February 23, 2009
My faith in government institutions and the people working for them has been restored by Martin Wolf of the FT when he pointed out an excellent paper "Why Banks Failed the Stress Test" by Andrew Haldane of the Bank of England. Reading this is a complete contrast to my experience at the FSA presentation on stress and scenario testing the other week (see earlier post).
The paper ends by putting forward five proposals for improving risk management:
- Better Scenario Definition – Regulators defining multi-factor scenarios for the industry that are truly representative of extreme tail events.
- Regular Scenario Evaluation – A common set of scenarios evaluated and reported upon to the regulators on a regular basis.
- Second-Round Stress – Making sure that the consequencies of stress testing for individual institutions can be evaluated for system-wide risk.
- Active Management of Risk – Ensuring that management take and can explain actions that provision for the risks identified, and do not simply passively report on risk levels.
- Transparency – Access to institutional stress testing results by regulators and potentially by the market as a whole through annual report and accounts.
In addition to solid content, Andrew Haldane writes a good story, and I love the usage of "regulatory camouflage" in the serious point below:
"…is that stress-testing was not being meaningfully used to manage risk. Rather, it was being used to manage regulation. Stress-testing was not so much regulatory arbitrage as regulatory camouflage."