At City Journal Gene Epstein provides a rather harsh critique of prize-winning economist Joseph Stiglitz (note: there is no Nobel Prize in economics; that’s a folk rendering; it’s a mistake I’ve made myself from time to time):
Stiglitz is a figure to be reckoned with, not just for his past impact on policy but for the influence that he might wield in future Democratic administrations. University Professor at Columbia, former chairman of President Bill Clinton’s Council of Economic Advisers, former chief economist at the World Bank, and author of more than 30 books, Stiglitz has been publicly associated with New York mayor Bill de Blasio and Senator Elizabeth Warren of Massachusetts, both of whom might bring him along if they win higher office. But how could an economist with his presumed sophistication publicly endorse the disastrous policies of Hugo Chávez?
Much of his criticism has to do with Dr. Stiglitz’s misdirected support for Hugo Chávez and the Chavistas in Venezuela but I want to concentrate on this:
As Stiglitz stressed in his Nobel lecture, “perfect competition is required if markets are to be efficient†(italics his). He was right that by challenging the idea of perfectly competitive markets, information economics signified a paradigm shift. But perfect-competition theory is so abstract that only economists blinded by their own mathematical proofs could possibly subscribe to it. Any institutions created and run by fallible human beings are bound to fail. Then again, these same human beings run the governments in which Stiglitz places such trust.
since it highlights my own views about economics and economists.
Economics is a science of human behavior. It’s not like physical sciences of physics or chemistry and it’s certainly not mathematics. Adam Smith, David Ricardo, and John Maynard Keynes weren’t mathematicians (although Keynes apparently had some talent for math). They were observers of human behavior. Observing human behavior dominated economics for its first 200 years.
In the 1970s economics took a turn, in my opinion a wrong turn. Econometrics came to dominate the field. Econometric is the branch of economics that uses mathematical notation, modeling, and statistical analysis.
The problem is that knowing a lot of economic theory, a little mathematics, and a little statistics doesn’t necessarily make you a good economist if you’re missing the observation and understanding of human behavior so necessary for the field. Your models can be completely internally consistent but if they don’t actually represent what happens in the real world, they’re useless.
The additional complication is that neither classical nor Keynesian economics deals well with very short timeframes or very small changes.
We can have pretty good confidence in some of the conclusions that classical economics has given us. People do respond to incentives. That doesn’t tell us how quickly they respond or how much they respond.
Additionally, there is absolutely no way to separate economics from politics.
So back to Mr. Epstein’s observations about Stiglitz. How much inefficiency does a tiny bit of imperfection introduce into markets? We have no idea. That’s something that can only be determined empirically and for some reason, like the Greek philosophers of old, economists seem to have an abhorrence of physical measurements. You can understand why. They’re terribly difficult to make.
Many economists outright reject empirical observations. I have had this argument with Boudreaux several times. If the empirical evidence disagrees with his “economic thinking” the numbers must be wrong, not his thinking. On occasion, he will actually be correct as sometimes data collection is flawed, and it is difficult, but even when it looks like the data collection is good, and you keep getting the same results, he would still cling to his theory rather than change.
Steve
Dave, your opinions on this match up pretty consistently with Karl Polanyi’s The Great Transformation.
A note: Perhaps the one and only econometrics course I took was rather skewed in its orientation, but it had little to do with macro or microeconomics, especially behavioral aspects. Rather, it was narrowly focused on modeling and analyzing the informational content and predictive power contained in securities price series, most especially public equities and futures prices. One might expect that at Chicago investment courses, but one would also expect application to Chicago economics courses. There was almost none.
“People do respond to incentives. That doesn’t tell us how quickly they respond or how much they respond.”
This is most assuredly true, and I don’t really understand someone of Stiglitz stature dying on the hill of perfection. Perhaps he is being mischaracterized. In any event, there is a lot of empirical discounting going around. I argue that tax incentives, money being fungible, act like any other stimulating incentive to returns on investment. Yet I’m routinely derided on that point. Only consumption apparently. Yet I do believe I’m the only one here who has made quite a substantial living making investment decisions routinely for 25 years now. That’s a reasonably observationally relevant empirical sample. I also point out (speaking of the irrelevance of econometrics) that when a change in the environment from “Let’s spread it around and You didn’t make that and We can’t do better than 2% and You want to poison the air” to pro-growth occurs one should not be surprised that a stalled 2% juggernaught (snicker) turns into a 3% or more economy. As I’ve pointed out, none of us knows what future Fed policy, a transitory trade adjustment, “overvalued” assets or fiscal irresponsibility may bring us, but empirically and for now, things under the current administration seem pretty good, especially in comparison to the past. The problems with GDP be damned.
O/T I don’t want to hijack this thread, but this piece by Jonah Goldberg, is worth the two minute read. Its salient points may be obvious, but he succinctly makes them and they are relevant points. They seem to have been largely lost in the public debate. I continue to maintain this is simply not an issue that can be adjudicated, and all the atmospherics indicate to me it deserves no other description than grotesque and irresponsible partisanship.
https://www.nationalreview.com/2018/09/brett-kavanaugh-allegations-partisan-politics/
There’s a relationship between my post this morning and the article to which you link. We don’t know how many accusations of rape are false. It may be 2% (the commonly cited figure) or it may be 20%. We simply don’t know but how we draw our conclusions really should vary depending on the actual facts.
“a stalled 2% juggernaught (snicker) turns into a 3% or more economy.”
Debt financed. The GOP could have done that while Obama was in office. Also, as Dave has been talking about GDP doesn’t matter a lot if we don’t look at the distribution.
On the rape issue there is a fair amount of research on the topic and we do know something about the people who make false accusations and the outcomes when they do.
https://qz.com/980766/the-truth-about-false-rape-accusations/
Steve
“a stalled 2% juggernaught (snicker) turns into a 3% or more economy.”
Debt financed. The GOP could have done that while Obama was in office. Also, as Dave has been talking about GDP doesn’t matter a lot if we don’t look at the distribution.
On the rape issue there is a fair amount of research on the topic and we do know something about the people who make false accusations and the outcomes when they do.
https://qz.com/980766/the-truth-about-false-rape-accusations/
Goldberg is just knocking down a straw man. Almost no one is saying he must be guilty because he is a man. People have said it is believable because men sometimes do such things when they are 17. Goldberg even acknowledges that is valid.
Steve
You’re conflating accusations with charges.
The resolution rate on rape charges is 36%. We cannot logically draw conclusions about the other 63% or, indeed, about the tens of thousands of accusations on which charges are never filed. They may be true, false, or indeterminate. We can make inferences based on premises or extrapolations of unknown veracity but we can’t draw conclusions.
That’s why I say we don’t know.
Note: when I say “we can’t draw a conclusion” that’s what I mean. I do not mean we can draw the opposite conclusion.
I agree, Dave. And that’s been my point on this all along. There are no facts. So we are left to speculate. An essential point of Goldberg is the falsity of the conclusion that because some men do, Kavanaugh did. Or that he must be the victim of a more general cause. That’s a mob reaction. Further, the circumstantials surrounding this just stink. Foggy or incomplete memory. Feinstein’s shameful handling. Katz is a cretin. The moving goal posts. On and on.
Ginsburg is not my cup of tea, but she was approved 97-3. But somehow we are supposed to believe that Republicans only nominate segregationists like Bork, or sexual predators like Thomas and Kavanaugh. It’s absurd on its face, but it has become the Democrat party’s go to move. Shameful.
Bang that drum, steve. Bang it baby.
It’s not that Republicans only nominate neo-confederates and sex abusers, it’s that Republicans tend to nominate fans of unrestrained power, among whom neo-confederate and sex abusers disproportionately fall.
All sitting Supreme Court justices and all prospective Supreme Court justices are statists. If they’re left-leaning statists, they believe in harnessing the power of the state to one set of ends and if they’re right-leaning statists to a different set of ends.
Last time I looked Congress was to legislate. The courts were to referee fidelity to the Constitution. That’s the real battle here. The left can’t win the legislative battle. So dirty up the court candidates to an activists liking.
Besides the debt being run up due to the tax cuts, corporate debt is also mounting.
http://conversableeconomist.blogspot.com/2018/09/corporate-debt-and-leveraged-loans.html
“The resolution rate on rape charges is 36%. We cannot logically draw conclusions about the other 63%”
Nope, we actually do know a lot from research into the topic. We know what kind of people make false accusations and in what kind of situations. You don’t need to resolve all cases to build up a picture of those who do that. You have such a large number of cases to work with that the numbers are meaningful. Also, I would dispute your claim that we cannot draw conclusions about the other 63%. In essence you are claiming that we need to throw out polling and statistical sampling. We can never know anything about anything unless we study and have results form 100% of the affected population.
Steve