Winner of the New Statesman SPERI Prize in Political Economy 2016

Wednesday 8 August 2012

One more time – good policy takes account of risks, and what happens if they materialise

From the Guardian's report of Mervyn King’s press conference today, where the Bank of England lowers forecast UK growth this year to zero.

Paul Mason of Newsnight suggests that the Bank of England should stop trying to use monetary policy to offset the impact of chancellor George Osborne's fiscal tightening, and call for a Plan B instead.
King rejects the idea, saying that Osborne's plan looked "pretty sensible" back in 2010. Overseas factors have undermined it, he argues.
Now Mervyn King had little choice but to say this, but he is wrong (and probably knows he is wrong) for a simple reason. Even if the post-2010 Budget forecast of 2.8% growth in 2012 had been pretty sensible, there were risks either side. There always are, although the nature of the recession probably made these risks greater than normal. It is what you can do if those risks materialise that matters.

Now if growth had appeared to be stronger than 2.8%, and inflation becomes excessive, the solution is obvious, well tested and effective – the Bank of England raises interest rates. But if growth looked like falling well short of 2.8%, the solution – more Quantitative Easing - is untested and very unclear in its effectiveness. (And before anyone comments, the government knows it has no intention of telling the Bank to abandon inflation targets.) With this basic asymmetry, you do not cross your fingers and hope your forecasts are correct. Instead you bias policy towards trying as far as possible to avoid the bad outcome. You go for 3.5% or 4% growth, knowing that if this produced undesirable inflation you could do something about it. That in turn meant not undertaking the Plan A of severe austerity.

So all the talk about how much austerity, or the Eurozone, or anything else, caused the current UK recession is beside the point when it comes to assessing the wisdom of 2010 austerity. Criticising the Bank of England for underestimating inflation in the past is even more pointless – do those making the criticism really think interest rates should have been higher two or three years ago? Even if the Euro crisis has been unforeseeable bad luck for the government (although I think excessive austerity is having its predictable effect there to), the government should not have put us in a position where we seem powerless to do anything about it.

If you are sailing a ship near land, you keep well clear of the coast, even if it means the journey may take longer.  So the fact that the economy has run aground does not mean the government was just unlucky. You do not embark on austerity when interest rates are near zero. Keynes taught us that, it is in all the textbooks, and a government bears responsibility when it ignores this wisdom. To the extent that the government was encouraged to pursue this course by the Governor of the Bank of England, that responsibility is shared.


  1. Absolutely right. But handing policy to a technocratic body like the Bank of England means we have no redress when they get it wrong.
    I try to make the case against allowing central bankers to have the power they do.

  2. I thought that James Meade had this issue pretty much in hand in the 60's when he used Bill Phillip's Moniac hydromechanical analog simulator as a teaching tool as discussed in my OECD insight article at:

    "Meade recognized the machine’s dynamic nature and the visibility of its flows as a powerful teaching tool. A favorite exercise at London School of Economics was to run an experiment on the impact of uncoordinated government intervention. One student (the “Chancellor of the Exchequer”) controlled taxation and public spending, a second managed monetary policy (“Head of the Bank of England”). Told to achieve a stable target level of national income while disregarding each other’s actions, they produced results that were invariably messy..."

    The article also links to Allan McRobie's wonderful video demonstrating the restored Moniac and his great comment showing the value of such simulators: "Let's turn off the Banking Sector for a moment..." It would be great fun for King and Osborne to have a go on Phillips machine. Probably lead to puddles on the floor.

    I'm not sure about your navigational analogy, as you have the option of slowing up or dropping anchor. Maybe sailing too close to the wind, so the sails stall. I use a Phillips-inspired fluid dynamics model, discussed in the article, and it clearly shows that Larry Summers comments about the US economy flying out of the recession 'too close to the stall' is correct. Applying austerity policies below a certain velocity of transactions, essentially a region of reversed commands, will lead to high sink rates and if carried to extreme, can literally stall the economy.

  3. "You do not embark on austerity when interest rates are near zero. Keynes taught us that, it is in all the textbooks, and a government bears responsibility when it ignores this wisdom."

    Something I find puzzling: If it is in all the textbooks, why the near unanimity in favor of austerity in Europe? I understand when zombie ideas reappear. What surprises me is the level of agreement.

  4. Very good post. I always enjoy your blog.

    I just posted a new blog post on Professor Arthur Laffer's op-ed in the Wall Street Journal and mentioned David Glasner's post on the op-ed and Nick Rowe's post on Milton Friedman's Thermostat.

  5. "One more time – good policy takes account of risks, and what happens if they materialise"

    Two words: Global Warming

    Ok, obviously you can't expect me to stay to two words. The foundation of finance is the risk-return tradeoff (and the NPV rule). When the risk is monumental, as with global warming, the expected return, discount rate, for insuring against it is negative, just like with any catastrophic insurance. Yet critics often want it to achieve the return of stocks or bonds -- and without counting full externality benefits.

    For more on this see:

  6. Guys, could you be polite enough to use the comment section to discuss the issues in the main post and not to boost you own blogs/articles?

  7. I am no apologist for the Bank, but they certainly "tested" QE in 2009 and it is reasonable to conclude it was effective then, correcting the crash in nominal demand exactly as they intended (5% NGDP growth Q2 '09 to Q2 '10).

    I would say also that the Bank's forecasts in 2010 (and early 2011) indicate they thought they *were* "well clear of the coast"; they were forecasting a circa 25% probability that real GDP growth would be above 4% in every quarter of 2012 in the August 2010 inflation report, when the broad outline of the deficit reduction strategy was already known.

    data here

    It seems more interesting to me to consider why the Bank did not "correct course" after the GDP figures started slowing in late 2010. They were getting high inflation + slowing output, so it seems they presumed demand stimulus remained sufficient, but we had supply-side problems. There were what, 15 MPC meetings between the announcement of the Osborne deficit strategy, and the resumption of QE in late 2011?

    1. Interesting data, particularly the low probability given to low growth. This looks odd, given the evidence around at the time that recessions following a financial crisis can be long lived. On QE, I think there are reasons for thinking its effectiveness declines with use. And the evidence on its effectiveness in 2010 was pretty weak, compared to what we know about the impact of interest rates and fiscal policy. So if I had been Governor, my advice would still have been - if the recovery is too strong, we can fix that, but if it is too weak, we are in uncharted waters.

    2. Never underestimate the overconfidence of a central banker! ;) This is how I read 2010, FWIW:

      a) King had long been impatient with fiscal deficits. There's an old FT interview from before the crisis where he almost complains that Brown is running deficits (2003 on) to boost demand, when the MPC could (and implied: should) be doing that with monetary policy.

      b) King has a "global rebalancing" grand plan, that he thinks we "need" to move demand from deficit to surplus countries. He wants to shift the UK to investment + exports, which he can only do with monetary policy.

      c) in early 2009 they stared deflation, a Great Depression, right in the face, and they beat it *with QE*. So fiscal policy is even less important.

      Or there's an institutional view: King wants demand policy to remain at the Bank. If the Treasury thinks it can do demand policy with deficit spending, we've rolled back the single most important reform of his working career: the shift of control over AD policy to the Bank under an independent MPC, and out of the hands of meddling politicians.

  8. I think a big problem is that the Bank worked with the wrong macroeconomic models (like every other official institution). They did not have debt and the financial sector inside their models and so underestimated the consequences of what had just happened. This made them think that the risk of a long-term malaise in the private sector was much lower than it was. Furthermore, they underestimated the likelihood of adverse external shocks.

    But I agree with Simon's main point: you have to decide what is the bigger risk - above-target inflation or deflation. On monetary policy, they largely took the right decision on this, focusing on avoiding the risk of deflation. But, because they did not realise how ineffective monetary policy was likely to be, they did not understand the risks of the recommended fiscal contraction, which they thought monetary policy could easily offset. (The Treasury thought the same thing.) That was a big error.

    One of the many conclusions I believe is that in crises of this kind the compartmentalisation of monetary and fiscal policy is a big mistake. It would have been far better to use them together.

    1. I think this view of the Bank has happy inflationists is at odds with their behaviour, because looking at the CPI rate is a very bad metric.

      One simple example: the CPI-CT was below 2% all through 2010; the Bank did not fully offset the VAT rise in that year; effectively forcing a disinflation to compensate for the fiscal contraction. Was that good policy?

      But more importantly, the Bank do not set policy which they expect to achieve inflation significantly above 2% on the forecast horizon which actually matters to them, which is looking two/three years out; see here (I hope Simon will forgive me the link bait).

      They are setting policy right now which they expect to achieve inflation in the 1.5%-2% range looking two years out. That is not the behaviour of a central banker running in the grand tradition of Gideon Gono.

      I think if Merv came out into an Inflation Report presser smoking a pipe, leant back, stuck his feet on the desk, and said he was going to print £100bn a month until we get 4% inflation, and then he'll carry on doing that for another two years just to piss off the the Daily Telegraph, that'd be pretty effective.

      When he comes out and says, well, we're shooting for a 1.5% CPI rate, but really, we're not very sure because of all the black clouds, and we'll buy the minimum amount of gilts the market expects us to... that's about as effective as you'd expect - not very.

    2. I regret that I do not understand this comment.

    3. Which bit, Martin? If Simon will again forgive my linkbait, I've done this post for you on how to judge the monetary policy stance under inflation targeting, and it seems very clear the Bank has no inflationary bias, in fact quite the contrary.

    4. @ Britmouse, The difference between us may be that you think the forecasts determined the policy, instead of the other way round.

    5. Martin, I guess we don't understand each other at all then :) "the forecasts" and "the policy" are synonymous terms, I don't know what you mean by one determines the other. The forecast *is* the current stance of monetary policy. That's what Svensson says.

      Do you mean the forecasts determine the setting of the *instruments* (Bank Rate, QE)?

      Or do you mean that the Bank has no ability to affect its own forecast? (Or possibly: has had no such ability at the ZLB?)

  9. This comment has been removed by the author.


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.