Ms. Jennifer J. Johnson
Secretary
Board of Governors of the Federal Reserve System
20th Street and Constitution Avenue, NW
Washington, D.C. 20551

Basel Committee Secretariat
Bank for International Settlements
CH-4002
Basel, Switzerland

Dear Ms. Johnson and Basel Committee Secretariat:

The Association for Financial Professionals (AFP) welcomes this opportunity to forward to you our comments on "A New Capital Adequacy Framework," the consultative paper issued by the Basel Committee on Bank Supervision of the Bank for International Settlements in June of 1999. This paper presents proposed approaches to revising the Basel Committee's 1988 Accord that developed a common capital adequacy framework for Belgium, Canada, France, Germany, Italy, Japan, Luxembourg, the Netherlands, Sweden, Switzerland, the United Kingdom and the United States.

AFP represents approximately 15,000 finance and treasury professionals who work closely with banks in arranging for credit and other financial services for more than 5000 organizations they represent. These organizations are generally drawn from the Fortune 1000 companies and the largest middle market companies. Many have extensive international operations and deal with both foreign and U.S. banks. We maintain information on the job responsibilities of our members. Although they encompass virtually all areas of financial management, over 3400 of our members indicate that their most important job responsibilities have to do with borrowing and bank relationship management.

Our members are interested in the Basel Committee's proposal because they have a strong stake in the preservation of economic stability, the maintenance of safe and sound banking systems, and the preservation of a variety of options for the purchase of financial services at the lowest possible cost. We believe the Basel Committee's proposal, and the directions it indicates for the future of bank capital regulation, offer significant promise that these objectives will be achieved in the future.

The Basel Committee's consultative paper presents three "pillars" which it envisions will constitute three different sets of tools to assure capital adequacy: (1) minimum capital requirements, including risk weights for different types of assets; (2) supervisory review of capital adequacy; and (3) market discipline. Although we endorse the use of all three pillars, we will confine our comments to the first pillar, and the overall framework of the Basel Accord. These areas have the most immediate impact on our members.

Minimum capital requirements are a necessary and useful tool for bank regulators to protect the government's interests. These include maintaining economic stability through the maintenance of a sound banking system and preserving the value of the deposit insurance fund reserves. The AFP recognizes these roles of bank capital regulation and supports vigorous pursuit of them. However, we hope the committee will also recognize that there are negative affects that would result from setting the capital requirements too high. The share of financial services provided by banks has been declining for some time. Recent technological changes, particularly the advent of the Internet, are likely to cause this trend to accelerate at an even faster rate.¹  It is our belief that the U.S. economy as a whole is best served by a strong financial services marketplace in which both banks and non-bank financial institutions remain viable competitors. In view of the ways in which the services offered by banks and non-bank financial institutions can substitute for each other, this goal will only be achieved if minimum capital requirements are high enough to protect the stability and soundness of the banking system, but not so high as to handicap banks in providing service to their customers.

The fact that more and more financial activity is taking place outside of the banking system means that future sources of financial instability are also more likely to come from outside the banking system. The extension of capital requirements to entities outside the banking system is probably not a reasonable alternative. However, in light of the trends discussed above we hope regulators are taking account of the effects of bank capital requirements on the relative position of banks in the financial services industry and the increased importance of using more broadly based tools in conjunction with bank capital regulation to promote economic stability.

The consultative paper does not adequately address how the minimum level of capital requirements is to be determined. Presumably, it is being set with the overall level of bank risk in mind. However, adjustments are then to be made for relative levels of risk. We would feel more comfortable about the relative adjustments if we were certain that they were not being made to account for risk that has already been factored into setting the minimum level of capital requirements.

The consultative paper proposes to change the way in which minimum capital requirements are adjusted for risk. The original Basel accords of 1988 entailed adjustments for the risk of different types of loans. If implemented, the proposed system would, for the first time, entail adjustments among different risk classes for commercial loans. We endorse this positive step forward. In any system that involves making adjustments for risk there will inevitably be some degree of arbitrariness and inaccuracy. However, if one starts with the presumption that an accurate measurement of risk and making adjustments for it are the ideal system, the systems being proposed will surely result in an improved system. Companies should experience fairer loan pricing. Well-managed companies will receive discounts for being better credit risks. Poorly managed companies will pay premiums.

The consultative paper suggests two ways such risk adjustments might be made. One is to specify that adjustments be made in the minimum capital requirements based on ratings the borrowers receive from an external rating agency. Under this method, risk "buckets" are created into which companies are placed. Each bucket corresponds to a rating of the rating agency. Depending on the bucket into which the company is placed, its required minimum capital ratio will be increased or decreased a specified amount. The second method envisioned allows banks to create their own internal rating system. This method would also involve the creation of risk buckets into which borrowers would be placed, with corresponding additions to or subtractions from the minimum capital requirement associated with each risk bucket.

Relying on the rating agencies for risk assessments has definite advantages. There are several widely respected and well-recognized rating agencies. Harnessing their expertise could be productive in achieving better capital regulation. However, there are certain problems with this proposal that should be recognized. Regulators must bear ultimate responsibility for the risk assessments made by the rating agencies. There must be an approved list of rating agencies that companies can use to get ratings. If this option is widely used, it will increase the demand for ratings by private companies. Since companies are charged fees to obtain a rating, this may prompt the entry of new companies into the rating business. Regulators should develop standards for the approval of new entrants into the rating business for purposes of bank capital regulation. If no new entrants are approved in the face of increased demands for ratings, the regulators will have created a valuable monopoly for the existing rating agencies. The fees for ratings would probably go up in such circumstances. On the other hand, if no standards are developed for the approval of new and existing rating agencies for purposes of bank capital regulation, and new rating agencies are automatically approved, bank regulators will lose control of the capital regulation process for which they are ultimately responsible.

The system of relying on internal ratings systems of the banks themselves also has advantages and disadvantages. Internal rating systems will probably be used mostly at larger banks. They require substantial work and thought to create. Smaller banks may not be able to afford the development costs that must be incurred in establishing such a system. For those banks that have the wherewithal to establish such a system, there are substantial advantages both for the banks and their customers. These can use internally generated information that is relevant to risk assessment but is not available to external rating agencies. The ability to create a finer gradation of risk buckets means that banks will have more precise discounts or premiums from the minimum capital requirement. This in turn should mean a system of loan pricing that is fairer and more precise in rewarding or penalizing customers for the way in which they manage their risk.

It is important to have fair standards for internal ratings systems and to have a system of monitoring such systems over time to be assured those standards are being maintained. This should be an integral part of the bank supervision process (Pillar II). If this is not done, regulators may be surrendering control of the bank capital regulation process for those banks that use internal rating systems.

In the proposed approach of placing individual borrowers into risk buckets and adjusting capital requirements for risk accordingly, regulators are ignoring the overall risk of a credit portfolio which could be quite different than the average of the individual risks of the borrowers. If probabilities of default for individual borrowers have low correlations with each other, the overall risk of the portfolio could be substantially less than the average of the individual risks. The overall risk of the credit portfolio is what is most important in setting capital requirements to protect the government's interests. If the overall risk of the portfolio is low then banks, and their customers, should receive benefits from this low level of portfolio risk. The consultative paper states that the reason the assessment of portfolio risk was not incorporated into the proposals presented in the paper is because methodologies for assessment of it are not yet advanced to the level where one could have confidence in such assessments. It is our hope that regulators, and the banks they regulate will continue working on this problem so that a more efficient and fairer system of bank capital regulation can be achieved.

As should be evident from our discussion above, both the internal and external ratings approaches have potential benefits and risks. We recommend both systems be implemented together. The potential risks of the two different approaches would be offset by such a policy. If the fees for ratings did become excessively high because of the demand for the services of an external rating company, the existence of companies that used an internal ratings system would constitute a useful alternative for many companies. Conversely, the existence of the external ratings option for banks that did not have the ability and resources to construct an internal rating system would give companies a much wider array of banks with whom they could negotiate loan agreements that take account of the risk charges banks incur to lend them money.

With respect to the use of an allowance for guarantors, the Committee proposes to expand eligible guarantors to those that attract a lower risk weight than the borrower.  We support this change. It gives borrowers more options to influence the capital burden banks must bear for lending to them.

In the original 1988 Accord, the committee gave zero or low risk weight to loans that were collateralized by cash, securities of an OECD central government or public utility, or a multilateral development bank. The Committee proposes to expand eligible collateral to all financial assets -- not just marketable securities -- that attract a risk weight lower than the underlying exposure. We support this change and hope the Committee will be able to go further in the future. The use of collateral, and guarantors, is as old as banking itself. Unlike newer credit risk mitigation techniques such as credit derivatives, supervisors in every country covered by the accord have had experience in assessing collateralised loans. We hope a system could eventually be developed to give allowances for readily marketable collateral, whether financial or non-financial.

We would like to comment on the basic thrusts of the efforts of the Basel Capital Accords. We view these thrusts as a desire to achieve uniformity in the burden of bank capital regulation among different countries so that one country's banks will not have a competitive advantage over those of another, and a desire to upgrade the quality of bank capital regulation by accounting for risk. We endorse both goals. In achieving these goals we hope the Committee will take account of the different institutional situations in different countries. Banks or regulators in some countries may have access to information or tools that would improve the quality of regulation that may not be present in other countries. These sets of information or tools should be used as long as their use does not change the relative burden of the regulatory system among different countries.

The Committee has already used this approach in its proposal to use both external and internal rating systems. The use of external rating agencies is quite common in the United States, Canada, and Great Britain. However, their use is very rare in continental Europe. The fact that rating agencies are not used in every country covered by the Basel Accord does not mean they should not be used as a regulatory tool. Their use as a regulatory tool in countries where use of external ratings is common offers the potential to improve the quality of regulation. We do not believe their use in countries where they can be used would put other countries at a competitive disadvantage. These countries will still have the benefit of internal ratings systems. In addition, if the use of external rating systems does prove to be a successful regulatory tool, such a demonstration may encourage non-external rating countries to develop such a system. 

We hope this will be the philosophical approach the committee adopts as it develops other regulatory tools.

Finally, we hope the Committee will recognize that, although there will only be twelve signatories to the agreements it develops, the agreements will ultimately be used by a much larger group of countries. More widespread usage of the agreements reached by the Committee will promote greater financial and economic stability worldwide and will promote fairer treatments of banks and their customers. This usage is particularly important to our members who do business worldwide with many institutions in many countries. We urge the committee to work with other international agencies to find ways to proactively promote the use of these standards.

In sum the Association for Financial Professionals believes:

• The use of standards such as those being developed by the Committee promote international financial and economic stability and fairer treatment of banks and their customers in the increasingly important international environment in which they operate;
   
• Consideration needs to be given to the rationale behind the development of the minimum capital requirement that must be adhered to by all banks before adjustments for risk are made. This needs to be done to adequately account for risk, assure economic and financial stability, and assure that banks continue to be competitive players in increasingly diverse and competitive financial services system;
   
• The use of both external rating agencies and internal rating systems to make risk adjustments is a positive approach which offers significant potential benefits for both banks and their customers;
   
• The use of both external rating agencies and internal rating systems needs to be monitored by bank regulators to assure that treatment given to different types of bank customers is consistent across different systems and that bank customers have a wide variety of choices in dealing with different banks from whom they can receive the benefits of proper risk management within their own organizations;
   
• The proposed expanded use of allowances for collateral and guarantors is a positive step; we hope its scope can be expanded in the future;
   
• It would be helpful if the standards could be expanded to take into account the overall risk of the credit portfolio as well as the risk of individual borrowers;
   
• Where the institutional situations in different countries allow different methods to be used to improve the quality of regulation, they should be used as long as the relative regulatory burden among different countries is approximately equalized;
   
• The standards being proposed by the Basel Committee will be a positive force for the maintenance of international financial and economic stability and fairer treatment of customers worldwide. Their use by countries outside of the Basel Committee's membership should be proactively encouraged by the Bank for International Settlements and other relevant international agencies.

Sincerely,