This statement is also available here as an Adobe PDF. 

Statement No. 248
For Information Contact:
Richard J. Herring
215-898-5612
Robert E. Litan
816-932-1179

After a decade of deliberation and negotiation, the Basel II framework for capital adequacy is about to be implemented in 2008 in Europe and much of the rest of the world.  Following agreement among the four federal banking regulatory agencies on July 20th of this year, the United States will phase in Basel II with a short lag for eleven large or internationally active "core banks." (The date of implementation for the remaining banks is uncertain.)

Basel II grew out of concern that the prior Basel I structure failed to address the new and widening range of risks of various bank activities.  Basel II is built on three Pillars:  (1) minimum regulatory capital requirements for credit risk, operational risk and market risk; (2) the supervisory review process; and (3) market discipline and disclosure.  The minimum capital requirements are determined by either external ratings from ratings agencies for smaller banks or by outputs from the larger banks' own internal ratings models.

The turmoil triggered by problems in the subprime mortgage market in the United States has challenged the Basel II framework in several important ways. 

First, these events raise significant questions about the wisdom of setting capital requirements based on external ratings. Ratings of structured debt based on subprime mortgages have proven to be excessively optimistic and slow to adjust to adverse market events.  Although the ratings agencies in the past frequently have been wrong and slow to adjust their ratings for individual securities issued by particular firms or countries, they have recently overrated an entire class of securities, underscoring the need to worry about the systemic impact of ratings agency errors.  This has rightly undermined confidence in the credibility of agency ratings generally and has reconfirmed longstanding concerns about the conflicts of interest arising out of issuers paying the agencies for their ratings.  By tying capital requirements to external ratings, regulators discourage banks from making their own independent evaluations of risk and amplify the consequences of rating agency errors. 

Second, recent events raise similar concerns about regulators using outputs from banks' internal models to set minimum capital requirements.  As the subprime correction unfolded, it became clear that even the most sophisticated model-builders "got it wrong."  They lacked sufficient data observed over an appropriate variety of macroeconomic conditions to develop reliable models, and the models themselves may not have been adequate. Almost every financial model is subject to these problems.  Recent events also have revealed difficulties in modeling liquidity risk.  Indeed, liquidity risk intensified credit risk and market risk.  Suspicions have also surfaced that some firms may have used different (and more favorable) models to value their own securities positions for reporting purposes than the models they used to value similar assets when offered as collateral by counterparties. This raises questions about whether complex financial models can be manipulated to provide desired outcomes.

Third, the turmoil in subprime lending has spread to other securities and has persisted for several reasons: inadequate documentation of the underlying mortgages, as well as the opacity of both the securitized instruments that allocate their cash flows and the investment vehicles that purchased those instruments. Sometimes these asset-backed securities are themselves sliced and diced and their cash flows resold. This process weakens the link between the source of risk and the bearing of loss. Indeed, regulators have long decried their inability to know how risk is ultimately redistributed.  The turmoil reveals that private market participants are also in the dark.  The disclosures envisioned under the market discipline provisions of Pillar 3 do not address this fundamental issue of the transparency of risk transfers.

The Shadow Financial Regulatory Committee has long advocated a more effective alternative approach to prudential regulation for banks: one that combines supervisory and market oversight. This better approach entails prompt corrective action in the event of financial problems and, for large banks, the mandatory issuance of subordinated debt, which would add market-generated information to risk assessment.  Investors are quicker to recognize changes in risk and risk premiums than are regulators. The recent market turmoil proves the inadequacy of Pillar 1 capital charges and the inadequacy of the Pillar 3 disclosures, if insufficient at-risk claimants and market discipline are present.  This places a much heavier burden on the Pillar 2 supervisory oversight.  Supervisors would be greatly aided if they could draw on market discipline exercised through the pricing of regularly issued uninsured subordinated debt.