Tuesday, 13 September 2011

ECB - What A Mistake to Make

You'd think the ECB would tread carefully, given that they're dancing around a Doomsday Machine. But, as I'll show you, they haven't been. Rather, they've been stomping around the market to clumsily that history may well regard them as a direct catalyst for this latest and most dangerous round of financial destabilization.

If you've seen it mentioned before by the press or by the Street, I apologize – I must have missed it. Possibly it's just too awful to mention.

To understand what happened, you need to reacquaint yourself with the timeline of the meltdown of peripheral European sovereign bonds. Here, then, is what happened to the average premium of 10yr bonds for Eurozone vulnerables (for this purpose: Greece, Ireland, Portugal, Spain, Italy), relative to generic 10yr Euro sovereigns.
What was happening as those premia surged higher? Cast your mind back to late-June to mid-July – it's a quaint and safe time compared to the horrors we confront today.

Risk premia for the Eurozones vulnerables had been rising since April, mainly because it wasn't at all clear that Greece would warrant the Eu12b tranche of bailout funds needed to keep them going through September. In a pattern with which we are now familiar, the ECB and Germany are squabbling, and across the full range of European institutions, no-one can agree on just what degree of necessity and/or coercion of financial institutions would constitute a Greek default. During June Eurozone finance ministers met and failed to act, waited a fortnight, then met again and failed to act.

By the second week of July, bond market panic has spread to Italy, pushing the premium on Italian bonds to a then-unheard-of 2.73% (it's just under 3.5% now). This is the point at which the European Banking Authority releases the results of this year's new and improved bank stress tests. By their calculations, eight out of 90 banks surveyed need more capital: when private analysts rework the numbers marking to market bond holdings, they think probably 27 banks fail.

Finally, on July 21, a emergency summit of Eurozone leaders agreed a second bailout for Greece, carrying the headline total of Eu109 billion. As the Europols headed for the beach, it seems in hindsight very unlikely any of them really understood what they'd agreed. Still, no matter, the deal's the thing and within a week the premium on 10yr Eurovulnerables had fallen from 8.2% to 6.5%.

Very roughly, that's the background to the ECB's cock-up. Publicly, there was the July 7th decision to raise rates by 25bps to 1.5%. (Possibly they were distracted by the succession struggle in which the French government was blocking confirmation of Mario Draghi as the successor to Trichet unless it was guaranteed a seat on the ECB's six-man exec committee.)

However, a more destructive mistake was to come. It is sometimes said, and in print, that the ECB doesn't disclose details about its bond-buying. That's true inasmuch as we can't see exactly what it's buying. But you can track the overall amount weekly through changes in its financial statements (which you can find here).

Anyway, here's what happened: as the chart showed, the ECB took advantage of the decline in premia during the next couple of weeks to start dumping bonds. In the week to August 5th, the ECB sold a net Eu14.6 billion of their bond position. As the chart below shows, this level of selling was absolutely unprecedented in the ECB's recent history. It represented a sale of 11.2% of the entire amount of bonds bought since the start of the bond-buying exercise in May 2010. The ECB was getting its retaliation in first.  
 think that decision to take advantage of the post-agreement relaxation to quietly reverse out of its bond position is a mistake of historic proportions. One can say either that the effect was catastrophic, or one can say that the timing was stupendously unlucky. During the week in which the ECB was extricating itself from its bond position, European equity markets imploded. While the ECB was selling, the Euro Stoxx 50 lost 11.1%, and in the next three days lost a further 9.3% - the biggest crash since Lehman Brothers, and about one and half times the crash suffered by the S&P during the same period. The subsequent loss of economic and financial confidence of course makes all policy options much worse, since the Eurozone's debt problems are also, more fundamentally, a growth problem. Who knows what price will ultimately be paid?

And, of course, it was an expensive mistake also for the ECB purely in terms of protecting their bond position. In the next week (to Aug 22) they bought Eu22.1b, the week after than Eu12.7b, the week after that Eu5.7b, the week after that Eu12.7billion. At which point Juergen Stark resigned 'for personal reasons'.  


  1. Interesting analysis: I checked on the ECB website: They claim that the reduction in assets of 14.6bn is due to accounting changes and was offset by an increase in "other assets".


    Securities of euro area residents denominated in euro/other securities (asset item 7.2) decreased by EUR 14.5 billion, while other assets (asset item 9) increased by EUR 17.3 billion, due mainly to an accounting reclassification by one Eurosystem central bank

  2. I've had a look at this, and I'm not sure this is the full picture either. If you look at changes in the ECB's net lending to credit institutions, it also fell by Eu 77.65 billion during this week. Now the ECB's note makes play of a refinancing operation which was going on at the same time. . .. though if you read it carefully, it's not at all clear why this refi - which took place three days before balance sheet day - should have had a significant impact on the net picture. In the end, we are left with the clear arithmetic that ECB tightened hard that week, and, as they acknowledge, the current account positions of credit institutions with ECB contracted by Eu 48.5 billion during that week. I can't read this as anything other than evidence that they were squeezing hard at exactly the wrong time.