Winner of the New Statesman SPERI Prize in Political Economy 2016

Tuesday 17 October 2017

The lesson monetary policy needs to learn

It seemed obvious to write a post about the Peterson Institute’s recent conference on ‘Rethinking Macroeconomic Policy’, but nowadays I find it more efficient to let Martin Sandbu do the job. We agree most of the time, and he does these things better than I do. It allows me to write something only in the unlikely event that I disagree, or if I want to take the discussion further.

I only have one quibble with Martin’s column yesterday. I think Bernanke’s suggestion that following a large recession (where interest rates hit their lower bound) central banks revert to a temporary price level target is rather more than the tweak he suggests. In addition, as Tony Yates pointed out, level of NGDP targets do not resolve the asymmetry problem that Bernanke’s suggestion is designed to address.

I also thought I could illustrate Martin’s final point that “admitting one has got things badly wrong is a prerequisite for doing better” by looking at some numbers. If we look at consumer prices, average inflation between 2009 and 2016 was 1.1% in the Euro area, 1.4% in the US and 2.2% in the UK. The UK was a failure too: average consumer price inflation should have been higher than 2.2%, because we had a large VAT hike and depreciation that monetary policy rightly saw through. If we look at GDP deflators we get a clearer picture, with 1.0%, 1.5% and 1.6% for the EZ, US and UK respectively.

You might think errors of that size are not too bad, and anyway what is wrong with inflation being too low. You would be wrong because in a recovery period these errors represent lost resources that, as the Phillips curve appears to be currently so flat, could be considerable. Or in other words the recovery could have been a lot faster, and interest rates could now be well off the lower bound everywhere, if policy had been more expansionary.

What I really wanted to add to Martin’s discussion was to suggest the main problem with monetary policy over this period, particularly in the UK and the Eurozone. It is not, in my view, the failure to adopt a levels target, or even the ECB raising rates in 2011 (although that was a serious and costly mistake). In 2009, when central banks would have liked to stimulate further but felt that interest rates were at their lower bound, they should have issued a statement that went something like this:
“We have lost our main instrument for controlling the economy. There are other instruments we could use, but their impact is largely unknown, so they are completely unreliable. There is a much superior way of stimulating the economy in this situation, and that is fiscal policy, but of course it remains the government’s prerogative whether it wishes to use that instrument. Until we think the economy has recovered sufficiently to raise interest rates, the economy is no longer under our control.”

I am not suggesting QE did not have a significant positive impact on the economy. But its use allowed governments to imagine that ending the recession was not their responsibility, and that what I call the Consensus Assignment was still working. It was not: QE was one of the most unreliable policy instruments imaginable.

The criticism that this would involve the central bank exceeding its remit and telling politicians what to do is misplaced. Members of the ECB spent much of the time telling politicians the opposite, Mervyn King did the same in a more discreet way, while Ben Bernanke eventually said in essence something milder than the above. Under the Consensus Assignment we have invested central banks with the task of managing the economy because we think interest rates are a better tool than fiscal policy. As such it is beholden on them to tell us when they can no longer do the job better than government.

A better criticism is that a statement of that kind would not have made any difference, and we could spend hours discussing that. But this is about the future, and who knows what the political circumstances will be then. It is important that governments acknowledge that the Consensus Assignment no longer works if central banks believe there is a lower bound for interest rates, and this has to start by central banks admitting this. Economists like Paul Krugman, Brad DeLong and myself have been saying these things for so long and so often, but I think central banks still have problems fully accepting what this means for them.       


  1. Thanks, very good post !

    And it might even be a bit worse. At least in 2009 there was some fiscal stimulus, it wasn't enough but it was much better than nothing. But the next time there is a recession, budget hawks will have a field day pointing to price level targeting and arguing that even a small stiumulus is unwarranted...

  2. As SW-L rightly says, fiscal policy under existing arrangement is the “prerogative” of government (i.e. a bunch of people called “politicians” whose combined grasp of economics is no match for that of a garden snail). I.e. the decision as to how much fiscal stimulus to impart should be taken by the same sort of well qualified people as decide on monetary stimulus.

    And in fact under the system advocated by Positive Money and others, the decision as to how much fiscal stimulus to implement is left to experts. And that does not, repeat not, mean that strictly political decisions like what % of GDP is allocated to public spending cannot remain with democratically elected politicians, as Positive Money explains. For more on that, see:

  3. " Mervyn King did the same in a more discreet way, "

    After 'coming to his {economic?} senses?

    His 'policy;]' was, in the best M. Norman {and neoclassical} tradition, to take no action.
    M.A. King said in his opening remarks: {Thursday 11 September 2008} :
    "In the UK we face a difficult but, temporary, period during which inflation will remain high for a while and output growth at best weak. . . . But provided we do not impede the required adjustment we will come through this temporary period and resume a path of normal economic growth with inflation close to target.
    Provided we focus on bringing inflation back to target, our present difficulties will prove to be temporary. Inflation will fall back, and growth will resume.”

  4. Re SW-L’s first sentence above I hope he does actually “…write a post about the Peterson Institute’s recent conference on ‘Rethinking Macroeconomic Policy’”.

    Pete Peterson is a billionaire who devotes large sums to opposing deficits. I.e. Peterson has done about the same amount of catastrophic economic damage as Ken Rogoff and Carmen Reinhart. I look forward to a suitably scathing article by SW-L.

  5. Is this an error which is related to the independence of the CB? In other words if CBs were not independent would they act "sensibly" in the terms you outline; to wit manage policy according to where the interest rate cycle is and recognise that fiscal policy needs to come into play at a certain point.

    It seems to me that it is not how things were managed before the fashion for independent CBs so we cannot say that the move to independence has made any difference as to how these things are managed or is this the case?

  6. S.W-L. may well feel that "I find it more efficient to let Martin Sandbu do the job". However for those of us who do not subscribe to the exorbitant Premium version of the FT, a summary of Sandbu's arguments would be very helpful.


  7. All economists are stupid. Banks don't loan out savings. Remunerating IBDDs destroys money velocity.

    1. Central bank's deposits only affect the supply and turnover of new money. They do not impact the flow of monetary savings into investment outlets. That can only be accomplished by the non-bank's pooled savings, and the idiosyncratic investment owners. Thus income velocity is a subset of money turnover, viz., the transactions velocity of circulation (a loop-back flow).

  8. "As such it is beholden on them to tell us ..." My old microprinted OED gives the only sense of "beholden [to]" as "having a moral debt [to someone]". This is a rare and indeed exotic word and it's very unlikely that new usages have emerged since the 1930s. Simon may be thinking of the equally recherché "behooves", "be under a moral obligation to [do something]". First citation is from 1175. That's Oxford for you.

  9. «interest rates were at their lower bound,»

    Ah I notice this time our blogger writes again more precisely "interest rates" rather than "the interest rate", and "lower bound" rather than "zero lower bound", but I think that can be improved still:

    * Investment is sensitive to the difference between "interest rates" and expected profits, according to JM Keynes.
    * Investment and consumption depends on both the price of credit and its *availability*, which is a function of credit policy, e.g. regulations on collateral. It is amazing that "collateral" is not part of "standard macro" or its "microfoundations" :-).

    «they should have issued a statement that went something like this: “[ ... ] Until we think the economy has recovered sufficiently to raise interest rates, the economy is no longer under our control.”»

    As I previously remarked, the central bank is not the Loyal Opposition, and treads carefully before undermining government policy. Even so, both B Bernanke and M King at various times said that, of course in euphemistic ways to cover their backs, because it is not the central bank's job to run a campaign of denigration of government policy.

    It is the job of opposition parties to persuade voters that the government has made credit policy powerless and that fiscal policy should be used.

  10. Agree with all, but having been ringside to the IMF board discussion, it was depressing to observe the huge reluctance to embrace fiscal policy. There is the theoretical justification for this, partly should not be discretionary and also best left to itself. Only when OECD did a 180 degree and Blanchard scale out of the closet did fiscal stimuli get acceptance, but then it was almost too late. Just sad


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.