Euro-MTNs remain predominantly an investor-driven market. The publication of the latest proposed update of the Capital Accord from Basle (the New Accord) potentially affects a significant number of investors. Therefore, as risk weightings change, both investors and issuers will have to modify their approach to the market. The New Accord seeks to modernize the original Basle Capital Accord, which was published in 1988. Since then the world of banking has changed substantially, especially in the area of risk management. The New Accord acknowledges this by shifting the focus away from pre-defined risk buckets towards an internal ratings-based approach. Whereas this route was initially envisaged to be open only to major sophisticated banks, it is now likely that most reasonably competent banks will be able to take advantage of it. Indeed, banks will be encouraged to develop and use a more sophisticated internal approach by virtue of requiring less capital than based on the standardized approach. Although some banks might well be required to hold less capital in respect of their credit risks, the chances are that this will be largely offset by the new requirement to hold capital against operational risk, defined as "the risk of direct or indirect loss resulting from inadequate or failed internal processes, people and systems, or from external events." The New Accord anticipates that operational risk will constitute on average approximately 20% of the overall capital requirement of a bank. As a result we do not expect that banks in general will face either a material reduction or increase in the amount of regulatory capital that they will be required to hold. The New Accord is very clear on this point. The Committee has strongly reiterated its aim that the overall level of capital in the banking system should neither increase nor decrease, and intends that a bank with an average risk profile should not see its overall capital requirements change significantly. One possible exception to this might come on the supervisory side where the Committee wants individual regulators to have the power to require capital levels in excess of the minimum 8%. This is currently the case in the UK, but hardly anywhere else. This could mean that some banks will be required to find more capital, although most major banks already operate comfortably above the minimum level. Although the focus will shift towards an internal ratings-based approach, the New Accord does contain a revised standardized approach, based on the familiar risk buckets. As expected from the June 1999 draft proposals, this seeks to introduce a greater degree of granularity primarily by introducing the use of external credit ratings from what the accord refers to as external credit assessment institutions (ECAIs), essentially the rating agencies, but, in the case of sovereign exposures, potentially also export credit agencies. The New Accord still divides the world of credit into major counterparty classes: sovereigns, non-central government public sector entities (PSEs), multi-lateral development banks, banks and securities firms, corporates and asset securitizations. These are broadly in line with the June 1999 proposals. The one major change is the introduction of a 50% risk weighting for single-A rated corporates. There was no logical rationale to the original jump straight from 20% to 100%. Another change both from the June proposals and the current regime is a more preferential treatment of multi-lateral development banks (MDBs). Provided they meet certain criteria, including a triple-A rating, MDBs will migrate from the current 20% to 0%. The New Accord specifically names those MDBs that will (from the current perspective) qualify: the World Bank Group including IBRD and IFC, Asian Development Bank, African Development Bank, European Bank for Reconstruction and Development, Inter-American Development Bank, European Investment Bank, Nordic Investment Bank, Caribbean Development Bank and Council of Europe Development Bank. The immediate significance of these improvements in risk weightings for the bonds of certain lower-risk issuers (i.e. higher-rated corporates, higher-rated ABS tranches and MDBs) is likely to be very muted, and will vary depending on residual maturity and the extent to which banks are already active buyers of an asset class. Other things being equal, we would expect a gradual but small degree of spread tightening on these assets as the 2004 implementation approaches, with longer residual maturities having the most potential. Even without these changes in the standardized approach, it is reasonable to assume that an internal ratings-based approach will also favour lower-risk assets. The retention of two options for banks is slightly surprising, and is one of the best indicators of the degree to which this whole process is political as well as regulatory. Option 1 derives the risk weighting of a bank from that of its sovereign of domicile. Option 2 derives it from the bank's own external credit rating. Although not stated, the only logical argument in favour of option 1 is that all banks constitute a contingent liability on their domestic governments (and ultimately taxpayers). Although there are numerous examples of banks being bailed out by governments, this is not the sort of thing that regulators like to make explicit, and in any case we do not believe it should apply to all banks in a system. After all plenty of banks (mostly smaller ones) have also been allowed to fail. On the face of it, those regulators with lots of small unrated banks in their system (Germany, Italy, France) or some large lower-rated banks (Japan), have a clear incentive to choose option 1, under which all their banks will retain 20% risk-weightings. This proposal, however, appears to have one major flaw: it potentially contravenes basic European Union law, which precludes discriminating between entities on the basis of their EU nationality. Given that the New Accord will find its way into EU national legislation via an EU Directive, this is indeed a major problem, and option 2 will have to be adopted. The impact on banks' cost of funding will be ameliorated to a certain extent by the provision of a 20% risk weighting for short-dated (three months or less) exposures, typically found in the inter-bank market. Furthermore, a widely applied internal ratings-based approach should also have a mitigating effect, if individual banks can demonstrate to their regulators that losses on exposures to other banks are very low. One other area of interest arises from the Committee's evident concern about banks' exposure to non-bank equity investments, which it says "have the potential to pose significant risk to a banking group since they may create incentives to bolster the financial condition of the commercial enterprise". Under the New Accord, such significant minority and majority investments, subject to materiality levels (15% of capital to one investment and 60% of capital to the aggregate of such investments) will be deducted from the bank's capital. Investments under the materiality threshold will be weighted no lower than 100%. This could accelerate the existing trend amongst the major German banks to dispose of their industrial participations, and trigger a similar process in other jurisdictions. The New Accord covers many details, not all of which can be addressed in a short article like this. But overall we believe that the New Accord is a significant step forward in the regulatory framework under which banks operate. Although much of the simplicity and transparency of the current accord will be lost, this is to be replaced by a far closer correlation of the regulatory capital banks are required to hold with the risks that they undertake. Combined with closer supervisory scrutiny and better disclosure, the other two pillars of the New Accord, this should ensure the continuation of banking's special status as the world's most regulated industry and its relative stability as a source of high quality debt products for investors. While the final implementation date in 2004 seems a long way away, some investors are starting to modify their investment criteria already. For example some banks (as investors) will only buy single-A rated banks for three years or shorter, to ensure they will be 20% risk weighted. Others are now demanding a premium for longer maturities to compensate for the potential change in risk-weighting. We have seen a similar pattern in the ABS market ahead of the proposed changes in that asset class. So far, what we have not seen is risk-weight sensitive investors buying highly-rated corporate paper in anticipation of a potentially beneficial reduction in risk weighting. However, as implementation draws nearer this development will also become inevitable. But investors will need to be aware of the potential impact of ratings downgrades, which tend to be more prevalent in the corporate sector (for example the current situation with telecoms ratings) than in the bank or ABS sectors. While the changes are still proposals, it is only at the margins that we are seeing an influence. But it does mean that we are heading for an exciting time for both investors and issuers.
September 14, 2001
