Comment: Banking mistakes will come around again

Sounding like Eeyore, the lugubrious friend of Christopher Robin for whom good news is always another excuse for gloom, has long been part of a central bank supervisor's business. And actually it makes perfect sense for anybody who has had a bird's eye view of the banking industry over several economic cycles. Come what may, clangers will be dropped as growth gives way to recession, and cast iron propositions agreed during a lending boom turn out to be as full of holes as a colander.

Another inevitability is that during the upturn, lots of clever people deliver rational arguments that say the cycle has changed, banks have learnt and the next downturn will be less ferocious in character.

Michael Foot, the Bank of England's executive director in charge of supervision, gave another timely reminder yesterday that this is the point in the cycle at which, in the worst cases, banks think they can walk on water.

Some of our biggest banks, organisations that in theory make a living out of judging risks correctly, have in the past been very bad at doing just that.

But there has been progress since the lending howlers exposed by the last recession. These days banks have much better systems for pricing and assessing risks. The old approach was often not much better than sticking a thumb in the air to see which way the wind was blowing.

Even so, statistical analysis can itself contain traps for bankers trying to look forward at the likely impact of the next downturn. This is especially the case if the assumptions on which their work is based are flawed. There are also important areas of lending, such as to large companies and to governments, where subjective judgements will always be required.

Even in a world of low inflation growth there is still plenty of scope for bad lending. Fraud, technical innovation - which can make some markets disappear altogether - and just the simple lending error all await the uncautious banker. It is quite possible that the next cycle of banking mistakes will be less ferocious than in the 1980s, but it is hasn't gone for good. Of that we can be certain.