Basel II Developments in the United States

I would like to thank the Institute of International Bankers for the opportunity to offer some remarks today. I am honored to be on a panel with such esteemed colleagues. My remarks will touch on the latest developments with the Basel II capital revisions, including the U.S. implementation process and progress on home-host issues.

Background
As some of you know, we are in our seventh year working on Basel II, and still several years away from full implementation. Some might bemoan this long formative period, but we must all remember that an international undertaking of this magnitude takes substantial time to complete. Part of the reason for the long gestation period is that we have been purposely transparent about the proposals, taken substantial time to listen to industry and other public comments, and worked diligently to make the framework better in response to those comments. In addition, this is an international agreement for which a certain "common denominator" had to be found to accommodate the variety of banking and financial systems involved. From the perspective of the Federal Reserve, we consider the work to be well worth it, given the objectives.

As you know, in the United States we have decided to implement only the advanced approaches of Basel II and make it mandatory for only the largest, internationally-active banks. We are moving along a parallel path to amend current Basel I rules for the vast majority of U.S. financial institutions. Broadly speaking, in developing Basel II we are striving to establish higher standards for internal risk management at complex banking organizations, including capital adequacy, and to improve both the supervisors' and the public's understanding of banks' risk taking and risk management. Indeed, one of the main reasons we started down the Basel II path was to better align regulatory capital with what banks are actually doing in practice to manage risks and the level of risk exposures. Over the past two decades, major banking organizations have become ever larger and more complex; most national banking systems look quite different than they did nearly two decades ago when Basel I was developed. Market changes and increased sophistication in risk-management techniques require that we now update that initial Basel framework, as it applies to the largest institutions. A fundamental premise of Basel II is that, for these major banks, neither supervisory nor market discipline can be effective unless banks' own systems can be relied upon to measure and manage risk taking and capital adequacy.

One sometimes hears that the advanced approaches of Basel II are "too complex" for anyone to understand, and the mathematical formulas in various drafts of the framework can look like a foreign language to some readers. But remember the complexity in Basel II reflects today's sophisticated risk-management practices that employ advanced quantitative methods. Most of you in this room understand that mathematics and statistics are an integral part of how you manage your businesses. Thus, it is only sensible to implement a framework at our largest institutions that is in line with what bankers themselves are doing. And since those practices change rapidly, it is necessary to have flexibility for incorporating new and better practices without having to conduct fundamental restructuring. I believe the framework provides a useful and credible basis for improving risk-management practices today and allowing for future improvements. We consider this vitally important because banking will remain a highly dynamic industry. Supervisors will have to be especially attentive to changing practices and ensure that Basel II does not inhibit adoption of new banking practices and financial instruments.

Basel II Developments in the United States
You are most likely aware that over the past several months the U.S. banking agencies have been conducting additional analysis of the results from the fourth Quantitative Impact Study--known as QIS4. As the interagency statement issued on April 29 indicated, results from QIS4 were more widely dispersed and showed a larger overall drop in regulatory capital requirements for the QIS4 population of banks than the agencies had expected. The agencies' reaction to the QIS4 results, taking additional time to understand and react to the information provided by the banks, shows how seriously we are taking Basel II implementation. This was the impetus for deciding to delay the notice of proposed rulemaking (NPR) for Basel II. Analyzing the data used in QIS4 is vitally important, because ultimately the success of Basel II will depend on the quality of data and models that banks use to implement the framework.

I would like to point out that QIS4 was the most recent in a series of assessments about the framework's potential impact. In conducting QIS4, we understood that banks lacked full information about what will eventually be expected of them--one example is lack of precision about downturn estimates for LGD (loss given default). Understandably, the results were less than perfect. But we commend the bankers for their extensive efforts and thank them for their patience in producing results under less-than-perfect circumstances. In the future, we expect to have better assessments of the framework's impact on U.S. banks before it is fully implemented. And as we move forward in this process, we expect that additional information provided by U.S. supervisors--through rulemaking and guidance--would allow bankers to produce regulatory capital results that are more accurate and likely more in line with our expectations.

One of the most important findings from QIS4, obviously, relates to the overall decline in regulatory capital requirements. Capital serves as an important backstop against risk-taking and we need to ensure that an adequate level of capital is produced by the Basel II framework. To be frank, the Federal Reserve would not be comfortable qualifying any bank based on the results of QIS4, if Basel II were to be applied today. But as I noted above, QIS4 does not represent the final version of Basel II in the United States and we realize that bank data and risk-management systems required by Basel II are not yet fully developed and implemented as expected by the framework. To be sure, by the time that Basel II "goes live," bankers would have significantly more information regarding our expectations and more robust default and loss-severity data. Banks will also have to adhere to the qualification standards set by supervisors.

But given some of the uncertainty that still remains about the impact of Basel II, it is important that we continue to effectively use the safeguards designed in the Basel II process. As I just mentioned, QIS4 was one of a series of studies conducted with banks to test the methodology and identify implementation issues. The process also requires banks to successfully complete four quarters of running their Basel II systems in parallel with the capital requirements of Basel I, before they can qualify for the next stage. The current framework then requires banks to demonstrate continued reliability of their process for two years, during which floors on minimum regulatory capital exist to prevent an unintended drop in capital. The public comment process and transparency in our implementation process would provide added checks and balances.

I would also like to emphasize that market participants are also expected to play an important role as Basel II is implemented. Not just supervisors will be expected to evaluate information provided by institutions. As Basel II rolls out, bankers would be expected to disclose where their institution is in its stages of implementation, how the institution assesses and manages risk, and relevant information for third parties to ascertain the effectiveness of the bank's risk management. Pillar 3 market discipline, therefore, will complement supervisory efforts to maintain adequate capital.

So where do we go from here? The U.S. agencies learned some valuable information from the follow-up work pertaining to QIS4 results this summer. Right now we are still evaluating the results from this follow-up work, discussing the options to respond to those findings, and determining how the implementation process will proceed in the United States. Obviously, we understand that the industry and others are interested in details about the next steps in the United States, and we hope to provide a public communication sometime in October. Additionally, the agencies are moving forward on an advanced notice of proposed rulemaking for revisions to Basel I rules, the details of which should be publicly available very soon. We expect that the timeframe to adopt revisions to current rules will overlap with the Basel II proposals, so that the industry can review them side-by-side.

Progress on Home-Host Issues
With such a large group of international bankers in attendance, I would be remiss in not offering a few thoughts on home-host issues relating to Basel II implementation. Understandably, internationally active banks are keenly interested in cooperative arrangements among supervisors for evaluating capital adequacy under Basel II. Several years ago, the Accord Implementation Group (AIG) was formed as a subcommittee of the Basel Committee for supervisors to discuss the tactical issues associated with cross-border supervision. Most recently, and in cooperation with the representatives of the Core Principles Liaison Group (non-Group of Ten supervisors), a joint working group was established to consider enhancing guidelines for information sharing among home and host supervisors. This working group is looking to clarify the roles and responsibilities of home and host supervisors as well as supervisory expectations of banks with regard to communications between their head office and their foreign offices. The working group is also hoping to address communications between home and host supervisors about enterprise-wide risk models and their use in assessing capital adequacy in host jurisdictions, as well as mechanisms for supervisors to work together for consistency and to avoid duplication. Obviously, whatever guidance is issued would be supplemented further as implementation plans and qualification processes become more fully developed.

I do think it is also helpful to remember that a number of the home-host issues pertaining to Basel II are actually variations of long-standing challenges faced by bankers and supervisors in the context of cross-border banking. That is, many of the issues under discussion are not necessarily new, but just made more complicated by the framework. However, we do understand the concerns of bankers surrounding Basel II and how it will actually be applied in various jurisdictions. The good news is that supervisors are already building upon their existing relationships to accommodate differences in national jurisdictions and minimize the burdens placed on banks. The Federal Reserve has a long history of working constructively with other national supervisors. To date, we have met many times about Basel II implementation and how proposed rules and practices might be reconciled. While such cooperation and information sharing can go a long way to reducing the burden placed on bankers, at the end of the day we must still realize that national jurisdictions still matter. We can work to harmonize differences, but it is probably not practical to imagine that they will be eliminated entirely.

Conclusion
I hope that I have offered some useful information today about Basel II implementation in the United States and the Federal Reserve's position on certain issues. Clearly, the QIS4 results have generated some concerns and an element of uncertainty still exists about the full impact of Basel II in the United States. We at the Federal Reserve clearly see some important issues that will require further analysis, but believe they would be best resolved by gathering better-quality data and information that can only come as banks progress in implementation of Basel II requirements. To facilitate that progress, it is important that the United States move forward with the Basel II NPR to provide greater clarity with respect to expectations.

Additionally, QIS4 estimates were not based on the finished models that banks will use to implement Basel II, and several years lie between now and when U.S. banks would operate under Basel II. Perhaps more importantly, the Federal Reserve is unequivocally committed to maintaining adequate levels of capital in our banking system and at individual institutions, in order to maintain safety and soundness.

One of the most important aspects of Basel II pertains to improved risk management, and having banks move forward in that area can only bring benefits. Obviously, in moving ahead we would adhere to the established protocols and procedures for implementing domestic rules in the United States--including additional comment periods and opportunity for the banking industry, Congress, and others to express their views. At this point, we believe, the best way for the industry and others to provide useful comments is to see the agencies' next set of proposals.

Return to top

2005 Speeches