, rating agencies, firms that develop models to analyze catastrophic risks, regulators, and academic experts. However, we did not verify the accuracy of data provided by these entities. Some entities we met with voluntarily provided information they considered to be proprietary, and therefore we did not report details from such information. In other cases, companies decided not to provide proprietary information, and this limited our inquiry. For example, we did not obtain any reinsurance contracts representing either traditional reinsurance or reinsurance provided through issuance of risk-linked securities.

To describe catastrophe risk and how insurance and capital markets
provide for insurance against such risks, we examined a variety of
documents, including books on insurance and reinsurance; academic
articles and essays; and analyses done by the Insurance Information
Institute, the Insurance Services Office, modeling firms, and the
Congressional Budget Office. We also interviewed officials from insurance
companies, reinsurance companies, the California Earthquake Authority
(CEA), the Florida Hurricane Catastrophe Fund (FHCF), modeling firms,
and university finance departments and schools.

To describe how risk-linked securities, particularly catastrophe bonds, are structured, we examined catastrophe bond-offering circulars, investment bank documents, reinsurance company analyses, rating agency reports, academic studies, futures exchange documents, and analyses prepared by the American Academy of Actuaries. We also met with officials of investment banks, insurance companies, reinsurance companies, rating agencies, modeling firms, a futures exchange, investment advisors, and the American Academy of Actuaries.

« AnteriorContinuar »