Everything that needs to be said has already been said.
But since no one was listening, everything must be said again.

Andre Gide

Saturday, May 22, 2010

Greg Mankiw on Fiscal Policy

Just over one year ago, when "fiscal stimulus" was all the rage, I asked what I thought was a simple question: How do the proponents of fiscal stimulus know that it works? More precisely: what evidence can we bring to bear on the question of whether a large increase in government spending during a peacetime economic crisis stimulates GDP in a welfare-improving manner? (I allowed for the possibility that such a spending program may have desired redistributional aspects, but redistribution is patently not what people talk about when they speak of the benefits of fiscal stimulus--it is something that is supposed to make us all better off).

In case you're interested, you can refer to:

Does Fiscal Spending "Work?"
Believing in Fiscal Stimulus.
Believing in Fiscal Stimulus 2.

I appeared to have hit a nerve with some people on this topic. One chap named Bruce Wilder was particularly amusing (that is, if you find appalling ignorance and santimonious drivel amusing). This "debate" had the side benefit of leading me to think about a theory of religion. In any case, the record of this exchange can be found here:

Bruce Wilder on Andolfatto.
Religiousity in Macroeconomics and the Sad Case of Father Wilder.

As far as I was (am still) concerned, it all boiled down to personal beliefs (religion). There is simply not enough data (as far as I am aware of) that could lead any honest (agnostic) scientist to come down strongly on one side of the debate or the other. It led me to question to academic integrity of strong proponents of fiscal stimulus like Paul Krugman and Brad DeLong. (I do not feel quite the same way for strong opponents, as they are usually more honest about their beliefs: they simply do not want the government to interfere in the lives of its citizens, period. However, this may simply reflect a defect or bias on my part).

And then along comes Greg Mankiw, a self-described Keynesian. He recently published this in the Federal Reserve Bank of St. Louis Review: Questions About Fiscal Policy.

Mankiw is a gifted writer. I encourage you to read it. He begins the paper with a great analogy. But what I really loved was this line:

I am actually a believer in Keynesian theory; much of my research is in that field. But even as a believer in many aspects of Keynesian theory, I appreciate that you cannot approach this subject matter without showing some humility about what we, as economists, can truly be confident about.

My only quibble is why he only included economists in that worthy sentiment.

Other than that, all I have to say is:



  1. Is Mankiw a genuine Keynesian, or one of these "neo-Keynesians" I keep hearing about? If the former, than I would suggest he read Hazlitt's "Failure of the New Economics." If the latter, than I don't know what to say.

  2. Oh, there is lots of data.

    See, for example, Miguel Almunia, Agustín S. Bénétrix, Barry Eichengreen, Kevin H. O'Rourke, and Gisela Rua (2009), "From Great Depression to Great Credit Crisis: Similarities, Differences and Lessons." http://www.nber.org/papers/w15524

    You have to close your eyes, put your fingers in your ears, and work really hard to pretend that there isn't lots of data.

    But why would anybody want to do so?

  3. Prof J: You'll have to ask Mankiw that question. I would guess that he is not hardcore in any particular school of thought. You'll note that he remarks that he is a believer in "many aspects" of Keynesian theory. Presumably, this means that he disagrees with other aspects of the theory. Exactly what this means though, I'm not sure.

    Brad: You should have stopped with your reference, which I highly appreciated. What you have to say next reveals something about you that is not entirely flattering. I suppose this is why you prefer to make such remarks behind the bold cloak of anonymity.

  4. I think the new key puzzle in macroeconomics is why output remains relatively low for such a long time following a financial crisis. This is (according to some) the puzzle of the great depression; of Japan in the 90s; and perhaps of the U.S. right now.

  5. I have just finished reading the paper linked to by Brad, above. It is an interesting addition to the literature on fiscal and monetary stimulus. I think that one must be very careful in interpreting the results though.

    They instrument for fiscal stimulus by using defense spending during the period 1925 - 1939. I'm going to check with my history books, but Europe and Asia were gearing up for war during the part of that period. Moreover, defense spending and fiscal stimulus are practically not equivalent, so I think it is not a good instrument for trying to calculate a precise point estimate for a fiscal multiplier.

    Second, the dependent variable is real GDP growth. I mean total GDP, which includes government expenditure. Seems like you should look at net real GDP at least - real GDP less real government expenditure. I can expand the government like crazy (for a while) and get some sweet looking growth in real GDP. That doesn't mean fiscal stimulus "worked" in any sense.

    I'm afraid the only way to get a real understanding of whether stimulus works is to get into the details of any particular "package" and look at the micro effects on targeted industries. I know Bush's "airdrop" is straight up stimulus, so that's fine, you can check an aggregate effect there (pst... people put the money in the bank!) but most are not this straightforward.

    In short, I'm not buyin' what they're sellin'. But, really, I have the same complaints about Robert Barro's work (the defense spending issue). I just don't think it is a very accurate way of getting an answer to these timely questions.

  6. Mo: is it not the case that in some recent financial crisis episodes, economies recovered relatively quickly (and without fiscal stimulus)? Scandinavia in the early 1990s comes to mind, but I may be wrong. In any case, prior to the Great Depression, the US economy always recovered relatively quickly from financial crises.

    Prof J: I have skimmed through the paper cited by Brad. It looks like a valiant attempt at measurement and should be taken seriously. The authors are also very careful to highlight the shortcomings of their methodology (not that this would have been notice by Brad, who appears to have made up his mind even before this paper came out).

    I believe that this is a very tricky issue to sort out conclusively. We simply do not have enough of the right kind of data. Even proponents complain that fiscal stimulus was not tried on a large enough scale in Great Depression U.S. or Japan 1990s (hence the emphasis on wartime experiences).

    Moreover, so many details are likely to matter: The composition of purchases, the fiscal situation, the size of government, the financial market, etc., etc.

    Having said this, I think that some progress is being made and will continue to be made. Until we know for sure though, I recommend a healthy dose of Mankiw humility for all of us.

  7. Do we have any recent polling data on the attitude of academica and/or professional economists to discrete fiscal stimulation?

    Strikes me that there is a solid consensus behind the use of counter-cyclical monetary policy but absolutely no consensus whatsoever on the benefits of discrete fiscal stimulation.

  8. No doubt, proposed policies are operating under conditions on extreme uncertainty and limited precedent.

    But the fair question should not be "Is fiscal policy effective or not?" but "Is fiscal policy helpful or harmful?" If fiscal policy is not harmful on the short term, potentially helpful, and not catastrophic on the long term, then we have good reason to go ahead and try it.

    When evaluated on these terms it seems fiscal stimulus wins out.