Sunday, December 30, 2012

DeLong on Murphy

Wow - I forgot the bet was over 10 percent inflation by January 2013! Brad writes:
"The most terrifying thing of all is that being completely, comprehensively, unmistakably, fundamentally, fatally, totally wrong has not led Robert Murphy to rethink or modify any of his analytical positions or ideological beliefs by even one iota.

I mean, one would expect an announcement on his weblog like: "I have been totally wrong, about everything. I am closing down this weblog for five years to avoid misleading readers while I intellectually retool. You will find me sitting at the feet of Paul Krguman, chanting 'om mani padme hum' until I achieve enlightenment."

Not gonna happen..."
I think it's fair to say that he hasn't changed his perspective on how the economy works to date (he certainly hasn't dropped dismissing Krugman), although he has acknowledged how badly this has gone at a couple points.

Even more embarrassing for Bob is that he lost a potential bet he seemed curious/eager about with an MMT commenter on his blog about 5 percent inflation too.

One of the nice things about Brad DeLong is that he regularly lays out where he was wrong in the past and has changed his mind. I'm curious if any Austrians who were predicting things similar to Bob have thoughts along these lines as well. Bob is not alone, after all. Peter Boettke recently suggested that the crisis confirms Hayek's position. Is there any grounds for these sorts of claims? Should they be revising any viewpoints in the way that DeLong did much earlier in the crisis?

Often you'll hear Austrians say that we're at the zero lower bound because the central bank is pinning us there with easy money policy. That explanation begs a lot of questions itself, of course, but you certainly can't say that and then also explain how low inflation is consistent with your theory. The Keynesian story, of course, has an answer for both the low inflation and low interest rates.

What's interesting is that in Bob's defense of himself (a year ago!) he cites the eurozone crisis and the flight to quality. This is quite close to admitting the Keynesian position, if only he would make that leap. A flight to quality in the face of revised expectations about returns to capital investments of any sort in an uncertain future is precisely the mechanism here. That will give you what we observe today - not the realization of a discoordinated capital structure followed by expansionary monetary policy.

Brad DeLong doesn't think Bob can come to terms with all this. I'm not so sure. Bob was able to embrace Sraffa over Hayek. And if you thought the lefties liked Keynes, wait until you see what they think of Sraffa! Bob knows when to aknowledge the other side has a point without abandoning the good points that the Austrians make. So will he do it this time?


  1. How does he know that? I had assumed David R. Henderson mentioned it at EconLog, perhaps in the comments, of his recent post on predictions. But I don't see it there.

    Was DeLong marking off his calendar for this day?

  2. Replies
    1. Seeing AP Lerner's offer to you made my day :)

    2. This comment has been removed by the author.

    3. Why? I don't understand why you think it's more embarrassing for me that I thought inflation would break 5% than it would break 10%. Shouldn't I be more embarrassed about the latter?

      To put it another way: It would be really embarrassing if I were willing to bet David it would break 10%, but I weren't willing to bet someone else it would break 5%. I really don't get why this tickles you.

    4. I was meaning that it was one of your MMT pesters. Then again, that AP Lerner would have won drives home how far off the assumption about inflation was.

  3. Economists really should bet more. Putting money and/or reputation at stake keeps things honest. We shouldn't be pointing fingers at Bob for being wrong, we should be applauding him for betting.

    1. *like*

      I'm not sure betting is always the solution, but I like the last sentence. This post wasn't meant to make fun of Bob, but it's definitely worth asking what ought to be revised as a result of the loss.

    2. Nah. Bryan Caplan bets all the time, but it doesn't change his views in the least.

      (Then again, this is the man whose book claimed children do not require a lifestyle change at the 2011 Kauffman conference and then didn't attend in 2012--because his wife had another child. I expect him to double down with Still More Selfish Reasons to Have More Children)

    3. Ken -
      Just because you don't have to change for your kids doesn't mean you don't have to change for your wife!

      Those are words of wisdom there, kiddies: pay attention!

  4. Daniel,

    A bit of a digression but I have a question about inflation within the Keynesian framework.

    When I is depressed during a recession then this leads to falling Y. The govt steps in and increases G, which allows Y to rise back up again. What previously was Y = C + I now is Y = C + I + G .

    In a traditional Keynesian model G is spent on things like roads and bridges which may add economic value but don't normally raise revenue. So the govt has spent money on wages for labor and capital goods that otherwise would not been used and this govt spending (directly and via the multiplier) leads to an increase in Y.

    My question is this: G adds to nominal incomes in the economy, but the goods produced by G (roads and bridges etc) don't normally lead to additional goods being produced for this additional money to be spent on. Compared to pre-recession times there is now the same E chasing fewer goods. Why is this not inflationary in the Keynsian model?

    1. Huh? Y is always C + I + G (+ NX, but let's work closed economy).

      All that's happening is dI (and generally dC) have declined and you would need dG to increase to keep Y the same.

      At time t I produced n units, expecting to sell them at t+1 for m dollars. dC goes down as people are laid off, and I can sell only n-x units for m dollars. Revenues: (n-x)*m, which means dI must decline.

      dG is an increase in unemployment benefits, food stamps, medical care, etc. Money is fungible, so some of those people who receive benefits spend that money--maintain some savings (I; recall, in the GDP equation S=I) they would otherwise have spent. Now I sell n-x+y (y<x) units for m dollars, and my I goes up.

      But unless the trillion-dollar-platinum-coin is circulated, I still sell less than I planned to sell.

      Under no condition is that inflationary.

  5. OK, so I agree that if G increases just sufficiently in t+1 to offset the fall in sales revenue due to the decline in demand then E will match expectations, and tyhe changed G will not be inflationary.

    However if dI falls won't that mean a decline in private investment for any given level of Y (where G is adjusted to keep Y steady) ?

    For example: dI decreases between t and t+1. G is increased in t+1 to maintain Y. The lower dI in t+1 leads to less goods being produced , but the increase in G means that Y (and E) remains the same. The same E being spent on less private goods seems like it should lead to a rise in prices.

    1. Lower dI would lead to less of an increase in goods, not less goods produced, so if dG perfectly countered dI the economy would be stationary. If dG exceeded dI as it did during WWII, then you could see inflation or rationing but dG is almost never that large.

    2. And as long as dG grew less than the potential growth of the economy, say 3%, no inflation would result.

  6. Food for thought: fundamentals-based investors were losing out on real "bets" left and right leading up to the tech stock crash of 2000. From the mid-90s they kept pointing out that "price-to-eyeballs" and other fanciful metrics were unjustified, and positioned themselves accordingly, but many went out of business before their arguments were empirically validated shortly thereafter. And of course, there's Irving Fisher's infamous "permanent plateau" remark just before the 1929 crash.

    Being early is famously indistinguishable from being wrong.


All anonymous comments will be deleted. Consistent pseudonyms are fine.