October 2017

Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31        
Blog powered by Typepad

« Rose Director Friedman | Main | Knowing What And Knowing How »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

great post. may I add that state of the art monetary macro a la woodford studies optimal monetary policies for inessential sequence economies (hahn), that is, for time-series which are just pseudo-dynamic reformulations of arrow-debreu economies which of course fnd no place for money at all. insane isn't it? this also hints to the problems with the taylor-rule. even though I have little doubt that taylor is right, that economic peformance would have been better if central banks followed his rule, this rule is no theory and lacks proper foundations since it is build on aggregate reduced-form models with representative agents. computational macro drives me nuts! here, I prefer the "hyperformalism" of radner, etc.

amv

Note well that Friedman in a 1975 letter to Hayek insisted that statistical testing would be able to adjudicate these things (a letter which ultimately betrays Friedman's complete ignorance of Hayek's picture of economic science / explanatory strategy in economics).

Samuelson in the 1940s and 1950s thought the same thing -- then confessed in the 1980s that his expectations about that statistics would be able to do had been utterly falsified.

I don't remember Friedman every finally confession what Samuelson acknowledged.

Pete writes:

"unfortunately, this situation cannot be resolved via sophisticated statistical testing. In short, statisitcal testing cannot by itself ajudicate between competiting hypothesis derived from the different formalistic models."

"And there is also a unique Austrian or market process strand in price theory that is vital --- a strand that has the logic of Arrow, the universality of Becker, and the sensibilities to evolutionary adjustment of Alchian. The ideas of Menger, Bohm-Bawerk, Mises, Hayek and Kirzner would be a central part of price theory."

Boettke, I have the perception that these 5 great authors that you cite are truly the most rigorous thinkers of the Austrian school, but while they are razor sharp in logic I had more difficulty understanding them (particulary Mises) compared to the more graphic/formal authors. I started studying economics around early 2004, when a read Human Action, The Road to Serfdom, Competition as a Discovery Procedure, etc, I learned a loot in those days but I noticed that I only understood what Mises really meant in 2008, after some "maturing" period (for example, only in 2008 that I really understood the relation between uncertainty and human action). The problem is that much is implicit in Mises (Hayek is more explicit than Mises and Kirzner is more explicit than Hayek). For comparison, in mathematical economics everything looks like that the theory is "in the book", while for the logical economists like Mises, the book represents only a constellation of manifestations of a underlying/implicit logic. It is a very tough nut to crack, I think that that is one of the main reason that so few people are Austrians (around 100?).

Peter

Enjoyed the post and I agree that John Taylor has done some interesting work within the fledgling "what went wrong" literature.

But where was John Taylor in 2005 in the midst of the housing boom when his Taylor Rule was being ignored by the Greenspan Fed. I'll tell you where he was, he was in Jackson Hole Wyoming at the KC Fed's Monetary Conference giving a hagiographic comment on Alan Blinder's paper: "Understanding the Greenspan Standard". In this paper he concurred with Blinder that "when the score is toted up, we think he has legitimate claim to be the greatest central banker who ever lived". Taylor's paper ends with "That Alan Greenspan's leadership is grounded in substance has been a huge positive for policy". No criticism of the Fed until 2008 when it was obvious to the world that the Global Financial Crisis had commenced.

Meanwhile, other mainstream economists and fellow attendees at the Jackson Hole conferences such as Claudio Borio and William White, using a mix of Austrian and Minskian insights, were not afraid to trash the Fed even as Greenspan himself sat in the audience.


@knapp: Greenspan even left the room as White started is lecture!

I agree with knapp and amv. Things might have been better if the Taylor Rule had been adhered to, but I do not see it as having prevented fully what happened. The housing bubble began in the late 1990s, and a major element of it was the spread of flakey derivatives based on the bubble, this latter having nothing to do with following the Taylor Rule or not doing so. We had a huge bubble, and its crash was going to cause a big mess, which it has.

Regarding the matter of statistical testing, well, it should be done, but it should also be kept in its proper place, given the numerous problems. There are alternatives to that for some things, especiallly regarding fundamental assumptions.

So, we know from repeated economic experiments that assuming rational expectations as people like Lucas and Fama and Woodford do without batting any eyes, is simply empirically wrong. People do not have ratex in general. Nevertheless, there was Lucas a few weeks ago defending Fama and all sorts of nonsense in an Economic Focus column in The Economist where he was taking to task the earlier cover story on problems in economic theory, all with an argument that implicitly assumed ratex, which the recent events have highlighted as being utter nonsense.

Great and very insightful posting. Awesome list of recommendations.Thanks a lot for this.Write really good, I feel so cool to see the next time you see your new article.

The comments to this entry are closed.