Sterilising Silvio | FT Alphaville

Sterilising Silvio

(More salubrious title: did the ECB just rescue Italy from a liquidity trap, or keep it there?)

Key points:

  • Italian government bonds outstanding: €1,597bn
  • Spanish government bonds outstanding: €576bn
  • ECB bond purchases outstanding: €74bn
  • Typical size of a single ECB purchase: €25-100m
  • Italy/Spain holders looking to get out at a good price: many

We could go on…

One last set of numbers though. Eventually the ECB also supported Greece, Portugal and Ireland ‘on a massive scale’, holding up to 20 per cent of their respective total debt stocks. If the ECB went on to hold that much of Italy, it would hold about €320bn (Spain €115bn).

So we have a problem. Actually, two problems. Whether the ECB can sterilise Spanish and Italian bond-buying if they get to a large size, and whether that size will itself become the big liquidity threat to Italian or Spanish bond markets over time as supply is sucked out. Hanging over all of it is the Securities Markets Programme’s long-running tendency to give long investors who want to sell an exit route at a good price, rather than encouraging them to maintain exposure.



But sterilisation’s a good place to start. Because it really shows the irony here — the ECB could buy all of Spain and Italy’s government debt if it wanted, by fully putting its balance sheet to work, but it won’t. Ostensibly this is of course about the ECB not tolerating any increase in the monetary base through unsterilised buying — “we do not do QE” — but when you think about it, it’s also a bet that the bank can happily avoid a liquidity trap arising from sterilisation.

Currently, the ECB offers seven-day deposits each week to absorb liquidity created by that €74bn of debt purchases. Bank demand for the deposits has ebbed and flowed (at times, ebbed severely) as and when liquidity conditions fluctuate. Ultimately though, sterilisation has been a manageable-sized operation until now, with a foreseeable evolution. After all, at this point Greece, Portugal and Ireland bonds have all failed under the ECB’s provision of so-called “liquidity”and are in bailouts, so there wasn’t much inventory left on the loose out there to sterilise in future buying.

Italy, however…

Well, you can see the numbers above. It’s not an imaginative leap to consider billions of fresh Italian bond purchases coming in per week, especially because there are already billions of debt coming fresh to secondary markets each month from Italy’s heavy auction schedule. There are probably ways to extend sterilisation if the ECB wanted — they’ve hinted at using 30-day debt certificates before, and there’s always the model of the Fed’s Term Deposit Facility for longer-term deposits. But is it too much to argue that this would only perpetuate liquidity traps in the eurozone? It’d create places for banks to park cash but not encourage them to put it into (eg.) peripheral debt. It’s maybe a small concern (and Eonia has long come under policy rates, so it’s not a new one either) but we think the potential scale of ECB buying of Italy renews it somewhat.


Bond liquidity

And for what it’s worth — we wonder if the same bad liquidity will sweep the Italian market in the wake of the ECB intervention. It’s maybe true, as Paul Krugman argues, that the bank might as well give bond purchases a shot if Italy has a short-term liquidity crisis. On the other hand, there are those effects of the ECB being used as investors’ rubbish bin to offload peripheral debt while they go and find something more liquid, as well as the fact that liquidity was the great psychological appeal of Italian debt. Once it’s gone…

Lastly, this is all kind of repeated in the idea of the EFSF buying up Italy and Spain in a month or two’s time. It’s well known that EFSF capacity for bond purchases is far less than the ECB, assuming the purchases were just supported from its capital. But, however EFSF purchases are financed, you’re substituting a liquid asset for Italian and Spanish debt taken off the market — either direct EFSF bonds assuming financing is as normal, or indirectly if there’s some means for the ECB to back the EFSF via buying those bonds, thus creating sterilisation issues again.

It’s probably getting way ahead of ourselves. But since the logic of ECB intervention into Italy has to come to terms with the possible expansion into tens –hundreds– of billions of purchases, we’d be very interested in feedback.

Related links:
What the Modest Proposal asks of the ECB – Yanis Varoufakis
ECB bond purchases won’t really help – Megan Greene
The slippery slope to non-sterilisation – FT Alphaville (2010)