Eonia went a bit doolally at the end of January.
Many blamed a lack of front-loading in bank liquidity management as they watched Europe’s key overnight lending rate drift above one per cent for the first time since June 2009.
Here’s an additional explanation.
Barclays Capital analysts Laurent Fransolet and Giuseppe Maraffino recently looked at the Irish central bank’s use of emergency lending assistance (ELA) to the country’s banks. ELA takes collateral not eligible at operations by the ECB. The credit risk falls on national central banks.
According to BarCap, a recent fall in ELA (or perhaps a shifting of the risks elsewhere might be more apt) may account for Eonia’s gyration.
It appears to have gone like this. Irish banks issued around €14bn of government-guaranteed floating-rate notes on January 25 and 26, according to BarCap. As far as we can make out, Bank of Ireland issued most of these January 26 FRNs (from the list below, about €9.5bn):
By contrast, we can’t find any FRNs by Anglo Irish, INBS, EBS or AIB. We wonder who else issued, then — but let’s move on.
These FRNs were then rolled by ‘Irish banks’ (or other banks which bought the notes off them) into two open market operations by the European Central Bank — either the three-month Longer-term refinancing operation (LTRO) opened on January 26, or a main refinancing operation (MRO) that settled on February 2.
The result — on the days the notes were issued, ELA lending fell by a proportionate amount (‘other assets’ in the Eurosystem central banks’ financial accounts fell €13bn). But at the same time – net autonomous liquidity factors rose €12bn, something the market didn’t anticipate. Consequently, with excess liquidity already low, Eonia pushed higher.
Which shows you two things.
First, ELA lending — which can involve national central banks creating money out of thin air — has surprisingly large effects on ECB liquidity.
Second, there is clearly a scramble to get out of ELA and into other (more sustainable?) sources of funding for Irish banks.
BarCap put it well:
From a presentational point of view, this is better for Ireland and the ECB. It is also more advantageous for the Irish banks, since the cost of the ELA is much higher than the cost of the ECB OMOs (1% currently), the collateral value of the paper is much higher as well (0.5% rather than probably 30-40% haircuts), and potentially, this paper can be sold in the open market.
Quite. ELA is supposed to be highly short-term in nature. An experts report into the collapse of Fortis in 2008 (which included ELA funding) darkly hinted at conflicts with Article 101 of the EU Treaty, which is a provision against public authorities interfering with competition in the internal market via aid or subsidies. The experts added that ELA that extended more than a few weeks would indicate that solvency was the real problem, not liquidity.
Needless to say, ELA to Irish banks has been more persistent than it was for Fortis, it seems. And speaking of solvency…
On Wednesday, Ireland’s bad bank Nama announced that it would eventually acquire loans totalling €88bn in nominal value for €37bn. That’s a 58 per cent discount.
Nama added that it would acquire €12bn (nominal value) of further loans from Bank of Ireland and Allied Irish Banks. But not the epicentre: Anglo Irish. Even then, Anglo’s chairman argued on Thursday that Ireland’s banking system would eventually require €50bn of recapitalisation, not the €35bn pencilled in by the country’s bailout.
Puts all the shuffling of the ELA billions into some context, really.
Related links:
ELA — All Anglo? – Corner Turned
Ireland’s secret liquidity is unbelievably cheap – FT Alphaville
Buiter on Europe’s secret liquidity operations – FT Alphaville