The debate over bank capital requirements could get lost in its own complexity. Here’s a simple suggestion to clarify matter (in my column from July).
When chief executives of banks start lining up to complain about something, regulators know they must be on the right track. The recent increase in whining is a response to the suggestions being made by regulators that they trust the complex risk-weighting approach used to set capital ratios under Basel III about as far as they can throw it.
Instead, there is a groundswell of regulatory support for the simpler leverage ratio – which expresses a bank’s equity as a percentage of its total assets – to act as counterbalance to the more complex risk-weighted approach.
However, at the risk of complicating an already complex issue, there might be a third way, that could simultaneously address the concerns of both sides and shed more light on what is really going on…
Over the past few weeks, bank chief executives have been falling over themselves to criticise leverage ratios. Anshu Jain at Deutsche Bank said recently that leverage ratios “not only fail to tell us what we need to know, but they may also encourage risky behaviour”.
Over at Barclays, Antony Jenkins said they “are a much cruder tool and need to be interpreted with care to avoid unintended consequences”.
But last week, the Federal Reserve said it would set a higher bar for big US banks than the 3% leverage ratio mandated as a backstop under Basel III.
In Europe, regulators in Denmark, Switzerland and the UK have all made positive noises about leverage ratios as a result of their mistrust of more complex risk-weighted ratios.
The problem is simple. We know from Goodhart’s law – coined by Professor Charles Goodhart in the 1970s – that the moment any single measure becomes a target, it ceases to be reliable as a measure. This is because whatever it is measuring will be gamed or distorted in pursuit of hitting that target.
This gaming was on display in the run up to the crisis. The capital regime for US banks was defined by leverage ratios, which encouraged them to pile into riskier assets (such as sub-prime mortgages) in the pursuit of higher returns. Meanwhile, European banks tended to have much higher leverage but to invest in lower risk assets.
Having both a leverage ratio and a risk-weighted approach acts as a checking mechanism. Goodhart told me last week: “If you have both simultaneously, it makes it much harder for banks to game the system one way or the other.”
Setting aside the accounting issues about what exactly should be defined as equity and how you calculate assets, the problem with this dual approach is that it will often sound a false alarm.
A bank with a high capital ratio on a risk-weighted basis – such as a Deutsche Bank – could easily have a low leverage ratio if a large part of its assets are in lower-risk assets such as government bonds or prime mortgages.
The real concern should not be the divergence between capital on a risk-weighted basis and capital under a simple leverage ratio, but the divergence between the two different approaches to risk-weighted assets: the standardised approach and the banks’ own internal models.
Right now, lots of European banks are stretching their own models to breaking point to “optimise” their RWAs (this can also be described as “RWA minimisation”) and so reduce the amount of equity they need to fund themselves.
Applying a leverage ratio on top of this does not reveal the extent of this gaming. However, forcing banks to disclose their capital ratios under a standardised approach – in addition to how they currently disclose them – would.
Under this “triangulation” approach, the leverage ratio would act as a backstop (its intended function), while the two different RWA numbers would act as a check and balance on each other.
If the capital ratio based on the RWAs that a bank has calculated itself were significantly lower than the ratio based on the standardised approach, it would flag up an amber warning to regulators and investors.
At the same time, it would shift the responsibility on to banks, forcing them to explain any discrepancies in their capital ratios and why they don’t add up.
–This article first appeared in the print edition of Financial News dated July 8, 2013.