This is the accessible text file for GAO report number GAO-03-738T entitled 'Insurance Regulation: Preliminary Views on States' Oversight of Insurers' Market Behavior' which was released on May 06, 2003. This text file was formatted by the U.S. General Accounting 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. Testimony: Before the Subcommittee on Oversight and Investigations, Committee on Financial Services, House of Representatives: United States General Accounting Office: GAO: For Release on Delivery Expected at 3:00 p.m. EDT: Tuesday, May 6, 2003: Insurance Regulation: Preliminary Views on States' Oversight of Insurers' Market Behavior: Statement of Richard J. Hillman Director, Financial Markets and Community Investment: GAO-03-738T: Madam Chairwoman and Members of the Subcommittee: I appreciate the opportunity to provide you with preliminary observations from our work on state insurance regulators' oversight of market activities in the insurance industry. As you know, Chairman Oxley requested that we review the market conduct activities of state insurance regulators. We are nearing completion of this work, and we plan to issue a report on this subject in the near future. As you requested, this testimony provides information on two important tools state insurance regulators use to oversee the market activities of insurance companies--market analysis and market conduct examinations. Market analysis is generally done in the state insurance departments. It consists of gathering and integrating information about insurance companies' operations in order to monitor market behavior and identify potential problems at an early stage. Market conduct examinations, which are generally done on site, are a review of an insurer's marketplace practices. The examination is an opportunity to verify data provided to the department by the insurer and to confirm that companies' internal controls and operational processes result in compliance with state laws and regulations. My focus today is on (1) the states' use of market analysis and examinations in market regulation, and (2) the effectiveness of the National Association of Insurance Commissioners' (NAIC) efforts to improve these oversight tools and encourage the states to use them.[Footnote 1] To address these objectives, we collected data and interviewed officials from nine state insurance departments--Arkansas, California, Indiana, Maryland, Michigan, Missouri, New Mexico, Ohio, and Oregon, and at NAIC's Kansas City Headquarters. We also reviewed NAIC's past and current efforts to improve the market regulation program. We collected and analyzed data from NAIC on all states, including the number of licensed companies in each state, the number and types of examinations conducted, and the resources allocated to these activities. We also asked 40 companies, 20 each from among the largest 300 property and casualty firms (based on direct written premiums) and the largest 300 life companies (based on asset size) about their market conduct examination experiences between 1999 and 2001. Most of these companies are national companies, selling in most or all of the states. However, because our sample was not statistically valid, our results cannot be projected to all insurers. In summary, we found that while all states do some level of market analysis, few states have established formal market analysis programs to maintain a systematic and rigorous overview of companies' market behavior and to more effectively identify problem companies for more detailed review. The way state insurance regulators approach and perform market conduct examinations also varied widely across the states. While NAIC has developed a handbook for market conduct examiners, states are not required to use it, and we found that it is not consistently applied across states. Moreover, the handbook is not intended to provide guidance for some important aspects of market conduct examinations--for example, how often examinations should be performed or what criteria states should use to select companies to examine. We also found that the number of market conduct examiners differed widely among states and that there were no generally accepted standards for training and certifying examiners. These differences make it difficult for states to depend on other states' oversight of market activities. Most of the states that we visited told us that they felt responsible for regulating the behavior of all companies that sold insurance in their state. With anywhere from 900 to 2,000 companies operating within each state, the pool of companies is simply too large for any one insurance department to handle. Attempts to do so are neither efficient nor effective. Moreover, since many states do not coordinate their examinations with other states, some large multistate insurance companies reported being examined by multiple states, while other companies were examined infrequently or never. We also found that since the mid 1970s, NAIC has taken a variety of steps to improve the consistency and quality of market conduct examinations. However, despite the NAIC's long-standing efforts and some limited successes, progress toward a more effective process has been slow. Recently, NAIC has increased the emphasis it places on market analysis and market conduct examinations as regulatory tools that could improve states' ability to oversee market conduct. With more consistent implementation of routine market analysis, states should be better able to use the resources they already have available to target companies requiring immediate attention. Also, by consistently applying common standards for market conduct examinations, states should be able to rely on regulators in other states for assessments of an insurance company's operations. These improvements should in turn increase the efficiency of the examination process and improve consumer protection by reducing existing overlaps and gaps in regulatory oversight. However, if NAIC cannot convince the various states to adopt and implement common standards for market analysis and examinations, current efforts to strengthen these consumer protection tools are unlikely to result in any fundamental improvement. While we focus on the states' use of market analysis and market conduct examinations, market regulation includes several other important regulatory tools, including complaint handling and investigation, policy rate and form review, agent and company licensing, and consumer education. Most states have functioning programs addressing each of these four regulatory areas. Ideally, all regulatory tools, including market analysis and market conduct examinations, should work together in an integrated and interrelated way. Lack of General Agreement on Standards for Market Analysis and Market Conduct Examinations Results in Wide Variations Among States: In the absence of generally accepted standards, individual states decide how they will do market analysis and perform market conduct examinations. While all states do market analysis in some form, few have established formal programs that look at companies in a consistent and routine manner. States also have no generally agreed upon standards for how many examinations to perform, which companies to examine and how often, and what the scope of the examination should be. As a result of the lack of common standards for market analysis and the lack of consistency in the application of the guidelines for examinations, states find it difficult to depend on other states' oversight of companies' market behavior. Few States Do Systematic and Routine Market Analysis: NAIC and some states have a growing awareness that better market analysis can be a significant tool for monitoring the marketplace behavior of insurance companies and deciding which insurers to examine. All states perform some type of market analysis. In many states, however, it consists largely of monitoring complaints and complaint trends; and reacting to significant issues that arise. Three states that we visited--Missouri, Ohio, and Oregon--have established a proactive market analysis program. These programs for market analysis have established processes for monitoring company behavior to identify trends, companies that vary from the norm (outliers), and potential market conduct problems. In general, an established program would have dedicated staff and protocols for gathering data and conducting analysis at the department offices. Each of the three states with an analysis process that we visited approached market analysis in a different way. Ohio's program consisted of special data calls to obtain extensive information from selected company files, and using computerized audit tools to analyze specific aspects of companies' operations relative to norms identified by peer analysis and to state law. For example, Ohio did 184 "desk audits" in 2001 using data requested from companies doing business in the state.[Footnote 2] Missouri relied on routinely collecting market data from all licensed companies. Missouri has developed a market data report that companies submit as a supplement to their annual financial reports. This data is then used to evaluate market trends and conditions, as well as to identify individual companies that were outliers. Oregon's newly established program involved maintaining files on companies in which all available data was collected to facilitate a broad and ongoing review of company behavior. Both Ohio and Oregon told us that their market analysis programs were still in an experimental stage of development. When properly done, market analysis can allow states to focus attention on the high-risk companies rather than selecting companies for examination based primarily on criteria such as market share, which does not directly correlate to market behavior problems. Missouri officials added that market analysis is not a substitute for market conduct examinations but should interact and be integrated with the examination process. We Found Variations in the Way States We Visited Performed Examinations: Each state has between 900 and 2,000 licensed insurance companies. Because in general states do not currently depend upon other states' regulation of companies' market behavior, most states feel a responsibility for overseeing all the companies selling in their state.[Footnote 3] The impossibility of examining so many companies requires regulators to identify and prioritize which companies they will examine. The states we visited used a variety of factors to choose companies for a market conduct examination. The most commonly used factors for choosing from among the companies deemed eligible for a market conduct examination were complaints, market share, and time since the last examination. Some states chose to do market conduct exams for only a subset of licensed companies, even though other companies could comprise a majority of the insurers selling in the state.[Footnote 4] For example, of the states we visited, Arkansas focused primarily on domestic companies--that is, on companies chartered in their state. In Arkansas, 245 of 1,668 licensed companies in 2001 were domestic. As a consequence, 1,423 non-domestic companies, or 85 percent of all the companies licensed in Arkansas in 2001, were not examined in Arkansas in spite of the fact that they may or may not have been examined by some other state. All the states we visited limited the scope of their examinations to customers from within their particular state. That is, examiners looked only at files of state residents. Moreover, most states further limited the scope of their examinations by focusing on only one or a few of a company's area of operations. While some states still do comprehensive market conduct examinations, the trend is to conduct targeted examinations of limited scope and in a specific area of concern. State officials we interviewed indicated that targeted examinations are being used more often because these examinations do not take as long as comprehensive examinations, allowing states to conduct more. Of the 9 states we visited, Arkansas, Missouri, and New Mexico continued to conduct some comprehensive examinations as well as targeted examinations. Arkansas officials told us that they believed comprehensive examinations were important because such examinations provided the greatest assurance that companies were complying with insurance laws and regulations. According to NAIC, 49 states and the District of Columbia reported performing some market conduct activities in 2001. Of these, 15 completed only targeted examinations, 4 did only comprehensive examinations, and 22 completed some of both types of examination. The remaining nine did not complete any market conduct examinations in 2001. The requirements for and level of training for examiners also varied widely among the states. Each of the states we visited provided some type of training for their examiners. However, there are no generally accepted standards for what constitutes adequate training for a market conduct examiner across the states. Several levels of certifications for market conduct examiners are available, but only 2 of the states we visited, Oregon and New Mexico, required their examiners to certify or become certified in a specified period. States Vary in the Emphasis Given to Market Conduct Examinations: As can be seen in table 1, there is considerable variation in the number of examinations completed in 2001 by the states we visited. Variation in the number of examinations consistent with the size of the insurance market would be expected. However, as shown in the table, the number of examinations completed bore little relationship to the size of the insurance market in each state. This comparison should not necessarily be taken as an indicator of the relative regulatory performance of the nine states we visited, because during another year the ranking of the states could be different. However, together with the variations in how states select companies for examinations and how they do them, this added variability helps further explain why the states may be reluctant to depend on other states to examine companies selling insurance to their citizens. Table 1: Market Conduct Examinations Completed in 2001 Relative to Various Measures of the Size of the Insurance Market in Each State: State: California; Market conduct examinations completed in 2001: 80; Total premium volume in 2001[1] ($ in millions): 95,368; Total number of licensed agents and brokers in 2001: 220,506; Estimated state population in 2001: (In thousands): 34,600. State: Ohio; Market conduct examinations completed in 2001: 5; Total premium volume in 2001[1] ($ in millions): 39,663; Total number of licensed agents and brokers in 2001: 154,100; Estimated state population in 2001: (In thousands): 11,390. State: Michigan; Market conduct examinations completed in 2001: 0[2]; Total premium volume in 2001[1] ($ in millions): 37,840; Total number of licensed agents and brokers in 2001: 86,739; Estimated state population in 2001: (In thousands): 10,006. State: Missouri; Market conduct examinations completed in 2001: 29; Total premium volume in 2001[1] ($ in millions): 20,656; Total number of licensed agents and brokers in 2001: 91,695; Estimated state population in 2001: (In thousands): 5,637. State: Maryland; Market conduct examinations completed in 2001: 26; Total premium volume in 2001[1] ($ in millions): 20,517; Total number of licensed agents and brokers in 2001: 72,039; Estimated state population in 2001: (In thousands): 5,386. State: Indiana; Market conduct examinations completed in 2001: 4[3]; Total premium volume in 2001[1] ($ in millions): 19,208; Total number of licensed agents and brokers in 2001: 83,277; Estimated state population in 2001: (In thousands): 6,127. State: Oregon; Market conduct examinations completed in 2001: 15; Total premium volume in 2001[1] ($ in millions): 10,750; Total number of licensed agents and brokers in 2001: 46,573; Estimated state population in 2001: (In thousands): 3,473. State: Arkansas; Market conduct examinations completed in 2001: 19[4]; Total premium volume in 2001[1] ($ in millions): 6,919; Total number of licensed agents and brokers in 2001: 41,268; Estimated state population in 2001: (In thousands): 2,695. State: New Mexico; Market conduct examinations completed in 2001: 8; Total premium volume in 2001[1] ($ in millions): 6,045; Total number of licensed agents and brokers in 2001: 28,910; Estimated state population in 2001: (In thousands): 1,831. Sources: State Insurance Departments. NAIC's 2001 Insurance Department Resources Report. U.S. Census. Note: Does not include follow-up exams or desk audits. [1] Total premium volume for life, health, and property/casualty insurance. [2] Michigan did a limited review of market conduct issues as part of its 37 financial examinations. [3] Three of these were multistate examinations. [4] Arkansas also examined 65 funeral homes that sold prepaid funeral insurance. [End of table] In addition to the variation in examinations completed, some states have dedicated very few resources to market analysis and market conduct examinations. NAIC's 2001 Insurance Department Resources Report does not even break out department staff assigned to market analysis, although financial analysts are separately identified. In addition, 14 states, or 27 percent, did not report having any market conduct examiners on staff, although 4 of the 14 did report using full-time contract examiners. Ten states, or nearly 20 percent of all states, did not report having any market conduct examiners at all. Interstate Coordination and Communication Is Inconsistent and Infrequent: Our review of the nine states indicated that the practice of sharing examination information with other states, when it occurred, varied substantially from state to state. Some states coordinate their examination plans with other states or review other states' examination reports prior to going into a company, while other states do not. Even in states where some coordination occurs, other states' examination results do not generally affect examination plans. Oregon officials told us that there is a need for more interstate collaboration and reliance on examination results from other states. More coordination of market conduct examination plans, efforts, and results could improve regulation and, at the same time, reduce the regulatory burden on companies. Many insurance companies, particularly the largest ones, report that they undergo frequent, sometimes simultaneous, market conduct examinations. We asked 40 of the largest national insurance companies to provide information about their market conduct examination experience for the years 1999 to 2001. Of the 25 companies that responded, 19 were examined a total of 130 times by multiple insurance regulators during the 3-year period. Six were examined once or twice during the period, and just over half the responding companies were examined between one and five times. However, three companies were each examined 17 or more times during the 3 years, with one company receiving 20 examinations--an average of seven nearly every year.[Footnote 5] These results appear to be consistent with concerns expressed by the insurance industry about excessively frequent and possibly duplicative market conduct examinations. One of the most common complaints from the 25 insurers that responded to our questionnaire was that states did not coordinate their examinations with other states. Some companies reported that, on occasion, multiple states had conducted on-site examinations at the same time. The companies told us that such examinations create difficulties for them and limited the resources they had available to assist the examiners. For example, one insurer wrote, "It takes an insurer a tremendous amount of effort to prepare for and deal with individual state insurance department's exams (every one is different, plus states generally do not accept others exams in place of another similar exam being done). The duplication of effort is wasteful by the states.": In contrast, six companies, or nearly one-quarter of those responding, had not been examined by any state during the period. Of these six companies, two were last examined in 1997 and the other four did not report having any market conduct examinations. These companies--like others that reported--are large multi-state insurance companies. Since in many states a primary criterion for selecting a company for examination is market share, these responses suggest that the proportion of medium-size and small insurers that rarely, if ever, receive a market conduct examination may be much higher. Groups of states, as well as NAIC, have taken actions to improve the coordination and efficiency of the market conduct examination process. One effort involves improving the sharing of examination information by providing notice of upcoming examinations and sharing results through NAIC's Examination Tracking System. However, the Examination Tracking System is incomplete and often ignored by the state regulators, in part, because it has been inconvenient and difficult to use for scheduling and reporting the results of market conduct examinations. As a result, states are not fully utilizing the system. NAIC's survey of states' use of the Examination Tracking System concluded that no more than 66 percent of the states, or 36 states, consistently reported their market conduct or combined market conduct/financial examination schedules to NAIC. Moreover, only 31 percent of the states reported back to NAIC when the examination had been completed. Another avenue of coordination being pursued by NAIC and some states is joint, or collaborative, examinations. Based on our review of nine states and of NAIC information, some states do conduct collaborative examinations. For example, Ohio officials told us that they had started to conduct collaborative examinations with Illinois, Nebraska, and Oregon. Indiana officials indicated that they had recently completed an examination of a large insurer jointly with another state. Such efforts, however, have not been consistent among states, nor is there a policy or standard procedure about when or how such examinations should occur. Furthermore, while collaborative examinations could reduce the total number of duplicative exams and may result in somewhat more efficient use of regulatory resources, they still require that each state send examiners into the company. In effect, collaborative examinations are a way for multiple states to do a market conduct examination of a company at the same time. Such an examination may be to the benefit of the company. However, if each state's examiners still ask for samples of files for only their own state's insurance consumers, the benefit may be reduced. NAIC Has Identified Market Analysis and Examinations as Areas Needing Significant Improvement: The NAIC identified the need for uniformity in market conduct regulation as early as the 1970s. Since then NAIC has launched a number of market conduct efforts intended to identify and address the issues and concerns caused by the lack of uniformity in states' market conduct examination processes, and more recently in the market analysis area. Although progress has been slow in establishing more uniformity in market conduct regulation, NAIC has had some successes. One of the earliest was the development of the market conduct examination handbook containing guidance on conducting and reporting examination results. In general, most states use the handbook as an examination guide, but they can still choose not to follow the handbook in an examination or to modify it. For example, although the handbook lays out the steps for conducting an exam, such as notice of an exam, use of sampling techniques, and preparation of an examination report, each state can go about those steps differently. Moreover, the handbook in not intended to cover some aspects of examinations, including examination frequency and company selection criteria. One challenge to establishing voluntary uniform national standards for examinations and examination processes is that states are free to adopt the NAIC's model laws, regulations, and procedures; to modify them to meet their perceived needs and conditions; or even to ignore them entirely. Once NAIC as an organization agrees on recommendations that would create more uniform regulatory statutes, two additional challenges to uniformity remain. First, when proposed changes affect state law, state legislatures must approve the recommendations without significant changes. Second, each state insurance department must successfully implement the recommendations. These challenges to establishing voluntary uniform national standards for examinations can clearly be seen in the number of states adopting the model laws and regulation that NAIC identified in 1995 as the essential elements for a market conduct examination program. By 2003, only nine models had been adopted by more than half the states, while two models had been adopted by five or fewer states. Achieving uniformity in market regulation will be a difficult process for NAIC and the states. However, a similar problem that existed in solvency regulation over a decade ago was solved by creating the Financial Regulation Standards and Accreditation Program. The program's overall goal was to achieve a consistent, state-based system of solvency regulation throughout the country. The program was designed to make monitoring and regulating the solvency of multistate insurance companies more consistent by ensuring that states adopt and adhere to agreed-upon standards, which establish the basic recommended practices for an effective regulatory department. To be accredited, states had to show that they had adopted specific solvency laws and regulations that protected insurance consumers, established defined financial analysis and examination processes, and used appropriate organizational and personnel practices. While the quality of regulation is still not consistent, the Accreditation Program has improved financial regulation across the states. As a result, states are now willing, in most cases, to depend on the solvency regulation of other states. While the process used by state insurance regulators to oversee solvency could provide a model for oversight of market conduct as well, there are structural differences in market regulation that will undoubtedly affect the ultimate design of an improved market conduct oversight system. These differences will have to be addressed by NAIC and the states in order to move forward. First, market conduct oversight involves many different activities and operations of insurance companies. This fact has broad implications for regulatory consistency and mutual dependence, including requirements for the necessary training of market conduct examiners and analysts. Second, regulators told us that life insurers tend to use a company- wide business plan and organizational structure. That is, a life company's operations tend to be relatively consistent across the entire company. Property-casualty insurers, on the other hand, tend to use a regional business model and organizational structure. As a result, a property-casualty insurer's operations could differ, perhaps substantially, from region to region. Clearly, the life insurer model is more directly amenable to domiciliary-state oversight than the property-casualty model. However, any regional or state-by-state variances in a company's operations and procedures would reduce the effectiveness of domiciliary-state oversight. Some aspects of market conduct oversight will always be state (or region) specific because of the differences between life and property-casualty insurers, but also because there will always be differences between some of the specific laws and requirements of individual states. As a result, even when greater uniformity of regulatory oversight is achieved, it is likely that states will always have to devote some attention to the activities of insurers not domiciled in their state. Nevertheless, if a state insurance department knew that the domiciliary state was doing consistent market oversight on the company with agreed-upon processes, appropriate scope, and well-trained examiners and analysts, the level of attention needed, even for a property-casualty company, could be substantially lessened. Finally, even to the extent that properly designed and competently performed market conduct oversight can effectively monitor and regulate insurance company practices, it will extend to the sales practices of insurance agents only to the extent that the company takes responsibility for and exercises control of the behavior of the agents that sell its products. Preliminary Observations: In the current environment of market regulation, most insurance regulators believe they need to oversee the market behavior of all companies selling insurance in their state because they cannot depend on the oversight of the other states. State regulators think this way in part because important elements of market regulation are characterized by a lack of even the most fundamental consistency. Formal and rigorous market analysis is in its infancy among state regulators, and whether, when, and how states do market conduct examinations vary widely. As a result, state regulators are now using the resources that they have in the area of market analysis and examinations inefficiently. Regulators from different states examine some insurers often, while other insurers are examined infrequently or not at all. More importantly, because market analysis is weak, regulators may not be finding and focusing on the companies that most need to have an examination. We support the goal of increasing the effectiveness of market conduct regulation through the development and implementation of consistent, nationwide standards for market analysis and market conduct examinations across the states in order to better protect insurance consumers. The emphasis placed on these issues by NAIC has increased substantially over the last 3 years. We believe that NAIC has taken a first step in the right direction. Much work, however, remains, as NAIC and the states have not yet identified or reached agreement on appropriate laws, regulations, processes, and resource requirements that will support the goal of an effective, uniform market oversight program. Such a program, consisting of strong market analysis and effective market conduct examinations, will facilitate the development of an atmosphere of increasing trust among the states. However, at present it remains uncertain whether the NAIC and the states can agree on and implement a program that will accomplish this goal. Madam Chairwoman, this concludes my statement. I would be pleased to answer any questions you or other members of the subcommittee may have at this time. FOOTNOTES [1] The National Association of Insurance Commissioners is comprised of the insurance commissioners of the 50 states plus the District of Columbia, Puerto Rico, and the United States Territories. The commissioners promulgate model (recommended) laws and regulations for consideration by the states and provide support services for the state insurance departments. NAIC meetings also provide a venue for discussion of issues that are of interest to all. [2] A desk audit involves a review of company files at the department without physically going to the company. [3] Not all licensed companies in a state are actively selling insurance. For example, some companies with existing business may be going out of business although still servicing existing customers (in run-off) or in liquidation. These companies may still have some active policies in the state, but are not selling any new business. [4] States generally have the authority to do a market conduct examination on any company that sells insurance in the state. [5] We did not verify the companies' responses with state regulators. Moreover, we have no basis for evaluating the states' reasons for selecting specific companies to examine.