Winner of the New Statesman SPERI Prize in Political Economy 2016

Thursday 18 April 2013

An Understandable Mistake

No, I’m not talking about coding in excel - as someone who has in the past done plenty of empirical work, my overriding reaction is empathy with the researchers concerned. There - not so much by the grace of god but because no one bothered to check what we did - go us. What I’m talking about is the weak global recovery and a primary reason for it. But there is a link, which I will come to at the end.

To be honest, this post is really just to encourage you to read this Vox piece by Kose, Loungani and Terrones, which comes from the recent IMF WEO report (pdf). It tells a story in pictures (particularly comprehensive and clear pictures) that I and others - most notably Paul Krugman - have been telling for some time. In the past I have often used cyclically adjusted primary deficits as a summary measure of fiscal stance, because they are readily available and comprehensive. However as an indication of fiscal demand impact they are not ideal - just think of the standard Ricardian Equivalence experiment. So occasionally I have looked at just government spending, which is what Kose et al do in their analysis. The key chart is below.

What the chart shows is that government spending in the advanced economies has grown at a much slower rate in this recovery than in previous recoveries, and of course this recovery has been significantly slower as a result. In contrast, government spending in the emerging economies has been as rapid during the recovery as before the recession, and they have recovered rapidly from recession. Go into detail within the advanced economies group, and the pattern is clear: the greater the contraction in government spending relative to previous recoveries, the slower the recovery has been.

Their analysis also shows us one of the reasons why this happened. The advanced economies started the recovery with debt to GDP almost twice its average level in previous recoveries. At the start of the recession this was not the predominant concern, and we saw some attempt at countercyclical policy in the US, the UK and Japan, as the chart shows. But this was short lived in the UK, and was also reversed in the US and Japan, as debt concerns took over.

Kose et al also illustrate why this was a mistake. Interest rates hit the zero lower bound, so monetary policy could not offset what fiscal policy was doing. Yes, we have had Quantitative Easing (QE), but I think Jonathan Portes puts it rather nicely here. “It seems to me the idea that you risk the government’s credibility by borrowing an extra couple of percent of GDP for investment but there would be no risk to credibility by doing something [QE] which nobody in modern times in an advanced developed country has ever tried and that is generally considered to be last ditch strategy is an odd conclusion.” (The link is to a nice site by the way - on it you will also find Andrew Scott saying: “the right response for those governments wanting to hold on to AAA, whether it be Japan, US, Germany or the UK, is for government debt to show huge increases into triple digits versus GDP.”)

So we have one clear reason why the recovery from the Great Recession has been weak, and it also explains why it has been weaker in some countries than others. I’m sure its not the only reason, but is anyone seriously arguing anymore that it is not an important factor explaining our weak recovery? So the real mistake that Reinhart and Rogoff made was to push the high debt issue at the wrong time. It was I believe an honest mistake: high government debt is a concern, if only (but not only) because it makes politicians do the wrong thing in a serious recession. And of course the mistake would have happened anyway (in part because of Greece, and partly because of those who see reducing the size of the state as the overriding priority) - academics probably overestimate the importance of the research that politicians use as cover.  But when history tells the story of why the Great Recession was so prolonged, charts like the one here will be what is shown.


  1. Very interesting article. I have a question.

    QE (which I interpret as increasing the money supply by non-conventional means) is apparently politically acceptable but seen as having a high risk of not working, while fiscal policy (which I take to mean direct govt investment) funded by budget deficit would work but is political unacceptable because of the fear of debt.

    Why has there not been more of a move to combine these 2 policies and have direct govt investment funded by monetary expansion rather than debt ?

    1. There have been plenty of “moves to combine” the two policies: i.e. just fund deficits with printed money. Keynes saw no problem with the idea. Milton Friedman and Warren Mosler likewise. Positive Money advocates the “print” policy as do backers of Modern Monetary Theory (like me).

      What is utterly bizarre here is that no one seems to be aware that Keynes, Friedman and others mentioned above have said or advocated the above point. That is economists keep repeating the mantra that a bigger deficit must mean more debt. It’s nonsense, of course: deficits (to repeat the point) can be funded simply by printing money. Frances Coppola’s latest post on her blog is just one very recent example of this “bizarreness”.

      The above makes a bit of a nonsense of Jonathan Portes’s claim that the above “print” or “combined” policy is “generally considered to be last ditch strategy..”.

    2. Well, there is one problem with just "printing money" which is that when it gets into the economy, that's it.

      If you print money when demand is well below potential supply, you don't have to worry about inflation at that point - because when you are operating below supply, if demand rises you can always increase supply at constant unit costs - but later on, when the economy begins to recover and demand starts to edge up to supply, you do have to worry about inflation.

      Too much M0 slopping around can push demand above long-term aggregate supply, and when it does each unit of new production starts to cost more - shift-work, overtime payments, delayed maintenance and so on. That can become wage/price inflation if let run too long.

      So you would like a way to get the excess M0 back out of the economy to halt that process. If you pushed that M0 into the economy by buying less liquid assets with it, then you just turn round and sell the less liquid assets, and so suck some liquidity out of your now-overheating economy.

      But if you financed stimulus with literal money printing, you could find it a bit more difficult to remove liquidity from the economy when you need to.

    3. I think the CB could always remove money from the economy just by selling assets for cash

    4. "I think the CB could always remove money from the economy just by selling assets for cash"

      Right. I am talking about the case where they "just print money" and don't increase the amount of illiquid assets that they can later sell for cash.

    5. Jon,

      I agree it’s “more difficult to remove liquidity from the economy” under the print option, or at least it seems to be. But to some extent that is only because we are not used to doing it and haven’t devoted much thought to the matter.

      Adjusting VAT isn’t difficult. It was reduced to 15 per cent in December 2008 and returned to 17.5 per cent on 1 January 2010. In January 2011 the standard rate increased to 20 per cent.

      But other things could equally well be adjusted: National Insurance contributions, and benefits and local authority spending could be adjusted from time to time.

      Re the conventional option, having the CB buy and sell assets, the CB does not need to have bought assets in order to subsequently withdraw money from the economy: it can simply announce it is willing to borrow at above the going rate. But that’s obviously still a concession to your philosophy.

      The problem with the “buy/sell assets” idea is that it skews the economy: it’s very directional. The “print” option is much less directional, but as you rightly say, withdrawing money is a problem.

    6. Of course they could also just tax people and burn the money they receive. Same affect on the money supply.

  2. If one wants to read a lot of of views on R&R including Simon then read here!

    simon always high quality articles. Well done

  3. Reading this, a couple of thoughts occurred to me. A developing economy is, well, developing, so Government spending can translate pretty directly into growth. New highways, bridges and railways can open up new areas for exploitation, while a developed economy has less scope for infrastructure expansion.

    Secondly, although Government spending may enable expansion in a developing economy, there are other factors driving it, such as rapid population growth, rapid urbanisation and just the fact that GDP/head is still quite low - for example, Brazil's GDP/head is about one third that of the UK - so introducing more technology into a developing economy will probably produce a higher percentage growth than in an economy which is already close to technological saturation.

    Finally, a developing economy with a relatively high growth rate means growing tax revenue, and economic expansion enables an emerging economy to outgrow its debt at a faster rate than a slower growing developed economy.

    To me, the contrast between developed and developing economies today resembles the contrast between the US and Europe in the C19th. So although Government spending has some explanatory power, I'm guessing other factors have more. The argument about Government spending has more force when applied to pairs of developed or pairs of developing economies.

  4. Excellent post.
    Politicians don't review all the data and then make a decision. They decide, then look for data to back their belief.
    We learned all this in the wake of the Great Depression, but a great misinformation campaign has been conducted by the malefactors of great wealth. I am more suspicious about incentives and motives. The incentives to support austerity are large. Those people can cash in. Those who oppose austerity may be correct, but they won't be rewarded as handsomely.

    -jonny bakho

  5. This is not to impugn the motives of R-R, though 'cross-checking', replicating work has been routine even in the days of the calculator (e.g., Friden, Wang). And accepted that in their paper itself they made no claim to causal influence.
    Yet your point on incentives and motives has been something of a bother, moderating influence in analysing most reports, especially empirical. That increased caution stems from the indifference or cynicism of well-recognised economists with regard to work or public statements or positions that would be otherwise considered as shoddy, atypical. Feldstein, Hubbard, and even Mankiw are a few names that come to mind. Absent any 'Economist Hippocratic Oath', and given incentives...

  6. derrida derider21 April 2013 at 09:08

    Shorter Jon Livesey:
    "Its far better that we ruin millions of lives by unemployment than that we risk inflation in the distant future".

    People are responding to this by arguing this is not in fact the empiric choice we face. They may be right - I'm not a macro guy and therefore not a dogmatist on stuff like this :-). But what we should be vigorously objecting to here is not the empiric claim but the implicit social welfare function. A bit of future inflation is a very small price to pay for dragging the world out of depression - if there is currently some downward slope on the short run Phillips curve then let's exploit it, and bugger the flatness of the long run one.

  7. derrida derider21 April 2013 at 09:21

    Oh, and as a civil servant (not a UK one), my career-long observations are that Anonymous above is quite right. Decision-based evidence making is much more common in real world policy circles than evidence-based decision making - Iraq was not an isolated case, for instance. I know - for my sins I've been involved in making such evidence for my masters, though fortunately none with such consequences.

    Now it's unfortunate civil servants answerable to scurvy politicians do such things sometimes, but for respected academics to do so is REALLY inexcusable. I think the sequence of events here ought simply to destroy R&R's careers. But of course that won't happen.


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.