Economics
Bad economists
Sep 15th 2010, 14:27 by R.A. | WASHINGTON
THERE was an interesting little discussion in a corner of the blogosphere this week over whether there was some difference between the concepts of "honesty" and "intellectual honesty". Noah Millman offered what seems to me like a good summary of the distinction:
"Intellectually honest" means you make arguments you think are true, as opposed to making the arguments you are "supposed" to make and/or avoiding making arguments that you think are true that you aren't "supposed" to make.
Advocates, by contrast, make the best arguments they can think of for the position that they are obliged to take by their position. They are still supposed to be honest "“ they are not supposed to actually lie. But they are not expected to follow their own consciences with respect to the arguments they make or the positions they advance.
To put this another way, intellectually honest folks are out there trying to find the right answer to a question. The reason I find Casey Mulligan's contributions to the New York Times' Economix blog so offensive is that he seems to me to be intellectually dishonest. He's not trying to find the right answer. Rather, he's certain he knows the answer, and he cites any piece of corroborating data he can find, no matter how flimsy, to justify his view.
A new post this morning provides a perfect example. Mr Mulligan thinks that fiscal stimulus can't boost the private sector, and he wants to find some support for this belief. He therefore considers this year's temporary census hiring, which added over half a million workers to the federal payroll for a few months in the spring in summer. If stimulus worked, Mr Mulligan says, this hiring should have had a multiplier effect on private hiring.
Did it? To find out, Mr Mulligan charts total employment and employment ex-census. He says the census-driven spike in the former should produce an echo of a spike in the latter. He eyeballs it, and concludes:
In fact, the spike..., if any, is pretty subtle.
Science! Obviously, if one were actually interested in seeing what impact census hiring had on demand, one would probably make an effort to control for lots of other variables that might obscure monthly shifts (actually, if one were really interested in this question, one would probably stick with published research involving careful statistical analyses of larger samples of events, but no matter). Mr Mulligan nods in this direction:
If every 10 people hired for the census created an additional six jobs, then the only way to explain what actually happened is to contend...that May and June had extraordinary and negative employment events that were offsetting the private sector benefits of census hiring and thereby obscuring an otherwise obvious spike in the red line.
And then he...entirely fails to follow up on the point! So wait, were there any offsetting employment events? Well, it took me all of five minutes to learn that in May and June, government employment ex-census fell by nearly 50,000. Local governments alone cut 30,000 jobs during those two months.
Is this enough to offset the census multiplier? I don't know. I don't know that it needs to be; there were other shocks to the economy during that period, not the least of which was the battering financial markets took thanks to the flaring up of Europe's debt crisis. The main point is that this is a terrible way to estimate the impact of a given policy, and no serious person would attempt to defend an argument in this way.
And that's the problem. Maybe I've got Mr Mulligan all wrong, and he really is interested in learning how fiscal stimulus operates in an economic climate like the present. But that hardly reflects better on him. In that case, this tenured economics faculty member is methodologically incompetent rather than merely hackish.
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
LOL! Mulligan is a bur under RA's saddle.
"To put this another way, intellectually honest folks are out there trying to find the right answer to a question."
Are you sure you want to stick with this definition, because it applies to you as well. You can't be an evangelist for pure Keynesian econ and expect to be taken seriously.
"this tenured economics faculty member is methodologically incompetent rather than merely hackish."
Translation: He ain't a Keynesian like me so he is stupid and dishonest.
Well, the academic tradition of trying to find evidence of what you know to be true has thousands of years of history to justify its use. Look at the Donation of Constantine for the most famous example of this method.
The other method of trying to look at all evidence and derive what you know to be true from that instead is a new upstart with barely two centuries of supporting work to it.
Maybe tonight I'll dig up my philosophy book and re-read "Plato's Cave."
I suggest others do so as well.
Regards
hedge, here's a better idea: check out Edward Feser's "The Last Superstition." He provides an intro to Aristotle and Plato and a good summary of philosophy since.
fundamentalist, while your posts are often illuminating, I find you calling someone out as an evangelist for an economic theory ironic.
fundamentalist:
This certainly seems to apply to Krugman. R.A. is a more interesting case. He has at least admitted that the Keynesian stimulus approach seemed to be right, but plausibly was wrong (or words to that effect). That is, R.A. at least knows he could be wrong. Also, he's willing to publicly criticize Krugman when he (R.A.) thinks it's warranted.
And his last criticism of Mulligan is, it seems to me, warranted as well. If Mulligan can't figure out how to properly examine the data, is that not methodological incompetence? Which side of Keynes he's on is not relevant to that question, and R.A.'s criticism was a fair one.
You know, Ryan, I agree with you about Mulligan, but isn't this exactly what YOU do all the time when talking about the Chinese exchange rate issue?
So an astrologer accuses another of dishonesty.
This is even beneath my trolling threshold.
fundy,
Thanks for the suggestion, but I'll pass. I'll stick with my college textbook. Now if I were a skittish person, I'd think you just want me to buy the book to stimulate the economy. ---
Just wondering...
Didn't we have a census 10 years ago?
Can't Mr. Mulligan look at those numbers too?
Regards
Attacking the intellectual honesty of your opponents may be honest (in terms of your personal sentiment about their arguments) but it certainly isn't intellectually serious.
Paul Krugman's attack on Paul Ryan's Roadmap is useful. He quickly went from the data to the personal attack. Trading charges of intellectual dishonesty overshadowed the reasonable conclusion that the specifics of the proposal would likely undergo tweaking in the legislative process. This does not mean the proposal itself was disingenuous.
As much as there might exist circumstantial evidence of a lack of integrity, attacking motives or hidden intentions is a straw-man argument. Simply saying, "Don't be lazy, do the math (or I'll do it for you!)" seems much more productive.
Why one or the other ? Why not an incompetent hack ?
Allow me to interpret what just happened:
Ryan dislikes Casey Mulligan's ideology. Mulligan, like pretty much every blogger in the world, likes to post data that supports his worldview. Ryan picks out a blog post (Mulligans's not trying to get published in an academic journal with this) wherein Mulligan employs a crude methodology; and Ryan uses this as an excuse to spew such invective as "offensive," "intellectually dishonest," "no serious person," "methodologically incompetent," and "hackish."
Sigh. If only The Economist's blogs were subjected to the same sober editing as the print magazine.
Its not just RA. Dean Baker has made similar observations here: http://www.cepr.net/index.php/blogs/beat-the-press/casey-at-the-blog-joy...
OneAegis, I have no problem being called an evangelist for Austrian econ. I am. But RA, and most economists want to maintain an air of objectivity as if they have no theory to promote. They want people to believe that they just look at the facts and let the facts lead them. RA, Krugman and others are the slaves of Keynesian ideology. They just refuse to admit it. But if they really cared about the facts, they would see that Keynesian economics has been tested many times and failed. Austrian econ, on the other hand, has endured many tests of its theory.
Rewt66: "If Mulligan can't figure out how to properly examine the data, is that not methodological incompetence? "
Read Full Article »