## William J McDonough: Completing the journey to the New Basel Accord
Remarks by Mr William J McDonough, President and Chief Executive Officer of the Federal Reserve Bank of New York, before the Twelfth International Conference of Banking Supervisors, Cape Town, South Africa, 18 September 2002.
* * *
## Introduction and Overview
As the chairman of the Basel Committee on Banking Supervision, I am honored to welcome you today to the Twelfth International Conference of Banking Supervisors. I would like to thank our hosts, Governor Mboweni and the staff of the South African Reserve Bank, as well as my colleague, Andrew Crockett, and the Bank for International Settlements, for organizing our gathering here in Cape Town. I'm sure you'll agree that they've done an excellent job.
It appears to me, and probably to many of you, that South Africans have discovered a wonderful niche for themselves as hosts of auspicious international gatherings, having just concluded the World Summit on Sustainable Development in Johannesburg a few weeks ago. So if you're visiting this beautiful country for the first time, I have a feeling that you may be back in South Africa someday in the future.
Participants at the United Nations summit in Johannesburg sought ways to improve the lives of people everywhere by focusing on, among other goals, alleviating poverty and promoting economic growth. Bank supervisors and central bankers certainly share those goals. Through our duty to ensure the safety and soundness of the financial system, we aim to preserve depositors' savings, to promote the circulation of credit, and ultimately to protect the health and growth of our economies. Indeed, we cannot alleviate poverty, we cannot promote economic growth unless our banks are well managed, our financial systems stable, and our markets transparent and open.
That the health of an economy depends to a great degree on whether its banks are financially sound and well managed is certainly not coincidental. Banks are charged with critical public trust to safeguard the public's wealth and to serve as a source of credit to consumers and businesses alike. When banks fail to discharge these tasks responsibly, the costs to society can be high and the impact on the economy devastating.
At the same time, the duty for supervisors to promote the safety and soundness of the banking system entails achieving a delicate balance. Ensuring the prudent management of banks does not mean that we should hinder their growth or profit. Rather, the supervisor's job is to promote the responsible pursuit of opportunity and innovation. Likewise, it cannot be our job to prevent every bank failure, for some failures remove poorly managed banks from the market and encourage other banks to improve their management as well.
Motivating all banks to improve their management of risk, and thereby strengthen the stability of the banking system, is one of the principal objectives that the Basel Committee set out for itself in undertaking a major revision of the international standards for regulatory capital. The three years since have been a tremendous journey for all of us as we seek fundamental reform of the capital adequacy framework.
Meeting here in Cape Town, it seems fitting to portray our work to date as an unfinished journey. This historic port has traditionally served as a way station and respite for weary seafarers who had weathered many storms, but had not yet reached their destination. So today we'll have a chance to rest near the Cape of Good Hope, to reflect on how far we've come, and to chart the rest of our way to the successful implementation of the New Accord. I'd like to share with you the latest news on where the New Accord stands and offer some insight into what you can expect to see in the future. I'll begin with my views on the milestones that we've achieved. Then, I'd like to bring you up to date on how we are resolving outstanding issues. It's wise to conclude any speech with thoughts about the future, so I'll close with some remarks on what we'll need to do to finish our journey - from the perspectives of the Committee and of other supervisors, including of those who may have adopted the 1988 Accord more recently.
## Milestones Achieved
Let me begin with the important milestones that we've achieved in this undertaking. As you know, we embarked on this voyage in the late 1990s because we realized that the original 1988 Basel Accord was becoming outdated. It was designed to be uncomplicated so that it could be applied to banks of various sizes and risk profiles in many markets and jurisdictions. Its subsequent adoption by more than 100 countries represented an important first step in the development of an international standard.
But it has since become clear that the 1988 Accord has been overtaken by advances in the financial sector - and in the broader economy. New technology, the globalization of financial markets, and innovative financial products and services have changed the way that banks monitor and manage credit, market, and operational risks in a manner that the 1988 Accord could not anticipate and does not address. Consequently, the internationally active banks for which the original Accord was intended are quickly "outgrowing" its limitations. Today it's quite clear that the 1988 Accord provides this class of banks with less meaningful measures of the risks they face and of the capital they should hold against those risks.
To ensure that the New Accord remains flexible, forward-looking, and fit for the service of internationally active banks of the twenty-first century, the Basel Committee established several goals for its work, goals that the industry has embraced.
- First, we intended to develop a framework that encompasses the "three pillars" necessary to support an effective system of regulatory capital: the appropriate measurement and minimum requirements, supervisory review, and market discipline.
- Second, we wanted to align the minimum requirements more closely with the actual underlying economic risks to which banks are exposed, which should help allocate capital resources effectively.
- A third goal was to encourage banks to refine their measurement and management of risk over time. By creating incentives in the New Accord for banks to re-evaluate and enhance their tools constantly, we expect that banks themselves will adopt a forward-looking perspective on risk.
I'm pleased to say that, through the Committee's efforts and the cooperation and support of other supervisors and the industry, it appears that the proposed framework will attain each goal. We can now count them among the milestones we've achieved.
## 1. Encompass the "three pillars" of capital adequacy
With regard to the first goal, the new framework will recognize the important roles that supervisors and markets can play in motivating the prudent management of risk and capital. Indeed, it is the Committee's view that it would be insufficient for any country to simply adopt the new minimum capital requirements and ignore the remaining two pillars. Both the involvement of supervisors in assessing a bank's risk management capabilities and the discipline that greater transparency and active markets bring to bear on management can reinforce a bank's desire to remain adequately capitalized and in control of its exposure to risks.
## 2. Improve the risk-sensitivity of the capital framework
The New Accord will likewise achieve the second and parallel goal to align regulatory capital requirements more closely with the underlying economic risks that banks face. It will do so by departing from the 1988 Accord's "one-size-fits-all" methodology that assesses credit risk based more on the type of borrower, offering instead a selection of methods to evaluate credit risk based more on the potential for loss. Under the new standardized approach, for example, banks with smaller loan portfolios or less advanced risk management systems will be better able to incorporate a borrower's creditworthiness into their capital management by relying on external assessments of credit quality. Larger or more sophisticated institutions, in turn, will be able to rely increasingly on their own assessments under the internal ratings-based approach.
Similarly, now that the industry is developing a more refined sense of operational risk - the risk of losses caused by failures in processes - the New Accord will "unbundle" this risk and allow banks to differentiate between, and manage separately, their exposures to credit, market, and operational risks.
## 3. Encourage banks to improve their management of risk
We've also attained the third goal: to go beyond a rules-based structure and instead encourage banks themselves to improve their abilities to identify, measure, and manage their risks appropriately. As you know, the most sophisticated approaches to credit or operational risk in the New Accord will permit banks to rely increasingly on their internal assessments of risk when determining how much capital to hold. At the same time, the New Accord will be flexible enough for banks to build their risk management functions on the best practices available in the industry, even as those practices evolve in the future. By promoting the improvement of the management of risk across the industry, I believe that we are laying a foundation for more stable markets.
## The milestones achieved square with banks' and supervisors' long-standing efforts
All of us can take great pride in meeting the goals that the Committee set out to embrace: the three pillars of capital adequacy, enhancing the risk-sensitivity of the capital framework, and creating incentives for banks to improve their management of risk. And we should recognize that these proposed revisions square well with the efforts that banks and supervisors themselves have undertaken, independent of the Committee's work. The banking industry has done much over the years to adopt a forward-looking approach to risk management - one that emphasizes flexibility in, and improvements to, the measurement and management of risk over time.
For our part, supervisors have come to focus much less on point-in-time assessments of financial condition and much more on evaluating the quality of internal processes and control structures that protect against future losses. By embracing a philosophy of supervision that centers on evaluations of risk management capabilities, we gain deeper insight into banks' likely resilience during downturns, and we are better positioned to focus management's attention on finding ways to do better.
Certainly the New Accord's inclusion of extensive economic incentives to adopt sophisticated risk management processes adds encouragement for their enhancement. Moreover, the New Accord's requirement of strong process controls as a predicate for using the more advanced approaches strongly supports our efforts to make evaluations of internal processes the heart of a more forward looking supervisory approach.
## The status of the New Accord
So it would seem that we've covered quite a bit of territory over the past three years. Yet a Chinese proverb reminds us that, "On a journey of one hundred miles, ninety is but halfway." I suppose the last miles of any marathon are the toughest to finish. The wisdom of that proverb is borne out by the tasks ahead of us to complete, and then to implement, the new capital framework. So I'd like to turn now to the status of the New Accord and of the issues we are resolving.
As you know, since the Second Consultative Document's release 21 months ago, the members of the Basel Committee have worked collaboratively and publicly with supervisors, banks, and others to revise the proposals so that they best serve the needs of modern banking. We've published and discussed thousands of pages of proposals and studies with the industry and the public. The size of and depth of detail in - those volumes demonstrates how hard we've been working and how great the challenge is.
## Challenges the Committee is resolving
Indeed, while our consultations over the years have reflected on some of the finest - and occasionally highly esoteric - points of contemporary banking, the Committee encountered a few central concerns expressed by the industry - and sometimes by supervisors as well. Some of the overarching issues include the complexity of the proposals and their potential to be "procyclical." Other respondents have weighed in on proposed charges against operational risk and other particular exposures. I'd like to share with you now the latest news on how we are resolving those challenges.
## Complexity of the New Accord
First, to the issue of complexity. Anyone who's dared to download the proposals in their entirety from the BIS website may have been surprised to find out how much paper it requires to print them out. The New Accord is truly weighty, in all meanings of that adjective. Naturally, as prudent supervisors, the members of the Committee know that it is easier to enforce a simple rule than a more complicated one. Wherever possible, the Committee has sought to provide simple alternatives that can be applied in many markets and settings. Consultations helped us to simplify or remove a good number of rules thought to be unnecessarily complicated.
However, it's important to recognize that we intend to apply the New Accord especially to large, internationally active banks, all of which are highly developed and complex institutions. One blanket won't cover a mountain range. In fact, the New Accord's complexity stems partly from the options it provides to address the wide range of risk profiles, strategies, and systems that banks maintain. Many options were created precisely because some banks felt disadvantaged under a proposed blanket rule. Yet each new option alters the balance between simplicity and complexity.
I think the best example of finding that balance lies in the treatment we've proposed for exposures to small and medium-sized enterprises. Equating loans to small and medium-sized businesses with loans to large corporations ignores critical aspects of the risks associated with each type of lending; it also could lead to results that seem unfair to banks that lend primarily to smaller businesses. So the Committee developed a special capital treatment for exposures to small and medium-sized enterprises that recognizes the unique characteristics of that form of lending, but admittedly makes the rules more complicated, and perhaps a page or two longer. Nonetheless, we believe that this treatment helps to align regulatory capital more closely with the underlying risks.
## Procyclicality
Another general concern expressed is that the New Accord's increased sensitivity to risk will reinforce behavioral patterns in banking organizations that may increase the cost of credit precisely at times when its supply is falling, namely during downturns in the business cycle. While we are working to address this concern about procyclicality, it is not a reason to shy away from creating better risk management tools.
Here, supervisors must articulate the need for banks to assess risks on the horizon. If banks fail to develop and manage to reasonable expectations about how their risk profiles may change over time, they may take on excess exposures today that they cannot administer responsibly tomorrow.
At the same time, we should remember that the New Accord sets out the minimum amount of capital necessary to comply with banking regulations. All supervisors expect their banks to operate at levels above the regulatory minima, and nearly all banks manage to their economic capital needs, not to their regulatory capital requirements. So although deteriorating credit quality might drive up the minimum capital requirement, a bank should already be operating above that level, which should dampen the effects of a sudden downturn in business conditions and asset quality.
Along those lines, the Committee recently agreed that banks adopting the "internal ratings based" approach to credit risk will be required under Pillar 2 to conduct meaningfully conservative credit risk stress testing. Stress testing should help banks to maintain adequate capital buffers in advance of potential downturns in the economy.
## The treatment of operational risk
Beyond those two broad concerns expressed about the potential for complexity and procyclicality, some respondents have called for changes to more specific capital charges in the proposals. One of the most interesting discussions centers on whether - or how - the New Accord should treat a bank's exposure to potential losses that result from inadequate or failed internal processes and systems, or from external events.
The endeavor to manage operational risk, as this group of risks has become known, has sparked significant research in the industry and beyond. Supervisors, like more and more banks, have adopted a view of operational risk management as a comprehensive practice comparable to the management of market or credit risk. While operational risk cannot be quantified with the same degree of precision as market or credit risk, the Committee believes that introducing a separate charge for exposures to
operational risk will bolster efforts to find better ways to address it. Failing to include an operational risk charge in the new minimum requirements could inadvertently derail the ongoing work to identify and manage this exposure.
Although we've seen encouraging progress in measuring operational risk - within both individual firms and industry working groups - the Committee recognizes that the industry has not settled on particular methodologies or principles. Accordingly, the New Accord will permit an unprecedented amount of flexibility to accommodate a spectrum of approaches to operational risk. For the most sophisticated institutions, the "advanced measurement approach" - or "AMA" - is really the ultimate expression of the goal of drawing on internal measures. In response to concerns that the proposed "floor" on the AMA stifled innovation, the Committee decided recently to eliminate it. Consequently, banks will be free to experiment with a great variety of methodologies, without undue constraint on their value to the firm. The AMA is meant to be a catalyst for innovation.
## Calibration of other particular charges and the overall level of capital
Of course, a good portion of the consultations have addressed rules and formulae used to determine capital charges for other categories of exposures and transactions. The Committee and its working groups have listened carefully to the views of market participants, academics, and other regulators to ensure that the resulting capital requirements make good economic sense and avoid burdening or disadvantaging banks unfairly. Based on the changes we are making, we expect that the Committee will achieve its goal of not raising in aggregate the capitalization required of the banking industry, though clearly and appropriately those banks that engage in higher risk businesses may see their requirements rise, and vice-versa.
## QIS 3
To ensure that we've gotten it right, the Basel Committee will launch a third Quantitative Impact Study - or "QIS 3" - on October 1, 2002, in conjunction with supervisors around the world. The industry has demonstrated tremendous support for this survey and will devote substantial resources to its completion. To date, 265 banks from nearly 50 countries have agreed to perform concrete and comprehensive assessments of how the Committee's proposals will affect them and then submit their findings by December 20, 2002. The inclusion of banks from so many countries will help to ensure that the New Accord's principles can be applied well to institutions in many countries.
With the public release of the QIS 3 instructions on October 1, the Committee will provide a full picture of its proposed framework and quantitative requirements. Assessment of the QIS 3 results will allow us to ascertain the need for adjustments prior to the release of an updated proposal for public comment in the second quarter of 2003. The Committee intends to finalize the New Capital Accord in the fourth quarter of 2003 such that it can be implemented in each country at year-end 2006. During this three-year period, banks and supervisors are expected to adapt and develop the systems and processes necessary to conform to the standards of the New Accord.
## What supervisors are doing to prepare
Just what will supervisors do during those three years? Let me conclude my remarks today with some thoughts on how the Committee and other supervisors will complete the journey to the New Accord. There are three areas of focus for us.
## 1. Evaluating banks' readiness
First, the member countries of the Basel Committee have begun to survey their banks' preparation for the New Accord. Some are gathering information on how banks plan to benefit from the new framework, while others have already evaluated selected banks' internal rating systems and databases to understand how close those banks may be to meeting the stringent requirements associated with the more advanced approaches. In all cases, we are working to understand the challenges that banks will face and are seeking ways to smooth the transition to the new framework.
## 2. Training and preparing supervisory staff
Likewise, supervisors in all of the Basel Committee's member countries are laboring to make sure that they, too, will be ready for the new rules, which is a second area of focus. From the supervisor's perspective, the New Basel Accord will require a heightened sense of the internal processes used to identify and quantify risk. This, in my mind, focuses our energy on exactly the right place - that is, how we can encourage improvements in risk management by critically evaluating current practices.
Our supervisory assessments will encompass several relatively new elements - importantly, validating the integrity, accuracy and consistency of the bank's process for assigning internal credit ratings. Since those ratings will feed directly into models used to manage credit risk, we will scrutinize more than ever their accuracy and integrity. If the initial inputs are wrong, the whole capital allocation system becomes faulty.
Additionally, we will cultivate further our staff's ability to evaluate advanced risk measurement and management methodologies for operational risk, especially because so much of that work centers on advanced statistical analysis.
Externally, the Committee anticipates providing more assistance, both formally and informally, to other supervisory agencies that are preparing for the added responsibilities we will all assume as the New Accord comes into force. Moreover, I am pleased to note that the Financial Stability Institute will continue its collaboration with the Basel Committee, especially in assisting supervisors globally in understanding and implementing all aspects of the revised Accord. The "FSI" anticipates that more than half of its 40 seminars and programs in 2003 will concentrate on components of the New Accord. In fact, shortly after the release of the Third Consultative Document next year, the FSI intends to offer three special seminars around the world to introduce other supervisors to the most important features of the new framework.
## 3. Working toward consistent implementation: Accord Implementation Group
The third area of activity will be to deepen the level of communication and cooperation among supervisors so that we implement the new rules more consistently across jurisdictions. A "level playing field" demands that international competition be driven by each bank's strengths, rather than by each country's rules.
Within the Basel Committee itself, we've established the "Accord Implementation Group" as a forum for sharing information on ways to assess risk management processes under the New Accord. The group is made up of senior line supervisors - that is, the people who actually assess banks' risk profiles - and is led by Nick LePan, the Canadian Superintendent of Financial Institutions.
## Applying the Basel Accord in the non-G-10 countries
Many other countries are similarly preparing for the final leg of our journey, and I applaud your persistence and dedication to this endeavor. At the same time, the Committee recognizes that some countries have only recently adopted the original 1988 Accord and may still be working to ensure a basic level of capital adequacy.
I would like to take a moment to address the supervisors of those countries and emphasize that the Basel Committee's efforts to revise the Accord are focused on the international banks that compete in the global capital markets and that have the greatest potential to create systemic risk. The need for a common approach to capital adequacy regulations for such banks is what drove the Committee to create the 1988 Basel Accord.
As I have noted, however, the original Basel Accord has in some sense become the victim of its success, having been adopted in a number of countries and applied to entire banking systems, including banks both large and small. Now, as the Basel Committee is clearly moving away from a single approach to capital, it is important for the Committee and supervisors worldwide to reflect on the choices available to regulate the capital of banks that are neither highly complex nor internationally active.
I believe that the differences between domestic banking systems make it impossible to conceive of a single capital Accord that would be appropriate for all banks in all countries. Rather, supervisors may need to consider different approaches. For some, retention of the current Accord, supplemented by
the second and third pillars, may be the best way forward. Other countries may elect the revised standardized approach of the New Accord. Still others may seek a hybrid of these two possibilities.
We should recognize that these might be perfectly valid choices for the treatment of non-complex domestic banks, depending on national circumstances. Attempts to force a uniform approach on such banks that typically do not compete across national borders are unnecessary and possibly counterproductive. Instead, our energy would be better spent on sharing ideas and developing a common understanding of the challenges that supervisors face in dealing with noncomplex domestic banking organizations.
Accordingly, at its meeting yesterday, the Basel Committee agreed to launch an initiative jointly with supervisors from many other countries to consider these issues and to develop practical guidance for supervisors on the issues associated with the capital regulation of smaller, non-complex banks.
Let me be very clear. The purpose of this initiative is not to develop one or even several specific "Basel-approved" approaches to such regulation. Rather, we wish to provide practical assistance to supervisors confronting the challenges ahead to develop an appropriate regulatory framework suitable for their national circumstances and domestic banking systems. We wish to support your efforts to enhance the adequacy of capital in a manner that conforms with your institutions' risk profiles, strengthens the management of risk, and fortifies the stability of your financial systems.
## Concluding thoughts on the New Accord
Clearly, all of us have come a long way since 1988, regardless of whether we were the first or simply the most recent supervisors to adopt the original Accord. As we develop new approaches to capital adequacy, none of the milestones we've achieved, and none of our success in resolving open issues, would have been possible without the dedication that you, the members of the global supervisory community, have demonstrated. You have been enormously supportive of our work and have undertaken considerable pains to share your views and ideas. The members of the Committee and I are particularly grateful to our friends and colleagues in the rest of the supervisory community for your advice and counsel. We know that none of us will be able to claim individually the success of the New Accord as our own.
Instead, it is dependent on all of us to work hard and promote the virtues and benefits of the New Accord in our own countries. Press releases and some encouraging words from supervisors will not suffice. Success will demand our mutual and dogged determination to encourage our banks, to prepare our staff, and to cooperate with each other. As we complete the final miles of our marathon journey, I am confident that the New Accord will be worthy of our efforts.
Thank you.