FRBSF Economic Letter
2003-22 | August 1, 2003
Disclosure as a Supervisory Tool: Pillar 3 of Basel II
International efforts are underway to improve the regulation and supervision of banking institutions to reflect advances in financial risk management techniques. In April 2003, the Basel Committee on Banking Supervision (BCBS 2003a), headquartered at the Bank for International Settlements in Switzerland, released for public comment the new Basel Capital Accord, which will replace the 1988 Capital Accord. These international agreements among banking regulators attempt to set regulatory capital requirements that are comparable across countries. On July 11, the Federal Reserve issued an interagency advance notice of proposed rulemaking, seeking public comment on the implementation of the new Basel Capital Accord in the U.S.
The new Accord, popularly known as Basel II, rests on three “pillars”: Pillar 1 focuses on making bank regulatory capital requirements more risk sensitive, while Pillar 2 emphasizes refinements of current bank supervisory processes regarding capital adequacy issues. The purpose of Pillar 3 is to complement the other pillars by presenting an enhanced set of public disclosure requirements focusing on capital adequacy. Banking institutions, like all firms, are monitored by their customers, trade counterparties, and investors in their securities. This type of monitoring is part of what is generally known as “market discipline,” which is increasingly viewed as complementary to the monitoring efforts of government supervisors; see Kwan (2002) for further discussion. The principle underlying Pillar 3 is that improved public disclosure of relevant information should enhance market discipline and hence its potential usefulness to bank supervisors. This Economic Letter reviews recent policy developments regarding disclosure by banking institutions, focusing on the disclosure requirements in the proposed Basel Accord.
In order for market discipline of banking institutions to be effective, banks must be sufficiently transparent; that is banks must provide a sufficient amount of accurate and timely information regarding their conditions and operations to the public. Improved public disclosures of such information lead to increased transparency and should lead directly to more effective market discipline.
As described in a study by the Board of Governors of the Federal Reserve System (BGFRS, 2000), bank disclosure standards in the United States are a byproduct of the demands of market participants and regulatory agencies as well as of the choices made by bank management. The core disclosure requirements for banks with publicly traded equity are set by the Securities and Exchange Commission (SEC) as well as the Financial Accounting Standards Board. These banks must meet the standards set for all publicly traded firms. For example, SEC disclosure rules require publicly traded firms to file the annual Form 10-K, which includes audited financial statements, and the quarterly 10-Q financial statements that are unaudited. In addition, SEC filings must be made in connection with special circumstances that can affect the reporting firm, such as the intention to issue new debt or equity securities.
In addition, all banks, whether publicly or privately owned, are required to file quarterly regulatory reports, such as the bank-level Call Reports, and much of this information is made publicly available. The reports contain detailed information regarding bank balance sheets and earnings. Also, agreements between banks and their supervisors, such as formal enforcement actions and cease-and-desist orders, are public documents that disclose specific steps bank management must take.
Although this overall regulatory reporting structure leads to a great deal of public disclosure, banks have a large degree of flexibility in meeting SEC disclosure requirements and thus maintain some control over what information is disclosed. A clear example is the case study reported in BGFRS (2000) regarding SEC requirements for disclosure of market risk exposures, defined as potential financial losses due to adverse movements in securities market prices. The most commonly used tool for reporting such risks are value-at-risk (VaR) estimates that summarize the potential losses that might occur with a specified probability over a given time horizon. In the case study, bank VaR disclosures were found to vary in detail across banks and to have an unclear connection with actual trading performance during the turbulent third quarter of 1998. Even though such heterogeneity is present in these types of public disclosures, the academic literature still suggests that market participants can assess bank risks accurately. For the case of VaR disclosures, Jorion (2003) found that VaR numbers in quarterly and annual reports of publicly traded commercial banks provide reasonable predictions of the variability of their trading revenues.
Overall, bank disclosure appears to be improving over time, guided by public and private sector efforts. A recent survey of the disclosure practices of internationally active banks by the BCBS (2003b) found that they have expanded the nature of their disclosures. Using a set of 104 questions addressing qualitative and quantitative disclosures, the survey indicated that banks disclosed 63% of the survey items in 2001, compared to 57% in 1999. The most commonly disclosed items were on banks’ capital structures, accounting policies, and market risk models.
In the United States, the private sector Working Group on Public Disclosure proposed enhancements to the public disclosure of market and credit risk information by large banking institutions. For example, the Group recommended that disclosures should explain how such risks within a firm change over time and how they evolve with innovations in a firm’s risk management practices; see Supervisory Letter 01-06 issued by the BGFRS (2001) for further details.
In an effort to continue this trend and to improve the ability of bank supervisors to use market discipline for their own monitoring purposes, the BCBS has made financial disclosure a key component of the newly proposed Basel Capital Accord.
Pillar 3 addresses the issue of improving market discipline through effective public disclosure. Specifically, it presents a set of disclosure requirements that should improve market participants’ ability to assess banks’ capital structures, risk exposures, risk management processes, and, hence, their overall capital adequacy.
The proposed disclosure requirements consist of qualitative and quantitative information in three general areas: corporate structure, capital structure and adequacy, and risk management. Corporate structure refers to how a banking group is organized; for example, what is the top corporate entity of the group and how are its subsidiaries consolidated for accounting and regulatory purposes. Capital structure corresponds to how much capital is held and in what forms, such as common stock. The disclosure requirements for capital adequacy focus on a summary discussion of the bank’s approach to assessing its current and future capital adequacy.
In the risk management area, the focus is on bank exposures to credit risk, market risk, risk from equity positions, and operational risk. For credit risk, which is defined as the potential losses arising from borrowers not repaying their debts, banks must provide a qualitative discussion of their risk management policies, the key definitions and statistical methods used in their risk analysis, and information on their supervisor’s acceptance of their approach. The quantitative disclosures include total gross credit risk exposures after accounting for offsets and without taking account of credit risk mitigation efforts. These exposures also must be reported in disaggregated form by exposure type (such as loans or off-balance-sheet exposures), by geographic region, by industry or counterparty type, and by residual contractual maturity. Impaired loans and past-due loans also must be reported by geographic region and industry type.
A key element of the new Accord is banks’ ability to use their own credit risk models and internal rating systems to set regulatory capital requirements. Credit risk models are tools for assessing the potential losses from aggregate fluctuations in loan repayments by borrowers, and internal rating systems provide these models with indicators of how likely different borrowers are to repay their loans. Given the Accord’s increased reliance on these internal components for setting regulatory capital requirements, the proposed disclosure requirements focus on a number of related informational areas. For example, banks must provide a description of their internal rating systems and the amount of credit exposure in each rating category. Banks that opt to use the most sophisticated capital methods based on their own models also must report key model parameters, such as default probabilities, credit exposures in case of default, and losses given default; see Lopez (2001) for further discussion. Furthermore, banks must report historical results regarding actual losses.
For market risk, banks must provide a general qualitative disclosure of their management policies, the statistical methods used in their models, and their model validation and stress-testing procedures. The quantitative disclosures include capital requirements for interest rate risk, equity risk, foreign exchange risk, and commodity risk, as well as various VaR measures and a comparison of these measures with actual outcomes. For the interest rate risk arising from loan portfolios, banks must report its nature qualitatively, their models’ key assumptions, and their earnings sensitivities to upward and downward movements in interest rates. For risk from equity positions, banks must disclose the types, amounts and nature of investments in public and private firms, as well as their total gains or losses, whether realized or not.
Finally, banks must disclose various elements of their operational risk exposures. Operational risk is commonly defined as the risk of monetary losses resulting from inadequate or failed internal processes, people, and systems, or from external events. Aside from a general qualitative discussion of their approach to managing such risks and supervisory approval, banks that choose to use the advanced measurement approaches permitted under the new Accord must describe their modeling approaches as well as the operational risk charge before and after any reductions resulting from the use of insurance.
The public disclosure requirements in Basel II are far-reaching and are intended to improve market discipline and its usefulness to bank supervisors. Yet, how well the requirements might work in practice depends on how they are implemented.
For example, the disclosures must not be overly burdensome on the reporting banks, but they must be accurate. The BCBS has made an effort to see that the relatively narrow focus of Pillar 3 on bank capital adequacy does not conflict with broader accounting requirements. The BCBS intends to maintain an ongoing relationship with international accounting authorities and to promote consistency among disclosure frameworks. Note that Pillar 3 disclosures need not be audited by an external auditor, unless otherwise required by accounting standard setters, securities regulators, or other authorities, but management should ensure that the information is appropriately verified.
Another key implementation issue is the frequency of disclosures. Qualitative disclosures of summaries of bank risk management policies are to be reported annually, and that should be sufficient since they are not likely to change often. However, many variables, such as regulatory capital ratios, are to be reported quarterly. More frequent disclosures have not been proposed and could be overly burdensome for some institutions at this point, but given the nature of certain bank business lines, such disclosures may become commonplace in the future.
Another important implementation issue is determining whether a specific piece of information is proprietary. The BCBS has determined that specific items that may prejudice a bank’s proprietary or confidential information need not be disclosed. However, the bank must disclose more general information about the subject matter, together with an explanation of why those specific items were not disclosed.
Along the same lines, determining whether specific disclosure items are material is another challenge. Under the current guidelines, banks should decide which disclosures are material based on commonly accepted principles; that is, information is material if its omission might change or influence the assessment or decision of a user relying on that information. This definition is consistent with International Accounting Standards and with many national accounting frameworks.
In summary, many implementation details are addressed within the Pillar 3 disclosure requirements, but many specific questions and additional issues will arise during the actual implementation process. However, it is reasonable to assume that as these issues are worked out, the improved disclosure by banks should facilitate market discipline, contribute to supervisory monitoring efforts, and enhance the stability of the national and international banking systems.
Jose A. Lopez
[URLs accessed July 2003.]
Basel Committee on Bank Supervision. 2003a. “Overview of the New Basel Capital Accord: Consultative Document” (April).
Basel Committee on Bank Supervision. 2003b. “Public Disclosures by Banks: Results of the 2001 Disclosure Survey.” Basel Committee Publications No. 97 (May).
Board of Governors of the Federal Reserve System. 2000. “Improving Public Disclosure in Banking.” Staff Study #173.
Board of Governors of the Federal Reserve System. 2001. Supervisory Letter 01-06: Enhancements to Public Disclosure.
Jorion, P. 2003. “How Informative Are Value-at-Risk Disclosures?” The Accounting Review 77, pp. 911-931.
Kwan, S.H. 2002. “The Promise and Limits of Market Discipline in Banking.” FRBSF Economic Letter 2002-36 (December 13).
Lopez, J.A. 2001. “Modeling Credit Risk for Commercial Loans.” FRBSF Economic Letter 2001-12 (April 27).
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120