President Obama’s proposal to reduce the size of banks and split off the risky operations will, if implemented, reduce the risk of future systemic banking crises. However, it will still leave a number of problems in place. It is intended to cut the number of banks that are too big to fail, but size is not everything.

It is worth considering how systemic banking failures arise, and there are a number of ways. One is that if many banks are exposed to a common counterparty and that counterparty fails, then a crisis can occur. This is not due to any particularly large losses, but because of uncertainty over who has what then the music stops and the well-connected bank fails. This is exactly what happened when a small German bank called Herstatt failed in 1974. Admittedly we now have the Basel Committee for Banking Supervision to avoid the same kind of liquidity freeze happening, but the important feature of this crisis is that it was caused by a small bank being linked ot many others. Banks are still linked to each other in a range of complex ways, and this will not change if they are forced to reduce in size.

A second cause of financial crises is exposure to a common shock. This the risk that all banks have similar risk exposures in a particular area, and that if a particular event occurs it will affect all banks in the same way. This is a major flaw in the existing Basel II regime, in that it required levels of capital adequacy and risk management for individual banks, but does not ensure sufficient diversification between the banks’ business models. If all banks are doing the same thing, then a shock that hurts them all will continue to do so no matter how large they are. Each individual bank may not be too big to fail, but does this matter if one insolvent giant is replaced by five insolvent replacements?