Missing pickets in Europe's new ring-fence

FT.com

In 2010, US banks had assets of €8.6 trillion. But those of the EU's were €42.9 trillion. In the US, bank assets were close to 80 per cent of gross domestic product. In the EU, they were 350 per cent. One half of the world's 30 biggest banks are headquartered in the EU. If the EU makes a mess of banking, it can explode the world economy. In brief, while individual US banks may be 'too big to fail', the EU has a banking sector that is not only too big to fail, but too big to save.

European banks have to be safer than American ones because the damage they can do is so much greater. Managing these risks is of overwhelming importance. Happily, the review of the structure of EU banking carried out by an expert committee chaired by Erkki Liikanen, governor of Finland's central bank, could, with important modifications, be a big step forward.

So what does the Liikanen report suggest? Here are four salient points.

First, it suggests the ringfencing of trading, not of retail, activities as the UK's Independent Commission on Banking (on which I served) recommended. Proprietary trading and either assets or derivative positions incurred in market-making would be assigned to separate legal entities. The latter could not fund themselves with insured deposits.

Second, it demands a hierarchy of 'bail-inable' debt instruments that must not be held by banks. Such instruments should also be part of remuneration of top management, in order to align their interests with those of creditors, not shareholders.

Third, the capital requirements on trading assets and real estate loans should take into account the shortcomings of risk-weighting. An extra capital buffer for trading book assets might be imposed.

Finally, the report makes a number of suggestions for improved governance. Apart from changes in the structure of remuneration a potentially important reform would be greater disclosure of risks.

These proposals will not help with today's crisis. It is too late for that. Nevertheless, they are welcome.

The suggested ringfencing of trading assets, though different in detail, represents an endorsement of the ICB's approach. The report argues that "the specific objectives of separation are to ... limit a banking group's incentives and ability to take excessive risks with insured deposits” and to "prevent the coverage of losses incurred in the trading entity by the funds of the deposit bank, and hence limit the liability of taxpayer and the deposit insurance system”. The ICB took a similar view. Again, the report seems close to the views of the ICB on bail-inable debt. I welcome plans to pay bankers in junior debt that must be held for long periods. Similarly, I agree with the scepticism on risk-weighting, given the experience of its limitations. Much higher unweighted equity requirements are needed.

As happened in response to the ICB report, the critics will argue that in the crisis universal banks showed no greater tendency to collapse than simpler entities. This is true, but irrelevant. There are at least four strong arguments for ringfencing of trading from retail activities inside universal banks. First, without the implicit guarantee given to deposits, the volume of trading activities that can be profitably financed will be far smaller. Second, it is through their trading activities that banks become so dangerously interconnected. Third, universal banks are far bigger than other banks and their failure correspondingly more dangerous. Finally, fully integrated universal banks are also far more complex than other banks, which makes them harder to resolve and to understand. Northern Rock was a quite different threat from Royal Bank of Scotland.

Nevertheless, I have concerns. The most important is over the threshold and the procedure for putting the trading activities into a ringfenced entity. This is to involve two stages. The trigger for the first stage would be that assets held for trading exceed €100 billion or 15-25 per cent of total assets. In the second, supervisors would determine the need for separation. Moreover, the European Commission is to calibrate the criterion for the share of a bank's total assets. But these thresholds are far too high and the room for coercing the commission too great. Hedging services for clients would also be exempt from ringfencing. In all, far too many activities would end up outside the ringfence.

The Liikanen report goes in the same direction as the ICB, though by a different route. Given reasonable flexibility, its recommendations should also be compatible with what the UK plans to do. All this is quite encouraging. I also like what it says about risk weighting, capital requirements, bail-inable debt, compensation and governance. But the holes left in the fence are too big to ensure adequate protection of retail banking, the taxpayer and the economy from trading activities. This is a step forward. But the next ones must be further forward, not backward. The EU must enjoy safer banking. It has no sane alternative.

Copyright the Financial Times 2012.