Tom Brown Raises the Bloody Shirt: Are Economists Real Scientists?

      He raises it in a context of an exchange he had with John Cochrane. To me it's impressive that he actually was able to have any kind of discussion with Cochrane as he usually seems to ignore comments he doesn't agree with.

     Tom really is a born diplomat-which is why he can always get a civil answer out of Sumner, which is no mean feat.

   "Somewhat related to your post above, I thought you might enjoy this: my comment to a John Cochrane post, and his response.

    "Maybe I'm wrong, but I detect a note of defensiveness on his part (and on the part of commentator "Rich B"). Which surprised me a little. I only asked in the same spirit that the moderator asked much the same kind of question to both participants during this debate. You'll note that Nye answers the question, whereas Ham really doesn't. Like the moderator, I'm really just interested in the epistemology."

   http://lastmenandovermen.blogspot.com/2015/08/sumner-on-government-and-culture.html?showComment=1438650530629#c6278762734468760317

  Let's look at the Tom-Cochrane exchange:

  "A return to economic growth depends on reforming the regulatory state. But the deeper and perhaps more important preservation of our political freedom depends on it even more."
"What objective evidence would convince you that you're wrong about this?"

  http://johnhcochrane.blogspot.com/2015/08/rule-of-law-in-regulatory-state.html?showComment=1438476327458#c3814923297137066799

 Here is ole Johnny Cochrane's reply:

 "That's an interesting comment. Well, let's start simple. I would be delighted to read a similar essay documenting that our regulatory state is competent, efficient, and a-political; that people all the time stand up to its decisions, win in court, and suffer no backlash; that political parties and the regulatory bureaucracy are unable to use legal and regulatory strong arms to silence their opponents. I'd be delighted to read point by point alternative interpretations of the events I passed on. "

  "At this point we seem to be just in a game of he said-she said. Cochrane says a return to growth requires a reform of the regulatory state-by reform he means scale back-and where Tom hypothetically raises the possibility that he's wrong."

   To me the question is why should we believe that we have to scale back regulations to return to growth? It seems to me that Cochrane's economic model simply states that to increase regulations as we have in teh Obama years must be bad for growth. He doesn't provide any empirical evidence yet he demands that Tom responds with evidence to disprove his how claim that is also not supported by evidence."

   It seems to me that a lot of arguments are really theoretical. However, the second part of his response does sound a little defensive-whereas the first part merely doesn't really answer the question-it remains just a battle of conjectures.

  "I'm also interested by the reaction, because in my first evaluation I thought the essay was far too long, far too loaded up with specifics, and that anyone reading the news for the last few years would put together the threads. This reaction suggests that perhaps quite the opposite is true, and the case needs to be made in hit-one-over-the-head detail and greater length."

  It seems that Tom's request for documented evidence that growth is being hurt by too much regulation is nitpicky and a demand for abstruse detail. Cochrane is right that his post was already very long and abstruse so he kind of hides behind his own abstruseness here. Pretty clever. 

  UPDATE: Actually though Tom later states that what he's after isn't evidence, but a question of epistemology which I agree is the right question. I want to clear that up as this makes it sound like he was asking for evidence here when that's not his point.

  More generally, Tom thinks that Cochrane's (non)response is symptomatic of a lot of economists falling short of the scientific ideal of Feynman.

  "Mike, I agree we need both theory (i.e. models) and evidence. Although I'm not a scientist, that's my understanding of how science proceeds. But you don't have to be a formal scientist to use the method. I use the method in my work and my life in general. I expect that my car mechanic does as well when he's fixing my car. I consider this method to be reliable. It's a system to minimize the chance that we fool ourselves and others. This is where I normally insert some Richard Feynman quotes.Like this one. Or this one."

   "Just to be clear, I don't know if John is correct or not. It's just that it strikes me that he's making a bold claim about reality, and he seems very confident in it. In his comment reply to my comment he brings up government competence for example. Well, what if the government were made 1% more competent (however you measure that), how much extra growth does that generate? Is the ONLY way to measure competence of the government by looking at changes in growth? IF that's the case, then there's no way to test this hypothesis, and thus the concept of government competence doesn't add anything in terms of an explanation, and by Occam's razor it should be eliminated from the model. I really just don't know."

  Well, no, to simply look at growth rates won't settle it. If you take Cochrane wholly at his word this is the trap you get caught in.

  "My impression (pure speculation) is that economics as a whole hasn't been able or willing to eliminate (through falsification) many hypotheses. Perhaps that's a sign of the immaturity of the field or difficulties with the data. Again, I just don't know. If I ask a biologist or physicist "How would you know if you're wrong?" I'd expect a matter of fact set of examples of conceivable evidence that would demonstrate they're wrong. If I ask a cult leader how he knows if he's wrong, I'd expect indignation, verbal gymnastics, indirection by pointing me to more evidence supporting his claims, snarky comebacks or questions pointed at me, or bold proclamations that he just KNOWS he's right and we're not being "open minded" enough to see it: I'd expect anything but a matter of fact laundry list of conceivable evidences that anyone can check against reality and hold them too. Also I expect a scientist to be honest about what amounts to pure speculation or at least degrees of confidence (0% to 100% for example). And again, this is not the response I expect from the cultist. I'd expect the cultist to express unwavering confidence in their own claims (or the claims they are trying to convince you of anyway)."

   "It's my policy to treat economics as if it's a science, not a cult. However, given the failure of the field as a whole to come to many consensuses and to let hypotheses multiply w/o falsifying many, I like to poke a little and see how these guys know what they claim to know. For example, I'm not asking Cochrane for more evidence here. I'm not questioning that (although you bring up some good points). I'm looking for a scientific type response rather than a cultist response."

  To be sure, economists understand that correlation doesn't prove causation which is why simply looking at GDP before and after we had Dodd-Frank won't tell you that much. Even if we had 10% or -10% GDP since, it wouldn't cinch it as how would you know that the correlation wasn't a coincidence?

   I have to say that no one is better at verbal gymnastics than Sumner. Feel free to contradict me here Tom. Overall, I think Tom is on the right track: this is a question not of evidence but of epistemology. Speaking of which, one economist who always seems to me to adhere to the standards of scientific discussion-David Glasner-has had some great posts over the years.

  Here he has one about the original Scott Sumner. I'm flatering you here Scott-it's about Milton Friedman who had many of Scott's virtues and vices. There was a fascinating discussion of epistemology for economists in the comments section. Here is W. Peden:

 "On the philosophical question, it’s possible to be both an instrumentalist and a falisficationist (and a positivist!) consistently, though there may not be many examples. It’s not clear that Friedman was any of these, but he was probably closest to a rather limited sort of instrumentalism. (Limited in that most instrumentalists, like Bas Van Fraassen, think that assumptions DO matter insofar as their predictive content is important.) The details of Friedman 1953 are notoriously muddled but the general spirit of empiricism and usefulness is one that economists should take to heart, and Friedman’s engagement with institutional, historical and econometric* issues was an example of good methodology in practice. My favourite economics is empirical economics practiced by people with a solid theoretical background, who know how to do interesting tests."

  http://uneasymoney.com/2015/01/22/nick-rowe-goes-bonkers-over-milton-friedman/

  Bas Van Fraassen-and the owner of his book on Amazon-ought to give W. Peden a cut. His HT just led me to purchase two of Fraassen's books.

 http://www.amazon.com/Scientific-Representation-Bas-van-Fraassen-ebook/dp/B005H5O1V2/ref=mt_kindle?_encoding=UTF8&me=

 http://www.amazon.com/Laws-Symmetry-Bas-van-Fraassen-ebook/dp/B005NE53IM/ref=sr_1_1?s=digital-text&ie=UTF8&qid=1438686204&sr=1-1&keywords=van+fraassen+laws+and+symmetry

 Then for good measure, this book on theory and reality.

 http://www.amazon.com/Theory-Reality-Introduction-Philosophy-Foundations-ebook/dp/B003URRG0W/ref=sr_1_1?s=digital-text&ie=UTF8&qid=1438686259&sr=1-1&keywords=peter+godfrey-smith

    P.S. Again the similarities of Sumner and Friedman are striking. A large part of Glasner's post is that Friedman wasn't a great economist.

   "The point about Friedman is that he was both a public intellectual and a highly regarded academic economist. I am not disputing his importance as a public intellectual; I am saying that, although he was a great economist, his reputation as an economist is overrated. People like Don Patinkin and Harry Johnson thought so as well."

   (This is in the comments section not the body of the post.)

   The same thing can be said about Scott.