The idea put forward by Bank of England deputy governor Sam Woods in your editorial “Sieving the alphabet soup of banks’ capital requirements” deserves consideration (May 2). The current framework is not easily reformed as it rests on too many assumptions, and too many models providing opportunities for cherry-picking and tweaking. Change is needed.

The wisdom of the original plan relied on three “pillars” of capital calculations. All three start with historical default data which often have either insufficient detail or modelling risks. Add to this the hypothetical portfolio correlations and shifts during times of stress and it could be argued we have an improved but overly engineered mechanism. It would not, for example, be the way to programme an aeroplane’s computer.

But retaining the obligation to collect this data, which Basel relies on for risk management and supervisory scrutiny, simplifies the process.

Capital should be informed by default data, not be subservient to often inadequate data and complex models.

John Pattison
Toronto, ON, Canada