Willem Buiter wants you to familiarise yourself with the ELA.
That’s the Emergency Liquidity Assistance that the eurozone’s national central banks (NCBs) are able to provide their local banks under some legal fuzziness in the eurozone. The acronym has managed to grab a few headlines over in Ireland, but for the most part ELAs remains relatively unknown. Soo too, do the details of them.
In theory ELAs are pretty much independent since they’re not (technically) part of eurosystem operations. National central banks can offer them where and when they see fit — as long as they follow a few basic rules. For instance, the ELA has to be consistent with the EU’s Monetary Financing Prohibition. And according to ECB opinions, any ELAs extended should also be temporary, extended to illiquid institutions (not insolvent ones) and against decent collateral.
And here’s where the Citigroup strategist’s latest note comes in:
While there are some indications that the [Central Bank of Ireland] CBI has provided the ELA with “different collateral and larger haircuts” compared to the collateral used for the Eurosystem facilities, it is not clear if a penalty rate has been applied … Assuming, plausibly in our view, that the terms of ELA are at least no more favourable than those offered by the ECB, it is likely that the collateral offered would not be accepted by the ECB.
We can’t know the details, because the collateral holdings and terms of ELAs aren’t really published. Indeed, the fact that they’re going on at all usually only becomes known to outside observers after a generous timelag. It’s all very hush hush.
Here’s Buiter again:
Above we noted that, at least in the interpretation of the ECB, the monetary financing prohibition in the Treaty would be violated if ELA were granted to an institution that was not just illiquid, but insolvent. Of course, the distinction between the two concepts is notoriously difficult, and especially so during periods of high market stress and very volatile asset prices. Nevertheless, in the Irish case, it appears that the main beneficiaries of ELA were institutions whose solvency must at the very least have been in question even at the time ELA was provided …
Buiter’s not holding back then.
If you want a quantitative estimate of ELAs’ significance in the eurozone, he’s here to help too. Remember these are generally recorded either as ‘Other Assets’ or ‘Other claims on euro area credit institutions denominated in euro” at NCBs.
Here are Buiter’s figures:
Since the beginning of 2007 ‘Other claims on euro area credit institutions denominated in euro’ of the Eurosystem have increased by around €34bn to €45bn (as of January 14, 2011) and Other Assets surged by around €75bn to €297bn in that period.
Now there are a few consequences of these extra-monetary union affairs, according to Buiter. For a start, they obscure countries’ public indebtedness figures since ELAs are — crucially and in all legal likelihood — the responsibility of that country.
Which follows on nicely to another point; ELAs and NCB insolvency.
Some scary numbers from Buiter. At Ireland’s central bank the ELA already accounts for 24 per cent of total assets. So even if all the capital of the Irish central bank were available to absorb the losses of the €49bn ELA, likely losses could easily wipe out the bank’s €1.5bn capital and reserves, according to Buiter. (Buiter, we should note, says the CBI is not “supposed to be able to print money.”)
So, enter the eurosystem. Says Buiter:
The alternative to NCB default, should the sovereign be unable to bail out its central bank, would be a financial rescue by the rest of the Eurosystem — a pooling of the losses, perhaps even going beyond that, if Ireland’s share of the pooled losses were to exceed the CBI’s capital. In that case, the now insolvent NCB would have created liquidity (‘money’) independently of the ECB — a challenge to the very essence of monetary union. This would be equivalent to treating the ELA as backed by the full faith and credit of the entire Eurosystem. A monetary union with multiple independent centres of money creation will end up looking like the Rouble zone that survived the collapse of the Soviet Union at the end of 1991 for a bit, until it collapsed in a series of chaotic hyperinflations.
And then there’s the whole conflation of the ECB with those NCBs:
If the Irish banks cannot borrow additional amounts from the Eurosystem using collateral issued by or guaranteed by the Irish sovereign, why are they interested in borrowing from the CBI’s ELA? At best, the liabilities of the ELA are as good (in terms of credit risk) as the Irish sovereign. Are the counterparties of the Irish banks that get paid with the liquid assets provided to the Irish banks by the ELA, under the impression that these assets are Eurosystem liabilities rather than liabilities of the informal ELA subsidiary of the Central Bank of Ireland that is not part of the Eurosystem. Why does anyone accept payment in ELA liabilities? Is it because no-one can tell the difference between an ELA deposit at the CBI and a Eurosystem deposit at the CBI?
Unsurprisingly, the name of Buiter’s latest is “ELA: An Emperor without Clothes?”
The full note is available over here.
Buiter’s €2,000bn solution for the eurozone – FT Alphaville