This is the accessible text file for GAO report number GAO-05-536 entitled 'Risk Retention Groups: Common Regulatory Standards and Greater Member Protections Are Needed' which was released on September 14, 2005. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Financial Services, House of Representatives: August 2005: Risk Retention Groups: Common Regulatory Standards and Greater Member Protections Are Needed: GAO-05-536: GAO Highlights: Highlights of GAO-05-536, a report to the Chairman, Committee on Financial Services, House of Representatives: Why GAO Did This Study: Congress authorized the creation of risk retention groups (RRG) to increase the availability and affordability of commercial liability insurance. An RRG is a group of similar businesses that creates its own insurance company to self-insure its risks. Through the Liability Risk Retention Act (LRRA), Congress partly preempted state insurance law to create a single-state regulatory framework for RRGs, although RRGs are multistate insurers. Recent shortages of affordable liability insurance have increased RRG formations, but recent failures of several large RRGs also raised questions about the adequacy of RRG regulation. This report (1) examines the effect of RRGs on insurance availability and affordability; (2) assesses whether LRRA’s preemption has resulted in significant regulatory problems; and (3) evaluates the sufficiency of LRRA’s ownership, control, and governance provisions in protecting the best interests of the RRG insureds. What GAO Found: RRGs have had a small but important effect in increasing the availability and affordability of commercial liability insurance for certain groups. While RRGs have accounted for about $1.8 billion or about 1.17 percent of all commercial liability insurance in 2003, members have benefited from consistent prices, targeted coverage, and programs designed to reduce risk. A recent shortage of affordable liability insurance prompted the creation of many new RRGs. More RRGs formed in 2002–2004 than in the previous 15 years—and about three- quarters of the new RRGs offered medical malpractice coverage. LRRA’s partial preemption of state insurance laws has resulted in a regulatory environment characterized by widely varying state standards. In part, state requirements differ because some states charter RRGs as “captive” insurance companies, which operate under fewer restrictions than traditional insurers. As a result, most RRGs have domiciled in six states that offer captive charters (including some states that have limited experience in regulating RRGs) rather than in the states where they conduct most of their business. Additionally, because most RRGs (as captives) are not subject to the same uniform, baseline standards for solvency regulation as traditional insurers, state requirements in important areas such as financial reporting also vary. For example, some regulators may have difficulty assessing the financial condition of RRGs operating in their state because not all RRGs use the same accounting principles. Further, some evidence exists to support regulator assertions that domiciliary states may be relaxing chartering or other requirements to attract RRGs. Because LRRA does not specify characteristics of ownership and control, or establish governance safeguards, RRGs can be operated in ways that do not consistently protect the best interests of their insureds. For example, LRRA does not explicitly require that the insureds contribute capital to the RRG or recognize that outside firms typically manage RRGs. Thus, some regulators believe that members without “skin in the game” will have less interest in the success and operation of their RRG and that RRGs would be chartered for purposes other than self- insurance, such as making profits for entrepreneurs who form and finance an RRG. LRRA also provides no governance protections to counteract potential conflicts of interest between insureds and management companies. In fact, factors contributing to many RRG failures suggest that sometimes management companies have promoted their own interests at the expense of the insureds. The combination of single-state regulation, growth in new domiciles, and wide variance in regulatory practices has increased the potential that RRGs would face greater solvency risks. As a result, GAO believes RRGs would benefit from uniform, baseline regulatory standards. Also, because many RRGs are run by management companies, they could benefit from corporate governance standards that would establish the insureds’ authority over management. What GAO Recommends: To strengthen the overall regulation of RRGs, GAO recommends that state insurance regulators adopt consistent regulatory standards for RRGs. Moreover, GAO suggests that Congress consider (1) granting the partial preemption only to states that adopt the standards and (2) establishing minimum corporate governance standards for RRGs. www.gao.gov/cgi-bin/getrpt?GA0-05-536. To view the full product, including the scope and methodology, click on the link above. For more information, contact Richard J. Hillman at (202) 512-8678 or hillmanr@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: RRGs Have Had a Small but Important Effect on Increasing the Availability and Affordability of Commercial Liability Insurance: LRRA's Regulatory Preemption Has Resulted in Widely Varying Requirements among States and Limited Confidence in RRG Regulation: RRG Failures Have Raised Questions about the Sufficiency of LRRA Provisions for RRG Ownership, Control, and Governance: Conclusions: Recommendations for Executive Action: Matters for Congressional Consideration: Agency Comments and Our Evaluation: Appendixes: Appendix I: Objectives, Scope, and Methodology: Appendix II: Survey of State Regulators on Risk Retention Groups: Appendix III: Selected Differences between Statutory and Generally Accepted Accounting Principles as They Relate to Financial Reporting for RRGs: Appendix IV: Liquidated Risk Retention Groups (RRG), from 1990 through 2003: Appendix V: Comments from the National Association of Insurance Commissioners: Appendix VI: GAO Contact and Staff Acknowledgments: Tables: Table 1: Characteristics of States We Interviewed, Based on Years of Regulatory Experience and Number of RRGs Domiciled: Table 2: Differences in Regulatory Actions When Calculating Risk-Based Capital for Three RRGs, Modified GAAP Compared with SAP: Figures: Figure 1: RRG Gross Premiums Written in 2003, by Time (Years) in Business: Figure 2: Number of RRGs, by Business Area for Selected Years: Figure 3: Percentage of Estimated Gross Premiums RRGs Collected in 2004, by Business Area: Figure 4: Number of RRGs, by Formation Date: Figure 5: Number of RRGs, by Captive or Noncaptive Charter and State of Domicile, as of the End of 2004: Figure 6: Number of RRGs Chartered, by State, as of the End of 2004, and Amount of Direct Premiums Written by RRGs, by State, 2003: Figure 7: State Regulators' Opinion of the Adequacy of the Regulatory Protections or Safeguards Built into LRRA: Figure 8: Permitted Wording of Guaranty Fund Disclosure in LRRA: Figure 9: The Effect of Differences in Accounting for Acquisition Costs on Assets, Capital, and Surplus, GAAP Compared with SAP: Figure 10: Impact of Counting an LOC and Prepaid Expenses as Assets on the Balance Sheet, Modified GAAP Compared with SAP: Figure 11: Impact of Counting Acquisition Costs, LOCs, and Prepaid Expenses as Assets on the Balance Sheet, Modified GAAP Compared with SAP: Figure 12: Differences in the Calculation of Net Premiums Written to Policyholders' Surplus Ratio, Modified GAAP Compared with SAP: Figure 13: Differences in the Calculation of Reserves to Policyholders' Surplus Ratio, Modified GAAP Compared with SAP: Abbreviations: ANLIR: American National Lawyers Insurance: Reciprocal Risk Retention Group: BRICO: Beverage Retailers Insurance Company: Risk Retention Group: CEO: chief executive officer: CIC: Corporate Insurance Consultants: DIR: Doctors Insurance Risk Retention Group: FAST: Financial Analysis Solvency Tools: GAAP: generally accepted accounting principles: IRIS: Insurance Regulatory Information System: LOC: letter of credit: LRRA: Liability Risk Retention Act: NAIC: National Association of Insurance Commissioners: NPW:PS: net premiums written to policyholders' surplus: PMIC: Professional Mutual Insurance Company Risk Retention Group: RBC: risk-based capital: ROA: Reciprocal of America: RPG: risk purchasing group: RRG: risk retention group: RRR: Risk Retention Reporter: SAP: statutory accounting principles: SEC: Securities and Exchange Commission: TAC: total adjusted capital: TRA: The Reciprocal Alliance Risk Retention Group: TRG: The Reciprocal Group: VSC: vehicle service contract: Letter August 15, 2005: The Honorable Michael G. Oxley: Chairman, Committee on Financial Services: House of Representatives: Dear Mr. Chairman: In 1981, in response to recurring shortages of liability insurance, Congress passed the Product Risk Retention Liability Act, now known as the Liability Risk Retention Act (LRRA), which authorized the creation of risk retention groups (RRG) to increase the availability and affordability of commercial liability insurance.[Footnote 1] An RRG is a group of similar businesses with similar risk exposures, such as educational institutions or building contractors, which create their own insurance company to self-insure their risks on a group basis.[Footnote 2] Through LRRA, Congress first established a flexible framework that allowed states to develop their own standards for the formation and operation of RRGs. In light of the recent unavailability of affordable liability insurance, especially for medical malpractice coverage, interest in forming RRGs has greatly increased. In addition, some industry advocates now propose that RRGs be permitted to offer property coverage as well. However, the recent and notable failures of several large RRGs have raised questions about the adequacy of the RRG regulatory environment and whether existing safeguards, such as the requirement that each RRG provide copies of operational plans and annual financial statements to each state in which it operates, are sufficient to ensure that RRGs are operated and governed adequately to protect their insureds. LRRA, expanded in 1986, facilitates the creation and operation of RRGs in several ways.[Footnote 3] Most notably, LRRA partially preempts state insurance laws by allowing an RRG's formation and operations to be regulated primarily by the state in which it is chartered, its domiciliary state, even when it sells insurance in other states. LRRA largely limits the oversight role of insurance regulators from nondomiciliary states (all states other than the chartering state) to the right to receive copies of an RRG's operational plans and annual financial statements. In having only one regulator, RRGs differ from "traditional" insurance companies, which are subject to licensing and oversight by regulators in each state in which they operate. Additionally, LRRA prohibits RRGs from participating in state guaranty funds, which are available to settle the claims of insureds of a traditional insurance company should that company fail. LRRA's legislative history indicates a view that single-state regulation would provide adequate supervision of RRGs, largely because RRGs would be providing insurance coverage only to their own members, and not the public at large. While preemption is central to LRRA's objective of facilitating the formation and efficient interstate operation of RRGs, Congress also expressed a view that LRRA's prohibition on participating in guaranty funds would provide a strong incentive for RRGs to set adequate premiums and establish adequate reserves, as each RRG member would know that there would be no other source of funds (other than the RRG's own assets) from which to pay claims.[Footnote 4] RRGs are not the only mechanism by which businesses may establish self- insurance coverage. States also charter and regulate captive insurance companies, which are established by single companies or groups of companies to self-insure their own risks. Traditional insurance companies sell insurance to the general public and are licensed in all states in which they do business. In contrast, captive insurance companies largely insure only their owners, who have the ability to manage and retain their own risk. Thus, the degree of regulatory oversight required for captives is different than that which is required for commercial insurers. States chartering captives offer some regulatory relief to these companies, based on the presumption that the owners of captive companies have sophisticated knowledge about managing their own risks and are motivated to protect their own interests. The captive is licensed in only one state and operates under the captive insurance law of that domicile. However, should the captive choose to conduct business outside its state of domicile, it would be subject to the licensing and oversight of each state because captives that are not also RRGs do not benefit from the partial preemption. Many states have recognized RRGs as a form of captive and charter them under their captive regulations. In light of proposals to expand LRRA, and recent shortages of affordable liability insurance, you requested that we assess how well RRGs have achieved LRRA's legislative goals of making commercial liability insurance available and affordable. This report (1) examines the effect RRGs have had on the availability and affordability of commercial liability insurance; (2) assesses whether LRRA's partial preemption of state insurance laws has resulted in any significant regulatory problems; and (3) evaluates the sufficiency of LRRA's ownership, control, and governance provisions for protecting the best interests of the insureds. To ascertain the effect of RRGs on the availability and affordability of commercial liability insurance, we surveyed regulators in all 50 states and the District of Columbia and interviewed representatives from eight RRGs serving different markets. In addition, we obtained information from the National Association of Insurance Commissioners (NAIC) that estimated the share of the commercial liability insurance market that RRGs held in 2003.[Footnote 5] To determine if LRRA's partial preemption of state insurance laws has resulted in significant regulatory problems, we surveyed all state insurance departments to obtain information on their regulatory experiences and obtained more specific information from regulators in 14 states, including some that do not domicile RRGs. To understand the regulatory framework, especially the capitalization and financial reporting standards under which most RRGs are regulated, we compared the regulatory standards of the six states (Arizona, the District of Columbia, Hawaii, Nevada, South Carolina, and Vermont) that had chartered the most RRGs as of June 30, 2004. To assess the sufficiency of LRRA's ownership, control, and governance provisions in protecting the best interests of the insureds, we identified provisions in LRRA that relate to these issues and reviewed LRRA's legislative history to ascertain Congress' concerns about these issues.[Footnote 6] Since LRRA largely delegates the regulation of the formation and operation of RRGs to the domiciliary states, we reviewed the statutory provisions of the six leading domiciliary states to determine whether they addressed ownership, control, and governance, and interviewed their regulators to identify insurance departmental policies. To understand how the regulators implemented these statutes and policies, we reviewed the chartering documents of the three RRGs most recently domiciled by each of the leading domiciliary states. Finally, we identified whether factors related to the ownership, control, or governance of RRGs contributed or, in some cases, were alleged to have contributed to RRG failures. We conducted our review from November 2003 through July 2005 in accordance with generally accepted government auditing standards. Appendix I contains a more detailed description of our objectives, scope, and methodology. Results in Brief: RRGs have had a small but important effect in increasing the availability and affordability of commercial liability insurance for certain groups with limited access to insurance. In 2003, according to NAIC estimates, RRGs provided about $1.8 billion or 1.17 percent of all commercial liability insurance. While the overall impact on the liability market has been small, most state regulators we surveyed believed that RRGs have increased the availability and affordability of insurance for groups that have had difficulties obtaining affordable coverage such as healthcare providers, building contractors, and commercial trucking firms. According to state regulators and RRG industry representatives, members have benefited in several important ways by using RRGs to self-insure their risks.[Footnote 7] These benefits include controlling their costs by targeting their coverage to the specific needs of members and designing programs to reduce risks. The representatives indicated that RRGs might not always benefit from the lowest insurance prices but could benefit from prices that remained stable over time. In recent years, a shortage of affordable liability insurance also prompted the creation of many new RRGs. From 2002 through 2004, 117 RRGs were formed, more than the total formed over the previous 15 years. In particular, a shortage of affordable medical malpractice insurance prompted healthcare providers to form about three- quarters of the new RRGs. As a result, more than half of all currently operating RRGs provide insurance in healthcare-related areas. LRRA's partial preemption of state insurance laws has resulted in a regulatory environment characterized by widely varying state standards and limited regulator confidence in the system. In part, state requirements differ because some states charter RRGs as captive insurance companies, which operate under less restrictive regulation than traditional insurers. A captive charter offers RRGs several advantages: For example, initial capitalization standards are usually easier to meet because states allow captives to start their operations with less capital than traditional insurers and use letters of credit rather than cash to meet capitalization requirements. As a result of these and other advantages, the majority of RRGs have domiciled in six states--Arizona, the District of Columbia, Hawaii, Nevada, South Carolina, and Vermont--that allow them to be chartered as captive insurers rather than in the states where they conduct most of their business. In addition, states chartering RRGs as captives also vary in how they regulate RRGs on an ongoing basis. These variations exist because LRRA grants domiciliary states the discretion and authority to regulate the formation and multistate operation of RRGs and because as captives most RRGs are not subject to uniform, baseline standards, such as those set forth in NAIC's financial accreditation standards for regulation of traditional multistate insurers. For example, five of the six leading domiciliary states allow RRGs to use a modified version of generally accepted accounting principles (GAAP), rather than statutory accounting principles (SAP), when filing financial statements. As a result, some nondomiciliary state regulators have difficulty interpreting these reports, especially since traditional insurers must file using SAP. Additionally, only 8 regulators of 42 responding to a particular survey question considered that LRRA's provisions adequately protected RRG insureds, often because of their concerns about a lack of uniform, baseline standards and perception that they needed additional regulatory authority. Finally, some evidence exists to support regulator assertions that some domiciliary states may be creating lenient regulatory environments to encourage RRGs to domicile in their state. For example, in the past 4 years, two leading domiciliary states have allowed RRGs to relocate their charters, even though the RRGs were subject to unresolved regulatory actions in their original state of domicile. Because (other than requiring that owners must also be insureds) LRRA does not impose minimum characteristics of ownership and control for RRGs or establish minimum governance requirements, RRGs can be operated in ways that do not consistently protect the best interests of their insureds. While RRGs were authorized for the purpose of providing self- insurance, LRRA does not explicitly require that all of the insureds contribute toward the capitalization of the RRG. Consequently, some of the six leading domiciliary states do not expect RRG insureds to contribute anything more than insurance premiums, and some regulators are concerned that members without "skin in the game" will have less interest in the success of their RRG. In addition, even though LRRA's legislative history indicates that the single-state regulatory framework was premised, in part, on insureds having adequate incentives to exercise control over their RRG, some of the six leading domiciliary states do not expect insureds to have the ability to elect their governing body (such as a board of directors). Because not all insureds actually participate in the formation or financing of their RRGs, some regulators are concerned that those RRGs may be operated for the financial benefit of the "entrepreneurs" who do provide the financing. An entrepreneur could be an individual insured or the company hired to manage the daily operations of the RRG. LRRA also does not include provisions regarding the management of RRGs, such as governance protections to counteract potential conflicts of interest between management companies and insureds. In contrast, Congress previously addressed a similar situation within the mutual fund industry by passing legislation intended to minimize potential conflicts of interest between mutual fund shareholders and management companies. The circumstances surrounding more than half of past RRG failures we examined suggest that management companies or managers have promoted their own interests at the expense of the insureds--for example, by charging excessive management fees or promoting transactions unfavorable to the RRG. Regulators knowledgeable about these failures said that the insureds likely were more interested in obtaining affordable insurance than assuming the responsibilities of owning an insurance company. Consequently, even though an insured's insurance policy may have stated that the RRG lacked guaranty fund coverage, the insureds may not have been fully aware of this restriction or the consequences of lacking such protection. Further, LRRA does not require RRGs to disclose to prospective claimants, those who submit claims for loss, that the RRGs would not benefit from guaranty fund protection should they fail. This can be of special consequence to certain claimants--consumers who purchase extended service contracts from the insureds of RRGs--because contracts issued by these insureds take on the appearance of insurance when, in most cases, they are not. This report contains recommendations for the states, as well as matters for congressional consideration that, if implemented, would create a more consistent regulatory framework for overseeing the chartering and management of RRGs, provide more reliable information about the financial condition of RRGs, and provide RRG members needed protections to help ensure that companies managing RRGs operate in the insureds' best interests. In addition, enhancing the availability and contents of the guaranty fund disclosure would provide RRG insureds, as well as consumers who purchase extended service contracts from RRG insureds, a better understanding of the lack of guaranty fund coverage. Finally, these recommendations would strengthen NAIC's ability to achieve its goals of improving the quality and consistency of state insurance regulation. We requested comments on a draft of this report from NAIC. The Executive Vice President and CEO, National Association of Insurance Commissioners, provided written comments on a draft of this report. NAIC generally agreed with our approach and methodology and our description of the regulation of risk retention groups. NAIC's comments are discussed later in this report and are reprinted in appendix V. NAIC and several of the states also provided technical comments, which we incorporated as appropriate. Background: In the legislative history, RRGs were described as essentially insurance "cooperatives," whose members pool funds to spread and assume all or a portion of their own commercial liability risk exposure--and who are engaged in businesses and activities with similar or related risks.[Footnote 8] Specifically, RRGs may be owned only by individuals or businesses that are insured by the RRG or by an organization that is owned solely by insureds of the RRG.[Footnote 9] In the legislative history, Congress expressed the viewthat RRGs had the potential to increase the availability of commercial liability insurance for businesses and reduce liability premiums, at least when insurance is difficult to obtain (during hard markets) because members would set rates more closely tied to their own claims experience. In addition, LRRA was intended to provide businesses, especially small ones, an opportunity to reduce insurance costs and promote greater competition among insurers when they set insurance rates.[Footnote 10] Because RRGs are owned by insureds that may have business assets at risk should the RRG be unable to pay claims, they would have greater incentives to practice effective risk management both in their own businesses and the RRG. The elimination of duplicative and sometimes contradictory regulation by multiple states was designed to facilitate the formation and interstate operation of RRGs.[Footnote 11] "The (regulatory) framework established by LRRA attempts to strike a balance between the RRGs' need to be free of unjustified requirements and the public's need for protection from insolvencies."[Footnote 12] RRGs are not the only form of self-insurance companies; "captive insurance companies" (captives) also self-insure the risks of their owners. States can charter RRGs under regulations intended for traditional insurers or for captive insurers. Non-RRG captives largely exist solely to cover the risks of their parent, which can be one large company (pure captive) or a group of companies (group captives).[Footnote 13] Group captives share certain similarities with RRGs because they also are composed of several companies, but group captives, unlike RRGs, do not have to insure similar risks. Further, captives may provide property coverage, which RRGs may not. Regulatory requirements for captives generally are less restrictive than those for traditional insurance companies because, for example, many pure captives are wholly owned insurance subsidiaries of a single business or organization. If a pure captive failed, only the assets of the parent would be at risk. Finally, unlike captive RRGs, other captive insurers generally cannot conduct insurance transactions in any state except their domiciliary state, unless they become licensed in that other state (just as a traditional company would) and subject to that state's regulatory oversight.[Footnote 14] In contrast to the single-state regulation that LRRA provides for RRGs, traditional insurers, as well as other non-RRG captive insurers, are subject to the licensing requirements and oversight of each nondomiciliary state in which they operate. The licensing process allows states to determine if an insurer domiciled in another state meets the nondomiciliary state's regulatory requirements before granting the insurer permission to operate in its state. According to NAIC's uniform application process, which has been adopted by all states, an insurance company must show that it meets the nondomiciliary state's minimum statutory capital and surplus requirements, identify whether it is affiliated with other companies (that is, part of a holding company system), and submit biographical affidavits for all its officers, directors, and key managerial personnel.[Footnote 15] After licensing an insurer, regulators in nondomiciliary states can conduct financial examinations, issue an administrative cease and desist order to stop an insurance company from operating in their state, and withdraw the company's license to sell insurance in the state. However, most state regulators will not even license an insurance company domiciled in another state to operate in their state unless the company has been in operation for several years. As reflected in each state's "seasoning requirements," an insurance company must have successfully operated in its state of domicile for anywhere from 1 to 5 years before qualifying to receive a license from another state.[Footnote 16] RRGs, in contrast, are required only to register with the regulator of the state in which they intend to sell insurance and provide copies of certain documents originally provided to domiciliary regulators. Although RRGs receive regulatory relief under LRRA, they still are expected to comply with certain other laws administered by the states in which they operate, but are not chartered (nondomiciliary states), and are required to pay applicable premium and other taxes imposed by nondomiciliary states.[Footnote 17] In addition to registering with other states, LRRA also imposes other requirements that offer protections or safeguards to RRG members: LRRA requires each RRG to (1) provide a plan of operation to the insurance commissioner of each state in which it plans to do business prior to offering insurance in that state, (2) provide a copy of the group's annual financial statement to the insurance commissioner of each state in which it is doing business, and (3) submit to an examination by a nondomiciliary state regulator to determine the RRG's financial condition, if the domiciliary state regulator has not begun or refuses to begin an examination. Nondomiciliary, as well as domiciliary states, also may seek an injunction in a "court of competent jurisdiction" against RRGs that they believe are in hazardous financial condition.[Footnote 18] In conjunction with the regulatory relief Congress granted to RRGs, it prohibited RRGs from participating in state guaranty funds, believing that this restriction would provide RRG members a strong incentive to establish adequate premiums and reserves. All states have established guaranty funds, funded by insurance companies, to pay the claims of policyholders in the event that an insurance company fails. Without guaranty fund protection, in the event an RRG becomes insolvent, RRG insureds and their claimants could be exposed to all losses resulting from claims that exceed the ability of the RRG to pay. Finally, in terms of structure, RRG and captive insurance companies bear a certain resemblance to mutual fund companies.[Footnote 19] For example, RRGs, captive insurance companies, and mutual fund companies employ the services of a management company to administer their operations. RRGs and captive insurers generally hire "captive management" companies to administer company operations, such as day-to- day operational decisions, financial reporting, liaison with state insurance departments, or locating sources of reinsurance.[Footnote 20] Similarly, a typical mutual fund has no employees but is created and operated by another party, the adviser, which contracts with the fund, for a fee, to administer operations. For example, the adviser would be responsible for selecting and managing the mutual fund's portfolio. However, Congress recognized that the external management of mutual funds by investment advisers creates an inherent conflict between the adviser's duties to the fund shareholders and the adviser's interests in maximizing its own profits, a situation that could adversely affect fund shareholders. One way in which Congress addressed this conflict is the regulatory scheme established by the Investment Company Act of 1940, which includes certain safeguards to protect the interests of fund shareholders. For example, a fund's board of directors must contain a certain percentage of independent directors--directors without any significant relationship to the advisers. RRGs Have Had a Small but Important Effect on Increasing the Availability and Affordability of Commercial Liability Insurance: RRGs have had a small but important effect on increasing the availability and affordability of commercial liability insurance, specifically for groups that have had limited access to liability insurance. According to NAIC estimates, in 2003 RRGs sold just over 1 percent of all commercial liability insurance in the United States. However, many state regulators, even those who had reservations about the regulatory oversight of RRGs, believe RRGs have filled a void in the market. Regulators from the six leading domiciliary states also observed that RRGs were important to certain groups that could not find affordable coverage from a traditional insurance company and offered RRG insureds other benefits such as tailored coverage. Furthermore, RRGs, while tending to be relatively small in size compared with traditional insurers, serve a wide variety of organizations and businesses, although the majority served the healthcare industry. Difficulties in finding affordable commercial liability insurance prompted the creation of more RRGs from 2002 through 2004 than in the previous 15 years. Three-quarters of the RRGs formed in this period responded to a recent shortage of, and high prices for, medical malpractice insurance. However, studies have characterized the medical malpractice industry as volatile because of the risks associated with providing this line of insurance. RRGs Have Represented a Small but Increasing Part of the Commercial Liability Insurance Market: RRGs have constituted a very small part of the commercial liability market. According to NAIC estimates, in 2003 a total of 115 RRGs sold 1.17 percent of all commercial liability insurance in the United States. This accounted for about $1.8 billion of a total of $150 billion in gross premiums for all commercial liability lines of insurance.[Footnote 21] We are focusing on 2003 market share to match the time frame of our other financial analyses of gross premiums. While RRGs' share of the commercial liability market was quite small, market share and the overall amount of business RRGs wrote increased since 2002. For example, RRG market share increased from 0.89 percent in 2002 to 1.46 percent in 2004.[Footnote 22] However, in terms of commercial liability gross premiums, the increase in the amount of business written by RRGs is more noticeable. The amount of business that RRGs collectively wrote about doubled, from $1.2 billion in 2002 to $2.3 billion in 2004. During this same period, the amount of commercial liability written by traditional insurers increased by about 21 percent, from $129 billion to $156 billion. In addition, RRGs increased their presence in the market for medical malpractice insurance. From 2002 through 2004, the amount of medical malpractice written by RRGs increased from $497 million to $1.1 billion, which increased their share of the medical malpractice market from 4.04 percent to 7.27 percent. According to State Regulators, RRGs Have Filled Voids in Markets, Allowing Numerous Groups to Obtain Benefits of Coverage: Despite the relatively small share of the market that RRGs hold, most state regulators we surveyed who had an opinion--33 of 36--indicated that RRGs have expanded the availability and affordability of commercial liability insurance for groups that otherwise would have had difficulty in obtaining coverage.[Footnote 23] This consistency of opinion is notable because 18 of those 33 regulators made this assertion even though they later expressed reservations about the adequacy of LRRA's regulatory safeguards.[Footnote 24] About one-third of the 33 regulators also made more specific comments about the contributions of RRGs. Of these, five regulators reported that RRGs had expanded the availability of medical malpractice insurance for nursing homes, adult foster care homes, hospitals, and physicians. One regulator also reported that RRGs had assisted commercial truckers in meeting their insurance needs. Regulators from states that had domiciled the most RRGs as of the end of 2004--Arizona, the District of Columbia, Hawaii, Nevada, South Carolina, and Vermont--provided additional insights.[Footnote 25] Regulators from most of these states recognized that the overall impact of RRGs in expanding the availability of insurance was quite small. However, they said that the coverage RRGs provided was important because certain groups could not find affordable insurance from a traditional insurance company. All of these regulators cited medical malpractice insurance as an area where RRGs increased the affordability and availability of insurance but they also identified other areas. For example, regulators from Hawaii and Nevada reported that RRGs have been important in addressing a shortage of insurance for construction contractors. The six regulators all indicated (to some extent) that by forming their own insurance companies, RRG members also could control costs by designing insurance coverage targeted to their specific needs and develop programs to reduce specific risks. In contrast, as noted by the Arizona regulator, traditional insurers were likely to take a short- term view of the market, underpricing their coverage when they had competition and later overpricing their coverage to recoup losses. He also noted insurers might exit a market altogether if they perceived the business to be unprofitable, as exemplified in the medical malpractice market. Regulators from Vermont and Hawaii, states that have the most experience in chartering RRGs, added that successful RRGs have members that are interested in staying in business for the "long haul" and are actively involved in running their RRGs. RRG representatives added that RRG members, at any given time, might not necessarily benefit from the cheapest insurance prices but could benefit from prices that were stable over time. Additionally, as indicated by trade group representatives, including the National Risk Retention Association, RRGs have proved especially advantageous for small and midsized businesses. In order to obtain more specific information about how RRGs have benefited their membership, we interviewed representatives of and reviewed documents supplied by six RRGs that have been in business for more than 5 years, as well as two more recently established RRGs. Overall, these eight RRGs had anywhere from 2 to more than 14,500 members.[Footnote 26] They provided coverage to a variety of insureds, including educational institutions, hospitals, attorneys, and building contractors. The following three examples illustrate some of the services and activities RRGs provide or undertake. * An RRG that insures about 1,100 schools, universities, and related organizations throughout the United States offers options tailored to its members, such as educators' legal liability coverage and coverage for students enrolled in courses offering off-campus internships. According to an RRG representative, the RRG maintains a claims database to help it accurately and competitively price its policies. Members also benefit from risk-management services, such as training and courses on sexual harassment and tenure litigation, and work with specialists to develop loss-control programs. * An RRG that reported that it insures 730 of the nation's approximately 3,000 public housing authorities provides coverage for risks such as pesticide exposure, law enforcement liability, and lead- based paint liability. The RRG indicated that while premium rates have fluctuated, they are similar to prices from about 15 years ago. The RRG also offers risk-management programs, such as those for reducing fires, and also reported that as a result of conducting member inspections it recently compiled more than 2,000 recommendations on how to reduce covered risks. * An RRG that primarily provides insurance to about 45 hospitals in California and Nevada offers general and professional coverage such as personal and bodily injury and employee benefit liability. The RRG also offers a variety of risk-management services specifically aimed at reducing losses and controlling risks in hospitals. According to an RRG official, adequately managing risk within the RRG has allowed for more accurate pricing of the liability coverage available to members. RRGs Have Remained Relatively Small in Size Compared with Traditional Insurers but Serve a Wide Variety of Markets: Generally, RRGs have remained relatively small compared with traditional insurers. Based on our analysis of 2003 financial data submitted to NAIC, 47 of the 79 RRGs (almost 60 percent) that had been in business at least 1 year, wrote less than $10 million in gross premiums, whereas only 644 of 2,392 traditional insurers (27 percent) wrote less than $10 million. In contrast, 1,118 traditional insurers (almost 47 percent) wrote more than $50 million in gross premiums for 2003 compared with six RRGs (8 percent). Further, these six RRGs (all of which had been in business for at least 1 year) accounted for 52 percent of all gross premiums that RRGs wrote in 2003. This information suggests that just a few RRGs account for a disproportionate amount of the RRG market. Additionally, RRGs that wrote the most business tended to have been in business the longest. For example, as measured by gross premiums written, of the 16 RRGs that sold more than $25 million annually, 14 had been in business 5 years or more (see fig. 1). Yet, the length of time an RRG has been in operation is not always the best predictor of an RRG's size. For example, of the 51 RRGs that had been in business for 5 or more years, 27 still wrote $10 million or less in gross premiums. Figure 1: RRG Gross Premiums Written in 2003, by Time (Years) in Business: [See PDF for image] [A] This figure compares the amount of gross premiums written in 2003 by the 79 RRGs that had been in business for at least 1 year. Of the 79, 51 had at least 5 years of business experience, and 28 had between 1 and 5 years. [End of figure] According to the Risk Retention Reporter (RRR), a trade journal that has covered RRGs since 1986, RRGs insure a wide variety of organizations and businesses.[Footnote 27] According to estimates published in RRR, in 2004 105 RRGs (more than half of the 182 in operation at that time) served the healthcare sector (for example, hospitals, nursing homes, and doctors). In 1991, RRGs serving physicians and hospitals accounted for about 90 percent of healthcare RRGs. However, by 2004, largely because of a recent increase in nursing homes forming RRGs, this percentage decreased to about 74 percent.[Footnote 28] In addition, in 2004, 21 RRGs served the property development area (for example, contractors and homebuilders), and 20 served the manufacturing and commerce area (for example, manufacturers and distributors). Other leading business areas that RRGs served include professional services (for example, attorneys and architects), and government and institutions (for example, educational and religious institutions). Figure 2 shows how the distribution of RRGs by business area has changed since 1991. Figure 2: Number of RRGs, by Business Area for Selected Years: [See PDF for image] [A] The RRG numbers that RRR projected for 2004 are based on the number of RRGs the journal identified operating as of the end of September 2004. For each year, we show only the four business areas with the highest number of RRGs, and group all other areas in a fifth category. [End of figure] Additionally, according to RRR's estimates, almost half of all RRG premiums collected in 2004 were in the healthcare area (see fig. 3). The professional services and government and institutions business areas accounted for the second and third largest percentage of estimated gross premiums collected, respectively.[Footnote 29] Figure 3: Percentage of Estimated Gross Premiums RRGs Collected in 2004, by Business Area: [See PDF for image] Note: Gross premium data estimates are based on information RRR collected from RRGs during a 2004 survey. RRR reported projections for all of 2004 in October 2004. [End of figure] In looking at other characteristics of RRGs, according to an NAIC analysis, the average annual failure rate for RRGs was somewhat higher than the average annual failure rate for all other property and casualty insurers. Between 1987 and 2003, the average annual failure rate for RRGs was 1.83 percent compared with the 0.78 percent failure rate for property and casualty insurers.[Footnote 30] Over this period, NAIC determined that a total of 22 RRGs failed, with between no and five RRGs failing each year.[Footnote 31] In comparison, NAIC determined that a total of 385 traditional insurers failed, with between 5 and 57 insurance companies failing each year. Although the difference in failure rates was statistically significant, it should be noted that the comparison may not be entirely parallel. NAIC compared RRGs that can sell only commercial liability insurance to businesses with insurers that can sell all lines of property and casualty (liability) for commercial and personal purposes.[Footnote 32] Moreover, because NAIC included all property-casualty insurers, no analysis was done to adjust for size and longevity. Recent Market Conditions Have Prompted the Creation of Many RRGs, Especially to Provide Medical Malpractice Insurance: In creating RRGs, companies and organizations are generally responding to market conditions. As the availability and affordability of insurance decreased (creating a "hard" market), some insurance buyers sought alternatives to traditional insurance and turned to RRGs.[Footnote 33] In response, more RRGs formed from 2002 through 2004 than in the previous 15 years (1986-2001). This increase is somewhat similar in magnitude to an increase that occurred in 1986-1989 in response to an earlier hard market for insurance (see fig. 4).[Footnote 34] The 117 RRGs formed from January 1, 2002, through December 31, 2004, represent more than half of all RRGs in operation as of December 31, 2004. Figure 4: Number of RRGs, by Formation Date: [See PDF for image] Note: This figure represents the number of RRGs formed during different periods, regardless of whether they are currently active or not, and includes only those RRGs for which NAIC has data. According to NAIC data, four companies either formed as RRGs or converted to RRGs after the passage of the Product Liability Risk Retention Act of 1981. [End of figure] More specifically, RRGs established to provide medical malpractice insurance accounted for most of the increase in RRG numbers in 2002- 2004.[Footnote 35] Healthcare providers sought insurance after some of the largest medical malpractice insurance providers exited the market because of declining profits, partly caused by market instability and high and unpredictable losses--factors that have contributed to the high risks of providing medical malpractice insurance.[Footnote 36] From 2002 through 2004, healthcare RRGs accounted for nearly three- fourths of all RRG formations. Further, 105 RRGs were insuring healthcare providers as of the end of 2004, compared with 23 in previous years (see again fig. 2). These RRGs serve a variety of healthcare providers. For example, during 2003, 23 RRGs formed to insure hospitals and their affiliates, 13 formed to insure physician groups, and 11 formed to insure long-term care facilities, including nursing homes and assisted living facilities. However, the dramatic increase in the overall number of RRGs providing medical malpractice insurance may precipitate an increase in the number of RRGs vulnerable to failure. Studies have characterized the medical malpractice insurance industry as volatile because the risks of providing medical malpractice insurance are high.[Footnote 37] Finally, many of the recently formed healthcare-related RRGs are selling insurance in states where medical malpractice insurance rates for physicians have increased the most.[Footnote 38] For example, since April 30, 2002, the Pennsylvania Insurance Department has registered 32 RRGs to write medical malpractice products. In addition, since the beginning of 2003, the Texas Department of Insurance has registered 15 RRGs to write medical malpractice insurance, more than the state had registered in the previous 16 years. Other states where recently formed RRGs were insuring doctors include Illinois and Florida, states that have also experienced large increases in medical malpractice insurance premium rates. LRRA's Regulatory Preemption Has Resulted in Widely Varying Requirements among States and Limited Confidence in RRG Regulation: LRRA's regulatory preemption has allowed states to set regulatory requirements that differ significantly from those of traditional insurers, and from each other, producing limited confidence among regulators in the regulation of RRGs. Many of the differences arise because some states allow RRGs to be chartered as captive insurance companies, which typically operate under a set of less restrictive rules than traditional insurers. As a result, RRGs generally domicile in those states that permit their formation as captive insurance companies, rather than in the states in which they conduct most of their business. For example, RRGs domiciled as captive insurers usually can start their operations with smaller amounts of capital and surplus than traditional insurance companies, use letters of credit to meet minimum capitalization requirements, or meet fewer reporting requirements. Regulatory requirements for captive RRGs vary among states as well, in part because regulation of RRGs and captives are not subject to uniform, baseline standards, such as the NAIC accreditation standards that define a state's regulatory structure for traditional companies. As one notable example, states do not require RRGs to follow the same accounting principles when preparing their financial reports, making it difficult for some nondomiciliary state regulators, as well as NAIC analysts, to reliably assess the financial condition of RRGs. Regulators responding to our survey also expressed concern about the lack of uniform, baseline standards. Few (eight) indicated that they believed LRRA's regulatory safeguards and protections, such as the right to file a suit against an RRG in court, were adequate. Further, some regulators suggested that some domiciliary states were modifying their regulatory requirements and practices to make it easier for RRGs to domicile in their state. We found some evidence to support these concerns based on differences among states in minimum capitalization requirements, willingness to charter RRGs to insure parties that sell extended service contracts to consumers, or willingness to charter RRGs primarily started by service providers, such as management companies, rather than insureds. Most RRGs Have Domiciled in States That Charter Them as Captives but Have Conducted Most of Their Business in Other States: Regulatory requirements for captive insurers are generally less restrictive than those for traditional insurers and offer RRGs several financial advantages. For example, captive laws generally permit RRGs to form with smaller amounts of required capitalization (capital and surplus), the minimum amount of initial funds an insurer legally must have to be chartered.[Footnote 39] While regulators reported that their states generally require traditional insurance companies to have several millions of dollars in capital and surplus, they often reported that RRGs chartered as captives require no more than $500,000.[Footnote 40] In addition, unlike requirements for traditional insurance companies, the captive laws of the six leading domiciliary states permit RRGs to meet and maintain their minimum capital and surplus requirements in the form of an irrevocable letter of credit (LOC) rather than cash.[Footnote 41] According to several regulators that charter RRGs as captives, LOCs may provide greater protection to the insureds than cash when only the insurance commissioner can access these funds. The insurance commissioner, who would be identified as the beneficiary of the LOC, could present the LOC to the bank and immediately access the cash, but a representative of the RRG could not. However, other state regulators questioned the value of LOCs because they believed cash would be more secure if an RRG were to experience major financial difficulties. One regulator noted that it becomes the regulator's responsibility, on a regular basis, to determine if the RRG is complying with the terms of the LOC. In addition, in response to our survey, most regulators from states that would charter RRGs as captives reported that RRGs would not be required to comply with NAIC's risk- based capital (RBC) requirements.[Footnote 42] NAIC applies RBC standards to measure the adequacy of an insurer's capital relative to their risks. Further, RRGs chartered as captives may not be required to comply with the same NAIC financial reporting requirements, such as filing quarterly and annual reports with NAIC, that regulators expect traditional insurance companies to meet.[Footnote 43] For example, while the statutes of all the leading domiciliary states require RRGs chartered as captives to file financial reports annually with their insurance departments, as of July 2004, when we conducted our survey, the statutes of only half the leading domiciliary states--Hawaii, South Carolina, and Vermont--explicitly require that these reports also be provided to NAIC on an annual basis.[Footnote 44] In addition, when RRGs are chartered as captive insurance companies they may not have to comply with the chartering state's statutes regulating insurance holding company systems. All 50 states and the District of Columbia substantially have adopted such statutes, based on NAIC's Model Insurance Holding Company System Regulatory Act.[Footnote 45] As in the model act, a state's insurance holding company statute generally requires insurance companies that are part of holding company systems and doing business in the state to register with the state and annually disclose to the state insurance regulator all the members of that system. Additionally, the act requires that transactions among members of a holding company system be on fair and reasonable terms, and that insurance commissioners be notified of and given the opportunity to review certain proposed transactions, including reinsurance agreements, management agreements, and service contracts. For 2004, NAIC reviewed RRG annual reports and identified 19 RRGs that reported themselves as being affiliated with other companies (for example, their management and reinsurance companies). However, since only two of the six leading domiciliary states, Hawaii, and to some extent South Carolina, actually require RRGs to comply with this act, we do not know whether more RRGs could be affiliated with other companies.[Footnote 46] The Hawaii regulator said that RRGs should abide by the act's disclosure requirements so that regulators can identify potential conflicts of interests with service providers, such as managers or insurance brokers. Unless an RRG is required to make these disclosures, the regulator would have the added burden of identifying and evaluating the nature of an RRG's affiliations. He added that such disclosures are important because the individual insureds of an RRG, in contrast to the single owner of a pure captive, may not have the ability to control potential conflicts of interest between the insurer and its affiliates. (See the next section of this report for examples of how affiliates of an RRG can have conflicts of interest with the RRG.) Because of these regulatory advantages, RRGs are more likely to domicile in states that will charter them as captives than in the states where they sell insurance. Figure 5 shows that 18 states could charter RRGs as captives. The figure also shows that most RRGs have chosen to domicile in six states--Arizona, the District of Columbia, Hawaii, Nevada, South Carolina, and Vermont--all of which charter RRGs as captives and market themselves as captive domiciles.[Footnote 47] Of these states, Vermont and Hawaii have been chartering RRG as captives for many years, but Arizona, the District of Columbia, Nevada, South Carolina, and five additional states have adopted their captive laws since 1999.[Footnote 48] In contrast to an RRG chartered as a captive, a true captive insurer generally does not directly conduct insurance transactions outside of its domiciliary state. Figure 5: Number of RRGs, by Captive or Noncaptive Charter and State of Domicile, as of the End of 2004: [See PDF for image] Note: States, with the exception of Maryland, provided us information about their captive laws as part of our survey. We did not independently verify the information provided or whether RRGs domiciled in the state were chartered as captives, although we updated some of the survey results to reflect states that have adopted captive statutes since the time of our survey. In addition, (1) the State of Maine indicated that while it had a captive law, the question of whether or not an RRG could form under it had not been formally considered and (2) the State of Kansas indicated that while it could charter RRGs as captives, its captive law does not explicitly permit RRGs to be chartered as captives. [End of figure] However, states of domicile are rarely the states in which RRGs sell much, or any, insurance. According to NAIC, 73 of the 115 RRGs active in 2003 did not write any business in their state of domicile, and only 10 wrote more than 30 percent of their business in their state of domicile.[Footnote 49] The states in which RRGs wrote most of their business in 2003--Pennsylvania ($238 million), New York ($206 million), California ($156 million), Massachusetts ($98 million)--did not charter any RRGs. Texas, which chartered only one RRG, had $87 million in direct written premiums written by RRGs. For more information on the number of RRGs chartered by state and the amount of direct premiums written by RRGs, see figure 6. Figure 6: Number of RRGs Chartered, by State, as of the End of 2004, and Amount of Direct Premiums Written by RRGs, by State, 2003: [See PDF for image] Note: The numbers displayed on some states in the map represent the number of RRGs domiciled in the state. For 2003, RRGs also wrote business in places such as Puerto Rico, Guam, and Canada (less than $17 million). [End of figure] Inconsistent Regulation of RRGs Resembles Earlier Regulation of Traditional Insurers, Which Suffered from Lack of Uniform, Baseline Standards: The current regulatory environment for RRGs, characterized by the lack of uniform, baseline standards, offers parallels to the earlier solvency regulation of multistate traditional insurers. Uniformity in solvency regulation for multistate insurers is important, provided the regulation embodies best practices and procedures, because it strengthens the regulatory system across all states and builds trust among regulators. After many insurance companies became insolvent during the 1980s, NAIC and the states recognized the need for uniform, baseline standards, particularly for multistate insurers.[Footnote 50] To alleviate this situation, NAIC developed its Financial Regulation Standards and Accreditation Program (accreditation standards) in 1989 and began the voluntary accreditation of most state regulators in the 1990s. Prior to accreditation, states did not uniformly regulate the financial solvency of traditional insurers, and many states lacked confidence in the regulatory standards of other states. By becoming accredited, state regulators demonstrated that they were willing to abide by a common set of solvency standards and practices for the oversight of the multistate insurers chartered by their state. As a result, states currently generally defer to an insurance company's domiciliary state regulator, even though each state retains the authority, through its licensing process, to regulate all traditional insurance companies selling in the state. NAIC's accreditation standards define baseline requirements that states must meet for the regulation of traditional companies in three major areas: First, they include minimum standards for the set of laws and regulations necessary for effective solvency regulation.[Footnote 51] Second, they set minimum standards for practices and procedures, such as examinations and financial analysis, which regulators routinely should do.[Footnote 52] Third, they establish expectations for resource levels and personnel practices, including the amount of education and experience required of professional staff, within an insurance department.[Footnote 53] However, NAIC does not have a similar set of regulatory standards for regulation of RRGs, which also are multistate insurers. According to NAIC officials, when the accreditation standards originally were developed, relatively few states were domiciling RRGs as captive insurers, and the question of standards for the regulation of captives and RRGs did not materialize until NAIC began its accreditation review of Vermont in 1993. NAIC completely exempted the regulation of captive insurers from the review process but included RRGs because, unlike pure captives, RRGs have many policyholders and write business in multiple states. NAIC's accreditation review of Vermont lasted about 2 years and NAIC and Vermont negotiated an agreement that only part of the accreditation standards applied to RRGs.[Footnote 54] As a result of the review, NAIC determined that RRGs were sufficiently different from traditional insurers so that the regulatory standards defining the laws and regulations necessary for effective solvency regulation should not apply to RRGs.[Footnote 55] However, NAIC and Vermont did not develop substitute standards to replace those they deemed inappropriate. Subsequently, other states domiciling RRGs as captives also have been exempt from enforcing the uniform set of laws and regulations deemed necessary for effective solvency regulation under NAIC's accreditation standards. As a result, some states chartering RRGs as captives do not obligate them, for example, to adopt a common set of financial reporting procedures and practices, abide by NAIC's requirements for risk-based capital, or comply with requirements outlined in that state's version of NAIC's Model Insurance Holding Company System Regulatory Act.[Footnote 56] In contrast, while NAIC's standards for the qualifications of an insurance department's personnel apply to RRGs, they do not distinguish between the expertise needed to oversee RRGs and traditional insurance companies. Because half of the 18 states that are willing to charter RRGs as captives have adopted captive laws since 1999, few domiciliary state insurance departments have much experience regulating RRGs as captive insurance companies. Further, in response to our 2004 survey, only three states new to chartering captives--Arizona, the District of Columbia, and South Carolina--reported that they have dedicated certain staff to the oversight of captives.[Footnote 57] However, the State of Nevada later reported to us that it dedicated staff to the oversight of captives as of June 2005. The importance of standards that address regulator education and experience can be illustrated by decisions made by state insurance departments or staff relatively new to chartering RRGs. In 1988, Vermont chartered Beverage Retailers Insurance Co. Risk Retention Group (BRICO). Launched and capitalized by an outside entity, BRICO did not have a sufficient number of members as evidenced by the need for an outside entity to provide the capital. It failed in 1995 in large part because it wrote far less business than originally projected and suffered from poor underwriting. Further, according to regulators, BRICO began to write business just as the market for its product softened, and traditional licensed insurers began to compete for the business. As a result, the Vermont regulators said that Vermont would not charter RRGs unless they had a sufficient number of insureds at start-up to capitalize the RRG and make its future operations sustainable. More recently, in 2000, shortly after it adopted its captive statutes, South Carolina chartered Commercial Truckers Risk Retention Group Captive Insurance Company. This RRG, which also largely lacked members at inception, failed within a year because it had an inexperienced management team, poor underwriting, and difficulties with its reinsurance company. The regulators later classified their experience with chartering this RRG, particularly the fact that the RRG lacked a management company, as "lessons learned" for their department. Finally, as reported in 2004, the Arizona insurance department inadvertently chartered an RRG that permitted only the brokerage firm that formed and financed the RRG to have any ability to control the RRG through voting rights. The Arizona insurance department explained that they approved the RRG's charter when the insurance department was operating under an acting administrator and that the department would make every effort to prevent similar mistakes. According to NAIC officials, RRGs writing insurance in multiple states, like traditional insurers, would benefit from the adoption of uniform, baseline standards for state regulation, and they plan gradually to develop them. NAIC representatives noted that questions about the application of accreditation standards related to RRGs undoubtedly would be raised again because several states new to domiciling RRGs will be subject to accreditation reviews in the next few years.[Footnote 58] However, the representatives also noted, that because the NAIC accreditation team can review the oversight of only a few of the many insurance companies chartered by a state, the team might not select an RRG. Variations in RRG Reporting Requirements Have Impeded Assessments of Their Financial Condition: As discussed previously, states domiciling RRGs as captives are not obligated to require that RRGs meet a common set of financial reporting procedures and practices. Moreover, even among states that charter RRGs as captives, the financial reporting requirements for RRGs vary. Yet, the only requirement under LRRA for the provision of financial information to nondomiciliary regulators is that RRGs provide annual financial statements to each state in which they operate. Further, since most RRGs sell the majority of their insurance outside their state of domicile, insurance commissioners from nondomiciliary states may have only an RRG's financial reports to determine if an examination may be necessary.[Footnote 59] As we have reported in the past, to be of use to regulators, financial reports should be prepared under consistent accounting and reporting rules and provided in a timely manner that results in a fair presentation of the insurer's true financial condition.[Footnote 60] One important variation in reporting requirements is the use by RRGs of accounting principles that differ from those used by traditional insurance companies. The statutes of the District of Columbia, Nevada, South Carolina, and Vermont require their RRGs to use GAAP; Hawaii requires RRGs to use statutory accounting principles (SAP); and Arizona permits RRGs to use either.[Footnote 61] The differences in the two sets of accounting principles reflect the different purposes for which each was developed and each produces a different--and not necessarily comparable--financial picture of a business. In general, SAP is designed to meet the needs of insurance regulators, the primary users of insurance financial statements, and stresses the measurement of an insurer's ability to pay claims (remain solvent) in order to protect insureds. In contrast, GAAP provides guidance that businesses follow in preparing their general purpose financial statements, which provide users such as investors and creditors with useful information that allows them to assess a business' ongoing financial performance. However, inconsistent use of accounting methodologies by RRGs could affect the ability of nondomiciliary regulators to determine the financial condition of RRGs, especially since regulators are used to assessing traditional insurers that must file reports using SAP.[Footnote 62] In addition, the statutes of each of the six domiciliary states allow RRGs, like other captive insurers, to modify whichever accounting principles they use by permitting the use of letters of credit (LOC) to meet statutory minimum capitalization requirements. Strictly speaking, neither GAAP nor SAP would permit a company to count an undrawn LOC as an asset because it is only a promise of future payment--the money is neither readily available to meet policyholder obligations nor is it directly in the possession of the company. In addition to allowing LOCs, according to a review of financial statements by NAIC, the leading domiciliary states that require RRGs to file financial statements using GAAP also allow RRGs to modify GAAP by permitting them to recognize surplus notes under capital and surplus. This practice is not ordinarily permitted by GAAP. A company filing under GAAP would recognize a corresponding liability for the surplus note and would not simply add it to the company's capital and surplus.[Footnote 63] See appendix III for more specific information on the differences between SAP and GAAP, including permitted modifications, and how these differences could affect assessments of a company's actual or risk- based capital. Variations in the use of accounting methods have consequences for nondomiciliary regulators who analyze financial reports submitted by RRGs and illustrate some of the regulatory challenges created by the absence of uniform standards. Most nondomiciliary states responding to our survey of all state regulators indicated that they performed only a limited review of RRG financial statements.[Footnote 64] To obtain more specific information about the impact of these differences, we contacted the six states--Pennsylvania, California, New York, Massachusetts, Texas, and Illinois--where RRGs collectively wrote almost half of their business in 2003 (see fig. 6). Regulators in Massachusetts and Pennsylvania reported that they did not analyze the financial reports and thus had no opinion about the impact of the accounting differences, but three of the other four states indicated that the differences resulted in additional work. Regulators from California and Texas told us that the use of GAAP, especially when modified, caused difficulties because insurance regulators were more familiar with SAP, which they also believed better addressed solvency concerns than GAAP. The regulator from Illinois noted that RRG annual statements were not marked as being filed based on GAAP and, when staff conducted their financial analyses, they took the time to disregard assets that would not qualify as such under SAP. The Texas regulator reported that, while concerned about the impact of the differences, his department did not have the staffing capability to convert the numbers for each RRG to SAP and, as a result, had to prioritize their efforts. Further, NAIC staff reported that the use by RRGs of a modified version of GAAP or SAP distorted the analyses they provided to state regulators. One of NAIC's roles is to help states identify potentially troubled insurers operating in their state by analyzing insurer financial reports with computerized tools to identify statistical outliers or other unusual data. In the past, we have noted that NAIC's solvency analysis is an important supplement to the overall solvency monitoring performed by states and can help states focus their examination resources on potentially troubled companies.[Footnote 65] NAIC uses Financial Analysis Solvency Tools (FAST), such as the ratios produced by the Insurance Regulatory Information System (IRIS) and the Insurer Profile Reports, to achieve these objectives and makes the results available to all regulators through a central database.[Footnote 66] However, NAIC analysts reported that differing accounting formats undermined the relative usefulness of these tools because the tools were only designed to analyze data extracted from financial reports based on SAP. Similarly, when we attempted to analyze some aspects of the financial condition of RRGs to compare them with traditional companies, we found that information produced under differing accounting principles diminished the usefulness of the comparison (see app. III). Lack of Uniform, Baseline Regulatory Standards Has Concerned Many Regulators: The lack of uniform, baseline regulatory standards for the oversight of RRGs contributed to the concerns of many state regulators, who did not believe the regulatory safeguards and protections built into LRRA (such as requiring RRGs to file annual financial statements with regulators and allowing regulators to file suit if they believe the RRG is financially unsound) were adequate.[Footnote 67] Only 8 of 42 regulators who responded to our survey question about LRRA's regulatory protections indicated that they thought the protections were adequate (see fig. 7).[Footnote 68] Eleven of the 28 regulators who believed that the protections were inadequate or very inadequate focused on the lack of uniform, regulatory standards or the need for RRGs to meet certain minimum standards--particularly for minimum capital and surplus levels. In addition, 9 of the 28 regulators, especially those from California and New York, commented that they believed state regulators needed additional regulatory authority to supervise the RRGs in their states. While RRGs, like traditional insurers, can sell in any or all states, only the domiciliary regulator has any significant regulatory oversight. Figure 7: State Regulators' Opinion of the Adequacy of the Regulatory Protections or Safeguards Built into LRRA: [See PDF for image] Note: In addition, seven regulators responded that they had no opinion on this question, and one regulator did not respond at all. [End of figure] In addition, the regulators from the six leading domiciliary states-- Arizona, the District of Columbia, Hawaii, Nevada, South Carolina, and Vermont--did not agree on the adequacy of LRRA safeguards. For example, while the regulators from the District of Columbia, Hawaii, Nevada, and Vermont thought the protections adequate, the regulator from South Carolina reported that LRRA's safeguards were "neither adequate nor inadequate" because LRRA delegates the responsibility of establishing safeguards to domiciliary states, which can be either stringent or flexible in establishing safeguards. The other leading domiciliary state--Arizona--had not yet formed an opinion on the adequacy of LRRA's provisions. The regulator from Hawaii also noted that the effectiveness of the LRRA provisions was dependent upon the expertise and resources of the RRG's domiciliary regulator. While many regulators did not believe LRRA's safeguards were adequate, few indicated that they had availed themselves of the tools LRRA does provide nondomiciliary state regulators. These tools include the ability to request that a domiciliary state undertake a financial examination and the right to petition a court of "competent jurisdiction" for an injunction against an RRG believed to be in a hazardous financial condition. Recent cases involving state regulation of RRGs typically have centered on challenges to nondomiciliary state statutes that affect operations of the RRGs, rather than actions by nondomiciliary states challenging the financial condition of RRGs selling insurance in their states.[Footnote 69] Finally, in response to another survey question, nearly half of the regulators said they had concerns that led them to contact domiciliary state regulators during the 24 months preceding our survey, but only five nondomiciliary states indicated that they had ever asked domiciliary states to conduct a financial examination.[Footnote 70] However, according to the survey, many state regulators availed themselves of the other regulatory safeguards that LRRA provides--that RRGs submit to nondomiciliary states feasibility or operational plans before they begin operations in those states and thereafter a copy of the same annual financial statements that the RRG submits to its domiciliary state.[Footnote 71] Almost all the state regulators indicated that they reviewed these documents to some extent, although almost half of the state regulators indicated that they provided these reports less review than those submitted by other nonadmitted insurers.[Footnote 72] In addition, nine states indicated that RRGs began to conduct business in their states before supplying them with copies of their plans of operations or feasibility studies, but most indicated that these occurrences were occasional. Similarly, 15 states identified RRGs that failed to provide required financial statements for review, but most of these regulators indicated that the failure to file was an infrequent occurrence.[Footnote 73] Some Evidence Suggests That States Have Set Their Captive Regulatory Standards to Attract RRGs to Domicile in Their States: Some regulators, including those from New York, California, and Texas- -states where RRGs collectively wrote about 26 percent of all their business but did not domicile--expressed concerns that domiciliary states were lowering their regulatory standards to attract RRGs to domicile in their states for economic development purposes. They sometimes referred to these practices as the "regulatory race to the bottom." RRGs, like other captives, can generate revenue for a domiciliary state's economy when the state taxes RRG insurance premiums or the RRG industry generates jobs in the local economy. The question of whether domiciliary states were competing with one another essentially was moot until about 1999, when more states began adopting captive laws. Until then, Vermont and Hawaii were two of only a few states that were actively chartering RRGs and through 1998 had chartered about 55 percent of all RRGs. However, between the beginning of 1999 and the end of 2004, they had chartered only 36 percent of all newly chartered RRGs.[Footnote 74] The six leading domiciliary states actively market their competitive advantages on Web sites, at trade conferences, and through relationships established with trade groups. They advertise the advantages of their new or revised captive laws and most describe the laws as "favorable"; for example, by allowing captives to use letters of credit to meet their minimum capitalization requirements. Most of these states also describe their corporate premium tax structure as competitive and may describe their staff as experienced with or committed to captive regulation. Vermont emphasizes that it is the third-largest captive insurance domicile in the world and the number one in the United States, with an insurance department that has more than 20 years of experience in regulating RRGs. South Carolina, which passed its captive legislation in 2000, emphasizes a favorable premium tax structure and the support of its governor and director of insurance for its establishment as a domicile for captives. Arizona describes its state as "business friendly," highlighting the lack of premium taxes on captive insurers and the "unsurpassed" natural beauty of the state. However, in addition to general marketing, some evidence exists to support the concern that the leading domiciliary states are modifying policies and procedures to attract RRGs. We identified the following notable differences among the states, some of which reflect the regulatory practices and approaches of each state and others, statute: * Willingness to domicile vehicle service contract (VSC) providers: Several states, including California, New York, and Washington, questioned whether RRGs consisting of VSC providers should even qualify as RRGs and are concerned about states that allow these providers to form RRGs. VSC providers issue extended service contracts for the costs of future repairs to consumers (that is, the general public) who purchase automobiles. Until 2001, almost all of these RRGs were domiciled in Hawaii but after that date, all the new RRGs formed by VSC providers have domiciled in the District of Columbia and South Carolina.[Footnote 75] The Hawaii regulator said that the tougher regulations it imposed in 2001 (requiring that RRGs insuring VSC providers annually provide acceptable proof that they were financially capable of meeting VSC claims filed by consumers) dissuaded these providers from domiciling any longer in Hawaii. In addition, one of the leading domiciliary states, Vermont, refuses to domicile any of these RRGs because of the potential risk to consumers. Consumers who purchase these contracts, not just the RRG insureds, can be left without coverage if the RRG insuring the VSC provider's ability to cover VSC claims fails. (We discuss RRGs insuring service contract providers and consequences to insureds and consumers more fully later in this report.) * Statutory minimum capitalization requirements: Differences in the minimum amount of capital and surplus (capitalization) each insurer must have before starting operations make it easier for smaller RRGs to domicile in certain states and reflect a state's attitude towards attracting RRGs. For example, in 2003, Vermont increased its minimum capitalization amount from $500,000 to $1 million--according to regulators, to ensure that only RRGs that are serious prospects, with sufficient capital, apply to be chartered in the state. On the other hand, effective in 2005, the District of Columbia lowered its minimum capitalization amount for a RRG incorporated as a stock insurer (that is, owned by shareholders who hold its capital stock) from $500,000 to $400,000 to make it easier for RRGs to charter there. * Corporate forms: In 2005, one of the six leading domiciliary states- -the District of Columbia--enacted legislation that permits RRGs to form "segregated accounts." The other leading domiciliary states permit the formation of segregated accounts or "protected cells" for other types of captives but not for their RRGs. According to the District's statute, a captive insurer, including an RRG, may fund separate accounts for individual RRG members or groups of members with common risks, allowing members to segregate a portion of their risks from the risks of other members of the RRG.[Footnote 76] According to the District regulator, RRG members also would be required to contribute capital to a common account that could be used to cover a portion of each member's risk. The District regulator also noted that the segregated cell concept has never been tested in insolvency; as a result, courts have not yet addressed the concept that the cells are legally separate. * Willingness to charter entrepreneurial RRGs: RRGs may be formed with only a few members, with the driving force behind the formation being, for example, a service provider, such as the RRG's management company or a few members. These RRGs are referred to as "entrepreneurial" RRGs because their future success is often contingent on recruiting additional members as insureds. In 2004, South Carolina regulators reported they frequently chartered entrepreneurial RRGs to offset what they described as the "chicken and egg" problem--their belief that it can be difficult for RRGs to recruit new members without having the RRG already in place. Regulators in several other leading domiciliary states have reported they would be willing to charter such RRGs if their operational plans appeared to be sound but few reported having done so. However, regulators in Vermont said that they would not charter entrepreneurial RRGs because they often were created to make a profit for the "entrepreneur," rather than helping members obtain affordable insurance. (We discuss entrepreneurial RRGs later in the report.) Finally, the redomiciling of three RRGs to two of the leading domiciliary states, while subject to unresolved regulatory actions in their original state of domicile, also provides some credibility to the regulators' assertions of "regulatory arbitrage." In 2004, two RRGs redomiciled to new states while subject to regulatory actions in their original states of domicile. One RRG, which had been operating for several years, redomiciled to a new state before satisfying the terms of a consent order issued by its original domiciliary state and without notifying its original state of domicile.[Footnote 77] Although the RRG satisfied the terms of the consent order about 3 months after it redomiciled, the regulator in the original domiciliary state reported that, as provided by LRRA, once redomiciled, the RRG had no obligation to do so. The second RRG, one that had been recently formed, was issued a cease and desist order by its domiciliary state because the regulators had questions about who actually owned and controlled the RRG. As in the first case, the original domiciliary state regulator told us that this RRG did not advise them that it was going to redomicile and, once redomiciled, was under no legal obligation to satisfy the terms of the cease and desist order. The redomiciling, or rather liquidation, of the third RRG is more difficult to characterize because its original state of domicile (Hawaii) allowed it to transfer some of its assets to a new state of domicile (South Carolina) after issuing a cease and desist order to stop it from selling unauthorized insurance products directly to the general public, thereby violating the provisions of LRRA.[Footnote 78] More specifically, Hawaii allowed the RRG to transfer its losses and related assets for its "authorized" lines of insurance to South Carolina and required the Hawaiian company to maintain a $1 million irrevocable LOC issued in favor of the insurance commissioner until such time as the "unauthorized" insurance matter was properly resolved. South Carolina permitted the owners of these assets to form a new RRG offering a similar line of coverage and use a name virtually identical to its predecessor in Hawaii. Had these RRGs been chartered as traditional insurance companies, they would not have had the ability to continue operating in their original state of domicile after redomiciling in another state without the original state's express consent. Because traditional companies must be licensed in each state in which they operate, the original state of domicile would have retained its authority to enforce regulatory actions. RRG Failures Have Raised Questions about the Sufficiency of LRRA Provisions for RRG Ownership, Control, and Governance: Because LRRA does not comprehensively address how RRGs may be owned, controlled, or governed, RRGs may be operated in ways that do not consistently protect the best interests of their insureds. For example, while self-insurance is generally understood as risking one's own money to cover losses, LRRA does not specify that RRG members, as owners, make capital contributions beyond their premiums or maintain any degree of control over their governing bodies (such as boards of directors). As a result, in the absence of specific federal requirements and using the latitude LRRA grants them, some leading domiciliary regulators have not required all RRG insureds to make at least some capital contribution or exercise any control over the RRG. Additionally, some states have allowed management companies or a few individuals to form what are called "entrepreneurial" RRGs. Consequently, some regulators were concerned that RRGs were being chartered primarily for purposes other than self-insurance, such as making a profit for someone other than the collective insureds. Further, LRRA does not recognize that separate companies typically manage RRGs. Yet, past RRG failures suggest that sometimes management companies have promoted their own interests at the expense of the insureds. Although LRRA does not address governance issues such as conflicts of interest between management companies and insureds, Congress previously has enacted safeguards to address similar issues in the mutual fund industry. Finally, some of these RRG failures have resulted in thousands of insureds and their claimants losing coverage, some of whom may not have been fully aware that their RRG lacked state insurance insolvency guaranty fund coverage or the consequences of lacking such coverage. While RRGs Are a Form of Self-Insurance, Not All RRG Insureds Are Equity Owners or Have the Ability to Exercise Control: While RRGs are a form of self-insurance on a group basis, LRRA does not require that RRG insureds make a capital investment in their RRG and provides each state considerable authority to establish its own rules on how RRGs will be chartered and regulated. Most of the regulators from the leading domiciliary states reported that they require RRGs to be organized so that all insureds make some form of capital contribution but other regulators do not, or make exceptions to their general approach.[Footnote 79] Regulators from Vermont and Nevada emphasized that it was important for each member to have "skin in the game," based on the assumption that members who make a contribution to the RRG's capital and surplus would have a greater interest in the success of the RRG. The regulator from Nevada added that if regulators permitted members to participate without making a capital contribution, they were defeating the spirit of LRRA. However, another of the leading domiciliary states, the District of Columbia, does not require insureds to make capital contributions as a condition of charter approval and has permitted several RRGs to be formed accordingly. The District regulator commented that LRRA does not require such a contribution and that some prospective RRG members may not have the financial ability to make a capital contribution. Further, despite Vermont's position that RRG members should make a capital contribution, the Vermont regulators said they occasionally waive this requirement under special circumstances; for example, if the RRG was already established and did not need any additional capital. In addition, several of the leading domiciliary states, including Arizona, the District of Columbia, and Nevada, would consider allowing a nonmember to provide an LOC to fund the capitalization of the RRG. However, as described by several regulators, including those in Hawaii and South Carolina, even when members do contribute capital to the RRG, the amount contributed can vary and be quite small. For instance, an investor with a greater amount of capital, such as a hospital, could initially capitalize an RRG, and expect smaller contributions from members (for example, doctors) with less capital. Or, in an RRG largely owned by one member, additional members might be required only to make a token investment, for example, $100 or less. As a result, an investment that small would be unlikely to motivate members to feel like or behave as "owners" who were "self-insuring" their risks. LRRA also does not have a requirement that RRG insureds retain control over the management and operation of their RRG. However, as discussed previously, the legislative history indicates that some of the act's single-state regulatory framework and other key provisions were premised not only on ownership of an RRG being closely tied to the interests of the insureds, but also that the insureds would be highly motivated to ensure proper management of the RRG. Yet, in order to make or direct key decisions about a company's operations, the insureds would have to be able to influence or participate in the company's governing body (for example, a board of directors).[Footnote 80] A board of directors is the focal point of an insurer's corporate governance framework and ultimately should be responsible for the performance and conduct of the insurer.[Footnote 81] Governance is the manner in which the boards of directors and senior management oversee a company, including how they are held accountable for their actions.[Footnote 82] Most leading state regulators said they expect members of RRGs to exert some control over the RRG by having the ability to vote for directors, even though these rights sometimes vary in proportion to the size of a member's investment in the RRG or by share class.[Footnote 83] Most of the leading state regulators generally define "control" to be the power to direct the management and policies of an RRG as exercised by an RRG's governing body, such as its board of directors. However, regulators from the District of Columbia asserted that they permit RRGs to issue nonvoting shares to their insureds because some members are capable of making a greater financial contribution than others and, in exchange for their investment, will seek greater control over the RRG. The regulators noted that allowing such arrangements increases the availability of insurance and has no adverse effect on the financial solvency of the RRG. Further, the District of Columbia permits nonmembers (that is, noninsureds) to appoint or vote for directors. In addition, we found that even regulators who expect all RRG members to have voting rights (that is, at a minimum a vote for directors) sometimes make exceptions. For example, an RRG domiciled in Vermont was permitted to issue shares that did not allow insureds to vote for members of the RRG's governing body. The Vermont regulators reported that the attorney forming the RRG believed issuing the shares was consistent with the department's position that RRG members should have "voting rights" because under Vermont law all shareholders are guaranteed other minimal voting rights.[Footnote 84] While most regulators affirmed that they expect RRG members to own and control their RRGs, how these expectations are fulfilled is less clear when an organization, such as an association, owns an RRG. Four states- -Arizona, District of Columbia, South Carolina, and Vermont--reported that they have chartered RRGs that are owned by a single or multiple organizations, rather than individual persons or businesses.[Footnote 85] One of these states--the District of Columbia--permits noninsureds to own the organizations that formed the RRG. However, the District regulator said that while the noninsureds may own the voting or preferred stock of the association, they do not necessarily have an interest in controlling the affairs of the RRG. In addition, Arizona has permitted three risk purchasing groups (RPGs) to own one RRG.[Footnote 86] While the three RPGs, organized as domestic corporations in another state, collectively have almost 8,000 policyholders, four individuals, all of whom are reported to be RRG insureds by the Arizona regulator, are the sole owners of all three RPGs.[Footnote 87] Regulators Expressed Concerns That Some RRGs Might Be Operated to Make Money for an Entrepreneur, Rather Than to Provide Self-Insurance: The chartering of an "entrepreneurial" RRG--which regulators generally define as formed by an individual member or a service provider, such as a management company, for the primary purpose of making profits for themselves--has been controversial. According to several regulators, entrepreneurial RRGs are started with a few members and need additional members to remain viable. The leading domiciliary regulators have taken very different positions on entrepreneurial RRGs, based on whether they thought the advantages entrepreneurs could offer (obtaining funding and members) outweighed the potential adverse influence the entrepreneur could have on the RRG. We interviewed regulators from the six leading domiciliary states to obtain their views on entrepreneurial RRGs. In 2004, South Carolina regulators reported they firmly endorsed chartering entrepreneurial RRGs because they believed that already chartered RRGs stand a better chance of attracting members than those in the planning stages.[Footnote 88] They cited cases of entrepreneurial RRGs they believe have met the insurance needs of nursing homes and taxicab drivers. However, regulators from Vermont and Hawaii had strong reservations about this practice because they believe the goal of entrepreneurs is to make money for themselves--and that the pursuit of this goal could undermine the financial integrity of the RRG because of the adverse incentives that it creates. Vermont will not charter entrepreneurial RRGs and has discouraged them from obtaining a charter in Vermont by requiring RRGs (before obtaining their charter) to have a critical mass of members capable of financing their own RRG. In addition, the Vermont regulators said they would not permit an entrepreneur, if just a single owner, to form an RRG as a means of using LRRA's regulatory preemption to bypass the licensing requirements of the other states in which it planned to operate. Two of the other leading domiciliary states--Arizona and Nevada--were willing to charter entrepreneurial RRGs, providing they believed that the business plans of the RRGs were sound.[Footnote 89] Finally, some of the leading state regulators that have experience with chartering entrepreneurial RRGs told us that they recognized that the interests of the RRG insureds have to be protected and that they took measures to do so. For example, the regulators from South Carolina said that even if one member largely formed and financed an RRG, they would try to ensure that the member would not dominate the operations. However, they admitted that the member could do so because of his or her significant investment in the RRG. Alternatively, the regulator from Hawaii reported that the state's insurance division, while reluctant to charter entrepreneurial RRGs, would do so if the RRG agreed to submit to the division's oversight conditions. For example, to make sure service providers are not misdirecting money, the division requires entrepreneurial RRGs to submit copies of all vendor contracts. The Hawaii regulator also told us that the insurance division requires all captives to obtain the insurance commissioner's approval prior to making any distributions of principal or interest to holders of surplus notes. However, he concluded that successful oversight ultimately depended on the vigilance of the regulator and the willingness of the RRG to share documentation and submit to close supervision. LRRA Lacks Governance Standards to Protect RRG Insureds from Management Companies with Potential Conflicts of Interest: LRRA imposes no governance requirements that could help mitigate the risk to RRG insureds from potential abuses by other interests, such as their management companies, should they choose to maximize their profits at the expense of the best interests of the RRG insureds. Governance rules enhance the independence and effectiveness of governing bodies, such as boards of directors, and improve their ability to protect the interests of the company and insureds they serve. Unlike a typical company where the firm's employees operate and manage the firm, an RRG usually is operated by a management company and may have no employees of its own. However, while management companies and other service providers generally provide valuable services to RRGs, the potential for abuse arises if the interests of a management company are not aligned with the interests of the RRG insureds to consistently obtain self-insurance at the most affordable price consistent with long-term solvency. These inherent conflicts of interest are exemplified in the circumstances surrounding 10 of 16 RRG failures that we examined.[Footnote 90] For example, members of the companies that provided management services to Charter Risk Retention Group Insurance Company (Charter) and Professional Mutual Insurance Company Risk Retention Group (PMIC) also served as officers of the RRGs' boards of directors, which enabled them to make decisions that did not promote the welfare of the RRG insureds. In other instances, such as the failure of Nonprofits Mutual Risk Retention Group, Inc. (Nonprofits), the management company negotiated terms that made it difficult for the RRG to terminate its management contract and place its business elsewhere. Regulators knowledgeable about these and other failures commented that the members, while presumably self-insuring their risks, were probably more interested in satisfying their need for insurance than actually running their own insurance company. The 2003 failure of three RRGs domiciled in Tennessee--American National Lawyers Insurance Reciprocal Risk Retention Group (ANLIR), Doctors Insurance Reciprocal Risk Retention Group (DIR), and The Reciprocal Alliance Risk Retention Group (TRA)--further illustrates the potential risks and conflicts of interest associated with a management company operating an RRG.[Footnote 91] In pending litigation, the State of Tennessee's Commissioner of Commerce and Insurance, as receiver for the RRGs, has alleged that the three RRGs had common characteristics, such as (1) being formed by Reciprocal of America (ROA), a Virginia reciprocal insurer, which also served as the RRGs' reinsurance company; (2) having a management company, The Reciprocal Group (TRG), which also served as the management company and attorney-in-fact for ROA; (3) receiving loans from ROA, TRG, and their affiliates; and (4) having officers and directors in common with ROA and TRG.[Footnote 92] The receiver has alleged that through the terms of RRGs' governing instruments, such as its bylaws, management agreements with TRG (which prohibited the RRGs from replacing TRG as their exclusive management company for as long as the loans were outstanding), and the common network of interlocking directors among the companies, TRG effectively controlled the boards of directors of the RRGs in a manner inconsistent with the best interests of the RRGs and their insureds.[Footnote 93] As alleged in the complaint filed by the Tennessee regulator, one such decision involved a reinsurance agreement, in which the RRGs ceded 90- 100 percent of their risk to ROA with a commensurate amount of premiums--conditions that according to the regulator effectively prevented the RRGs from ever operating independently or retaining sufficient revenue to pay off their loans with ROA and TRG and thus remove TRG as their management company.[Footnote 94] Within days after the Commonwealth of Virginia appointed a receiver for the rehabilitation or liquidation of ROA and TRG, the State of Tennessee took similar actions for the three RRGs domiciled in Tennessee.[Footnote 95] The following failures of other RRGs also illustrate behavior suggesting that management companies and affiliated service providers have promoted their own interests at the expense of the RRG insureds: * According to the Nebraska regulators, Charter failed in 1992 because its managers, driven to achieve goals to maximize their profits, undercharged on insurance rates in an effort to sell more policies. One board officer and a company manager also held controlling interests in third-party service providers, including the one that determined if claims should be paid. Further, the board officer and a company manager, as well as the RRG, held controlling interests in the RRG's reinsurance company. A Nebraska regulator noted that when a reinsurance company is affiliated with the insurer it is reinsuring: (1) the reinsurer's incentive to encourage the insurer to adequately reserve and underwrite is reduced and (2) the insurer also will be adversely affected by any unprofitable risk it passes to the reinsurer. * PMIC, which was domiciled in Missouri and formed to provide medical malpractice insurance coverage for its member physicians, was declared insolvent in 1994. The RRG's relationship with the companies that provided its management services undermined the RRG in several ways. The president of PMIC was also the sole owner of Corporate Insurance Consultants (CIC), a company with which PMIC had a marketing service and agency agreement. As described in the RRG's examination reports, the RRG paid CIC exorbitant commissions for services that CIC failed to provide, but allowed CIC to finance collateral loans made by the reinsurance company to CIC. In turn, CIC had a significant ownership stake in the RRG's reinsurance company, which also provided PMIC with all of its personnel. The reinsurer's own hazardous financial condition resulted in the failure of PMIC. * In the case of Nonprofits, Vermont regulators indicated that essentially the excessive costs of its outsourced management company and outsourced underwriting and claims operations essentially contributed to its 2000 failure. The regulators said that the management company was in a position to exert undue influence over the RRG's operations because the principals of the management company loaned the RRG its start-up capital in the form of irrevocable LOCs. In addition to charging excessive fees, the management company also locked the RRG into a management contract that only allowed the RRG to cancel the contract 1 year before its expiration. If the RRG did not, the contract would automatically renew for another 5 years, a requirement of which the RRG insureds said they were unaware. Although LRRA has no provisions that address governance controls, Congress has acted to provide such controls in similar circumstances in another industry. In response to conditions in the mutual fund industry, Congress passed the Investment Company Act of 1940 (1940 Act). The 1940 Act, as implemented by the Securities and Exchange Commission (SEC), establishes a system of checks and balances that includes participation of independent directors on mutual fund boards, which oversee transactions between the mutual fund and its investment adviser.[Footnote 96] A mutual fund's structure and operation, like that of an RRG, differs from that of a traditional corporation. In a typical corporation, the firm's employees operate and manage the firm; the corporation's board of directors, elected by the corporation's stockholders, oversees its operation. Unlike a typical corporation, but similar to many RRGs, a typical mutual fund has no employees and contracts with another party, the investment adviser, to administer the mutual fund's operations. Recognizing that the "external management" of most mutual funds presents inherent conflicts between the interests of the fund shareholders and those of the fund's investment adviser, as well as potential for abuses of fund shareholders, Congress included several safeguards in the 1940 Act. For example, with some exceptions, the act requires that at least 40 percent of the board of directors of a mutual fund be disinterested (that is, that directors be independent of the fund's investment adviser as well as certain other persons having significant or professional relationships with the fund) to help ensure that the fund is managed in the best interest of its shareholders.[Footnote 97] The 1940 Act also regulates the terms of contracts with investment advisers by imposing a maximum contract term and by guaranteeing the board's and the shareholders' ability to terminate an investment adviser contract.[Footnote 98] The act also requires that the terms of any contract with the investment adviser and the renewal of such contract be approved by a majority of directors who are not parties to the contract or otherwise interested persons of the investment adviser. Further, the 1940 Act imposes a fiduciary duty upon the adviser in relation to its level of compensation and provides the fund and its shareholders with the right to sue the adviser should the fees be excessive.[Footnote 99] The management controls imposed on mutual fund boards do not supplant state law on duties of "care and loyalty" that oblige directors to act in the best interests of the mutual fund, but enhance a board's ability to perform its responsibilities consistent with the protection of investors and the purposes of the 1940 Act. RRG Members May Not Realize They Lack Guaranty Fund Protection: In addition to lacking comprehensive provisions for ownership, control, and governance of RRGs, LRRA does not mandate that RRGs disclose to their insureds that they lack state insurance insolvency guaranty fund protection. LRRA's legislative history indicates that the prohibition on RRGs participating in state guaranty funds (operated to protect insureds when traditional insurers fail) stemmed, in part, from a belief that the lack of protection would help motivate RRG members to manage the RRG prudently. LRRA does provide nondomiciliary state regulators the authority to mandate the inclusion of a specific disclosure, which informs RRG insureds that they lack guaranty fund coverage, on insurance policies issued to residents of their state (see fig. 8).[Footnote 100] However, LRRA does not provide nondomiciliary states with the authority to require the inclusion of this disclaimer in policy applications or marketing materials. For example, of 40 RRGs whose Web sites we were able to identify, only 11 disclosed in their marketing material that RRGs lack guaranty fund protection. In addition, 11 of the RRGs omitted the words "Risk Retention Group" from their names.[Footnote 101] Figure 8: Permitted Wording of Guaranty Fund Disclosure in LRRA: [See PDF for image] [End of figure] All of the six leading domiciliary states have adopted varying statutory requirements that RRGs domiciled in their states include the disclosure in their policies, regardless of where they operate. The statutes of Hawaii, South Carolina, and the District of Columbia require that the disclosure be printed on applications for insurance, as well as on the front and declaration page of each policy. By requiring that the disclosure be printed on insurance applications, prospective RRG insureds have a better chance of understanding that they lack guaranty fund protection. Regulators in South Carolina, based on their experience with the failure of Commercial Truckers RRG in 2001, also reported that they require insureds, such as those of transportation and trucking RRGs, to place their signature beneath the disclosure. The regulators imposed this additional requirement because they did not believe that some insureds would be as likely to understand the implications of not having guaranty fund coverage as well as other insureds (for example, hospital conglomerates). In contrast, the statutes of Arizona and Vermont require only that the disclosure be printed on the insurance policies. The six leading domiciliary state regulators had mixed views on whether the contents of the disclosure should be enhanced, but none recommended that LRRA be changed to permit RRGs to have guaranty fund protection. It is unclear whether RRG insureds who obtain insurance through organizations that own RRGs understand that they will not have guaranty fund coverage. Four states--Arizona, the District of Columbia, South Carolina, and Vermont--indicated that they have chartered RRGs owned by single organizations. When an organization is the insured of the RRG, the organization receives the insurance policy with the disclosure about lack of guaranty fund protection. Whether the organization's members, who are insured by the RRG, understand that they lack guaranty fund coverage is less clear. The Vermont regulators indicated that members typically are not advised that they lack guaranty fund coverage before they receive the policy. Thus, the regulators recommended that applications for insurance should contain the disclosure as well. The Arizona regulator reported that the insurance applications signed by the insureds of the Arizona-domiciled RRG owned by three RPGs did not contain a disclosure on the lack of guaranty fund coverage, although the policy certificates did. Further, he reported that the practices of RPGs were beyond his department's jurisdiction and that he does not review them. Not understanding that RRG insureds are not protected by guaranty funds has serious implications for RRG members and their claimants, who have lost coverage as a result of RRG failures. For example, of the 21 RRGs that have been placed involuntarily in liquidation, 14 either have or had policyholders whose claims remain or are likely to remain partially unpaid (see app. IV).[Footnote 102] Member reaction to the failure of the three RRGs domiciled in Tennessee further illustrates that the wording and the placement of the disclosure may be inadequate.[Footnote 103] In 2003, an insurance regulator from Virginia, a state where many of the RRG insureds resided, reported that he received about 150-200 telephone calls from the insureds of these RRGs and the insureds did not realize they lacked "guaranty fund" coverage, asking instead why they didn't have "back-up" insurance when their insurance company failed. He explained that the insureds were "shocked" to discover they were members of an RRG, rather than a traditional insurance company and that they had no guaranty fund coverage. According to the regulator, they commented, "Who reads their insurance policies?" Regulators in Tennessee also noted that insureds of the RRGs, including attorneys, hospitals, and physicians did not appear to understand the implications of self-insuring their risks and the lack of guaranty fund coverage. In 2004, the State of Tennessee estimated that the potential financial losses from these failures to the 50,000 or so hospitals, doctors, and attorneys that were members of the Tennessee RRGs could exceed $200 million, once the amount of unpaid claims were fully known.[Footnote 104] Other regulators, including those in Missouri, in response to our survey, and New York, in an interview, also expressed concern that some RRG members might not fully understand the implications of a lack of guaranty fund protection and were not the "sophisticated" consumers that they believe may have been presumed by LRRA. In addition, in response to our survey, regulators from other states including New Mexico and Florida expressed specific concerns about third-party claimants whose claims could go unpaid when an RRG failed and the insured refused or was unable to pay claims. The Florida regulator noted that the promoters of the RRG could accentuate the "cost savings" aspect of the RRG at the expense of explaining the insured's potential future liability in the form of unpaid claims due to the absence of guaranty funds should the RRG fail. In addition, regulators who thought that protections in LRRA were inadequate, such as those in Wyoming, Virginia, and Wisconsin, tended to view lack of guaranty fund protection as a primary reason for developing and implementing more uniform regulatory standards or providing nondomiciliary states greater regulatory authority over RRGs. Lack of Guaranty Fund Protection Also Has Consequences for Consumers Who Purchase Extended Service Contracts: Lack of guaranty fund protection also can have unique consequences for consumers who purchase extended service contracts from service contract providers. Service contract providers form RRGs to insure their ability to pay claims on extended service contracts--a form of insurance also known as contractual liability insurance--and sell these contracts to consumers.[Footnote 105] In exchange for the payment (sometimes substantial) made by the consumer, the service contract provider commits to performing services--for example, paying for repairs to an automobile. Service contract providers may be required to set aside some portion of the money paid by consumers in a funded "reserve account" to pay resulting claims and may have to buy insurance (for example, from the RRG they have joined) to guarantee their ability to pay claims.[Footnote 106] However, potential problems result from the perception of consumers that what they have purchased is insurance, since the service contract provider pays for repairs or other service, when in fact it is not.[Footnote 107] Only the service contract provider purchases insurance, the consumer signs a contract for services. The failure of several RRGs, including HOW Insurance Company RRG (HOW) in 1994 and National Warranty RRG in 2003, underscores the consequences that failures of RRGs that insure service contract providers can have on consumers: * In 1994, the Commonwealth of Virginia liquidated HOW and placed its assets in receivership. This RRG insured the ability of home builders to fulfill contractual obligations incurred by selling extended service contracts to home buyers. While settled out of court, the Commonwealth of Virginia asserted that the homeowners who purchased the contracts against defects in their homes had been misled into believing that they were entitled to first-party insurance benefits--that is, payment of claims.[Footnote 108] A Virginia regulator said that while his department received few calls from the actual insureds (that is, the home builders) at the time of failure, they received many calls from home owners who had obtained extended service contracts when they purchased their home and thought they were insured directly by the RRG. * In 2003, National Warranty Insurance RRG failed, leaving behind thousands of customers with largely worthless vehicle service contracts (VSCs). This RRG, domiciled in the Cayman Islands, insured the ability of service contract providers to honor contractual liabilities for automobile repairs. Before its failure, National Warranty insured at least 600,000 VSCs worth tens of millions of dollars. In 2003, the liquidators of National Warranty estimated that losses could range from $58 to $74 million.[Footnote 109] National Warranty's failure also raised the question of whether RRGs were insuring consumers directly, which LRRA prohibits--for example, because the laws of many states, including Texas, require that the insurance company become directly responsible for unpaid claims in the event a service contract provider failed to honor its contract.[Footnote 110] The failure of National Warranty also raised the question of whether RRGs should insure service contract providers at all because of the potential direct damage to consumers. Several regulators, including those in California, Wisconsin, and Washington, went even further. In response to our survey, they opined that LRRA should be amended to preclude RRGs from offering "contractual liability" insurance because such policies cover a vehicle service contract provider's financial obligations to consumers.[Footnote 111] At a minimum, regulators from New York and California, in separate interviews, recommended that consumers who purchase extended service contracts insured by RRGs at least be notified in writing that the contracts they purchase were not insurance and would not qualify for state guaranty fund coverage. Conclusions: In establishing RRGs, Congress intended to alleviate a shortage of affordable commercial liability insurance by enabling commercial entities to create their own insurance companies to self-insure their risks on a group basis. RRGs, as an industry, according to most state insurance regulators, have fulfilled this vision--and the intent of LRRA--by increasing the availability and affordability of insurance for members that experienced difficulty in obtaining coverage. While constituting only a small portion of the total liability insurance market, RRGs have had a consistent presence in this market over the years. However, the number of RRGs has increased dramatically in recent years in response to recent shortages of liability insurance. While we were unable to evaluate the merits of individual RRGs, both state regulators and advocates of the RRG industry provided specific examples of how they believe RRGs have addressed shortages of insurance in the marketplace. This ability is best illustrated by the high number of RRGs chartered over the past 3 years to provide medical malpractice insurance, a product which for traditional insurers historically has been subject to high or unpredictable losses with resulting failures. However, the regulation of RRGs by a single state, in combination with the recent increase in the number of states new to domiciling RRGs, the increase in the number of RRGs offering medical malpractice insurance, and a wide variance in regulatory practices, has increased the potential for future solvency risks. As a result, RRG members and their claimants could benefit from greater regulatory consistency. Insurance regulators have recognized the value of having a consistent set of regulatory laws, regulations, practices, and expertise through the successful implementation of NAIC's accreditation program for state regulators of multistate insurance companies. Vermont and NAIC negotiated the relaxation of significant parts of the accreditation standards for RRGs because it was unclear how the standards, designed for traditional companies, applied to RRGs. However, this agreement allowed states chartering RRGs as captives considerable latitude in their regulatory practices, even though most RRGs were multistate insurers, raising the concerns of nondomiciliary states. With more RRGs than ever before and with a larger number of states competing to charter them, regulators, working through NAIC, could develop a set of comprehensive, uniform, baseline standards for RRGs that would provide a level of consistency that would strengthen RRGs and their ability to meet the intent of LRRA. While the regulatory structure applicable to RRGs need not be identical to that used for traditional insurance companies, uniform, baseline regulatory standards could create a more transparent and protective regulatory environment, enhancing the financial strength of RRGs and increasing the trust and confidence of nondomiciliary state regulators. These standards could include such elements as the use of a consistent accounting method, disclosing relationships with affiliated businesses as specified by NAIC's Model Insurance Holding Company System Regulatory Act, and the qualifications and number of staff that insurance departments must have available to charter RRGs. These standards could reflect the regulatory best practices of the more experienced RRG regulators and address the concerns of the states where RRGs conduct the majority of their business. Further, such standards could reduce the likelihood that RRGs would practice regulatory arbitrage, seeking departments with the most relaxed standards. While it may not be essential for RRGs to follow all the same rules that traditional insurers follow, it is difficult to understand why all RRGs and their regulators, irrespective of where they are domiciled, should not conform to a core set of regulatory requirements. Developing and implementing such standards would strengthen the foundation of LRRA's flexible framework for the formation of RRGs. LRRA's provisions for the ownership, control, and governance of RRGs may not be sufficient to protect the best interests of the insureds. While acknowledging that LRRA has worked well to promote the formation of RRGs in the absence of uniform, baseline standards, this same flexibility has left some RRG insureds vulnerable to misgovernance. In particular, how RRGs are capitalized is central to concerns of experienced regulators about the chartering of entrepreneurial RRGs because a few insureds or service providers, such as management companies, that provide the initial capital also may retain control over the RRG to benefit their personal interests. Further, RRGs, like mutual fund companies, depend on management companies to manage their affairs, but RRGs lack the federal protections Congress and SEC have afforded mutual fund companies. As evidenced by the circumstances surrounding many RRG failures, the interests of management companies inherently may conflict with the fundamental interests of RRGs--that is, obtaining stable and affordable insurance. Moreover, these management companies may have the means to promote their own interests if they exercise effective control over an RRG's board of directors. While RRGs may need to hire a management company to handle their day- to-day operations, principles drawn from legislation such as the Investment Company Act of 1940 would strongly suggest that an RRG's board of directors would have a substantial number of independent directors to control policy decisions. In addition, these standards would strongly suggest that RRGs retain certain rights when negotiating the terms of a management contract. Yet, LRRA has no provisions that establish the insureds' authority over management. Without these protections, RRG insureds and their third-party claimants are uniquely vulnerable to abuse because they are not afforded the oversight of a multistate regulatory environment or the benefits of guaranty fund coverage. Nevertheless, we do not believe that RRGs should be afforded the protection of guaranty funds. Providing such coverage could further reduce any incentives insureds might have to participate in the governance of their RRG and at the same time allow them access to funds supplied by insurance companies that do not benefit from the regulatory preemption. On the other hand, RRG insureds have a right to be adequately informed about the risks they could incur before they purchase an insurance policy. Further, consumers who purchase extended service contracts (which take on the appearance of insurance) from RRG insureds likewise have a right to be informed about these risks. The numerous comments that regulators received from consumers affected by RRG failures illustrate how profoundly uninformed the consumers were. Finally, while opportunities exist to enhance the safeguards in LRRA, we note again the affirmation provided by most regulators responding to our survey--that RRGs have increased the availability and affordability of insurance. That these assertions often came from regulators who also had concerns about the adequacy of LRRA's regulatory safeguards underscores the successful track record of RRGs as a self-insurance mechanism for niche groups. However, as the RRG