Robert Jenkins: Basel II proved to be inadequate, so are the new rules really 'too severe'?

The pre-crisis Basel rules required less than 1 per cent of loss-absorbing capability for a mortgage with no money down

Do you think the pending changes in bank regulation are sufficient, insufficient or excessive? Banks take risks. When more risks go right than go wrong, the result is a profit. At other times a bank's equity capital absorbs the losses. If there is not enough capital there is one less bank – or one more call to the taxpayer.

Of course some risks are riskier than others. It follows that one should have more capital to support the riskier activities and less for the less risky. This in a nutshell is the basis for the Basel rules governing global banking. Sounds sensible? So how much capital should a bank have to support a given level of risk?

Take foreign government bonds. Do you think they are risky, very risky, or safe as houses?

You answered correctly: some are and some aren't. Greece is riskier than Germany, nicht wahr? So would you expect a bank to have equity capital available to absorb losses from its investment in sovereign debt? Yes? The Basel committee of the day did not think so. No capital was required for what was judged to be an asset without risk. Bad call: the writedown on Greek bonds was 74 per cent.

Now let us take a look at mortgage lending. Many of us own a house, or wish to. Do house prices always go up? No. Might they sometimes go down? Yes. How much might they go down: 5, 10, 15 per cent, or all of the above?

"All of the above" would be a reasonable reply. So, what level of loss did the pre-bust Basel rules implicitly assume? More simply, how much loss-absorbing equity do you think the banks had to have to support a loan for 100 per cent of the purchase price of a house? £5 per £100? £10 per £100? Answer: 86 pence for every £100 of loan. In other words, the Basel regulations required less than 1 per cent of loss-absorbing capability for a mortgage with no money down.

Of course, straight mortgage lending is not the sexy stuff for many financial giants. Some investment bankers made millions by originating and distributing an alphabet soup of securitisations. Remember CDOs squared? How much loss-absorbing capital do you think the old Basel rules required of banks to carry an investment in a debt obligation backed by a debt obligation backed by a pool of loans made to US sub-prime residential home owners? (No, that was not a misprint.) Oh, and assume that these easy-to-understand securities were rated AAA by the world's leading credit rating agencies. What could possibly go wrong? Well, not much, in the judgement of the day. Basel required banks to have 40p per £100 of such stuff. Many banks thought this was far too conservative and convinced the regulators that their "advanced" risk management systems warranted a reduction to 14p per £100. The ensuing losses proved somewhat in excess of that.

Of course that was then and this is now. We've all learned from our mistakes, right? Judging from the reaction of the banking lobby, you would say yes. "Onerous", "tough" and even "unreasonably severe" lace the language of regulatory pushback. Yet for your info, the new rules (the new improved version of Basel II, which was in turn the new improved version of Basel I) demand £1.35 of equity capital in support of every £100 of similarly rated CDOs squared – a loss-absorbing capability of less than 1.4 per cent for a risk which most rating agencies, regulators, bankers and investors have already misjudged.

Is it surprising that some regulators – such as the Swiss – have signalled their intention to go beyond the new Basel minimum?

Capital is there to absorb losses from risks we understand and risks we may not understand. Evidence suggests that neither risk-takers nor their regulators fully understand the risks that banks sometimes take. That's why banks need an appropriate level of loss-absorbing equity. Will the new Basel levels be enough? Time will tell. But if we are to err at all, surely it should be on the side of too much rather than too little capital. Many bankers believe they know better. But is that the way to bet?

Robert Jenkins is Adjunct Professor of Finance at London Business School and an external member of the Financial Policy Committee of the Bank of England. He writes in a private capacity