January 23, 2015


€800bn in sovereign QE and target linked. Solid, but no bazooka

We did warn our readers yesterday not to fall for headline numbers, and to look at the way the programme is linked to the inflation target in greater detail. Much of the initially exuberant reaction, especially from media commentators, was based on some arithmetic confusion. They found it hard add up to 1tn before yesterday's decision - and even harder to subtract from 1tr yesterday. Total sovereign debt purchases under this programme are around  €800bn - depending on assumptions about the share of private sector purchases under existing asset purchase programs, which are encompassed in the new program. This is more than the numerically inconsistent previous market expectation of €500bn. But a bazooka it is not. The programme is bang in line with the €1tn balance sheet increase target. Anything else would have been a surprise.

Now let's look at the single most important statement of yesterday's news conference, which Mario Draghi repeated time and again whenever he was pressed to be more specific. Emphasis ours. 

"They are intended to be carried out until end-September 2016 and will in any case be conducted until we see a sustained adjustment in the path of inflation which is consistent with our aim of achieving inflation rates below, but close to, 2% over the medium term."

Clearly the programme is linked to the inflation target. "Will in any case be conducted" is a stronger commitment than "intended". This is by far the best news. But obviously this begs the question of what happens if inflation is not back to target by the time - which it won't be. Note that the commitment is not linked to the target itself, nor to an indirect metric of this target, i.e. some measure of inflation expectation, but to the target path, a somewhat ephemeral concept. In late 2015 headline inflation is bound to rise to or above the core - unless oil prices keep on declining from current levels. The question to ask is whether the ECB's governing council will decide to renew the programme beyond September 2016 when headline inflation rates may have been around 1% for a sustained period, and expected to be around that level for some time.

The best news analysis this morning was from Federico Fubini in La Repubblica. He made a couple of pertinent observations. The first is that Ignazio Visco, governor of the Bank of Italy, has reservations about the agreement - because of the risk sharing element. The concern is that by accepting risk share, the ECB acknowledges a non-trivial probability of default. The other is that the GC members were still very much divided at the end of their working dinner on Wednesday. They did not settle until Thursday morning. Klaas Knots, apparently, played a pivotal role in this compromise.

The math of the programme

Here is the relatively simple math on the size of the programme, which has a duration of 19 months:

  • Of the €60bn per month, around €10bn is accounted for by private asset purchases, leaving €50bn in additional purchases.
  • Minus 12% agency debt - EIB, ESM - leaving €44bn for sovereign debt purchases per month.
  • Multiply 19 months is €836bn for the period.
  • From that you have to deduct the portion of the purchases of Greece and Cyprus - about €25bn - so this makes around €800bn net. 

And here is the math of risk-sharing.

  • 12% of additional NCB asset purchase is for agency bonds, subject to risk sharing - around €6bn, based on an assumption of €10bn in private sector asset purchases;
  • 8% of additional purchases done by ECB directly, €4bn a month, based on the same assumptions;
  • Total approximate size of risk shared portion is €10bn a month, or €190bn for the period (minus Greece and Cyprus);
  • Total approximate size of the non-risk shared portion: €760bn. This number includes the agency bonds.  

Of the additional asset purchases, around €50bn a month, 20% is subject to risk sharing. This means around €40bn a month, or €760bn in total, is demutualised. Should a government default on its central bank, it would have to use its capital buffer to make up the losses - which in the case of several central banks, like the Bank of Italy, should be sufficient. Draghi did not allow himself to get drawn into the negative equity debate. But note, if risk is shared, so are profits. We have heard calculations that these gains would be substantial in Ireland, for example.


Instead of quoting individual commentators, we will break this down by arguments. Among serious commentators - the subset not in Germany, and not blended by headline totals - the consensus was cautiously optimistic - a programme at the upper end of (real) expectations, with some uncertainty about whether it would work. Their scepticism is based on the following reasons. Compared to QE in the US and the UK, the programme

  • is smaller in relative size;
  • comes much later, as inflation expectations have already decoupled;
  • the housing market had a different structure than in the US and the UK, where it played a big role in the overall success of the programme;
  • is subject to risk demutualisation. 

Among the other (and larger) group, the loudest voices yesterday were those of Germany where the reaction was bordering on the hysterical. We dissect four categories of criticisms, all nonsense of course.

The first is that the program is illegal. Peter Gauweiler, MP, and the law professors are at it again. We noted the following extraordinary comment by Hans-Werner Sinn, who wrote that the 20% risk sharing element constitutes a eurobond. “This is tantamount to illegal government financing through the money-printing press. If the ECB buys sovereign bonds, the governments will issue new bonds, in effect drawing their funding from the printing press. This is prohibited by Article 123 of the EU Treaty and would require clearing by the German Constitutional Court”. Note that the Bundesbank sides with Draghi on this issue. We cannot see any relevance for the German constitutional court. The ECB is not subject to the court. And there is no German legislation through which the court could apply leverage.

The second category is currency debasement. We thought Draghi dealt with this "hyperinflation" argument with an appropriate sense of humour with his "statute of limitation" remark. Give us a time frame please!

The third is even more absurd - the criticism by various German banking associations that QE would destroy the savings culture in Germany. That's not a bug. That's a feature. If you run persistent current account surpluses of such extreme degree in a monetary union, this is what happens. 

The fourth point, marginally but at least not completely insane, is whether the programme will have fiscal and other policy externalities. Will it slow down reforms? Will governments be using the central bank profits to expand fiscal policy? Since nobody really expects the programme to work miracles on its own, it would in our view be very unlikely that any European government would draw the conclusions that they can now end the reform process. And in any case, it is not the ECB's job to incentivise governments to do structural reforms. That would be a violation of the treaties. 

We loved the way the Italian Huffington Post summarised the decision. 

"Mercati, analisti, broker, economisti, imprenditori e sindacati, tutti d'accordo. C'è però chi dice no. Sono i falchi di Berlino"


Our other stories

Our coverage today focuses entirely on the QE announcement. We are doing that math on the actual size of the programme - as opposed to the headline size. We dissect whether it is really as open-ended as many believe, and we categorise the reactions to it across the eurozone. A big decision for Europe, with some caveats.

Eurointelligence Professional Edition

For premium access, please log in or register 
for a free 14 day trial for the Eurointelligence Professional edition. The best independent intelligence on the eurozone anywhere in a fast and easy to read format.

A message from Wolfgang Münchau

Welcome to the eurointelligence.com homepage.

Since 2006 we have been providing our readers from central banks, European and international institutions and the financial sector with our daily morning newsbriefing, each morning, at 9am CET, Mondays to Fridays. We are independent from governments and institutions, so you get our honest, sharp and frequently humorous take on the news and the debate. The subjects we are currently focusing on are the now clearly visible economic slowdown, the sustainability of the Italian debt, whether the economic recovery in Spain is for real and our assessment of the impact of the geopolitical tensions in Eastern Europe in particular. Naturally we also follow the ECB and the European institutions as they work through the crisis.

You might ask, is the eurozone still important, since the acute phase of the crisis is over for now? Unfortunately, few of the problems have been resolved. This is something you will quickly notice once you start reading the briefing on a daily basis: there is no better and quicker way to keep yourself ahead of events in the eurozone.

I would like to invite you to register for a free 14-day trial, without commitment, so you can judge for yourself.

Wolfgang Münchau
Director Eurointelligence