Constantly reading criticism of Professor Paul Krugman must be getting old. Admittedly, it is difficult to avoid, given that he is one of the most well known Neo-Keynesian economists who actively posts on the blogosphere, and probably one of the most influential overall. Today, however, he brings up a good argument, although he probably analyses it with a far too partisan method. He refers to an article published in National Affairs, by Jim Manzi, called “Keeping America’s Edge”. Manzi relies on evidence which suggests that since 1970 “Europe” (including the former Eastern Bloc) suffered a loss in productivity equal to a change from 40% of global productivity to 25%. Since the article was published, there has been a great deal of criticism revolving around the statistics in question. Apparently, Jim Manzi’s statistics are not as accurate as he would like you to believe.
Paul Krugman uses the news as a way of critiquing conservatives who consistently grapple for evidence to prove theories that they accept as correct without question. He is undoubtedly correct on this point (not so much on his second conclusion, which is that the evidence suggests that Europe has not had the productivity decline that the “right-wing” suggests—it depends on the specific European country in question, and Europe should not be treated as a homogenous mass with a single macroeconomic policy). However, I fear that by attacking the right-wing, he overlooked a much more important lesson.
To this end, I responded by saying:
Dear Dr. Krugman,
Although I partly agree with your conclusion—that assertions “ingrained on the right” are not often enough questioned—
I do not think that that should be the real moral of the story.
The moral of the story is that historians, economists, and all professions and areas of studies which rely on empiricism, should do a better job at verifying statistics. More time should be allotted to making sure that the evidence presented is accurate and correct. This is especially true when most of these fields are positivist in nature—poor statistics collection can lead to poor conclusions.
I am not sure if a “praxeological”, or non-positivist, is inherently superior, and I am not sure that I am willing to make that argument (nor do I think that the argument is especially pertinent or relevant to my general point). I digress however, fearing that I am spinning a broken record.
— Jonathan Finegold Catalán
Although I look to avoid an argument over positivism versus non-positivism, or antipositivism, as it is neither my area of expertise nor a subject which is really relevant to my overall argument, I admit that it could be taken as a jab or a cheap shot. I do feel that there is no room for positivism in the scholarship of economics, but I recognize that the majority of professionals probably disagree with me.
But, I feel that my general conclusions are correct. That is, that all social scientists and scientists who rely on empiricism should place more effort in guaranteeing the accuracy of their statistics. Bad statistics can lead to bad conclusions, and while Jim Manzi’s inaccuracies were recognized early down the road, society is not always as fortunate. Sometimes, it could take several decades for empirical evidence to be reviewed and corrected.
Although this doesn’t seem to be a case of inaccurate statistics so much as it is a case of terrible misuse of accurate statistics. Your same point applies, of course. And I agree with you – the issues you raise here aren’t really relevant to the validity of positivism.
I find it particularly interesting that you discuss the field of economics as an empiric science, which is especially noteworthy (although not personally displeasing) given your dedication to the Austrian school. Would you say then that you agree that economics is an a posteriori science, or maintain the Austrian view (at least, as depicted by Hans-Hermann Hoppe in his Austrian Theory of the Trade Cycle) that it is a priori but should be examined with empiric means?
Charles Lehman,
I personally believe that science should be approached with a non-positivist (specifically, “a priori”) methodology. I just recognize, as someone who is consistently working with mainstream economists, that my belief is not widely recognized as true. So, I was willing to drop the point when commenting on Professor Krugman’s blog post, and focus on the more important fact that all scientists who rely on empiricism should work on their accuracy.
But, to answer your question, although I am personally inclined to agree with Ludwig von Mises, Friedrich Hayek and Murray Rothbard, I have been influenced by my mainstream education. So, perhaps more than the first two (and even more than Rothbard) I do rely on empiricism. On the other hand, I recognize the weaknesses of such an approach. In any case, when it comes to topics such as these I am still an amateur (and, I am not familiar with Hans-Hermann Hoppe’s methodology).
Thnaks for your detailed attention to this debate.
A few quick comments are:
1. In the linked blogpost, I think you will find that Professor Krugman’s criticisms of my data are in demonstrable error, in that he misidentified my data source, and then misread the data table that he did cite. I think you will find his reply to this pretty unsatisfying.
2. You will also see that I did not treat Europe “as a homogenenous mass” but was very careful to define specific dimensions on which European nations are consistently and systematically different than the U.S. in describing the ‘European model’, and provide the data to back this up.
3. You will also find, I believe, that I express a deep skepticism for any non-experimental methods for establishing non-obvious causal relationships through empiricial analysis of human society.
Link: http://corner.nationalreview.com/post/?q=NTZmNDIyNWQyNGI4YzdhNjRjMzg5MjFjOWFkNjhiZDk
Best regards,
Jim Manzi