Thursday, July 15, 2010

Assault of Thoughts, Special Stimulus Edition - 7/15/2010

"Words ought to be a little wild, for they are the assault of thoughts on the unthinking" -JMK

- David Henderson ably explains why it's hard to make definitive statements about the effect of the stimulus because of our inability to identify a counterfactual, and then goes on to make definitive statements about the stimulus despite his inability to identify a counterfactual. See my comments in the comment section - it's the first one. [UPDATE: David Henderson grants the point in the comment section of his blog, which is not surprising at all because he clearly employed the logic w.r.t. the CEA, he clearly understands the logic, and he clearly is a genuine guy that is more than happy to apply the logic to himself. It was a momentary oversight that happens a lot in quick blogging. I just found the stark juxtaposition between what he was dishing out and what he was claiming surprising. I agree with Henderson probably the least of the three Econlog bloggers, but it's not my intent to critique him for the sake of critiquing him. Sorry for the delay in updating this, David! Thanks for visiting the blog.]

- Greg Mankiw has a better critique of the same issue (Henderson cites Mankiw - he should have said "Mankiw's dun' good" and left it at that). All we have available to us is to try to understand how fiscal and monetary policy effect the macroeconomy under given circumstances and then apply what we learn or think we've learned. We can't point to one policy and one downturn and say "this is what happened". We can infer what happened based on what we know about fiscal and monetary policy, but ultimately that's all it ever is, and we've just gotta be straightforward about that. It doesn't mean you can't have a firm sense of what policy will do - that's fine to have. Just don't defend that on the basis of a single event for which you have no counter-factual.

- Brad DeLong shares some of Jean-Baptiste Say, and suggests that Say "gives what is the first thorough explanation I have ever seen of irrational exuberance, overtrading, excessive leverage, disappointment, panic, flight to quality, excess demand for high-quality and liquid assets producing excess supply of goods and services and labor, falls in incomes, and the fall in incomes causing the initial problem to snowball...". I have to take his word for it, because fancy-pants DeLong quotes Say in French. Nevertheless, I've seen English selections before too that confirm this view of Say. DeLong concludes that Keynes was ultimately unfair to Say. I don't think this is exactly new news. I personally view Keynes's version of Say's Law as a heuristic device rather than a probing history of economic thought. In that sense, the fact that Keynes's Say's Law was a strawman has never bothered me that much. It was - but it was also a mindset that the Classical School slipped into or was at risk of slipping into. In that sense, it was still an important backdrop for the General Theory - it was a way of saying "insofar as I am different, these are the grounds on which I'm different", even if the Classics didn't conform exactly to what has been called the "strong version" of Say's Law. If Keynes were presenting a review of Say's works, it would obviously be inappropriate, but I think you have to consider the context.

- David Indiviglio makes an important observation: the administration's forecast of inflation may have been way off, but it's forecast of GDP was pretty much on target. What does this mean? Well, it may mean that Keynesians have the relationship between stimulus and output right (which is pretty good, since that's the primary mechanism through which stimulus works!), but that the underlying relationship between GDP and unemployment is behaving unpredictable. Harry Holzer, of Georgetown University and the Urban Institute, makes this point a lot - that Okuns Law is not functioning as we expect it to. This is an important point for guys like David Henderson who point to that unemployment forecast, forget everything they just cautioned us about when it comes to counter-factuals, and claim the stimulus didn't work. What is the stimulus designed to impact? Demand. And how does that work on the economy? Through output and prices. The impact on output followed predictably (I talk about why nothing surprised me about the performance of output in response to stimulus here). But something is breaking the expected link between output and employment. This is actually a very important critique that Hayek made of Keynes - that he didn't spend enough time on the link between output and employment. Recent events seem to suggest that that relationship is worth a closer look.

10 comments:

  1. "If Keynes were presenting a review of Say's works, it would obviously be inappropriate, but I think you have to consider the context."

    No, it is still inexcusable. If Keynes were an honest agent he would have simply stated "this is what I don't like about the classical school - or some elements of the classical school," but that wasn't nearly as provocative (and career enhancing) as attacking Say.

    ReplyDelete
  2. As for GDP, I'm confused about that claim. There have been a number of stories in the press about the Obama administration being significantly off in its estimates. See for an example of such: http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aNaqecavD9ek

    ReplyDelete
  3. 1. Your article is from August of 2009. Yes, by the summer people realized it would be worse. But this is in the summer of 2009. Look at BEA real GDP growth rates and look at the CEA's estimates. They match pretty close in early 2010, and in the last quarter of 2009, CEA actually underestimates GDP growth. I don't see how you can determine if they were off using an article from a year ago. Look at the data from this year and compare it to their numbers (1.) GDP didn't do so bad, (2.) unemployment prediction was way off. The question is, why? Okun's Law is not holding right now and its unclear why.

    2. What exactly do you think was an "attack" on Keynes's part? I reread the passage in the GT and it's not that harsh. What part are you thinking is harsh, exactly? This is also problematic because there are two Say's. The 1803 Jean-Baptiste Say does seem to believe Say's Law. The 1825 Say doesn't seem to.

    ReplyDelete
  4. Daniel,
    I notice that you didn't give me credit for granting your point. Do you believe in updates?
    David

    ReplyDelete
  5. I most definitely believe in updates! My apologies!

    ReplyDelete
  6. Nice discussion of the "stimulus" here: http://online.wsj.com/article/SB10001424052748703394204575367421573463984.html?mod=googlenews_wsj

    ReplyDelete
  7. Thanks, Daniel. I appreciate your "genuine" comment.
    Wow! You disagree with me more than you disagree with Bryan even though he's a more-radical libertarian than I am. Interesting. It would be fun to talk about why sometime.
    Best,
    David

    ReplyDelete
  8. It's only an impression.

    I think what I like is Bryan's speculative approach - I feel like I can enjoy thinking through his ideas perhaps because I never take any of them as seriously as he intends readers to! Speculative thought, however radical, is always useful in my mind. I think the fact that he is more neoclassically minded also makes it easier for me to embrace some of what he writes.

    Honestly you're also a lot like me - you can be a blunt blogger. It may just be that I feel we bump heads more often because of that style, although it's very likely that on the issue itself I may indeed disagree with Bryan more often.

    Like I said - just an impression :) Thanks for visiting - please do so often!

    ReplyDelete
  9. For example - "libertarians need to get more fertile - we need to breed out other political groups".

    It's entirely possible that the reason I can say "I enjoy reading Bryan Caplan" is that something in my head hasn't quite clicked into "he really, truly, is serious about this" mode. If it did, I might be saying "wow, that Bryan Caplan is a real weirdo" :)

    ReplyDelete

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