Michael W. Taylor
© 2004 International Monetary Fund
Financial Sector Regulators Be Independent?]
This Economic Issue is based on IMF Working Paper WP/02/46, Regulatory and Supervisory Independence and Financial Stability, March 2002. Brenda Szittya prepared the text. Citations for the studies reviewed are provided in the original paper, which readers can purchase from IMF Publication Services ($10.00) or download from www.imf.org.
In nearly every major financial crisis of the past decade—from East Asia to Russia, Turkey, and Latin America—political interference in financial sector regulation helped make a bad situation worse. Political pressures not only weakened financial regulation generally, they also hindered regulators and the supervisors who enforce the regulations from taking action against banks that ran into trouble. In so doing, they crippled the financial sector in the run-up to the crisis, delayed recognition of the severity of the crisis, slowed needed intervention, and raised the cost of the crisis to taxpayers.
Increasingly, both policymakers and policy analysts are recognizing the need to shield financial sector regulators from political pressure to improve the quality of regulation and supervision with the ultimate goal of preventing financial crises. Surprisingly, however, few analyses have systematically discussed why independence for the financial regulatory agency might be desirable and how it might best be achieved.
This pamphlet investigates why financial sector regulators and supervisors might need a substantial degree of independence—not only from the government but also from the financial services industry—to fulfill their mandate to achieve and preserve financial sector stability. It also looks at the need for keeping regulators accountable as they exercise the (often) far-reaching powers delegated to them by their government.
Regulators and supervisors in nations around the world are charged with managing the health of banks and other financial institutions and preserving the stability of the financial system. Governments regulate financial institutions for two main purposes. The first is consumer protection. This is much the same reason they regulate public utilities and telecommunications: to provide a framework of rules that can help prevent the excesses and failures of a market left entirely to its own devices. Second, regulation in the financial sector has the additional goal of maintaining financial stability, a clear public good that justifies a more elaborate framework of regulation and supervision.
Financial sector supervision, in particular, is more rigorous and intensive than supervision in other regulated sectors. Banking supervisors engage not only in off-site analysis of banks' performance but also in extensive on-site inspections, and they intensify their monitoring and may intervene when banks fail to meet minimum requirements designed to ensure their financial soundness. Supervisors can even, in extreme cases, take ownership rights away from the owners of failed or failing financial institutions.
Banking regulation developed out of the concern of central banks to ensure financial stability. In many parts of the world, the central bank is the agency responsible for regulating banks, while, in others, it is a separate agency. In the nonbank financial sector, such as securities markets, insurance, and pensions, regulation has usually been conducted either by a central government ministry or by a specialist agency answerable to a ministry. The need for independent regulatory agencies has not, however, featured very strongly in public debates. In recent years, this has begun to change, driven by two important reasons.
In many of the world's recent financial crises, policymakers in the countries affected have sought to intervene in the work of regulators—often with disastrous results. It is now increasingly recognized that political meddling has consistently caused or worsened financial instability.
In her account of the Venezuelan banking crisis of 1994, former central bank president Ruth de Krivoy cited ineffective regulation, weak supervision, and political interference as factors weakening the banks in the period leading up to the crisis. In her book, Collapse, published in 2000, de Krivoy emphasized the need for lawmakers to "make bank supervisors strong and independent, and give them enough political support to allow them to perform their duties."
In East Asia in 199798, political interference in the regulatory and supervisory process postponed recognition of the severity of the crisis, delayed action, and, ultimately, deepened the crisis. In Korea, for example, a lack of independence impeded supervision. While the country's commercial banks were under the authority of the central bank (the Bank of Korea) and the Office of Banking Supervision, Korea's specialized banks and nonbank financial institutions were regulated by the ministry of finance and economy. The ministry's weak supervision encouraged excessive risk taking by the nonbanks, which helped lead to the 1997 crisis. Korea subsequently reformed its supervisory system, both to give it more autonomy and to eliminate the regulatory and supervisory gaps.
In Indonesia, banking sector weaknesses stemmed from poorly enforced regulations and from supervisors' reluctance to take action against politically well connected banks, especially those linked to the Suharto family. When the crisis hit, central bank procedures for dispensing liquidity support to troubled banks were overridden, it has been claimed, on the direct instructions of the president. Even after Suharto's fall, political interference continued in the bank restructuring effort. Indonesia's Financial Sector Action Committee, which was composed of several heads of economic ministries and chaired by the coordinating minister, intervened intrusively in the work of the Indonesian Bank Restructuring Agency and undermined the credibility of the agency's restructuring effort.
The lack of independence of financial supervisors in Japan's ministry of finance weakened the Japanese financial sector and contributed to prolonged banking sector problems. Although there was probably little direct political pressure on the ministry to allow weak banks to continue operating, the system lacked transparency, and implicit government guarantees of banking sector liabilities were understood to be widespread. Following a decline in the ministry's reputation in the late 1990s, the Japanese government created a new Financial Services Agency to oversee banking, insurance, and the securities markets, in part as an attempt to improve the independence of supervision.
A second development that has encouraged interest in regulatory independence is the success achieved by independent central banks in fighting inflation. Since the late 1980s, more and more countries have freed their central banks from political control because evidence was growing that independent central banks had a successful record of achieving monetary stability—in other words, controlling inflation. Central bank independence is seen as essential to counter the natural preference of politicians for expansionary economic policies that promise short-term electoral gains at the risk of worsening inflation in the long run. Making central banks independent frees them from political pressure and thus removes the inflationary bias that could otherwise unsettle monetary policy.
The incentives for politicians to rescue failing banks are similar to those for inaction in the face of inflation. The decision to close a failing bank is usually unpopular. Politicians eager to avoid a necessary closure may be tempted to pressure bank supervisors to organize a bailout or to excuse the failing bank from regulatory requirements, even at the risk of worsening the problem and increasing the long-term costs of resolving it. This similarity strengthens the case for regulatory and supervisory independence in the financial sector. Bank regulatory independence is to financial stability what central bank independence is to monetary stability, and the independence of the two agencies can be mutually reinforcing. Both agencies also provide a public good—financial stability—which sets them apart from other sector-specific regulatory agencies.
Still, banking supervisory authorities differ from central bank authorities in one important way. When banking supervisors revoke a failing bank's license, they are using the coercive power of the state against private citizens. When central banks conduct monetary policy, they have no such coercive power. However, the unique power enjoyed by financial regulators should not be used as an argument against granting them independence. Instead, governments should fully accept and take the implications of that power into account—both by establishing strong accountability mechanisms to prevent abuse and by recognizing the need to employ highly qualified supervisors of demonstrable integrity and to pay them adequately.
Growing recognition of these factors, together with a growing trend toward the creation of "integrated" financial supervisory agencies that regulate banks, securities markets, and insurance companies—which forces policymakers and legislators to rethink institutional arrangements—has focused attention on the case for agency independence. Increasingly, a consensus is forming that regulatory and supervisory agencies should be independent—that they should not be embedded in the executive hierarchy and thus subject to political pressure. In theory, independent regulators can decide on and conduct market interventions shielded from political interference and improve regulatory and supervisory transparency, stability, and expertise. And, indeed, growing evidence suggests that independent regulators have made regulation more effective, have led to smoother and more efficient operation of the market, and are a distinct improvement over regulatory functions located in government ministries.
But independent regulatory agencies have potential problems of their own. As the Nobel Prizewinning economist George J. Stigler pointed out in a seminal article in 1971, agencies tend to respond to the wishes of the best-organized interest groups. When regulators are free from political control, the risk of "regulatory capture" by other groups—in particular, the industry they regulate—grows. Agencies that suffer from such capture come to identify industry interests (or even the interests of individual firms) with the public interest. And industry capture can undermine the effectiveness of regulation just as political pressure can. Regulators may, for example, formulate rules so as to minimize industry costs rather than strike an appropriate balance between those costs and public benefits. They may also apply rules inconsistently and exempt individual firms from regulatory requirements.
Agency independence entails another important concern, namely, the need for accountability. An independent agency might pursue an agenda of its own, going against the wishes of the political majority. Some worried analysts have branded independent regulatory agencies the "fourth branch of government," implying that they are outside the control of the traditional three branches that keep mature democratic systems in equilibrium through a system of checks and balances. Although such fears appear to be exaggerated, they nevertheless demonstrate the need for proper forms of accountability to balance the advantages of agency independence with the disadvantages.
Achieving both political independence and independence from the regulated industry, as well as accountability, is essential. Political independence, however, remains the prime concern from the point of view of financial stability, given the vested interests that many national governments still have in the banking system, and, therefore, in bank regulation and supervision, as well as the dismal track record of political interference in regulatory and supervisory arrangements.
In passing the laws that create regulatory agencies, politicians properly set and define regulatory and supervisory goals. But once those laws are in force, regulators must be free to determine how to achieve these goals—and should be held accountable if they fail to achieve them. Independence has four dimensions: regulatory, supervisory, institutional, and budgetary.
Regulatory independence in the financial sector means that regulators have wide autonomy in setting, at a minimum, prudential rules and regulations that follow from the special nature of financial intermediation. These rules and regulations concern the practices that financial institutions must adopt to maintain their safety and stability, including minimum capital adequacy ratios, exposure limits, and loan provisioning. Regulators who are able to set these rules independently are more likely to be motivated to enforce them. They are also able to adapt the rules quickly and flexibly in response to changing conditions in the global marketplace without having to go through a lengthy, high-pressure political process.
In some countries, the laws governing financial institutions and the financial regulatory agency are so detailed as to leave little room for independent rule setting by the supervisory agency. In others, the laws merely brush a broad framework, leaving much greater scope for regulatory discretion. Regardless of a country's particular legal traditions, however, independent regulators should be given ample discretion to set and change regulations within the broad confines of the country's constitution and banking law.
Supervisory independence is crucial in the financial sector. It is also difficult to establish and guarantee. Supervisors work quite closely with financial institutions, not only inspecting and monitoring them but also enforcing sanctions and even revoking licenses. Much of their activity takes place outside direct public view, and interference with their work, either by politicians or by the industry, can be subtle and can take many forms. Steps to protect supervisors' integrity include offering legal protection (for example, repealing laws that, in some countries, allow supervisors to be sued personally for their work) and providing financial incentives that allow supervisory agencies to attract and keep competent staff and discourage bribery. Crafting a rules-based system of sanctions and interventions also lessens the scope for supervisory discretion—and thus for political and industry interference. To protect supervisors from political or industry intimidation during a lengthy court process, banking law should also limit the time allowed for appeals by institutions facing sanction. Independent supervisors, not a government agency or minister, should be given sole authority to grant and withdraw licenses because they best understand the financial sector's proper composition—and because the threat to revoke a license is a powerful supervisory tool.
Institutional independence, or the agency's status outside the executive and legislative branches of government, has three critical elements. First, senior personnel should enjoy security of tenure: clear rules, ideally involving two government bodies, must govern their appointment and, especially, dismissal. Second, the agency's governance structure should consist of multimember commissions composed of experts. And, third, decision making should be open and transparent to the extent consistent with commercial confidentiality, enabling both the public and the industry to scrutinize regulatory decisions.
Budgetary independence depends primarily on the role of the executive or the legislative branch in determining the agency's budget and how it is used. Supervisors should not be subjected to political pressure through the budget. Senior personnel should have the budgetary freedom to staff the agency as they see fit and to respond quickly to emerging agency needs. If funding must come from the government budget, the supervisory budget should be proposed and justified by the agency itself, following objective criteria related to developments in the market. Some supervisory agencies are funded through industry fees, a practice that minimizes political interference but risks increasing dependence on—and attracting interference from—the industry. Industry fees, if used, should be determined jointly by the agency and the government. Also, fee-based funding may leave the agency strapped for funds during a crisis, because that is precisely when the industry is most likely to have difficulty paying the fee. Agencies should therefore be allowed to build up reserve funds as insurance.
Arguably, the simplest way to secure independence for bank regulators and supervisors is to house them in an independent central bank. Because the concept of central bank independence is now generally accepted, financial sector regulators and supervisors could "piggyback" on the autonomy the central bank already enjoys.
The chief argument for housing financial sector regulators in the central bank is that banks are the instruments through which the central bank transmits monetary policy to the wider economy. The central bank is therefore naturally concerned with bank soundness as a precondition for effective monetary policy. And because the central bank also acts as a lender of last resort to banks in time of crisis, it should have access to information about the financial soundness of any bank that might apply to it for emergency liquidity assistance.
On the other hand, the arguments for keeping financial sector regulators separate from the central bank are equally strong. The first involves the potential for conflict of interest. A central bank that supervises the financial sector might be tempted to operate a lax monetary policy to keep banks healthy. This might succeed but raise inflation in the long run. The second argument involves the risk to reputation. If a bank were to fail, it not only would attract blame to the bank supervisors but also might undermine the credibility of the central bank itself. To avoid this reputational contagion, the central bank should keep the supervisory function at arm's length. A similar argument holds from the point of view of the supervisors, whose reputation can be damaged if central bank actions to stimulate the economy also tend to prolong the life of banks that they believe should be closed.
These arguments for and against combining financial sector regulators and the central bank are finely balanced, with neither arrangement a clear winner. In the case of developing economies, however—including economies in transition from central planning—several considerations appear to tip the argument in favor of combination. Central banks in many of these economies have been reformed to give them strong guarantees (sometimes constitutional guarantees) of independence. Often the governor of the central bank enjoys strong security of tenure, and the bank has its own dedicated funding sources. The central bank's budgetary autonomy and prestige also give it an advantage in attracting and retaining skilled and expert staff. Thus, with only a few exceptions, most transition economies have adopted this model.
A final point is that merely housing financial sector regulators and supervisors in the central bank will not automatically ensure their independence. Separate arrangements must be made not only to guarantee the independence of the central bank but also to guarantee the operational autonomy of the supervisory agency and the integrity of the supervisory function within the central bank.
In recent years, the debate about housing banking supervision inside or outside the central bank has been complicated in many advanced and emerging economies by the expansion of the financial sector into nonbank activities, such as securities and insurance. If banking regulation and supervision are housed in the central bank, the monetary authorities could, in principle, assume these new regulatory duties as well, ensuring their independence and helping build regulatory capacity by making the central bank's information technology, data collection, and human resource functions available to regulators. But this approach entails several serious drawbacks. It gives the central bank responsibility for a wide range of financial activities about which its staff cannot be expected to have special expertise. In addition, extending the central bank's regulatory responsibilities to nonbank financial institutions may be interpreted as extending its guarantee of financial assistance beyond banks. Most important, granting the central bank such extensive regulatory responsibilities may make it appear—and be—excessively powerful, raising serious issues of accountability.
In these circumstances, one alternative to centralizing all regulatory and supervisory functions in the central bank is to create an integrated financial regulatory authority as a separate agency responsible for banking, securities, and insurance. However, the history of these various subsectoral regulators and supervisors is often one of highly differing degrees of independence. The goal of the authorities in a unified agency must therefore be to take the opportunity provided by unification to ensure the highest level of independence for all rather than the lowest common denominator.
National political culture can also determine whether a country is able to achieve regulatory and supervisory independence. In many Western countries, long-standing political institutions such as a transparent political process, a system of constitutional checks and balances, and a free press support regulatory independence. But many other economies around the world still lack some of the institutions that would provide a secure basis for regulatory independence.
Again, an analogy with the central bank is useful. Experience has shown that not all legal arrangements that grant independence to central banks are equally effective. For example, independence granted to a central bank in a political system without clear checks and balances is fragile, because it can be withdrawn at little political cost. Only in a political system with at least two veto players with different preferences is central bank independence likely to thrive.
The parallel with financial regulatory agencies is straightforward, although it cannot yet be tested extensively because few independent agencies exist. However, studies so far show that checks and balances tend to better insulate financial sector supervisors and to encourage better and more prudent regulation. The fewer checks and balances there are, the greater the incentives for the government to override or preempt, at no cost to itself, supervisory actions directed at troubled banks, thus keeping such banks open and risking higher costs to society in the future. Underdeveloped checks and balances also make it easier to relax key prudential rules.
In a financial crisis, a political system with multiple checks and balances has, almost by definition, no unrepresented groups onto which to shift costs. Therefore, checks and balances can help strengthen weak regulation and head off a crisis at an early stage through supervisory intervention. True, negotiations among the different interests could slow the initial response to a crisis, ultimately increasing its costs. But this is less likely when all interest groups are represented, because it is to the advantage of most to avoid delay and thus limit those costs.
Although regulatory and supervisory independence is most effective in a nurturing political culture, such cultures need time to take root. Countries whose political systems lack checks and balances must therefore base their commitment to regulatory independence on the need to adhere to best international standards and practices in today's globalized system. If regulations diverge too far from international best practices, as embodied, for instance, in the Basel Core Principles for Effective Banking Supervision, and supervisory practices are considered weak, domestic and, more important, foreign investors might turn their backs, cutting the country off from the benefits of foreign investment in the financial sector.
Accountability of independent regulators and supervisors is the key to effective independence. Independence can never be absolute, because, even among "equal and independent partners," cooperation and coordination are essential. Moreover, political legitimacy demands that any independent regulatory agency be held accountable for how it uses the powers delegated to it by the legislature. Many analysts see this issue as a dilemma: if the regulatory agency is part of the state administrative apparatus, it cannot be independent; if it is independent, how and to whom should it be accountable?
This view, however, rests on a misunderstanding of agency independence. Properly designed, independence includes mechanisms for holding the agency accountable for carrying out its responsibilities while allowing it to remain free of interference in its operations. There is no trade-off between independence and accountability; the two concepts are complementary. Accountability is needed to make independence work. The greater the degree of independence, the more important accountability arrangements become.
In countries with longer traditions of independent agencies, the evidence suggests that independent regulators typically do not behave as an irresponsible "fourth branch of government." Indeed, regulators usually behave according to a kind of "dialogue model," doing their best to be informed about the intentions, wishes, and opinions of the political leadership and to anticipate their reactions to new policy proposals. In other words, independent regulatory agencies remain subject to some form of political control—a type of self-censorship, as it were.
Accountability is better served, though, if this loose control is supplemented with more formal arrangements. Setting up proper accountability arrangements is a balancing act. At least seven separate criteria for genuine accountability require careful attention. Perhaps most important, the independent agency needs a clear legal basis. Its powers and functions should be set out, preferably by statute, to lessen the potential for disputes between the regulatory agency and other government agencies or the court system. The agency should also have a clear, public statement of its objectives—for example, preserving the stability of the financial system and the soundness of individual banks and protecting depositors and other financial services customers. A clear mandate makes it easier to measure the agency's performance against that mandate. A public mission statement protects the agency against claims by politicians and by the supervised institutions that it has not carried out its mandate.
The agency's relationships with the executive, legislative, and judicial branches must be clearly defined. The issues on which, and the form in which, it must inform or consult the ministry of finance or seek its approval must be spelled out. The procedures by which the legislature holds the agency to account for using the powers delegated to it must be carefully defined. And its exercise of those powers should be subject to judicial review.
Although officials of the independent agency must have security of tenure, accountability requires that the law spell out who has final responsibility for appointment, reappointment, and dismissal of senior officials and what procedures should be followed. The agency must also be answerable for the way it manages its budget, either in advance, during the budgetary appropriations process, or afterward, by a review of its accounts. And its decision making must be transparent to the extent permitted by the need to preserve the business confidentiality of the regulated institutions.
Finally, for those rare occasions when the agency's independence must be overridden (during a financial crisis, for example), the override mechanisms and the circumstances in which they may be triggered must be defined.
Time and again in the past decades, national and regional financial crises have been deepened and worsened by political interference in financial sector regulation and supervision. Gradually, policymakers and policy analysts alike are recognizing that regulators charged with preserving a financial system's stability can carry out their mandate most effectively when they are independent of political influence.
The case for regulatory independence in the financial sector can be based on analogies with two areas where the issue of regulatory independence has already been largely settled: the regulation of public utilities and telecommunications, and the conduct of monetary policy by the central bank. Mounting evidence indicates that independent regulators have increased the efficiency and effectiveness of regulation and helped markets operate more smoothly and efficiently. And the concept of central bank independence has gained near-universal acceptance over the past two decades, as independent central banks have shown that they can successfully pursue monetary stability. Regulatory and supervisory independence in the financial sector complements central bank independence in achieving and preserving the twin goals of monetary and financial stability.
Even though independent regulatory agencies have proved themselves in practice, creating them and keeping them independent and accountable poses difficulties that cannot be ignored. An agency that is independent of political control may become subject instead to pressure from the industry it regulates. And freedom from political control can lead to serious deficiencies in political accountability. Making financial sector regulatory independence work requires attention to all four aspects of independence: regulatory, supervisory, institutional, and budgetary. The arguments for and against jump-starting independence by housing the agency in the central bank need to be considered and the best course decided upon. Whether the country's political institutions, especially its system of checks and balances, are adequately developed to safeguard the agency's independence must also be determined. Finally, the proposed arrangements must be tested against the seven essential criteria laid out above to ensure the proper balance between effective independence and political accountability.