Can eurozone banks be strengthened?
There was progress towards a eurozone rescue deal during the IMF's annual meeting in Washington, according to those present. But they caution that a workable proposal cannot yet be banked.
There were two positive developments.
- First a de facto deadline for agreement on a rescue package was set. This phased deadline will be the European Union council meeting in Brussels at the end of October and the G20 government heads meeting in Cannes at the beginning of November.
- Second, the outlines of the rescue deal, with three main elements, were sketched.
Here are the elements.
There would be a haircut or writedown of Greek sovereign debt of 50%. This appeared to have been leaked at the end of the week by the Greek Finance Minister, Evangelos Venizelos, as per his reported remarks to Socialist MPs that a 50% "haircut" was the best of three available options.
The Ta Nea newspaper quoted Mr Venizelos as saying: "It is very dangerous for us to ask for it; this requires agreed and co-ordinated effort by many."
For what it's worth, the Greek authorities subsequently denied that such a restructuring of Greek debt is under active consideration, though that's not what European Union officials tell me privately.
The next element would be that the EU's bailout fund, the European Financial Stability Facility, could acquire more firepower than its revised financing limit of €440bn.
This would be done not by enlarging the facility - which is seen as politically and practically impossible, because that would require fresh approval by reluctant parliaments in eurozone member states.
Instead the EFSF would do something a bit different from its original conception. The EFSF would take on the main risk of lending to governments struggling to borrow from normal commercial sources - governments like Italy - and it would do this by providing what's known as first-loss capital or equity.
In this way, the EFSF would make it less dangerous for the European Central Bank to lend alongside it.
So the EFSF's 440bn euro could be "leveraged" or enhanced by say 1.5 trillion euro of loans from the ECB. There would therefore be around 2 trillion euro of standby credit available for eurozone governments shunned by the market.
Many analysts believe 2 trillion euro of lending power is the bare minimum needed by the EFSF for it to have credibility, given the massive refinancing needs of Italy over the coming three years. So this reconfiguration of the EFSF looks clever.
That said the wheeze could be too clever by half: it would involve much more risk being heaped on eurozone taxpayers, so MPs in eurozone member states may go ballistic if deprived of a vote on the reform.
Finally, eurozone finance ministers and central bankers in Washington have broadly accepted the need for the recapitalisation and strengthening of the eurozone's banks to be accelerated and enlarged - or so I am told.
Which is positive mood music.
Even so, there would be massive technical and political obstacles in the way of actually injecting sufficient capital into Europe's banks so that they can absorb potential losses from the putative 50% writedown of Greek sovereign debt, and possible losses on loans to other financially over-stretched eurozone countries.
It is the gulf between how investors and creditors see these banks, on the one hand, and how they are seen by eurozone ministers and regulators, that makes the banks so hard to fix.
Here's the thing. Big lenders to banks, especially the US money market funds, are reluctant to provide vital finance to eurozone banks, especially French and Italian ones, for fear that they will be brought to their knees by their loans to eurozone states with excessive debts.
This market view of the weakness of these banks has to an extent been corroborated by a recent IMF analysis (in its latest Global Financial Stability Report), which argued that the potential losses for eurozone banks from the sovereign debt crisis could be as high as 300bn euro - and that losses on this scale could see 22% of European banks suffering erosion of their capital resources of at least 50%.
The IMF did concede that these were fairly imprecise estimates - and it did not have access to data on how many European banks had already recognised some or all of these losses. But the number floating around Washington for the size of future losses that the banks might incur from the eurozone crisis was 200bn euro.
Regulators v governments v banks
So here's another thing. The recent health checks by European regulators of European banks - the so-called stress tests - found that most of them were tickety boo. According to the coordinating body for the tests, the European Banking Authority, just eight banks (mostly small ones you've never heard of) needed to raise a paltry 2.5bn euro in aggregate.
So how on earth do eurozone governments force the banks perceived by investors to be vulnerable - banks like Unicredit of Italy, or BNP Paribas, Soc Gen and Credit Agricole of France - to raise expensive capital, when their respective regulators have already ruled that they don't need to raise capital?
An effective cordon sanitaire around the banking system, that would remove the risk of a full-scale banking crisis, almost certainly requires tens of billions of euros of capital to be injected into big Italian and French banks, for example, because they have been finding it harder and more expensive to borrow in recent weeks, and their shares have plummeted.
But goodness only knows how this can be made to happen.
There could be a political decision to recalibrate the stress tests, so that instead of stipulating that no bank's capital should fall below 5% of assets under stressed conditions, the new minimum capital requirement would be 7% or 8% of assets.
That's a notion that was knocking around Washington, I am told. But I am not sure it would work, even if European governments agreed to do it.
For example, lifting the capital bar would force Soc Gen to raise a fair chunk of new capital from commercial investors or the French state, but BNP Paribas would need to find only a tiny amount of new capital and Credit Agricole would not have to raise a bean.
So two giant French banks would still be seen by some investors and creditors as fragile.
Paradoxically, Royal Bank of Scotland and Barclays - which are seen as less vulnerable by investors than the French banks - would have to raise a good deal of new capital. Which, in the case of Royal Bank of Scotland, would pose an additional strain on the UK's public-sector balance sheet at an awkward time (to put it mildly).
To repeat, the mechanics of strengthening European banks look fraught with difficulties. But unless the banks are fixed, there will remain too big a risk that a financial crisis could turn the current global economic slowdown into something more akin to depression than recession.
So presumably a way to mend them will be found.