Why Conflicts of Interest Cannot Be Eliminated

The quest for synergies, the benefit from combining multiple activities within a financial institution, is one of the driving forces behind the development of financial institutions. However, synergies and conflicts of interest are a package deal. You cannot eliminate one without eliminating the other. The information approach to financial markets reveals the nature of this problem, which is at the core of the function of any financial institution.

Reliable information is required for markets to perform their economic function of channeling funds to businesses and households with the greatest productive opportunities. While this is essential, the asymmetric relationship between lenders and borrowers creates a major impediment. The party on one side of any financial contract has less accurate information than the party on the other side, creating the possibility that the party with more information may be able to take advantage of the situation.

In the simplest example, business managers have better information about their firm’s returns and risk than do the purchasers of its securities or a bank’s loan officers. Such an information asymmetry can lead to problems of adverse selection and moral hazard. Adverse selection occurs before a transaction is consummated; for example, managers who want to divert funds for their own pay or perquisites or further engage in risky ventures will be the most likely to seek external funds. If a bank or an investor grants them funds, their selection will make it more likely that their overall set of investments will perform below what was expected. Furthermore, investors who are unable to screen out bad from good investments may decide not to invest. Moral hazard occurs after the financial transaction has taken place when managers may, in the absence of an incentive to do otherwise, misallocate funds because bankers or investors have not been able to adequately monitor their behavior. Moral hazard may also discourage investors who may reduce lending.

While individual investors may conduct their own screening and monitoring of managers, they will face substantial problems. Investors may invest in information and become well informed, but they will become subject to free-rider behavior by others who observe their decisions and freely benefit from their investment, thus discouraging further investment. Also, an investor may not be an efficient collector of information if there are economies of scale and scope. Consequently, investors may turn to delegate the task of screening and monitoring to specialist financial institutions. Thus, analysts in investment banks can advise numerous clients on what securities to buy; auditors may opine on accuracy of the accounts presented by management, and rating agencies will rank the relative riskiness of securities for thousands of investors. These are fee-based services that permit investors to decide where to place their funds, but they may delegate investment as well as information collection to intermediaries that pool funds from the public in a diversified portfolio designed with private information they have collected. Yet, the problem of monitoring is inescapable, since investors using the services of financial institutions need to monitor them to ensure that they are providing accurate information and are appropriately investing their funds. Hence to be successful, financial institutions must convince or signal to their customers that they are adequately monitoring their employees.²

Economic theory typically treats financial institutions as though each type of institution solved one kind of informational asymmetry; however, institutions that have learned how to manage one information asymmetry often possess skills that may be used to handle others. Banks have long-term customer relationships, which they use to obtain information about firms’ resources, cash flows, and other characteristics and reveal more confidential information. Financial institutions gain cost advantages because they can exploit cross-sectional information across customers, becoming low-cost producers of information for complementary financial services.

Such synergies or economies of scope provide producers and customers substantial benefits, but they also create potential costs in the form of conflict of interests. For the financial industry, conflicts of interest may be defined as arising when a financial service provider, or an agent within such a service provider, has multiple interests that create incentives to act in such a way as to misuse or conceal information needed for the effective functioning of financial markets.

The combination of financial services in one intermediary creates conflicts of interest that may be exploited. If the firm or individuals within the firm can exploit conflicts of interest, they do so because they can benefit from the information asymmetries vis-à-vis customers. This behavior will obstruct the efficient allocation of funds to their most productive uses. But behavior that exploits a conflict of interest will, once recognized, reduce the reputation of an institution. Consequently, the existence of a conflict does not imply it will be exploited if the institution places a high value on its reputation. The implication is that public policy remedies may not be required to control the exploitation of conflicts. When they are necessary, government intervention to reduce conflicts needs to be balanced against any reduction in the economies of scope.

Conflicts of Interest in Investment Banking

Investment banks gain economies of scale and scope, by providing financial services that tackle informational asymmetries in the primary and secondary capital markets. They combine a large variety of activities that include floating new and seasoned securities, advising on mergers and acquisitions, serving as brokers or dealers, providing research, and making markets.

In the scandals that erupted after the collapse of stock markets in 2000, most attention has been focused on the conflict of interest between providing research to investors and underwriting initial public offerings. Research analysts identify and monitor companies for clients, helping to reduce the problems of adverse selection and moral hazard. The investment bank that serves as the lead underwriter for a syndicate acts as the delegated monitor for individual investors and the rest of the syndicate, by collecting the vital information on the firm issuing a new security, thereby gaining an information advantage. Its research analysts should be able to offer better recommendations to buy and sell stocks and better forecasts of performance. In addition, the lead underwriter usually is the dominant market maker. Information synergies from underwriting, research, and market making provide a rationale for combining these distinct financial services.³ An investment bank’s success at combining these activities will contribute to its reputation and profitability.

When an investment bank serves two clients, a conflict of interest may arise. Issuers will benefit from overly optimistic research while investors will want unbiased research. If the incentives at the investment bank for the provision of these two activities are not appropriately aligned, employees on one side of the firm will be tempted to distort information to the advantage of their clients and the profit of their department. To manage this conflict of interest, compensation must be set appropriately for research analysts. Setting compensation to produce the correct incentives is difficult because the information that analysts generate for investors is not a purely private good and cannot be limited to the firm’s clients. Hence, it is not surprising that brokerages find pricing their services difficult and do not often charge clients for research.⁴ The evaluation of analysts’ performance is also a problem because they provide multiple services. During the stock market boom, some stock prices moved far away from fundamentals, increasing the reputations of analysts who correctly picked these stocks at the expense of those who focused on accurate forecasts of earnings.

While compensation rules vary among firms, an analyst’s external reputation is important for compensation. Their external reputation is influenced by the Institutional Investor’s polls⁵ that rank analysts on buy/sell recommendations, earnings forecasts, reports, and overall service. Analysts’ reputation is important not just for brokerage customers; it is also used as a marketing tool for investment banks in the initial public offering (IPO) market. A survey of chief executive officers and chief financial officers whose firms issued IPOs in the 1990s reported that most considered the reputation of a research department in the choice of a lead underwriter.⁶ Analysts’ support is often considered part of an implicit understanding between underwriter and issuer. Lastly, over the longer run, analysts’ focus on the development of industries gives them a highly specialized knowledge that may assist underwriters in screening new companies.

When analysts are compensated by underwriting departments, there will be a strong conflict of interest potential. It will be most acute if the IPO market is highly profitable relative to brokerage because it will induce analysts to bias their reports in favor of issuers. In a soaring market, the short-term payoff for an analyst may outweigh the benefits of investing in a long-term reputation, being tempted to promote hot issues to expend the firm’s reputational rents.

The boom market of the 1990s created conditions that induced analysts to exploit this conflict of interest. IPOs were primarily technology stocks, and the quality of companies going to market changed dramatically.⁷ Most IPO companies had several years of established positive earnings in the 1960s to the 1980s, but the number of issuers with negative earnings soared to nearly 80 percent in 1999–2000. They may have had long-term potential, yet stock prices were moving away from their conventional relationships with fundamentals. Investors seem to have ignored these standard signals. They seem to have given more attention to target prices and other information, improving most optimistic analysts’ reputations.

When IPO markets boomed, media attention focused on analysts’ pronouncements. There appears to have been pressure on them to join in the optimistic promotion of stocks. The story of Henry Blodget is one of many striking examples. In 1998, most analysts held that Amazon.com was overvalued at US$240; Jonathan Cole of Merrill Lynch believed US$50 to be a reasonable price. Henry Blodget at Oppenheimer and Co. set a price target of US$400. When Amazon.com surpassed it, he was hailed as a guru; Cole departed and Merrill Lynch hired Blodget. Rewards for optimism spread through the industry. Studies of the movement of analysts from job to job to higher or to lower status brokerage houses have found that while analysts were rewarded for the relative accuracy of earnings forecasts, those who were more optimistic than the consensus were also more likely to experience favorable job separations.⁸ Promotions for optimism became more important by the late 1990s.

While analysts influence investors, and their buy/sell recommendations are treated as “new” information that moves the market,⁹ it is not clear that the analysts intended to deceive customers. While analysts made far more buy than sell recommendations in the boom, the ratio changed little after the collapse of the market. Furthermore, many research-only houses where there was no potential conflict also had far more buy than sell recommendations.¹⁰

What evidence is there that analysts tried to exploit conflicts of interest? Some answers are found in the differential behavior of analysts at underwriting and non-underwriting banks. Owing to their key position, lead underwriters’ analysts’ reports should carry extra weight and their predictions should be unbiased and more accurate than those of other analysts. Consequently, the market should react more to their announcements than to reports of other analysts. But, if they are biased, underwriter analysts will issue relatively more positive recommendations for firms that trade poorly in the IPO aftermarket. If the market recognizes the potential conflict of interest, then investors should discount underwriter analysts’ recommendations relative to non-underwriter analysts.

A 1999 study examined the “buy” recommendations of lead underwriters and other analysts for IPOs after the Securities and Exchange Commission’s (SEC’s) 25-day post-IPO “quiet period” from 1990 to 1992.¹¹ The findings showed that in the month after the quiet period, lead underwriters’ analysts made 50 percent more buy recommendations than other firms’ analysts for the same securities, suggesting some conflict of interest. But stock prices of firms recommended by lead underwriting banks declined during the quiet period, while other banks’ picks rose. The market recognized the difference in the quality of information. The excess return at the recommendation date is 2.7 percent for underwriters’ analysts and 4.4 percent for other analysts. Considering a two-year holding period from the IPO date, the performance of other analysts’ recommended issues was 50 percent better than the performance of underwriters’ recommendations. In addition, the same investment banks made better recommendations on IPOs when they were not the lead underwriter, implying that it is not a difference in analyst’s ability but an underwriter bias.¹² The finding that the market discounts analyst optimism is also supported by survey data. A study found that 86 percent of the professional money managers and buy-side analysts said that they discounted the recommendations and reports of analysts when there is an investment banking relationship between the bank and the company analyzed.¹³

The outbreak of the post-crash scandals did not lead to an extended inquiry and policy debate. Instead, remedies were abruptly imposed by the “Global Settlement” reached on December 20, 2002 by the SEC, the New York Attorney General, NASD (formerly the National Association of Securities Dealers), the North American Securities Administrators Association, the New York Stock Exchange, and state regulators with the 10 largest investment banks.¹⁴ The five key terms of the agreement were (1) firms are required to sever the links between research and investment banking, including analyst compensation for equity research and analysts accompanying investment banking personnel on road shows; (2) the practice of spinning of IPOs is banned;¹⁵ (3) each firm is required to make public its analyst recommendations, including its ratings and price target forecasts; (4) for a five-year period, brokerage firms will be required to contract with no fewer than three independent research firms to provide research to their customers. Regulators will choose an independent consultant “monitor” for each firm to procure independent research to ensure that investors get objective investment advice; and (5) total fines of US$1.4 billion will be levied that are partly for retrospective relief, independent research, and investor education.

To a considerable degree, this settlement attempts to socialize research. Taxing firms to fund independent research will create an incentive for them to reduce their own internal analysis and lower its quality. Separation between analysts and underwriters with some type of firewalls is appropriate, but complete separation is mistaken. The design of the Glass-Steagall Act of 1933 to separate commercial and investment banking ultimately failed, suggesting that a similar remedy here is misguided. The market already discounted lead underwriters analysts’ recommendations, making firms subject to some degree of market discipline. Firms will now have to have a separate staff for underwriting to perform the analysis, raising costs and losing some economies of scope. Market discipline was not permitted to take its course, and discipline by loss of reputation and litigation where conflicts were exploited by individual firms was preempted. The banning of spinning, however, is probably appropriate as it will ensure that insiders do not take advantage of outsider investors, exploiting the lack of information about how shares are distributed.

The global settlement relies heavily on separation and the socialization of research as remedies. An alternative approach would be to increase the ability of the market to discipline firms by increasing disclosure to investors of underwriting relationships, complementing this with supervisory oversight.

Conflicts of Interest in Auditing

To judge the usefulness of accounting statements, investors depend on auditors to overcome the informational asymmetry between shareholders and managers. Managers provide a set of accounts to demonstrate how they have used the firm’s resources, but this act does not eliminate the inherent agency problem. To ensure the reliability of these accounts, auditors can provide an independent assessment of the accounts prepared by managers and attest to the quality of information. The information collected by an audit firm has many potential uses beyond an audit report. Auditors have branched into new consulting services, notably tax advice, accounting, management information systems, and strategic advice, commonly referred to as management advisory services (MAS). These complementary services have economies of scope, but they create two potential sources of conflict of interest: (1) auditors may bias their opinions to limit any loss of fees in the “other” services, and (2) auditors may be called upon to evaluate tax and financial structures that were designed by their non-audit counterparts. Both conflicts may lead to biased audits, with the result that less information is available in financial markets. As in investment banking, individuals or firms can benefit from the exploitation of these conflicts of interest because of the reputation that auditing firms have built up with investors.

An audit opinion is expressed in a report attesting to whether the financial statements provide what is known as a “fair presentation” or “true and fair view” of the performance and position of a firm. To form judgments, investors require standards for comparisons, which were originally developed in the nineteenth century by professional accounting societies that evolved into the American Institute of Certified Public Accountants (AICPA). Until the New Deal, there was no federal regulation of accountants and auditors. The collapse of the stock market in 1929 led to the passage of the Securities Act of 1933, which required companies offering shares to the public to submit regular financial statements certified by an independent public or certified accountant. The Securities Exchange Act of 1934 created the Securities and Exchange Commission, which was given jurisdiction over the accounting profession and its rules. The SEC delegated its accounting and auditing rule-making authority to private standard-setting bodies with self-regulation and SEC oversight. In financial accounting, the task of setting standards was delegated to AICPA committees, which evolved into the Financial Accounting Standards Board (FASB). For auditing, the AICPA retained its standard-setting role in the Auditing Standards Board. The rules for auditing govern the conduct of an audit and the nature of the reports, and provide monitoring by peer reviews and the Public Oversight Board. As a result of the auditing scandals, much of this standard-setting and oversight has been taken over by the government. The Sarbanes-Oxley Act of 2002 established the Public Company Accounting Oversight Board (PCAOB). Under the SEC’s oversight, the PCAOB will register accounting firms and establish rules for auditing, conduct inspections and investigations with the power to hold disciplinary proceedings, and impose sanctions. The PCAOB has indicated its intention to take over the rule-making authority for auditing standards, while leaving accounting rules in the hands of the FASB.

Because of the information that they provide, investors depend heavily on auditors. Empirical evidence reveals that favorable opinions issued by audit firms with a strong reputation are valued more than those issued by firms with weaker reputations. One study found that the use of a big-six auditor (as the six largest U.S. accounting firms are called) reduces the rate of return required by investors for a firm and that this effect is almost three times larger for non-investment grade issuers, where information asymmetries are greater.¹⁶ Respect for audit reports is so great that there is acknowledged to be an “expectations gap.” Another study¹⁷ discovered that there is a large difference between the perception of what an audit opinion is intended to convey and what it actually accomplishes.¹⁸ Many users of audits believe that an unqualified audit opinion indicates that the entity is financially sound and that there is no fraud or illegal activity. These expectations are widely held, in spite of the fact that in the United States the audit opinion only indicates that management’s presentation of the financial information is a fair presentation of the position and performance of the company and conforms to generally accepted accounting principles (GAAP).

Audit failures, particularly those appearing to arise out of perceived conflicts of interest, caused a demand for reform. The conflict most frequently discussed in the popular financial press arises from an auditor providing non-audit services. As in the case of research and underwriting, the conflicts of interest arise because services were combined to exploit economies of scope. Auditing firms are natural consultants because they gather and assess a wide array of information that enables them to evaluate management’s accounting decisions. Such expertise has natural information synergies with consulting services. This alliance of activities was furthered as corporate accounting systems became computerized, and auditing firms soon specialized in computerized management information systems. In the early 1980s, auditing firms emerged as powerhouses of the consulting business, with the largest entering the ranks of the top 10 global consulting firms. The leading firm was Andersen Consulting, which was almost solely a systems-oriented consulting firm.¹⁹

While there are many potential conflicts of interest, they will not be exploited unless encouraged by the incentive structure. Most studies find little evidence for widespread exploitation. One such study found that audit clients have incentives to limit non-audit purchases from incumbent auditors.²⁰ Their conjecture is that a perceived reduction in auditor independence reduces audit credibility, adding agency costs for companies as the value of the auditors’ monitoring role is reduced. More recently studies have examined whether auditors’ fees for MAS are associated with abnormal accruals, used as a proxy for earnings management²¹ and biased reporting. Another study found that non-audit fees are positively associated with small earnings surprises and the magnitude of discretionary accruals.²² Using a U.K. data sample, where there is greater disclosure of audit and non-audit fees,²³ a further study saw no significant effect of abnormal accruals on audit fees or non-audit fees. In addition, they discovered that higher fees for non-audit services decreased abnormal accruals and interpreted this finding as evidence that non-audit services improve financial management. Others have also found no evidence that non-audit service fees impair auditor independence.²⁴ Auditors were more likely to issue qualified audits to clients that pay higher audit fees, consistent with a risk-based propensity to audit more; this finding was supported by a study that ascertained that risky clients have higher fees because of extra effort expended by auditors.²⁵

It appears that auditors expend greater effort to address aggressive or risky accounting decisions made by clients. Although conflicts of interest have long been a concern, there appears to be no systematic pattern of exploitation. Yet, the rapid growth of MAS activities, followed by audit failures, have produced demands that auditing be separated from non-auditing services. Congress responded to this perceived problem in Section 201 of the Sarbanes-Oxley Act of 2002, which makes it unlawful for a registered public accounting firm to provide any non-audit service to an issuer contemporaneously with the audit, including bookkeeping, financial information systems design, appraisals, internal audit outsourcing, management functions, and actuary, broker, dealer, investment advisor, investment banker, and legal services, and any other services that the PCAOB determines are impermissible. This drastic measure will eliminate not only potential conflicts of interests but also economies of scope. However, research suggests that conflicts of interest may not have been an overwhelming problem for audit firms. Instead the loss of auditor independence, the use of the partnership form for multiproduct firms, and litigation risk may have been the most important ingredients in the recent audit failures.

Auditing firms have been primarily organized as partnerships; and until the 1980s, the managing partners, governance structure, and profitability were dominated by the audit side of the firms. In the 1990s, MAS activities and revenues grew dramatically and audit profits were under increased competitive pressure and litigation risk.²⁶ The partnership structure meant that non-audit partners had to share their growing revenues with the audit side and incur increased risk. Power struggles erupted, and the battle was most intense at Arthur Andersen. In these struggles, audit partners were pressured to increase revenues, and the dominance of local offices by single clients played an important role. Many of Arthur Andersen’s largest failed clients, Enron, WorldCom, Qwest, and Global Crossing, were the largest companies in their local regions. The manager of a regional or city office would be wary of taking a negative stance on an audit that would risk losing the client. The loss of an Enron or WorldCom account would have been devastating to a local office and its partners even if it were only a small part of firm-wide revenues and profits. Whether this problem was systemic is unclear. For example, one investigation analyzed the influence of large clients on office-level auditor-reporting decisions and found no evidence of it influencing accruals.²⁷ This study, along with another,²⁸ found the evidence consistent with auditors reporting more conservatively for larger clients because these clients pose greater litigation risk and hence more reputational risk.

Concern over preservation of an audit firm’s reputation is the driving factor behind litigation risk. Class action lawsuits, filed on behalf of shareholders, beginning in the 1970s, claimed that declines in share prices were caused by faulty auditing.²⁹ Litigation defeats imposed financial penalties and higher insurance costs, plus a loss of reputation. As a result, audit firms focused attention on reducing litigation risk, assessing the risk of their practices and clients. The national offices of audit firms began to perform risk assessments of clients and practices to manage these costs. Firms adjusted their activities to protect themselves from litigation. Auditors and corporations sought and relied upon an increased codification of auditing and accounting standards. The adherence to these rules facilitated a legal defense of compliance with rules.³⁰ Auditors shifted their focus from opining on whether financial statements fairly present the “true” financial condition and performance of the company to compliance with the detailed GAAP rules. Managers were able to argue now that audit opinions should concentrate on compliance with the rules, shifting attention from performance and obscuring the true economic condition of companies. The focus on GAAP rules thus was another vital part of the debacle at Enron and other companies.

The decline of auditor independence also threatens the effectiveness of audits. In principle, firms have an audit committee of the board of directors that is supposed to monitor auditing to prevent any conflict of interest between the auditors and management. However, audit committees were rarely in complete charge. Executive officers have often become the primary decision makers, selecting the audit firm and negotiating the fees. This conflict of interest can only be remedied by a change in the governance structures.

Although the Sarbanes-Oxley Act addresses some issues, the emphasis on separation by function is not the key and may well lower the effectiveness of auditors. It is unlikely that the proscription of non-auditing services would have prevented the recent audit failures. The market can impose considerable discipline, as Arthur Andersen paid the ultimate price by its demise. To bolster the drive for reputation, some reforms are necessary. As Section 301 of the Sarbanes-Oxley Act recognizes, the corporate governance structure of companies needs to be altered so audit committees, not management, hire and determine the compensation of auditors. A fundamental change is also required to shift auditors away from focusing on the adherence to detailed prescriptive accounting rules. The continued focus on the codification of accounting and auditing standards will not improve the quality of auditors’ reports and may lead to more manipulative innovations to hide companies’ true conditions. Lastly, audit firms themselves need to adjust their internal governance and compensation structures to limit the problem of large-client dominance of local offices and of competition between audit and non-audit services. Firms can devise their own structures, but the PCAOB may help by monitoring and encouraging the development and use of best-practice compensation and performance measurement structures.

How to Resolve Conflicts of Interest

Can the market control conflicts of interest? Market discipline hits firms hard with pecuniary penalties and promotes the development of institutional structures that limit conflicts and signal the firm’s intent to the public. Litigation is an important part of market discipline. It is effective, as exploitation of conflicts is not uniform across the financial industry. While litigation may be the appropriate response to discipline specific firms and individuals as part of an overall market solution, legal liabilities and penalties need to be carefully designed, as demonstrated by the behavior of audit firms seeking to avoid the litigation risk from class action lawsuits.

The market may not be effective if it is unable to obtain sufficient information to punish firms that are exploiting conflicts of interest. To address this failure, there are four classes of interventions: (1) mandatory disclosure for increased transparency, (2) supervisory oversight, (3) separation by function, and (4) socialization of information. However, each of these remedies interferes with the combination of financial services, from which firms gain economies of scope, thereby imposing a potentially high cost on market efficiency. The most potent example of a misplaced remedy is the separation of commercial and investment banking by the Glass-Steagall Act.³¹ The separation imposed a high cost, and only after a long struggle was the Act reversed in 1999. Market discipline that forced institutional changes on banks worked fairly well before 1929. The repeal of Glass-Steagall resulted in a move back to a greater emphasis on disclosure and oversight than was originally recommended by contemporary experts.

Mandatory disclosure to increase transparency is the least intrusive remedy. Disclosure that reveals whether a conflict of interest exists may help investors to judge how much weight to place on the information delivered by each firm. Yet, mandatory disclosure may be insufficient. Financial firms may hide relevant information and disclosure may reveal too much proprietary information. These problems suggest that the more intrusive approach—some supervisory oversight—may be needed. Supervisors can observe proprietary information about conflicts of interest without revealing it to a financial firm’s competitors and can take actions to prevent financial firms from exploiting conflicts of interest.

If the market cannot get sufficient information from disclosure or supervisory oversight is ineffective, one may contemplate the more extreme solution of enforcing the separation of financial institutions by function. Separation by function has the goal of ensuring that “agents” are not placed in the position of responding to multiple “principals” so that conflicts of interest are reduced. Moving from less stringent separation of functions (different in-house departments with firewalls between them) to more stringent separation (different activities in separately capitalized affiliates or prohibition of the combination of activities in any organizational form) lessens conflicts of interest. Sometimes, firms may adopt these solutions independently—as did American universal banks in the 1920s—to signal that they are controlling conflicts. However, stringent separation of functions—as selected by the Glass-Steagall Act—may seriously reduce synergies of information collection, thereby preventing financial firms from taking advantage of economies of scope in information production.

The most radical response to conflicts of interest is to socialize the provision or the funding source of information. The argument for this approach is that information is a public good and so may need to be publicly supplied. Of course, the problem with this approach is that a government agency or publicly funded entity may not have the same strong incentives as private financial institutions to produce high-quality information, thus reducing the flow of essential information to financial markets. While conflicts may not be entirely prevented by mandatory disclosure and supervisory oversight, the case for separation by function or socialization of information is hard to make given the costs imposed on the financial system.

In evaluating remedies, it is important to remember that the many types of agents who provide information to the financial markets have a range of access to information. Analysts have the least access, and rating agencies have more; auditors probably have the most privileged private access followed by government regulators charged with supervisory oversight. This gradient of access to information should reflect the ability of agents to discover the true financial condition and performance of the firms that they observe. Agents’ ability to discover this information will also be determined by their compensation and the other incentives provided to them.

Although these agents provide some overlapping information, one is not a substitute for another. This lack of substitution is not solely because they provide different types of information or signals to the public. These agents are all subject to various pressures and conflicts of interest that may diminish their ability to perform their task of discovery. Analysts may be well compensated and have substantial research resources at their disposal, but they may be too favorable to the firms if their bank is an underwriter and they have the least access to proprietary information. Ratings agencies are more insulated from conflicts of interest and have better access to information, but enjoying an oligopoly, their research effort may be reduced. Auditors enjoy superior access to proprietary information and operate in a competitive industry, but the value of their opinions may be reduced by conflicts of interest and a litigation-risk-induced focus on rules rather than principles. Finally, regulators/supervisors may have the best access to proprietary information, yet their capacity to monitor is limited by the resources they have been allocated and political pressures for forbearance.

It is necessary to have multiple agents working to reduce the information asymmetries to ensure that capital markets are properly served. One group of agents may become less useful at one point in time, but maintaining the quality of information delivered by these different groups of agents engaged in overlapping work is more likely to provide sufficient monitoring of companies. Policies should be designed so that remedies increase the effectiveness of these agents rather than constrain or co-opt them.