Winner of the New Statesman SPERI Prize in Political Economy 2016


Sunday 11 May 2014

Yes, economic policy did fail

Tony Yates today criticises Paul Krugman’s argument that economics had the answer to how to respond to the crisis, but policy failed to follow the prescription. As I completely agree with Paul Krugman on this, let me say why I think Tony’s criticism is completely wrong. 

The argument he is criticising is that, following the recession, we had a demand deficit that fiscal stimulus could have tackled, but from 2010 policy went for fiscal contraction instead. How do we know that we had a demand deficit? Because interest rates went to their zero lower bound pretty well everywhere. Monetary policymakers had to go for largely untried and untested Quantitative Easing to try and plug the demand gap. Fiscal stimulus is a much more reliable method of achieving the same goal. That logic is, in my view, unassailable.

Tony’s first argument is that, despite this, we got lucky. If you look at inflation, “conventional demand-side fiscal policy was roughly on track”. So, when economists in twenty years time look back on the 2008/9 recession, Tony thinks they will describe it as a textbook example of how policy should deflate the economy in response to an inflationary shock. Policy did good - the recession was just what was needed to stop inflation going higher. Somehow I really doubt that is the story they will tell.

Tony defends this argument by invoking the social welfare functions implied by New Keynesian models, where inflation is much more costly than non-zero output gaps. At which point I do worry about what today’s macroeconomists actually believe. These social welfare functions may capture the costs of inflation - or at least the costs due to relative price distortions - but they certainly do not capture the costs of output gaps at all. We have a huge amount of evidence that tells us this. Just because we have not yet microfounded why people hate being unemployed so much does not mean they really don’t mind.

Tony acknowledges this, but then says “But, with the models thus binned, you are in the dark about what should be done.” That is just silly. New Keynesian models do not stand and fall according to the accuracy of the social welfare functions they imply. After all, many NK models have a labour market that clears. Economists use them not because they think the labour market actually clears, but because they give answers about output gaps and what to do about them that are not too far off. So there is absolutely no problem using a NK model with an ‘ad hoc’ social welfare function where weights follow the evidence rather than the model. If we do that, and use other measures of inflation besides consumer prices (as theory suggests we should), then 2009-13 does not look like an optimal response to an inflationary shock.

Then Tony falls back on the argument that there is so much we still do not understand about the financial crisis, so how can anyone argue that the economics is clear.

“PK seems to be backing away from all these intractable debates about the detail, and saying that we can ignore it.  Big picture, demand was weak, public demand had to be stronger.  Politicians did not get this message clearly enough, and were able to ignore it.  End of story.  Well, maybe.  Maybe not.  Perhaps only great minds can see the wood for all these unfinished modelling trees.”

No, you do not need to be a great mind to do this, and I should know. You just need to assess whether the things we do not understand can seriously compromise what we do know, which was that we had deficient demand and we knew how to deal with that. The only possible factor was - briefly - the Eurozone debt crisis, but then a balanced budget fiscal expansion could have been used to avoid increasing debt.

So Tony’s argument that the reason for this “policy failure is that our economics profession had not yet come up with clear answers” is not tenable. As he knows, monetary policymakers cope with uncertainty all the time, but we still tell them (or at least most of us do) that you cut rather than raise interest rates in a recession. For exactly the same reason, you undertake fiscal stimulus, not contraction, when interest rates are at the zero lower bound. It really is that simple. 


23 comments:

  1. "following the recession, we had a demand deficit that fiscal stimulus could have tackled, but from 2010 policy went for fiscal contraction instead. How do we know that we had a demand deficit? Because interest rates went to their zero lower bound pretty well everywhere. Monetary policymakers had to go for largely untried and untested Quantitative Easing to try and plug the demand gap."

    The central bank could instead directly interact with the broad public in monetary expansions which would be much more efficient becuase the broad public have a higher propensity to consume and lower liquidity preference than QE counterparties.

    ReplyDelete
    Replies
    1. That's what QE does. The central bank buys mortgage backed securities and corporate bonds from the public.

      Delete
  2. "How do we know that we had a demand deficit? Because interest rates went to their zero lower bound pretty well everywhere."

    I don't understand this reasoning. Interest rates went to the zero lower bound because central bankers lowered the rates. Central banks lowered the rates because they saw deficient demand. So, as near as I can make it out, the reasoning is: if central bankers see deficient demand, then there is deficient demand. This gives central bankers a certain omniscience which I don't think belongs to them. BTW I'm not arguing against the fact there was "deficient" demand; just against this particular kind of reasoning.

    ReplyDelete
    Replies
    1. "How do we know that we had a demand deficit? Because interest rates went to their zero lower bound pretty well everywhere.

      I think you are right to associate the price of credit with demand in the real economy.

      I think there was demand, but the flow of funds system stopped working because banks needed to readjust their loan/deposit ratios, which makes the case even stronger for an injection of government credit.

      Delete
    2. Anonymous 01:16

      I think it's that they lowered rates as low as they can go and we didn't see inflation.

      Delete
    3. This comment has been removed by the author.

      Delete
  3. "Nerd savant" was what Krugman called the PhD bankers whose models couldn't fail in 2008.

    I think that term might need a dust down.

    ReplyDelete
  4. So economics did have the answers - how do we make sure that next time, the answers influence policy?

    ReplyDelete
    Replies
    1. By having a dictatorship.

      Paul Krugman seems to think that ISLM models also obeyed optimisation conditions and also had the answers and......I need a drink.

      Delete
    2. But they did! The Q about how we avoid this happening again is a very serious one, and I'll be addressing that in future posts.

      Delete
    3. The first step to any hope for sane policy in any subject is to destroy the republican party. That doesn't sound high-minded enough but it happens to be true.

      The first step to destroying the republican party is for the media to report, on page one, as straight news, the clear and objective fact that the republican party is insane.

      The first step to the media reporting that fact is for everyone capable of seeing it to call it out for what it is, avoiding the temptation to make it more fair or nuanced.

      Delete
    4. (That last post was clumsy - the media is always 20 years behind the obvious. We have to destroy the republican party without them.)

      Delete
    5. @steeve I'm be more concerned with the few dozen billionaires that have bought the Republican party than the party as a whole.

      Delete
    6. I've thought about that problem and have decided to be flexible - either one or three injections would be OK with me.

      Delete
  5. so,

    people like the pareto efficiency concept where optimally in any economic situation or goverment situation, everyone should be helped or at least one person should be helped and no one else should be hurt

    by hurt they mean lost money


    but all the times the rich are benefitting at the expense of the poor and middle class

    fractional reserve banking helps the bankers (interest on loans) more than the depositers, lower interest paid on depositers

    so that money is always flowing from poor middle class to rich

    so government intervention, especially fiscal policy, means investment

    which is redistribution of wealth from a lower multiplier place to a higher multiplier place

    must "hurt" the wealthy in that they will lose some of their ever increasing wealth, back to the people they took it from in the first place

    ReplyDelete
    Replies
    1. So next time (and the lack of real reform ensures that there will be a next time), what if we were to save the middle class and let the big banks wallow in their mess? For example, by allowing a trillion-dollar tax holiday to the lowest income brackets?

      I think this would have had a number of very good effects, including supporting inflation (thus allowing monetary policy to function), avoiding housing asset deflation, and killing off the vampire squids, which are otherwise pretty much unkillable.

      Delete
    2. Someone has to do something about Goldman Sachs and other investment bankers. First the industry gets a bailout for their incompetence from government (which basically means savings from low and middle income earliers). Now as a reward for their remarkable lack of talent they get their high salaries back. The answer is to tax them to the hilt. And if people say we need them, they are "good for the City and therefore the country" - call their bluff.

      Delete
  6. Isn´t the answer Kalecki 1943 Political Aspects of Full Employment. While a bigger recession than necessary may harm individual entrepreneurs, capital collectively has an interest in keeping the state small and the economy below full output, both goals consistent with depression austerity. It even has the perfect ideology to secure popular legitimacy for the programme: the economy as household.

    ReplyDelete
    Replies
    1. Yes macroeconomics doesnt exist...its the same as microeconomics....

      we can't spend money we don't have

      Would you run your household that way

      Etc

      Delete
  7. «After all, many NK models have a labour market that clears. Economists use them not because they think the labour market actually clears, but because they give answers about output gaps and what to do about them that are not too far off. So there is absolutely no problem using a NK model with an ‘ad hoc’ social welfare function where weights follow the evidence rather than the model.»

    That is yet the clearest Milton Friedman style statement that some Economists use certain models and microfoundations because they justify the "right" answers.

    To the point that as long as the models justify the "right" answers they may even tolerate basing part of them on the evidence instead of finding which assumptions deliver the "right" results. The "Lucas critique" seems to be just an excuse... :-)

    ReplyDelete
    Replies
    1. That is not what I said or what I meant. The justification has to be, and I think this is clear in Woodford for example, that adding a non-clearing labour market makes little qualitative difference to the answers, so it is OK to do some work with something simpler.

      Delete
  8. What annoys me is the combination of dogmatism, and then whining and studenty "oh, we can't really know anything 'cos of quantum". Mate, I can remember when There was No Alternative!

    ReplyDelete
  9. What is the principle that I should explain to policy makers that tells them to listen to SWL and PK and not other economists? It's not about this or that model, it's about how the reasonable non-economist sorts thru the mess that you call a science.

    ReplyDelete

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.