The new Basel Accord: Competition v Co-operation

Posted 29 Jun 2001 at 23:05 UTC by ringbark Share This

The new Basel Accord for Capital Adequacy runs to over 500 pages and has had 270 submissions commenting or criticising particular aspects of the new Accord. Most of those commenting are presenting, in a public arena, information which may be of advantage to their competitors. Why have they done this?

Capital Adequacy relates to the amount of capital a bank must have as a buffer to cover them in the event of unexpected loss. Under the terms of the 1988 Accord, this must not be less than 8% of the value of the financial institution's risk weighted assets. In general terms, lending to sovereigns denominated in their own currency (e.g. holding of US Govt bonds denominated in USD) attract a 0% weighting, residential mortgages backed by land attract a 50% weighting and everything else attracts 100%. This system is reasonably straightforward and works well up to a point, but it has considerable deficiencies which the new Accord seeks to address.

In general, a Bank wants to minimise the capital that they need to set aside because it is money that they cannot use and for which they incur a cost. On the other hand, a bank wants to stay in business and have investors with confidence that they will not fail. These two goals pull the policy of a bank in opposite directions.

The complex systems in the new Accord, which now offers three methods for calculating capital adequacy (Standardised, Foundation, Advanced) instead of the one in the 1988 Accord, result in different levels of required capital. On the other hand, local regulators (Federal Reserve, FSA, Reserve Bank of New Zealand, etc.) need to be satisfied before a Bank is allowed to use one of the more sophisticated models.

In the 270 or so submissions which have been received, issues have been raised by a great many financial institutions from all over the world. Many of those who have responded to the request for submissions are direct competitors to some of the others, and raise issues or make suggestions which, if implemented, would result in reduced requirements for capital.

Unfortunately, this will also mean that capital requirements for their competitors in similar markets will probably also fall, possibly by a larger amount. The shape and quality of Bank lending portfolios is disclosed up to a particular point as required by disclosure legislation, but the detail is kept confidential within the different banks. This makes it very difficult to be sure that different banks are interpreting Acts and Accords in the same way, making comparison very difficult.

The submissions, pointing to both major and minor defects in the proposed new Accord, are so numerous that the Basel Committee for Banking Supervision has this week delayed implementation of the Accord by one year to 2005.

The principle of the submissions providing information to the Basel Committee being beneficial to the Bank but also to their competitors is reminiscent of the Prisoner's Dilemma, long debated by philosophers and mathematicians. Certainly I will struggle to find the time to read all the submissions received. Many of the submissions run to dozens of closely argued pages, so I have focussed on the submissions by banks which are competitors to my own organisation. They operate in the same market, face largely the same issues and (we assume) have largely similar lending and loss experience to our own. Reading these has shown me that this seems to be the case, but they also raise issues which I had not necessarily noticed or previously deemed material.

urmmm..., posted 2 Jul 2001 at 13:38 UTC by matt » (Journeyer)

What does this have to do with free software? Maybe I'm just being dense...

Underlying concepts, posted 2 Jul 2001 at 18:40 UTC by ringbark » (Journeyer)

Although there's very little directly about free software, the concepts are the same. Should companies hold onto their own intellectual property in the shape of credit risk data and information, or should they put it into the public domain and let everyone benefit from what they are learning.

A related software issue is: will anyone produce open source software for assisting the calculation of the different key pieces of data in the new Accord?

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!

Share this page