Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 13 August 2012

ECB conditionality exceeds their mandate

To get a variety of views on this issue, read this post  from Bruegel . Here is my view.

We can think of the governments of Ireland or Spain facing a multiple equilibria problem when trying to sell their debt. There is a good equilibrium, where interest rates on this debt are low  and fiscal policy is sustainable. There is a bad equilibrium, where interest rates are high, and because of this default is possible at some stage. Because default is possible, a high interest rate makes sense – hence the term equilibrium.

Countries with their own central bank and sustainable fiscal policy can avoid the bad equilibria, because the central bank would buy sufficient government debt to move from the bad to the good. (See this pdf by Paul De Grauwe.) The threat that they would do this means they may not need to buy anything. Anyone who speculates that interest rates will rise will lose money, so the interest rate immediately drops to the low equilibrium.

How do markets know the central bank will do this, if that central bank is independent? They might reason that independence would be taken away by the government if the central bank refused. But suppose independence was somehow guaranteed. Well, they might look at what the central bank is doing. If it is already buying government debt as part of a Quantitative Easing (QE) programme, then as long as the same conditions remain the high interest outcome would not be an equilibrium.

Suppose instead that the central bank does not have a QE programme, and announces that it will only undertake one if the country concerned agrees to sell some of its debt to other countries under certain onerous conditions, and agreement is uncertain. We are of course talking about the ECB. Now the bad equilibrium becomes a possibility again. Perhaps the country will not agree to these onerous conditions. As Kevin O’Rourke points out, this possibility is quite conceivable for a country like Italy. Equally, based on past experience, the lenders may only agree if there is partial default. Neither of these things needs to be inevitable, just moderately possible – after all, interest rates are high only because there is a non-negligible chance of default. The ECB also says that even if the country and its potential creditors agree, it may still choose not to buy that country’s bonds. This throws another lifeline to the existence of a bad equilibrium.

So, we have moved from a situation where the bad equilibrium does not exist, to one where it can. As the good equilibrium is clearly better than the bad one, there must be some very good reason for the ECB to impose this kind of conditionality. What could it be?

The ECB’s mandate is price stability. So without conditionality, would there be an increased risk of inflation? One concern is that printing more money to buy government debt will raise inflation. But that does not appear to be a concern in the UK and US, for two very good reasons. First, the economy is in recession, or experiencing a pretty weak recovery. Second, central bank purchases of government debt are reversible, if inflation did look like it was becoming a serious problem.

What about the danger that by buying bonds now, when there is no inflation risk, governments will be encouraged to follow imprudent fiscal policies at other times when inflation is an issue. But why would the ECB buy government bonds in that situation? Buying bonds now does not commit the ECB to do so in the future. No one thinks the Fed will be doing QE in a boom. OK, what about all those ‘structural reforms’ that might not occur if the bad equilibria disappeared? Well, quite simply, that is none of the ECB’s business. It has nothing to do with price stability. If the ECB is worrying about structural reforms, it is exceeding its mandate.

Cannot the same argument – that an issue is not germane to price stability - be used about choosing between the good and bad equilibria? No. The bad equilibrium, because it forces countries like Ireland and Spain to undertake excessive austerity (and because it may influence the provision of private sector credit in those countries), is reducing output and will therefore eventually reduce inflation below target. The only ‘conditionality’ the ECB needs to avoid moral hazard is that intervention will take place only if the country in the bad equilibrium is suffering an unnecessarily severe recession. The ECB can decide itself whether this is the case by just looking at the data.

So, in my view, to embark on unconditional and selective QE in the current situation is within the price stability mandate of the ECB. To impose conditionality in the way it is doing is not within its mandate. Unfortunately, as Carl Whelan points out, this is not the first time the ECB has exceeded its mandate. As he also says, if the Fed or Bank of England made QE conditional on their governments undertaking certain ‘structural reforms’ or fiscal actions, there would be outrage. So why do so many people write as if it acceptable for the ECB to do this?


  1. "So why do so many people write as if it acceptable for the ECB to do this?"

    Perhaps because there is no central government in Europe (don't mention the Commission), and the parliament is almost powerless. No one has a real mandate. So, for Europe to exist, some institutions may have to exceed the little mandate they've got.
    (This is not meant to approve nor disapprove Draghi. One or another, his hands are not free, so it looks difficult to bear a judgement on him).

  2. So there's two central bank promises:

    1) the central bank will buy sufficient debt to create a "good" equilibrium.


    2) but not if inflation becomes a "serious problem".

    To the markets, this sounds like just the same-old bet on the output gap containing inflation.

  3. I often agree with your columns, but not this time.

    It is plausible that large-scale bond purchases do not entail immediate inflationary risks. However, they certainly entail the risk of significant losses to the Eurosystem if the finances of the relevant member state ultimately turn out to be unsustainable, requiring the kind of restructuring that Greece has gone through. This outcome is not only possible, but its probability depends directly on the seriousness with which national policymakers address their budget problems.

    Hence, if already the ECB ventures into the rather unusual business of transferring large amounts of risk from bondholders to taxpayers, it is well-advised to at least condition such transactions on sound fiscal policies being pursued. Of course, this implies a self-imposed limitation of the ECB's capacity to act, but given that the action is so far outside of its usual mandate, and so deep into the domain typically occupied by the fiscal authorities, it's only proper to tie the latter ones in, via ESM/EFSF conditionality.

    In this context, your comparison of the ECB with BoE and Fed is not convincing. First, both Fed and BoE policymakers have repeatedly insisted on the need for concrete and credible deficit reduction plans. In fact, Mervyn King went so far as to criticize Labour's plan as insufficiently ambitious, effectively endorsing the Conservatives ahead of the last election. Second, in the UK and US the central bank faces only one fiscal counterpart (and sovereign), distinguishing them fundamentally from the complexity of the ECB's situation. Put differently, whatever the BoE does in the gilt market will not entail restribution of risk between different nations in the way that the ECB's bond purchases do.

  4. Simon,

    Like Anders, I struggle I bit to agree with this post.

    You might be right that it is stretching the definition of price stability to include criticising countries for not undertaking structural reforms. But interpretation of the mandate is subjective. (And as Adam Posen remarked yesterday on Mervyn King and the BoE, it isn't useful to engage in "anguished religious ethics" on the definition of the central banks' responsibilities).

    In the case of the ECB, it needs to interpret to interpret its mandate in a way which will retain the consent/agreement of its funders - especially Germany.

    In the course of subjectively interpreting its mandate, the ECB seems to be saying: bond-buying is within our mandate; but over the long-term, if countries do not reform, it will not be in the mandate; and to give Greece, Italy etc enough of a sense that it will not be in the mandate, it is signalling that now, rather than simply assuming that the governments will understand that.

    Imagine that Bank of England, for example, faced a fiscal authority strongly hostile to QE (for some reason, it's just a hypothetical example). I'm sure it would interpret its mandate somewhat differently.

  5. It may be in breach of its mandate. It is certainly counter-productive. The ECB says two things. It will do "whatever it takes" to solve the problem. And it will only do things if it is satisfied with other measures which governments take. It is possible to say either of these but not both.
    The crisis is one of uncertainty. The ECB could end it by promising to buy bonds to hold interest rates down to a given level. But to have any worth, such an assurance must be unconditional.
    This ought to be raising a much wider debate about the way in which so much power has been handed to unaccountable central banks, one of the many mistakes of the pre-war world which we recognised, cured and have now brought back again.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.