Andrew Haldane, well respected expert in banking and bank regulation has recently delivered a devastating speech in the USA on the topic of complexity in bank regulation. In it, he says that not only has bank regulation become much more complex and expensive (by orders of magnitude) in recent decades, but these exceedingly complex analyses of risk within banks may in fact produce inferior results when compared to simpler regulatory strategies.
This echoes my own thoughts on the matter, and indeed I wrote to the Basel committee on this topic in 2008. The excessive complexity of risk analysis included within regulatory structures is unnecessary and only create loopholes for banks to exploit. Rather, blunt, robust regulation is what is needed.
Haldane goes on to give dozens of examples of risk analysis by banks and regulators where their complex models in key prudential areas are outperformed by simple (occasionally linear) assessments in predicting future outcomes.
This must be incredibly depressing for regulators and banks: they have employed hundreds of thousands of highly qualified staff, at massive expense, to conduct risk analysis within the current complex regulatory framework -and their outcomes are not just poor, but actually underperform simple linear models. To say that it calls into question the current, expensive prudential regime is an understatement.
Haldane is a senior figure in the Bank of England and a respected voice worldwide, his opinions have also been echoed in other quarters -including the former head of the FDIC (and of course myself -only 4 years earlier). His assessment carries weight and it is a welcome intervention in the ever expanding mess that is the current prudential framework.
Haldane is pessimistic that his advocated simpler approach to regulation (focusing on uncertainty itself, rather than pointlessly trying to measure risk to a certainty) will be adopted in international regulation. In particular, he feels the zeal for reform has been defused by Basel III (which is essentially Basel II with more capital). I hope he is wrong, because apart from too many loopholes, Basel II/III incorporate market influenced risk metrics that actively magnify financial crises. These should have been addressed instead of ignored (along with the absurd overcomplexity of these agreements).
My logic is simple. If Basel II did not cause the crisis, then what was the point of Basel III? If it did contribute to the crisis, then why is Basel III essentially the same thing only more-so?
Personally, I go further than Haldane. I don't believe that regulation should focus on trying to prevent bank failures (i.e. ensuring adequate capital for the banks risk profile). This is impossible given the size of the task and the uncertainty inherent in every lending decision. Rather, I think regulation should address the symptoms of bank failures and then simply let them happen. I would leave in place 100% deposit guarantees (for demand or short-notice accounts anyway) and have the deposit protection schemes administered and guaranteed by the Central Banks. In this way, depositors would be assured that their money was safe and bank runs would be averted. More controversially, I would also severely curtail inter bank lending or ownership. For many banks, this would mean a sea-change in their business models. Banking would still be international, but it would be banks themselves that would cross borders, rather than simply capital. The "distribute to originate" (sic) process of interbank lending would end.
This approach basically isolates every bank, leaving the only exposed parties to a bank failure as investors, investment lenders, and the future contributors to the deposit guarantee scheme (in that order). In such a situation, bank failures would cause investors to lose money, but no bank runs and no systemic crises. The cost of protecting depositors would be paid by all banks (and hence depositors) over the long term. Supervisors would withdraw from prudential regulation altogether and let investors worry about it.
This approach basically abandons all pretence of prudential regulation -whether it is complex (as at present) or simple (as advocated by Haldane). In my view, regulation is more for the reassurance of society than any real positive impact it has on risks taken by banks. Indeed, regulation and compliance has become so expensive that it would need to deliver a high degree of safety to justify its own cost -clearly it has not done so, banking crises are just as frequent now as they were in the era before regulation. The only reason the recent crisis was any less worse than the 1930s was state intervention, not expensive prudential regulation in the pre-crisis period.
http://www.bankofengland.co.uk/publications/Documents/speeches/2012/speech596.pdf
