Network News

X My Profile
View More Activity

On macroeconomic models

My interviews with Alan Blinder and Mark Zandi focused, probably to the great boredom of my readers, on the nature of the model they used to estimate the effects of different stimulus and financial policies. I did that because this question of models is an important one: If you believe the macroeconomic models, they say the stimulus has worked extremely well. If you don't, then it's harder to answer.

This isn't a debate I'm qualified to judge, so I wanted to make sure I linked to the other side as well. An admirably clear summation of the case against models can be found in this National Affairs essay by Greg Mankiw. It's a good, easy read.

A more opinionated and declarative case against models (and the Zandi/Blinder paper in particular) can be found in this post from Arnold Kling. The Kling post goes a bit far. Plenty of respected macroeconomists still use macroeconomic models, which is in itself a refutation of the idea that "the profession has decided that this macroeconometric project was a blind alley." Alan Blinder, for instance, is part of the profession, and so is the president of the Minneapolis Fed, who wrote this paper on macroeconomic modeling. It's also worth noting that the private sector relies extensively on these models, and it would be odd for them to give Moody's all that money if they thought there was no predictive value. Still, Kling's post is a good place to read the strong case against evaluating economic effectiveness this way.

If readers have other interesting links on the subject, leave them in the comments section. I'm oddly fascinated by this debate, and the implications are obvious: If the models are fairly predictive, than we should be listening to them and, say, directing money away from tax cuts and towards state and local aid. If they're not predictive, then we're obviously dealing with more of a jump ball.

By Ezra Klein  |  July 29, 2010; 12:12 PM ET
Categories:  Economics  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Blinder: Bank profits 'collateral damage in a war to save the economy'
Next: Lunch break

Comments

From Worthwhile Canadian Initiative, "What modern art and economic theory have in common:"
http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/07/what-modern-art-and-economic-theory-have-in-common-1.html

Posted by: trevindor | July 29, 2010 12:18 PM | Report abuse

I would say the best predictive model is listen to what Republicans say and then do the opposite.

Posted by: johnsonr1 | July 29, 2010 12:47 PM | Report abuse

I suspect Blinder and Kocherlakota mean somewhat different things when they say "macroeconomic models." Indeed, this is in part what Kling is talking about. I expect the private sector that relies on macro models is relying on models that are more of the Blinder sort than of the Kocherlakota sort. That, in fact, would be an interesting survey exercise.

Posted by: bdballard | July 29, 2010 12:47 PM | Report abuse

The problem with johnsonr1's suggestion is that what Republicans say is often so muddled that it would be hard to figure out what "the opposite" entails!

Posted by: bdballard | July 29, 2010 12:49 PM | Report abuse

Andrew Lo of MIT (business school) and Mark T. Mueller (physics department) attempt to create a taxonomy for what can and cannot be explained in this paper: http://web.mit.edu/alo/www/ (there are actually more levels inbetween those extremes).

Posted by: chrisgaun | July 29, 2010 1:42 PM | Report abuse

Sorry, the link only goes to his website. The paper is located here: http://web.mit.edu/alo/www/Papers/physics8.pdf

Posted by: chrisgaun | July 29, 2010 1:45 PM | Report abuse

oil spill? record unemployment?
giggle giggle, joy behar' and
da view. whoop whoop!!!

Posted by: simonsays1 | July 29, 2010 2:32 PM | Report abuse

The suggestion in Mankiw's article that lower tax rates give an incentive to invest in enterprises (with the implicit corollary that high tax rates stop people from investing) would seem to be undercut by the history of the Eisenhower years. Despite a top marginal income tax rate of 90%, the economy seems to have done pretty well back then.
And I recall reading recently that businesses were sitting on cash rather than hiring, and that business leaders gave as their reason the lack of demand for products and services. (Sorry - I didn't bookmark, so can't give you links. But here's a link to one of Ezra's posts from three months ago that still seems pertinent: http://voices.washingtonpost.com/ezra-klein/2010/04/its_the_demand_stupid.html )

Posted by: quickj | July 29, 2010 3:12 PM | Report abuse

If modeling doesn't work,we really don't need economists,at least not in policy-making positions. After all, if modeling doesn't work, then there is no more reason to adopt Suggested Policy A rather than than Suggested Policy B or Suggested Policy C, etc., because no theory has any real predictive value or, at least, no predictive value we can test after the fact.

Posted by: sltax1 | July 29, 2010 4:53 PM | Report abuse

If modeling doesn't work,we really don't need economists,at least not in policy-making positions. After all, if modeling doesn't work, then there is no more reason to adopt Suggested Policy A rather than than Suggested Policy B or Suggested Policy C, etc., because no theory has any real predictive value or, at least, no predictive value we can test after the fact.

Posted by: sltax1 | July 29, 2010 4:53 PM | Report abuse

The section where Mankiw uses the tax multiplier from Romer & Romer but the spending multiplier from Valerie Ramey is b.s. and Mankiw knows it. After Mankiw posted this prima facie suspicious claim on his blog the first time, Brad Delong about his discussions with David Romer and it seems clear that an apples to apples comparison cannot be made to the Ramey paper.
http://delong.typepad.com/sdj/2009/01/the-romer-view-of-tax-and-spending-multipliers-revisited.html

Given all the bawwwing that Mankiw does about the difficulties of measuring and modeling, he is suspiciously credulous in repeating this Romer/Ramey comparison.

Posted by: zosima | July 29, 2010 5:37 PM | Report abuse

--"If the models are fairly predictive, than we should be listening to them and, say, directing money away from tax cuts and towards state and local aid."--

How much money are you willing to steal, Klein, to satisfy what level of predictability? If Klinger found that you could shoot your grandma and shave a percent off unemployment, would you do it?

Posted by: msoja | July 29, 2010 8:35 PM | Report abuse

Klinger??? I meant Zandi. Or Blinder. Maybe even Krugman. But I think I got all three.

Posted by: msoja | July 29, 2010 8:36 PM | Report abuse

JOURNOLIST mean any to the Post?

Credibilty and reputation of free press forever stained on Post. Disgraceful!

Posted by: jkptak | July 30, 2010 12:18 AM | Report abuse

Kling responds once again: http://econlog.econlib.org/archives/2010/07/some_further_co.html

Klein is way over his head here...macro does not appear to be his cup of tea.

Posted by: WilliamFreeland | July 31, 2010 10:01 AM | Report abuse

The President of Minneapolis Fed, cited by Klein, says "by and large, journalists and policymakers—and by extension the U.S. public—think about macroeconomics using the basically abandoned frameworks of the 1960s and 1970s." He is referring to exactly the sort of econometric model Mr. Zandi tried to bring back to life "nearly twenty years ago." If such models worked, outfits like economy.com would win all the forecasting awards. They don't. And for Zandi to say "we" (meaning he) didn't know what the unemployment rate was in early 2009 because of "data lag" (one month)is a pathetic excuse for another truly lousy forecast, just like Zandi's upbeat views on housing a year earlier. Should we really make $864 billion policy choices on the basis of this mysterious black box? Blinder should be ashamed of associating with this statistical mumbo-jumbo.

Posted by: RobinPeter | August 1, 2010 11:21 AM | Report abuse

John Taylor had some interesting and insightful things to say about Blinder and Zandi's study -- http://johnbtaylorsblog.blogspot.com/2010/07/more-on-blinder-zandi-working-paper-on.html -- and he even "debated" Zandi on the PBS Newshour -- http://www.pbs.org/newshour/bb/business/july-dec10/economy_07-29.html

But why not simply rip the heart right out of the Keynesian dragon by demonstrating Keynes' and Keynesians' abysmal failure but dire need to refute "Say's Law" -- http://www.youtube.com/watch?v=rIgkbdT5V6w

Posted by: iamrj | August 5, 2010 4:16 AM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company