Economics, like all social sciences, is a soft science, though its devotees believe otherwise. It is also the most politicized of all the social sciences, including political science.

Does anybody know whether the basic form of the periodic table precedes subatomic theory?

I'm pretty sure the answer is yes -- Mendeleev or however his name is spelled reported the regularities in the characteristics of elements before there was a theory that explained it.

It's only a rule of thumb

See also this from Hilzoy.

There was some study done sometime by someone which demonstrated that economics majors behaved according to game theory predictions, when placed in these fuck-your-neighbor situations. All the other majors performed like regular people.

But that's my point -- that's data collection and analysis: he looked at the data he had, and organized it systematically, and when a real theory came along, it dovetailed beautifully with his useful, accurate, and reliable data collection and systemization. But without that theory, he wasn't going to be able to use his systemization to predict something outside the scope of the reliable data he already had (interpolate, yes; extrapolate, no).

4: Linked in the post already.

D'oh!

Friend: why didn't you pick up that $20 bill?
Economist: if it was truly a $20, someone would have already picked it up.

Usage question: Does 'Naturalism' in the title of the post successfully refer to 'What 19th Century naturalists did, collecting and cataloging ferns and the like'? Because I'm looking at it and starting to worry that it looks as though I'm referring to some philosophical position or other. And if it doesn't work, what word do I want there?

I didn't get it. Maybe put, "Or 19th Century Naturalism?"

Woohoo, an unfogged post that touches my area of expertise.
2- yes, it was a phenomenological grouping, as LB says.
As for chemistry vs. physics, I'd still argue that what most people consider chemistry- that is, catalytic inorganic, synthetic organic and biological, at least in terms of industrially relevant disciplines- is still highly indepenent of the physics. You learn that if you mix this with this under these conditions, you can make that, and if you want to make that, how do you figure out what to start with?
Now, there are attempts to rationalize what happens in terms of physics, and to predict new reactions, but it's still a lot of guesswork. In our advanced organic class, you'd often be asked to rationalize a product in terms of the physics and such- but it turns out that you can rationalize other products just as reasonably. In other words, you can make the outcome be whatever you want and you can justify it with the mathematics- so yes, it's exactly like economics.

6 presents a slightly over-simplified model of the scientific method, imho.

It's fairly common for there to be theories that codify certain generally observed phenomena and from which predictions are regularly derived but where there's no 'base-level' explanation for the success of that theory.*

If economic models were wildly successful then we wouldn't care that much that they rested on some unexplained and more basic set of psychological laws governing human behaviour. However, as we all know, economic models aren't really very successful at all.**

* If you're some form of instrumentalist [loosely speaking] about scientific theories you might think that there aren't any base-level explanations of this type. Just ways of codifying observations and generating predictions for future observations.

** Also, I can't listen to economists speak without hearing all kinds of shifty attempts to avoid this fact.

"Natural history"?

"Natural Philosophy". I'm not sure whether "Naturalism" means anything in any particular dialect of humanities jargon, but N. Philosophy is what the Enlightenment dudes in question themselves used, and I'm almost positive it hasn't since been co-opted by any other discipline.

15 was me.

Ooooooh. Economists and the ridiculous "policy science" people I studied under for grad school wanted to be scientists more than they wanted to breathe or eat or fuck. They thought that if they had enough numbers and maybe some error bars, other people would believe they were scientists too. They despised the rest of the humanities for not being technical and they despised scientists for not being imaginative and they were desperately longing for the day when enough of their theories would come together that they could actually predict something just once, just this time, so they could get some approval from anyone who's theory predicted something accurately twice in a row. They neeeeeddddd to be scientists.

There is the new subfield of behavioral economics that's trying to address this very problem. They haven't gotten very far, yet, though.

I'd say that economics is even worse than you claim. To the extent that it has a basis in fundamental models, those models are poppycock (rational behavior and such).

Speaking of which, Milton Friedman will not be making any more theories.

17: Yeah. Too bad for them about that whole "falsifiability" thing. (Disclaimer: I am not actually a crazy Karl Popper devotee.)

Oh, and it is just barely possible that I shouldn't have said something very close to that on the first day of our first policy science seminar to the department chair guy who loved his theories and already hated women, lawyers and engineers. It was a grim three years after that.

Megan: Do you have a direct link to the post LB referred to? I haven't found it.

No, he won't.

I was completely obsessed with Unsuggest for about 24 hours. Results at my URL.

If they had more members (say a million) I really think that someone could design a bot and come up with a map or graph of The American Mind.

As it is the average library there seems to be 70 books, which apprently means that a lot of people catalogued about 10 books and some several hundred. (Though I an not sure if the 100,000 members catalogued 7 million books in all, or 7 million different titles.)

Now, there are attempts to rationalize what happens in terms of physics, and to predict new reactions, but it's still a lot of guesswork. In our advanced organic class, you'd often be asked to rationalize a product in terms of the physics and such- but it turns out that you can rationalize other products just as reasonably. In other words, you can make the outcome be whatever you want and you can justify it with the mathematics- so yes, it's exactly like economics.

(And I suppose this also goes to 13). I don't mean to argue that economics isn't respectable (and maybe I should change the title because I don't really mean to argue that it's not science -- I mean something more like that it's not theory-driven science). What I mean is that, there's a sense in which chemistry is resting on theoretical rock -- well-understood physics. There may be (and presumably you know, so I should say 'are') areas where we can't easily make the connection between the theory and the phenomena we're trying to explain, so we collect data, and and work to explain the data we have in terms of the theory that we can rely on. But we certainly never ignore data in favor of theoretical predictions that something different should happen.

What seems to happen in economics is that a weakly supported underlying theory is privileged above the data that's actually being collected. This would be a problem even if the theory were solid, because it would mean that we were ignoring whatever other facts were necessary to explain the discrepancy. But it's a huge problem where the theory is itself handwavy rules of thumb.

If economics viewed itself more like 19C chemistry, allowing theory to develop out of the observed regularities in the data (like the periodic table), I'd have more respect for it. And I'm sure that many, many economists do view their field in this light. But the very common view that it's an area in which important truths can be deduced from first principles strikes me as completely lacking in validity.

(What I really want is for one of you philosophy of science types to pick this up, and get the argument I'm trying to make right.)

6: Actually, I think Mendeleev (also not sure of the spelling) did make some useful predictions with his periodic table--he pointed to some inconveniently-empty spots in his chart, and said "there ought to be more elements here and they ought to be kind of like this." And he was right, because he'd grabbed onto a meaningful pattern. He didn't know what the source of the pattern was, but he didn't have to.

It doesn't really matter where your mathematical model comes from, so long as you can test it. The economists use some fairly crude and silly intuitions to guide their model building, but then so have physicists, sometimes (electrons: not actually little planets orbiting a little sun). The trouble with economics is that they don't have a good way to test most things, and then they proceed to treat the model as true anyhow.

re: 15

Naturalism is often used to mean something like: http://en.wikipedia.org/wiki/Naturalism_%28philosophy%29

Yeah, I edited the title to invoke 'Natural Philosophers' rather than 'Naturalism'.

"if economics viewed itself more like 19C chemistry, allowing theory to develop out of the observed regularities in the data (like the periodic table), I'd have more respect for it."

The idea that science develops out of observed regularities in data is a fairly old-fashioned [17th century] view of the scientific method. That sort of Baconian view isn't necessarily how science actually works.

However, however you view what science does -- Baconian induction, Popperian falsification, whatever -- much of modern economics seems to fail to pass muster.

And, for what it's worth, sociologists of science would say that scientists ignore irregularities in the data all the time if they conflict with the mathematical model. That's not even necessarily a bad thing -- given the imperfections in our data collecting methods there may well be times when it's prudent to trust the model rather than the data.

Agreed- part of that paragraph wasn't clear- given a hypothetical reaction, you could frequently say that pathway a will lead to product A, and pathway b to B, and a and b are both reasonable so you could get either A or B. However, once you run the actual reaction, you measure whether you got A or B and then go back and say whether a or b was the correct arguement. No one (well, almost no one) will argue that b is the correct argument if you can observe A. (Of course, usually you get something like a 9:1 mix of A:B, so they're both valid to some extent.)
I take it that running experiments and making observations in economics is not quite as simple, so perhaps that's part of the problem?

The idea that science develops out of observed regularities in data is a fairly old-fashioned [17th century] view of the scientific method.

I pride myself on never having moved out of the 17th Century.

Seriously, I put this up in the hopes that someone better educated than I would run with it. I think the chemistry:physics as economics:psychology captures something important about what makes claims that important economic truths can be deduced from first principles, but my exposure to the philosophy of science necessary to run with it is limited. (I've heard of Popper, but that's about as far as I go.)

The problem with the chemistry:physics to economics:psychology observation is that psychology itself isn't a science.

26: nails it. The economists could be getting their models from Ambien fantasies for all that matters, just as it doesn't matter that Kekule dreamed up the benzene ring after an overdose of Wiener schnitzel.

If it can't be tested and accurate real-world predictions can't be made a math model is meaningless.

To 25: That would be better LB. The conceptual problem is that the whole purpose of economics is to create a model, which almost by definition is simpler then the reality it allegedly describes. So some papering over of complexity and individuality is necessary. In my somewhat studied opinion (undergrad + something of an L&E focus in LS) where economists tend to fall down is failing to look critically at how appropriate is adherence to there underlying assumptions.

I'm not sure who the worst culprits are - the free marketeers somehow assume away the inefficiencies of market power. Not to mention that the entire theory of efficiency now, equity later is, in practice, complete bunk because in any given time period, equity is always later. On the other hand, pro-regulatory people tend to overestimate the effects of incentives on behavior because of the dual assumptions of rationality and homogeneity of the target population.

To complete LB's analogy, the intractable problem is always going to be measurement of aggregate preferences whether you assume rationality or not - economics is unlikely to ever have the strong theoretical base provided to chemistry by our understanding of atomic physics.

Something bad happened to that sentence. There should be a "what's wrong with claims that important economic truths..."

32: Not so much the problem with the analogy, as the point of the analogy.

I think you're giving econ and chemistry short shrift. First, econ, which I'm not predisposed to defend: the unpredictability of individuals' behavior doesn't invalidate the idea of economic models any more than quantum effects prevent me from using a recipe to bake a cake. The number of individual actors is so vast that you only need a very crude model of their behavior before statistical effects overwhelm them and everything Just Works Out. That's not to say that very bad economic predictions don't get made — the systems are just way too complex relative to the amount of data collected — but I don't think you need to be able to count the turtles all the way down in order to make predictions. A parallel might be Newtonian physics: it's not exactly right, but it's good enough for nearly all situations you're likely to encounter.

Similarly, I think you're underestimating the explanatory power of chemistry before the advent of subatomic physics. Even without an understanding of the structure of the atom, chemistry could be (and was) developed into an internally coherent system that made predictions about the world. Now that we have a more detailed physical model we've rephrased those chemical systems in its terms. But of course there's a limit to that, too, whether it's quarks or strings or whatever, until we build a bigger collider.

Whatever point you're currently at, you're working with a set of axioms from which complex behavior arises. The lower the level you can get to the better, but not having reached the bottom turtle doesn't mean that your current level can't be a very, very good model.

There is the new subfield of behavioral economics that's trying to address this very problem. They haven't gotten very far, yet, though.

Yes, this is very cool stuff. They've done a lot of work in economics and in finance, but the whole field is only about 20 years old. The first decade and a bit were really only dedicated to compiling evidence against the orthodox models, which is really some fun research to read if you like to see theory emperically picked apart.

By now, there are some good economic models of how people respond to incentives based upon psychological studies, but the biggest problem seems to be that the math is messy. It's way harder to integrate Prospect theory (the sensible new ideas of responses to incentives) into a big elegant framework than the easy-peesy utility and universal indifference curves. There's some very cool predictive modelling using these new theories, but it's almost entirely simulation-based, which just doesn't get the respect of a cool new mathematical expression.

Sorry, I just love behavioral econ and finance.

A big chunk of the economics profession effectively fucked post-soviet russia by not worrying about the distribution of the soviet union's assets. It didn't matter to their models whether the assets ended up in the hands of criminals.

Economics is like meteorology - we wave around a lot of numbers, talk about lines (consumption functions or cold fronts, your choice) shifting, and how that will affect you, but really, we're guessing. Too many variables to nail down.

The number of individual actors is so vast that you only need a very crude model of their behavior before statistical effects overwhelm them and everything Just Works Out.

Bullshit. The model may be crude, but it has to be accurately crude, and my contention is that economists don't even have a model of human psychology that's crude in a reliably accurate way. The only way statistics makes everything Just Work Out is if we have the capacity to predict what most people are likely to do, and we often don't have even that.

Philosophers used to talk about the theoretical laws of one science being *reducible* to the laws of some other more basic science.

So, in principle, the laws of chemistry ought to reduce through some bridging rules [that explain how you translate the terms at one level of scientific description into those at another] into the laws of physics. It sounds like you are arguing that economics ought to reduce to psychology in the way that chemistry reduces to physics.

However, it's not really fashionable to believe in reductionism as a program any more. I don't think that many philosophers of science/metaphysicians believe that the laws of chemistry will ever reduce to laws of physics. Instead, people talk about supervenience.

That is, the laws of chemistry are in some sense dependent on the laws of physics even if no translation from one to the other is ever possible.

Economics then would supervene in some way on psychological phenomena.

The point being, that economists inability to correlate their 'laws' with psychologists laws isn't necessarily a failing of economics. If you don't believe in reduction, then that's a failing all sciences share.

However, it IS a failing if the laws and principles described by economists don't fit with the data. And here I think you have a point -- their failings (in terms of correlation with observation and the makings of predictions) can't be excused away by appeal to more basic 'laws' since there *are* no more basic laws to be had here. (Yet)

You can find a lot on this if you go to CT and search for "physics envy."

41: LB that's a little harsh on economists - they do a good job of saying what will likely happen if everything else remains equal, but a wretched job of evaluating whether everything else does in fact remain equal. And unfortunately, "Pop Econ" just neglects to mention the conditional entirely to avoid having to deal with the difficulty in demonstrating the condition precedent.

By the way, LB, this critique of economics was made by a nobel prize winning economist decades ago. Economists, as all of you are pointing out, have models that work well within sample (interpolation), but badly out of sample (by the way, it was economists who pointed out that fact to each other, not engineers who point it out to us). The Lucas critique says that if your theory isn't based on solid foundations (maybe psychology, though I'm not convinced psychology is the only possible option), then your model can't be expected to predict things well out of sample.

So what did economists do? we (I'm using "we" in the sense that I'm part of a tradition, maybe an apostolic sort of "we") went and tried to find microfoundations for everything. we have mathematical models that actually do explain the world fairly well in some cases (regardless of how proud some people are about being able to point out our failures). Inflation is one example -- we can explain inflation reasonably accurately. How did we get better with inflation? We looked at how individual firms and pricesetters should be expected to act given the world around them. The models aren't perfect, but they can explain surprisingly detailed facts about the distribution of price changes and the reaction of inflation to various shocks.

sure, we used rationality in those models, but it turns out that when you look at the data, firms and markets tend to work in rational ways (I don't want to get in a digression about asset price bubbles, it won't be productive).

finally, there is a difference between out of sample prediction accuracy and predicting the future. When was the last time a geologist told you the precise date an earthquake would occur? In fact, physicists will be the first to tell you that no matter how much we learn, it will be literally impossible to predict the future. In the same way, economists can't predict the future. I'm willing to say, if A happens, then B will happen, because I have models predicting this correlation, but that doesn't mean that I can predict that A will happen.

"Matural Philosophy" is quite good, or just philosophy, as in Hegel & Nietzsche. Economics is so close to this, or so obviously some blend of sociology & psychology & philosophy. Looked up Pareto, who is supposed to a decent mathemetician, and looked at the other that was obviously sociology. Of some kind. Veblen, Galbraith, hell, Hayek & Friedman. As Emerson said at CT, it really *feels* like the privileging analytic philosophy wants, and I suspect it is not disconnected.

Look, this shouldn't be hard. Marx and his legions of followers developed a pretty decent philosophy and set of tools, which capital has had a terrible time refuting as moral and natural philosophy, so they are trying to blind us with "science."

It is actually despicable propaganda. Just last night somehow was led to Piero Sraffa, who apparently reinvigorated the "labor theory of value." Uhh, why wasn't I told? I was told LBV has been utter nonsense for a century.

This is great, I'm recapitulating the history of the philosophy of science by getting stuff wrong.

But seriously, this:

It sounds like you are arguing that economics ought to reduce to psychology in the way that chemistry reduces to physics.

Is precisely what I'm arguing. That if you examine a mixture of reagents in a beaker, if you could observe each interaction at the subatomic level, and aggregate the descriptions of each interaction appropriately, that you would end up with a correct description of the chemical reaction arrived at through your knowledge of the laws of physics. Is there an easy way to walk me through what's wrong with this?

But in any case, this:

However, it IS a failing if the laws and principles described by economists don't fit with the data. And here I think you have a point -- their failings (in terms of correlation with observation and the makings of predictions) can't be excused away by appeal to more basic 'laws' since there *are* no more basic laws to be had here. (Yet)

is all that I really want to be able to stand by.

Ian D-B

... Stephen's little brother, the black sheep of the family?

45: Oh, I'm being rude to economists generally, when I should be rude to people who make poorly supported arguments based on economics specifically, and that's wrong of me. I did kind of say that in 25:

If economics viewed itself more like 19C chemistry, allowing theory to develop out of the observed regularities in the data (like the periodic table), I'd have more respect for it. And I'm sure that many, many economists do view their field in this light.

Should I edit that into the post as an update?

When everybody here is saying that economic models don't perform well, and this apparently has been accepted as fact, what exactly do you mean? Do you mean I can't predict the future with an economic model? If that's the condition for being a science, then physics and chemistry fail. Do you, on the other hand, mean simply, if A happens, can economics predict B? If that's all you want to know, then I think you're underestimating economics. Again, a few high profile failures, exchange rates for example, don't invalidate the whole field.

Milton Friedman himself propounded the velocity theory of money, which basically says that over the long term, inflation and money growth are correlated, after controlling for changes in payment technology. And in periods with little change in how payments are made, we find precisely that correlation. (also, I don't want to get into an argument about the velocity theory of money, and nobody else here wants to listen to it.)

49 cont.: Another way to say this is that I have no particular quarrel with economics or economists generally, particularly people working as you describe. My quarrel is with the political argument that "X is obviously going to happen. It's Econ 101."

49: yes, you should. there is an enormous amount of shitty economics in the public sphere. I get sick every time I read the WSJ opinion page. Economists in the ivory tower I think actually manage to be more realistic than the ones you hear out in public.

My quarrel is with the political argument that "X is obviously going to happen. It's Econ 101."

If your Econ 101 prof didn't tell you that everything you learned in Econ 101 was, on some level, bullshit, you had a crappy Econ 101 prof.

41: by "crude" I don't mean "inaccurate". I meant that they can be simple. It's enough to say "price of hamburgers goes down, people buy more". You don't need to assign variables to body image, cholesterol count, affection for pickles, etc.

42: It's outre to apply reductionism to the hard sciences? Really? Huh. I have to admit, my encounters with supervenience are solely from David Chalmers' use of it to write up what walked and quacked like Dualism, but, the reader was assured, was not (I was probably just too stupid to understand his claims).

Anyway, I remain surprised that translating one hard science to another isn't considered a viable approach. The laws of thermodynamics are certainly applied across physics and chemistry, and nuclear chemistry obviously has to bridge the two. And I think you'd be hard-pressed to say that biology can't be reduced to chemistry.

Seriously, they're really making that claim? Huh.

Econ 101 is wrong about just about everything. What they teach about asset pricing was disproved 15 years ago. They teach models of exchange rates that predict precisely the opposite of what happens. Econ 101 is useful despite its flaws because you elarn a specific type of analysis. It's not even econ 201 or 301 where you really learn models that work terribly well, it's econ 401 (and don't tell me physics is any better, they do the same thing when they teach outdated models of the atom and elide most of quantum theory because it's too complex. and what's more, physics changed 80 years ago, as opposed to the 20 or 30 years during which we've had modern economics).

If economics viewed itself more like 19C chemistry, allowing theory to develop out of the observed regularities in the data (like the periodic table), I'd have more respect for it.

Even though I'm on the empirical side (I'm a third year econ grad student), what I see the theory guys doing is trying to come up with models that explain the data. Really. It's difficult to appreciate the explanatory powers of these economic models without being familiar and up to date with the literature in the field.

47- If you want to get all deterministic, if you could observe and understand the chemical reactions of every person's brain you could predict how they would act and then your economic theories would be based on physics.

When everybody here is saying that economic models don't perform well, and this apparently has been accepted as fact, what exactly do you mean?

The example I referred to in the post is minimum wage laws. It's 'Econ 101' that raising the minimum wage will increase unemployment. If you look at the actual data that's been collected, this doesn't appear to be, you know, true, or anything like that. And yet people continue to argue on the basis of what the laws of economics demonstrate will happen, ignoring what the data shows (or denying that it shows what it does).

Do you mean I can't predict the future with an economic model? If that's the condition for being a science, then physics and chemistry fail.

Really? In the sense I mean, they do pretty well at predicting the future. At time T, I fire a cannon with muzzle velocity v and elevated at angle θ above the horizon. Physics allows me to predict with a fair amount of accuracy where and when the cannonball will land. (Even if I rely on the outmoded assumptions of Newtonian physics.) It is not clear to me that economics has demonstrated comparable ability to make predictions.

(cont'd): but maybe I'm completely wrong about my first assertion in 54 -- maybe X-sub-pickles is necessary. But you've got to develop a model of economic behavior at some point (and presumably before you start talking about synapses).

I've been trying to figure out economics from an unsympathetic but rational outsider perspective for decades. Recently I've been making this a project, and I have stuff up at my URL.

Hodgson's "How Economics Forgot History" is my most recent reading. His claim is that economics is a historical science rather than a theoretical science, and can never be predictive the way physics ideally is. History is contingent, and as a result you wil always have particular path-dependent economic systems which have to be studied according to their particular organization. Economics attempt at universal theories leads to empty theories which are inadequate to any actuality -- the way a general theory of vertebrates would lack much of the knowledge needed to understand birds or fish or mammals.

I am obviously over my head here. But this whole "Econ 101" blog discussion has been buffing me for a week. This is, to me, not merely about eliding some externalities in a freshman course.

It feels like there are ***profound*** epistemological questions involved here:Foucauldian level profound.

61:buffing s/b bugging

58: If you increase the money supply, prices on average will rise proportionately (if you give me a country and a date, I can also calculate that proportion using only past data).

also, 56 is right. The very newest stuff really is trying to fix the failures. I just read an interesting little paper explaining why exachange rate models actually work so poorly. It's something along the lines of Einstein explaining why Brownian motion is actually random and why we can never hope to predict it (and physicists know they can't predict it, but do we call them failures? No, we give einstein a nobel prize for the work).

I just read an interesting little paper explaining why exachange rate models actually work so poorly. It's something along the lines of Einstein explaining why Brownian motion is actually random and why we can never hope to predict it (and physicists know they can't predict it, but do we call them failures? No, we give einstein a nobel prize for the work).

But we'd start making fun of them if they continued producing predictive models in that area.

I've been staying out of this because it's nowhere close to my competence, but I think there's an emerging agreement that
(a) the questions about theory-reduction, supervenience, etc. are something of a red herring, since the real issue concerns the predictive success of economic models; and
(b) holding any discipline hostage to its intro course is a terrible idea.

Fair?

What really confuses me is the hostility to economics. People seem to get really upset about us. I mean, I know some of us are sanctimonious, but shit, look at how some scientists talk to the religious. We'll say non-economists are stupider than college freshmen, but at least we don't call everybody else delusional.

And really, we're not all bad people. We try to explain things about the world. As far as I'm aware, nobody else has any models that work any better. Are we physicists, no, but what's wrong with trying to be so? And why should people like megan get so hostile and superior?

64: Economists have been out of the business of predicting exchange rates for a long time. It was at least a decade ago that it was shown that most models of exchange rates performed no better than random noise, and since then, models have been forced to meet that standard or else be dumped. We don't try to predict Brownian motion.

And when you make claims like that, claims that we try to predict things that are random, what research are you referring to? The fact that somebody somewhere in a newspaper made some prediction does not mean that economics as a profession systematically tries to make those predictions.

why should people like megan get so hostile and superior?

That's not anything about economics, Ian; ask her about her parties.

55: Actually, it is quite different in physics. The reason is that the models have been superseded, but they were pretty good models. A less problematic example that quantum is that simple newtonion mechanics, while wrong in some details, is good enough to base mechanical and civil engineering on. As I understand it Econ 101 teaches models that never were that good, and it often teaches them as if it were true.

Yeah, it was the week when she both said east coast parties suck and economics can suck it that I stopped reading her blog.

66: Really confuses you? Try mathematizing the patriarchy. Try proving by objective methods that male dominance really does generate the most efficient societies, with the greatest satisfaction to the greatest number. Use algebra.

The Delphi worker, laid off after thirty years, losing his heath care and pension, yells:"I am not a commodity, a factor of production, some unit in partial equilibrium and comparative advantage." You don't get the hostility?

Fuck you.

65b- The problem is that people argue we should make policy based on Econ 101. No one is arguing that we should run the space program based on Physics 101.

63: But I'm being snippy, and I shouldn't be. I added an update to the post, trying to clarify that I'm really not making the claim that all economics is bunk, or that academic economists aren't likely to be doing useful, interesting stuff. What I'm arguing against is a very common political argument about the status of economics -- that it's a 'real science' like physics, and that simple application of Economics 101 is going to give you some useful policy answers, in the same way that simple application of Physics 101 will give you some useful information about how to point a cannon. And that appealing to real world data, where it contradicts the conclusions of Economics 101, is going to give you a wrong answer.

This argument is all over the place, and it is nonsense. And I think the reason that it's nonsense is that respectable economics is much more like natural philosophy -- collect data in a limited realm, make deductions based tightly on the data you collect -- than like their image of physics -- deduce the universe from a few simple rules.

69: Supply and demand models work pretty well most of the time. Look at the market for oil -- it tracks supply and demand pretty well. Is it a perfect model? No. But it's a great start. Econ 101 is in general not based on models that were never good.

Are we physicists, no, but what's wrong with trying to be so?

I think because, in public discourse, some economists--and more often, some people with undergraduate degrees in economics--like to pretend that they are physicists. And then they say "differential equations" and use math jargon as cudgel rather than as a well-worked out justification.

Also, people assume that dsquared is an economist, and he uses the c-word a lot.

72: part of the problem here is that I'm not sure if the target is actual economists or people who make half-assed pronouncements based on one course they took in college. The problem with the latter group is not that course.

Not at all, FL. teaching undergrads is one of the main things economics does, and if they're sending them out into the world seriously misinformed, the profession is doing harm.

Second, it's being claimed that the problems with Econ 101 go beyond the minimum simplifications necessary for introductory presentation, and have a right-wing free-market bias.

At this point it seems to me that economics itself is outgrowing its theory, but that the old theory is still dominant, and is especially dominant in Econ 101. Mainstream economists seem entirely untroubled by this, but a lot of the best new stuff seems to be rying to escape from the othodoxy. My guess is that in the next 20 years econ will either fossilize into irrelevancy or explode, and that in the latter case Econ 101 will be entirely transformed.

mechanical and civil engineering

Huh? I thought civil engineering was what my dad, a city planner, did; and that it did not have much to do with Newtonian mechanics.

74: no, you misread me. I meant they were never up to the standard of the classical physics models. I'm not claiming the economic models are garbage, by any means. What I'm saying is that the in physics the theory really has been proven out pretty well. You've never had anything like the same degree that in econ.

I'm not claiming that the average economist is particulary confused about this (although some seem to be), but it certainly seems many outside the field, and in particular econ undergrads, have a really unrealistic idea of exactly how far from those sort of standards of empirical proof econ is.

Hey Bob, fuck you too. I didn't lay off the delphi worker. I didn't say the delphi worker has no feelings, or that the guy or girl isn't important. I don't sit here and say fuck physicists just because they say people are just lumps of carbon, oxygen, hydrogen and some other crap. Physicists tell me I have no soul and that free will is clearly impossible. I don't say fuck you to them.

On top of that, Bob, a lot of economists do work to try to help the Delphi worker. They try to figure out ways of making a welfare state work. they try to figure out ways to make income less unequal, and make unemployment less common and volatile. Shit Bob, go read my work -- google "ian income distribution". go read my work and tell me I'm telling the Delphi worker he can suck it. I'm sorry for the Delphi worker, but fuck you, Bob.

Does anyone claim economics is like physics? I have never met any economist who made this claim. Ec 101 thinking is prevalent because, a) it has a theoretical underpinning that strikes people as in the most part sound, and b) it often has great explanatory power. Usually, if you institute a price floor or price ceiling, you distort the market and make it less efficient. Maybe this is sometimes false, but it's usually a good null hypothesis. I imagine that we all have rules of thumb like this for setting the null hypothesis. For example, many people will conclude from a gender or race skew in hiring that the hiring process is discriminatory. Is this always true? No. But there are settings in which it is the appropriate null hypothesis.

My hostility to economists is due to their frequent right-wing bias and their customary arrogance.

What I've said here about physics derives from the frequent claims (e.g. by Lazear) that economics is "a true science" unlike the other social sciences.

65(b): And Physics 101 is actually pretty useful in the areas it covers. Artillery is still aimed with the aid of Newtonian mechanics.

66: Ian -- My 73 was written before I saw your 66. The hostility to economics comes from people beating us to death with Economics 101 arguments, which get made even by people with enough economics training that they should know better. It means that my (and I think, others') reaction to someone bringing up econ in a political argument is (a) he's going to try and browbeat me with having knowledge of how things work that I don't, and (b) his argument will be bullshit.

This really isn't meant to be a criticism of the field of academic economics. It's meant to be a criticism of the school of argument that expects 'Basic principles of economics demonstrate that I'm right. The demand curve slopes down, dimwit' to be treated with the same respect that 'Basic principles of physics prove that I'm right. Energy is always conserved, dimwit" is.

But I don't mean to attack economics generally.

78: It's not nearly as pervasive as in mechanical engg., but it's there. `based on' is unfair though, so I'll retract it. If you are designing a curve in a road, for example, it's newtonian mechanics that tells you how to set the camber and other parameter so people drive smoothly around it. But yeah, there is a lot more to it than that!

I presume that dsquared is an economist

I thought he himself had denied this, making a distinction between "economics" and "finance" and declaring that his field was the latter.

though I'm clearly on the opposite side on things as most of the people here, I think we do agree fairly generally. a lot of econ 101 is wrong, and a lot of modern economics is trying to understand that. However, the majority of the people you hear in public discourse haven't made it past the first few classes. I don't know about the future of economics. It's got some serious trouble right now. Micro is flourishing, but macro is in a deep funk.

The hostility aside, I'll say this, I was reluctant to comment on this post because I generally don't like blog comment threads, and I prefer not to argue with the rabble outside the ivory tower, but I've actually found it very interesting and useful. So, thanks.

Physicists tell me I have no soul and that free will is clearly impossible. I don't say fuck you to them.

You should consider doing so. They're often fairly dweeby, and unlikely to be much of physical threat.

Reading 86, a cock joke is obviously overdue in this thread.

I generally don't like blog comment threads

You don't? But you've commented here before, or if not that, someplace else that I read -- your name's familiar.

Here is a statement by Lazear. I think that many points in this statement are false or problematic, and the statement as a whole is arrogant.

Economics is not only a science, it is a genuine science. Like the physical sciences, economics uses a methodology that produces refutable implications and tests these implications using solid statistical techniques. In particular, economics stresses three factors that distinguish it from other social sciences.
Economists use the construct of rational individuals who engage in maximizing behavior. Economic models adhere strictly to the importance of equilibrium as part of any theory. Finally, a focus on efficiency leads economists to ask questions that other social science ignore. These ingredients have allowed economics to invade intellectual territory that previously was deemed to be outside the discipline's realm.

Lazear

I do recognize that there are individual economists different than the rest (so-called heterodox economists in about six schools) but I'm reacting to what I perceive to be the dominant, modal, or mean economists as they function in American life.

You know, I think the economist-scorn, at least from scientists, is part of the "I AM KING NERD!" problem. My discipline is a "realer", more rigorous science (and uses harder math), therefore it is more challenging than your discipline, therefore I am both smarter and doing more useful work than you are, neener neener neener.

You see the same thing in sciences > humanities and math > computer science > computer engineering/IT.

baa: to continue your example though, the perfect competition models presented in most econ 101 are broken as designed, at least if I am recalling a talk i heard a few years ago correctly. I mean that they get the math wrong, not that they don't model reality very well (which is also true). It's not clear to me that the price floor stuff is actually of any real world use but it is a standard `truth' that is bandied about by freshment econ students. It's this sort of thing that, when it ends up being echoed in political speeches and policy documents, that is the root of `why people get pissed at economists' , I think.

Ian, I'm with LB though. I'm not trying to attach economists of the area in general. I do believe that there is some credence to the idea that particularly intro econ has some flaws (which of course the economists recognise) that can and have been misrepreseted to prop up poor policy decisions. It's the latter part that gets people in a tizzy, not the former. I believe the CT discussion is mostly about people feeling that there is a `not our problem' attitude taken to this by some economists, where others feel that actually, it is (partially) their problem. Make sense?

And why should people like megan get so hostile and superior?

I got hostile in retrospect, because at the beginning I was just bewildered and I was all, how are you claiming that you can predict anything at all? I'd just gotten out of engineering school, where it was well-established 18th and 19th century hydraulics, and the predictions were always right, again and again. The water level was as high as they said it would be, and the signal propagated in the time they said it would take. Then I get to my econ and policy classes, where the ideas say that something might happen that way because of something, but it is really hard to test once, much less do repeated trials. So I blurted out "this is great and all, and also interesting, but it sure isn't science or even consistent" and spent three more years getting reamed by the department chair.

That made me hostile. And, moving between departments makes it very clear who has the envy. The econ/policy guys were always discussing how their theories were provable, just like physics, and the physics guys were all, econ who?

And, I've given the wrong impression again, because... I really loved my econ classes. Loved them. Thought they were great, and an interesting framework, and fun math, and neat people. Econ was all sorts of great things, but it was pretty conspicuous that the people in it wanted it to have more validity than I thought it warranted.

89: I may have commented here once or twice, but on blogs in general, I comment once and then leave. As somebody said on Mankiw's blog once, it usually feels like screaming in a crowded room.

80,83: Consider my language and attitude a rhetorical stance, a bit of role-playing. Do not take 71 so personally. I did not mean it that personally. We do not know each other, you are a self-representation, etc. I have, umm, an attitude toward comment threads. I have an attitude toward comment threads.

That doesn't mean I did not have a purpose. The politeness of this conversation...

Digby last night discussed his new ally:Jim Webb. Kill Wall Street. The only war is class war.

I don't sit here and say fuck physicists just because they say people are just lumps of carbon, oxygen, hydrogen and some other crap. Physicists tell me I have no soul and that free will is clearly impossible. I don't say fuck you to them.

Wait, are you saying this as an illustration of what's wrong with McManus's argument, or are you actually serious?

I don't say fuck you to them.

Ian, I hereby join Tim in encouraging you to say "fuck you" to more people. With physicists, the best time to do this is when they reveal that they're deeply confused about the conceptual/normative/philosophical significance of some result.

96: While I recognize that a bad time to ask everyone to calm down is right after a new set of people has been insulted, I think in the interests of peace letting that pass would be an excellent thing. I sincerely doubt that there was any serious intent to describe all or most physicists as hostilely proselytizing atheists. (and am curious, combining your umbrage with your handle, do you work at NASA or someplace cool like that? or am I overspeculating?)

94: Why does Ian hate screaming? Is there anything more authentically American than the barbaric yawp? Also, per #95, recall that bob's a dirty hippie, or at least plays one on the blogs. But also old, so again, telling him to fuck off is fun and relieving.

Megan, you too. You're great until you start trying to be nice.

Remember the Unfogged rule: "Fuck 'em if they can't take a joke!"

LB, whatever made you think that "fuck you" might have been insulting?

Ian, I hereby join Tim in encouraging you to say "fuck you" to more people.

I, being a lover rather than a fighter, encourage you to say "fuck me" to more people.

Is there anything more authentically American than the barbaric yawp?

Football.

And to make it even clearer who I'm trying to harass and denigrate with this post, it's this guy, whoever he is and his unfortunately common ilk.

Ian, really. At least 25 per cent of my everyday blogs are economists. David Altig Fed-watching, for Gawds sake. I know nothing, but I can at least drop names like Walras and Sraffa. I am working on it. It is the most serious thing in the world, above feminism and politics, serious like Keynes's famous epigram.

I love economists. I consider all Marxists political economists. Brad DeLong bleeds for the Delphi worker, whether that worker be in Detroit or Shanghai.

But economics is not lit-crit. It is the number-one tool of oppression and exploitation. I will be rational when you get your foot off my neck.

I'm just not willing to scream along with idiots. the precise people you're upset by, LB, the people who will simultaneously tell me the phillips curve doesn't exist and that the Fed is destroying the country are just blathering idiots. that's what I can't handle.

Saying "I don't say fuck you to physicists" doesn't mean I'm against saying it, sure, fuck em all, I was just trying to be morally superior, it felt right in the moment.

Oh, no, I'm not umbrage; just as a possible misconception that would be illustrative of why physicsts get cranky at economists. Of the sort that are prone to care about philosophy of science issues, which as Megan pointed out above is actually vanishingly small.

And nope, I don't work at NASA, but the handle analysis is right on.

This blog is all about moral superiority. And anal sex, but mostly moral superiority.

I will be rational when you get your foot off my neck.

My gut says, "No you won't."

Shit, I missed the drama because I don't read any of mcmanus's comments.

Er, taking umbrage? Being umbraged? Umbragizing?

LB, you quit being nice too. I've had to warn you before.

Yeah, but it's not like anyone can hear what your gut says from all the way inside there.

113: Labs can usually pick it up, via the vibrations it produces in his viscera.

111: Have I ever mentioned my father's three imaginary cats, Ceremony, Umbrage, and Nostalgic?

"Please, don't stand on Ceremony." [offscreen yowling.]

"And don't take Umbrage." [More yowling.]

"And whatever you do, don't wax Nostalgic." [Ripping noise, extended yowling.]

Someone should paste that onto Mick Jagger's face. It's prettier w/o Apo, though. All cherry-red.

Oh man, Umbragizer would be a great handle.

A cynic might say that there is a certain sort of left-of-center person who dislikes economists because they have developed some very powerful (if flawed) insights into human behavior which make it impossible to dismiss arguments about the importance of markets, etc. as mere self-interest and which need to be considered, and will likely prevail, in debates in which tendentious appeals to sentiment would otherwise prevail. These people thus attempt to discredit economists by making various off-point arguments about the theoretical foundations of economics as a discipline, as opposed to actually engaging with an economist's argument.

Ian D-B, please tell me that you also have undying hatred for mvpy and happyjuggler on the aforementioned Mankiw's blog. Those seem to be exactly the people that LB and especially McManus hate. The ones who feel absolute certainty in the basic econ models for every single problem. Who feel that anyone who ends up poor is merely an unworthy speck who got caught under the treads of progress because they weren't clever or super enough to compete in the perfect market.

They're the people who deserve a hearty "Fuck You".

I'm always surprised to find out that unf's the funny one. Nice e-mail address.

110: I almost left at that.

109:You are right. I won't. I comment in opposition to, besides everyone and everything, certain, what, rules of discourse? The rules that say Amanda Marcotte is way too strident and Yglesias is #1 because of his equanimity and articulateness. The rules that say politics and economics demand a certain tone, a certain submissiveness.

Yeah, the Delphi worker vs the cold equations. The long ethical discussions, tormented Ezra Klein, about abortion. I think the "limits of discourse":what
can be said, how it can be said, how credentials are established, are at least part of the point.

It is my ature to attempt an alternative. Or my nature to be an asshole.

119: I've actually never read the comments at Mankiw's blog once. I feel like it would be about as stupid as trying to wade through a Drum comment thread.

As to economics and the delphi worker, what an I supposed to do, dot all the i's in my equations with hearts? Should I ask him nicely before I write down a model of equilibrium unemployment?

Also, do economists not get any credit for trying to measure to what extent inequality and exploitation are a problem, rather than just assuming there must be some big problem?

Last, Bob, what's your solution? More unions? Great, then when the black kid with no job who's not in the union tries to cross the picket line, maybe the union members can beat the shit out of him to enforce their cartel. Fantastic.

122: You know ogged guested at Drum's place, right? It's worth reading those comment threads.

Those threads were great. Man, do those guys suck.

More unions? Great, then when the black kid with no job who's not in the union tries to cross the picket line, maybe the union members can beat the shit out of him to enforce their cartel.

And hey, I may need to get hostile about this. A history of racism in many unions doesn't mean there's anything necessarily racist about the modern labor movement generally.

126, maybe the whole labor movement isn't racist, but it sure as shit is exclusionary. It protects those with jobs, and screws those without them. Europe has pretty convincingly shown us that worker protection ruins teenage labor force participation (and those kids weren't gonna go to college anyway).

"Last, Bob, what's your solution? More unions? Great, then when the black kid with no job who's not in the union tries to cross the picket line, maybe the union members can beat the shit out of him to enforce their cartel. Fantastic."

I wonder if this is getting into the territory of deliverance from economic expertise that gets people's hackles up.

One starts from a reasonable economic insight--that unions resemble commodity-cartels--and then concludes that the result will be injustice to those worse off.

And, sure, there are some simple analyses that will give you the simple answer that unions will exclude those who are worse off.

But this is exactly where the fact that other things are not equal tends to come in: is it really the case that in those times and places where unions have more power, the "black kid with no job" does worse? Or does he in fact find that his wages for a non-union job get better because of the effect of union bargaining? Or some other results altogether?

My point: this seems like a good case of 101-level analysis leading to right-wing conclusions, without having considered the complexities of particular situations and the actual empirical outcomes.

So this might be the kind of thing that earns people's ire. (I can't say for sure, since I'm not feeling ire-ish.)

"A cynic might say...."

No, I'm the cynic here. It would be a smug economists who would say that.

123:Of you and your profession, I guess I would for now only ask that you discredit the Mankiws and Laffers and Klings like it mattered. Maybe I should be ignored. I think they should be invisible.

God knows, as a socialist. I am not expecting quick and easy.

Give us tools. Webb wants to lower relative executive compensation from Washington. Help him.

It protects those with jobs, and screws those without them.

This is an argument that increased union participation necessarily increases unemployment? If I've understood that correctly, I don't know that I believe it -- to the best of my knowledge, the decades with the highest union membership in the US weren't notable for crippling unemployment.

But I shouldn't get into this argument now, I'm not prepped for it and I should really be getting work done. And I'll get all hostile and unpleasant.

necessarily

I'm pretty sure you're reading this modal into Ian's claim.

Well, was Ian then just cherry-picking the most convenient anti-union tidbit he could find?

128: I love to see analysis without evidence. I'm making a general theoretical point. to respond by saying, well the real world doesn't work like that, even if the theory is sensible, should require a bit of evidenc.

131: no, this is not a claim that unions increase unemployment necessarily. the argument is simply that people who have jobs will be protected. In essence, the rate of transition from jobs to unemployment is lower, as is the rate of transition from unemployment to a job. how those changes compare to each other will tell us about the ultimate effect on unemployment.

that having been said, we can say one thing convincingly -- outsiders will lose out.

I'd also note that the the tradeoff here between structural sclerosis and more stability in manufacturing jobs just doesn't seem worth it to me. there are ways of supporting labor without cartels.

122: Fair enough. I can never tell whether it's masochism or car-crash fascination that causes me to read comments at TAPPED, Marginal Revolution and Mankiw's place, just waiting to see how stupid it can possibly get.

But I'd also like to restate that a lot of the models in Econ 101 are pretty correct. It seems like people often sound like idiots when they use Econ 101 arguments because many public policy disagreement occur in those few instances where Econ 101 models fail. Price and wage stickiness along with asymmetrical information may mean that supply and demand curves don't work perfectly for labor markets, but thousands of firms can still use them to set prices. Governments can still use them to examine where tax burdens lie with some accuracy. The efficient market hypothesis clearly may not hold in all cases, but it's still damn hard to beat the market through skill. It's not a perfect science, admittedly, but the models are fairly accurate so long as they're applied correctly. Just because the basic models fail for a huge number of situations in the complicated real world doesn't mean they can't be accurately applied to an even larger number of everyday situations and provide essential insights into otherwise odd outcomes.

I gotta catch a plane, but I don't think unions are universally bad. However, the evidence, not theoretical models, pretty clearly shows that unions depress the rate of transitions from unemployment back to employment.

131: But LB, that's the kind of response that drives economists nuts. Just because high employment and high union membership were coincident doesn't mean there was a causal relationship. In fact, many economic historians point to the macro conditions that generated high employment as enabling high union membership (because the cost to being excluded was low, unions were seen as largely benign; I'm thinking post WW2 here).

Similary, a rise in employment following a rise in the min wage means nothing if we can't determine causation. This is what economists try to do: find the causal relationships when they aren't readily apparent. It turns out that conducting statistical tests on mathematical models developed from a few intuitive assumptions is awfully helpful in this task.

I'd also note that the the tradeoff here between structural sclerosis and more stability in manufacturing jobs just doesn't seem worth it to me. there are ways of supporting labor without cartels.

Only because everyone else is jumping him do I admit that I agree with Ian on this. I should, however, note that I do so without any expertise in the area and that I am irrationally hostile to unions (which I'm working on, LB; the animus is almost gone, there just isn't any pro-nimus yet).

that having been said, we can say one thing convincingly -- outsiders will lose out.

Outsiders in the sense of 'people with non-union jobs in heavily unionized industries'? Because my understanding is that they'll tend to do better than they would in the absence of unions. My father-in-law had a job like that for thirty years, working in a factory, half of which was unionized. The wages and benefits where he worked were very good, so as to dissuade the workers from organizing. Of course, that didn't keep them from firing him when he hit his fifties, but that wasn't the union's fault.

Wow, Milton Friedman died today.

Wow.

no, outsiders = those without jobs. anyway, much as i'd like to see what I can elicit by telling you all that unions are destroying america and turning us into a bunch of limp-wristed frogs, I gotta go.

137: But LB, that's the kind of response that drives economists nuts.

Look at what I was responding to -- first a casual association of the whole labor movement with racist violence, and then "It protects those with jobs, and screws those without them." I don't think you can say that I responded to a rigorous, emotionless economic argument with something invalid.

If I'd been presented with something more substantial, I would have worked on refuting it more carefully -- for what D-B said, I don't think more than my claim that high union participation isn't necessarily bad for workers as a class, look at this counterexample, was required.

LB, one tidbit, for your correlation causation issue: there has never been a time, not even now, when France had a higher rate of unionization than the US did.

134

you misread 128 if you thought that I was offering analysis without evidence, or claiming that "the real world doesn't work like that."

I was asking whether *you* were offering analysis without evidence, not proposing an alternative analysis. And I was asking whether the world *does* in fact work like the Econ 101 model, not claiming that it does not.

My only claim in 128 was: people here are complaining about deriving policy recommendations from simplified econ models. The claim that unionization is going to lead to the beating of poor black kids looks like a classic case of that. May be true, may not be true, but it sure can't be proven true by appeal to the over-simplified model.

If you've got a different argument via a lot of empirical data, or via a much more sophisiticated model, then bravo.

Further to 137:

Just because high employment and high union membership were coincident doesn't mean there was a causal relationship.

and

Similary, a rise in employment following a rise in the min wage means nothing if we can't determine causation.

Sure. But counterexamples like that mean that in the real world, we can't say that raising the minimum wage will raise unemployment, or increasing union participation will screw workers.

And Ian, if you're still around, come back by sometime and we'll mix it up about the labor movement. I love arguing with people who know stuff.

Although I have to admit that I'm entirely puzzled by 143. I'm sure it fits into the conversation in some obvious way, but I'm missing it.

145: This appears to suggest that I do not know stuff.

I think we can say things about the connection between the min wage and unemployment, but not by using things like hilzoy's charts. You'd need a good sample size that would allow you to control for other factors, but it could be done.

The models are fairly accurate so long as they're applied correctly....

That's one hell of a ringing endorsement.

Almost everyone knows stuff. I was identifying the combination of knowing stuff and disagreeing with me on something I have strongly held and bitterly defended opinions on, which usually means a good workout.

has there ever been a private corporation where the workers were compensated purely through shares in that corporation (as in, all of the shares, collectively)? basically, merging the union with a board of directors type function?

would that work? if the gov'ment hadn't busted up strikes in the early 20th cent., would we have ended up with companies like that?

I'm pretty ignorant on this stuff, except for the silly economic arguments they teach law students.

Friedman

The Tapped thread on Friedman, post by Klein. The comments are already fun. Galbraith was just "junk food for the mind." "Even Neo-Keynesians are glad Friedman moved Keynes out of central banking."

I am old enough to, IIRC, actually have watched Galbraith vs Friedman on Firing Line. Back when TV had intellectuals.

What I'm thinking of (but not remembering much about in detail) is the Amana community. Weren't they originally a communitarian religious organization, that was also somehow a company that produced appliances? I know the company split away from the community at some point, and I really don't know how it all worked, or if I don't have it completely garbled.

In France unions were replaced by organized leftist political parties which wrote the labor laws.

We killed Milton. Damn.

I've never heard of the Amana. It just seems like a natural solution, one that we might have come to with a slightly different history. If companies were like this, I could be a libertarian, I think.

154--

I do think this is crucial. You miss out on a lot if you compare union participation straight across without factoring in what are the things that govt regulation does in one context that unions do in the other (e.g. provide job security, a counterweight to arbitrary dismissal, destroy free enterprise through sclerotic stagnation, whatever--I'm not taking sides on good or bad right now).

It would be like comparing private insurance coverage here and there without noting that *they have national health*.

150 - I just came back from Argentina, where, after the economy crashed and much of the capital fled the country, workers formed cooperatives and took over a lot of the factories. Kind of cool, we'll see what happens.

not to get down on text, but don't practically all of the meta-level complaints about "Economics Simplified" that motivated this original thread apply four-fold to libertarians?

Like--it's political science 101, only class didn't meet for the first five weeks and they never did the reading after that. BUT now they can show you four clever arguments why all government is bad, none of which take into account any actual human experience with governing or being governed?

In fantasy land where workers owned controlling shares in all major corporations, I wouldn't be a libertarian for reasons derived from classic economics 101, which I never took.

I think a lot of people desire strong central governments, in great part, because large corporations make poor decisions. Or because they don't internalize the harms they cause, or something. Without that, live free or die, baby.

Haven't waded through the comments, but had this though:

Chemistry (and biology) do not actually boil down to the physics of the atoms. Think of this analogy: You can write the same computer program in many different languages. Studying chemistry is an organizational system up from studying particle physics.

158- I just liked the Argentine crisis because it made the steaks really cheap when I was there.

or political science 101, for that matter. If wealth were evenly distributed without much government control, why would we need a powerful government? I would still want firemen, roads, and train tracks, schools and hospitals. But not much else.

159: That's because libertarianism is what happens when you take Econ 101 misconceptions, and subtract math and add a mild attachment disorder.

162: Yeah, NattarGcM was telling me the same thing (see 42, 47). I still don't exactly follow, but I'm sure you're right.

165: It also helps if they've heard just enough about Nietzsche to say the word "ubermensch" even if they could never tell you what the hell "sublimate the will to power" means.

Is there a good argument against mild libertarianism apart from gross, intrenched inequalities, which have all disappeared in my fairy land? The need for environmental protection? Could a small government still protect the environment?

I don't think there's a good answer to that without a lot more definition of terms. What's 'small government'? 'Mild libertarianism'? What inequalities remain?

There's no sense in government serving any function that isn't necessary and most effectively served by the government. I suppose given an initial state of affairs in which everything was good, for some level of good, a lot of government could be dispensed with.

168: Most of post-WWII fundamental research in the hard sciences and medicine.

And the Intertubes, which, you know, hey.

170: Practically all of the current Western industrial world derives from the research and engineering of WW2, either directly or indirectly inspired by govenments.

What we need are more high-intensity wars with dedicated opponents.

sure, I'm being very silly. but my silly hypo is how libertarianism came into discussion, so I was wondering what anyone thought about it in the context of the hypo.

The ridiculous terms would be, bizarro USA, where unions were able to take control of major corporations through massive strikes in the 30s, 40s, and 50s. There basically isn't an investor-class; workers are investors in their own companies. We end up having something very similar to a European welfare state in terms of wealth distribution, only not through welfare. There's still unemployment, and a marginal homeless issue.

What would we need a large government for? The EPA? We would still need a safety net for the homeless. But what else?

Maybe this isn't at all interesting, except to make the obvious point that libertarians ignore the structural impact of government intervention in past labor disputes.

Dude, that's anarchosyndicalism, and it rocks! (Not that I think it would work, but I'm all for it.)

168, 170, 172: Yeah, any individual or firm behavior which has significant positive or negative effects on society really needs government intervention. Early education is a biggie, since society gets more total benefit from each individual being literate and moderately knowledgable than the individuals themselves. Without public education, people would probably invest too little in themselves early on.

Fundamental research investment is another cool example. Either the government needs to contribute funding, or you need a monopoly with too much money that gets a little silly and decides to fund a lot of useful but unprofitable research (like Bell Labs in the 50s and 60s, IBM in the 60s and 70s, and I'd argue that Google is doing something similar today).

Yes, the withering away of the state is a lovely fantasy, but who can take it seriously?

sweet! that's what I am, then. nobody tell any employers.

pwnd! by 178!

"but who can take it seriously?"

I dunno, it might have happened. Maybe Marx ruined it via observation bias. Anyway, since when do we take things seriously around here?

172: Yep, but I think you could possibly make a case that while we might have developed nifty consumer electronic gizmos and transportation system and industrialized farming and even the internet with limited/libertarian government, a lot of the more basic research would have been left by the wayside. Certainly no space program. I can think of a dozen individual sub-disciplines in physics alone that would be pretty a good fifty or sixty years behind but for DoD spending; fluid dynamics and accoustics, for instance, which are both totally the bitches of the Navy (which actally kinda sucks, at the moment, GO GO GO MASSIVE DEFUNDING POST-9/11!).

Granted, you could pretty much debate the fallout of the scenario indefinitely, and the libertarians can spin out bullshit rationales for why so-and-so technology or research could have been funded by market forces alone essentially ad nauseam.

Oh, holy crap, I just saw the kitten for the first time! I feel vaguely like I've been initiated into some kind of sacred mystery. Sweet.

Sigh...see, 178 was spam gibberish, advertizing some casino with lots of weblinks. But hey, most of my jokes misfire anyway.

180: Who do you work for, if you don't mind me asking? Some governmental organization or possibly a defense contractor?

I'm feeling Boeing here, but I have no good reason.

Labor union disputes have always come down to two parties -- one defending the poor black kid with no job, and the other dismissing him.

what percentage of the budget do we even spend on research grants? In my new anarcho-syndicalist party, I think we plan to keep that.

184 - says the king of Elephants! Oh wait, that's Babar. (That's how I read your name at first. I kind of loved it.)

184: Which side was which in last winter's MTA strike in NYC?

The answer is both; the "bitches of the Navy" thing no longer applies (although that's not the reason for my bitterness), but when it did, I was at a fairly large research institution that was ostensibly tied to a particular university. They had their own (fucking small) internal R&D budget, and they also funed some projects via the school itself, but mostly wound up finding sponsors for a great deal of the research either through a few defense contractors or the Navy directly. My dad still works there, and has for a very long time; a lot of the "okay who am I billing to this month?" problems that I had simply didn't exist for him in the 70s and 80s, when he'd be billing straight to one Navy-funded initiative for years at a time.

Er, both as in "governmental organization" and "the gubmint". Not Boeing, obviously.

And by governmental organization I mean defense contractor. Shit.

185: not nearly enough. Seriously, funding for basic research is in trouble nearly across the board. This has been brushed away by wonks with the promise of imaginary pink market unicorns delivering private sector interest in a way that historically almost never happens.

191 gets it exactly right, and it's been getting worse and worse for decades. Actually, arguably the death of Bell Labs was sort of the ultimate deathknell of good, long term, open-ended funding.

Which makes dinner conversation with my economist Federal deregulator ersatz mother-in-law singularly uncomfortable.

A perfect opportunity for the classic Freudian slip conversation:

"I meant to say 'please pass the salt' and accidentally said somehow 'You bitch! You ruined my life!'"

Possibly vaguely off-topic: Marx didn't contain this hermetically sealed thing that died with him -- there is still an active Marxist tradition of political economy, in every country in the world other than the United States.

A lot of the critiques of Marx are answerable, in Marxian terms, and have in fact been answered, and many of the limitations inherent in Marx having only been able to see a very early form of capital have been overcome in later writers, including contemporary ones -- for an American representative, one might check out David Harvey.

I say this just because people tend to view Marxism as a historical artifact only, but that's because of the American refusal to deal with it. Same thing with Freud -- in other countries, psychoanalysis is alive and well. But our all-wise university departments don't do it much, so it's "dead" or "discredited." Well, I've got news for you -- being ignored by academics and being discredited are not identical things.

If chemestry really comes down to physics, than doesn't psychology and economics also come down to physics, via the chemical processes that occur in people's brains? I just don't see the analogy as actually being "precise."

Regarding economic modeling. Any model is a simplification of reality which hopefully captures the essential features of whatever you are trying to study. So there are two issues. Are you correctly describing what will happen given the simplified situation? And have you included enough? I think the main problem with economic modeling is that it can be difficult to include enough to reasonably represent reality and still have a simple enough model that you can figure out how it will behave.

On further thought there is another big problem with economic modeling, that it is often hard to test the results. This makes it hard to devise reasonable models and hard to be confident in their predictions.

Good heavens. Yes, I agree entirely.

Everything comes down to physics. It's just that awesome.

The term is unfortunately ridiculously overloaded (just like "science"), which someone should probably do something about.

196: And it's impossible to falsify those result, or control for pretty much anything, at all, ever.

Following on 174, the guy I associate with anarchosyndicalism is Rudolph Rocker (who wrote a collection of essays on the subject), but some of the hard-leftier people may know other or better names.

In addition to Amana (which is now part of Westinghouse, I think), the Oneida Corporation (which probably made your parents flatware) also grew out of a religious collective that practiced group ownership, along with holding a number of remarkable religious beliefs and the belief in the spiritual value of fucking all the other memebers of the collective. The Oneida Community didn't last particularly long. The Palgrave Companion to North American Utopias has details, I believe, and is a fun book in many other respects.

Adam, I don't think that's true. Neo-Marxism is very studied here in the United States. And Freud's legacy is pervasive in modern psychology.

Which makes dinner conversation with my economist Federal deregulator ersatz mother-in-law singularly uncomfortable.

Oh, that can be fun. A friend of mine in college had two Eurocrats for parents. One was a head economist for the EC and the other was a trade negotiator who had to go out and defend the Common Agricultural Policy at every major trade talk. Apparently they could have some really fun dinner conversations, especially once the kids started taking sides.

195: Well, it's really "turtles all the way down". Yeah, it all comes down to vibrating superstrings in 57 flavors but doing the math needed to figure out what I'm thinking is tiresome.

195, 202:

See 162

194:Kotsko I don't think it is off topic at all. I was hoping you would have more, with some of you readings of post-moderns.

Reification, objectification, I don't know, listening to the Manifesto at CT last night it just felt that economics moving from Smith to Mills to Marshall etc, becoming ever more mathematical, abstruse, dominant, and largely at the service of capital was predictable 150 years ago.

Of course it was the intention of Marx-Engels to move economics from the natural order-natural rights foundation of Locke-Smith-Ricardo to a materialist scientific foundation, instruct the workers, etc. Is there a Hegelian thing going on, thesis-antithesis? Surely there is a Marxian explanation of modern academic economics.

The title of the post just felt so profoundly right.

195: Paul, it is a very fuzzy line. In some real sense, `all that is, is physics', but not in a very *useful* sense.

Some of chemistry as it is actually practiced, especially quantum chemistry, is essentially applied physics. Other bits of chemistry, not so much. You can certainly make an argument in principle that biology comes down to physics via chemistry as well, and you can keep tromping along to reach psychology via biology.

However, physics as it is practiced is successul by studying very simple things to very great depth. This mode of `reductionism', if you will, is really quite hopless when facing something as complicated as a biological system. So biologists are forced to work quite differently, which has both benifits and detriments.

I don't think it is very useful to think of most areas as coming down to physics for that reason. On the other hand, parts of chemistry really are quite that way.

Regarding empirical relations (rules of thumb), clearly in a certain sense they are less satisfactory then rules derived from well established fundamental physical laws. However they are hardly limited to economic modeling. For example large computer weather and climate forecasting models are full of empirical relations like more humidity means more clouds which are no more robust than economic modeling relations like higher prices means lower demand. So if you are going to completely discount economic models whose conclusions you find uncongenial because they use rules of thumb you have no grounds for complaint when others completely discount climate models whose forecasts (global warming) they find uncongenial for the same reason.

This calls for a Pratchettesque reference to "quantum", I think. Or maybe some mocking of Steven Wolfram.

200: In econ departments?

206: Ah, good, that feels much more familiar.

So if you are going to completely discount economic models whose conclusions you find uncongenial because they use rules of thumb

No, the idea is that I am going to completely discount economic models whose conclusions I find uncongenial because they use rules of thumb and they fail to successfully explain the relevant data. Where they use rules of thumb, and they correspond well to the actual world, I'm good.

In sociology departments.

"The Oneida Community didn't last particularly long."

Was that Owen? I need to Wiki before I guess. Been looking at Owen and Thompson, the Manchester dudes, Godwin, Wollstonecraft since I read a piece by Veblen directly connecting Marx to Manchester. Marxism to feminism. So much to read.

199:

Anarcho-Syndicalism 101

There are really quite a few sites.

196: Yes, this is the obvious problem with complex systems. This is also the reason that biologists don't work quite the same way that physicists do, for example.

The problem comes when you start letting the models drive rather than the real systems. I believe you see this sometimes in both economics & medicine (I'm sure elsewhere, also). That is, you put together a model that you can handle mathematically, and then try and go justify how the real world is kinda-sorta like it. I don't mean this is how the work is *presented*. I mean how it is actually done.

I think, for example, when exchange prediction was in vogue we saw a lot of that. When you start thinking about real systems you can't restrict your possible models to ones that allow you to do the calculations you'd like to do. If you're doing science, you have to allow for things that you can't handle, or indeed models in which it is provable you won't be able to do the calculations you'd like to. It is this sort of constrained thinking I am imagining when I speak of how it is `actually done' above.

Regarding modeling minimum wage increases. Econ 101 models don't apply very well because people are flexible and can potentially work many different jobs (but not at the same time). So someone may choose to work at a pleasant $6/hr job rather than at an unpleasant $8/hr job but still prefer the $8/hr job to being unemployed. So even if an increase in the minimum wage eliminates the $6/hr job this does not necessarily mean the worker will end up unemployed. Simple Econ 101 models do not account for this sort of thing.

Regarding empirical tests of the effects of minimum wage increases. The increases studied have tended to be pretty insignificant thus making any predicted small job loss hard to detect among all the other factors affecting employment. Similarly the effects of adding small amounts of CO2 to the atmosphere can be hard to distinguish from the noise. However I think even skeptics concede that at some point adding enough CO2 will cause bad effects. Similarly at some point raising the minimum wage will have bad effects. It is a question what you think you can safely risk.

Similarly the effects of adding small amounts of CO2 to the atmosphere can be hard to distinguish from the noise.

Not so much.

Define insignificant, small, and minimal- I think people agree that if the minimum wage were $50/hr it would have negative effects. The question is where the border between reasonable and unreasonable is, which isn't covered in intro courses- they just say that all increases are bad for simpicity and is just as wrong as saying there's never an effect.
Re: CO2, of course there's a cutoff because the carbon cycle is a buffered system, burning one log doesn't make a difference, but I wouldn't call what we've been doing small by any stretch of the imagination.

196 - That set off my "I hate models" rant, because they set an unnatural and unachievable standard for the rest of women to beat themselves with, which I posted at my own blog, because maybe everyone doesn't want to read two pages of my ranting. I don't especially hate economics models more than all complex models, but I still have plenty of hate for all of them. I used your comment as a straw man, but I'll take it down if want me to.

214: That's a poor analogy, since CO2 effects are both measureable, have reasonable data, and pretty good agreement between theory and the real world. It took a while (far longer than it should have, due to outside intereference, largely) but some good science has been done in this area. The same cannot be said about min. wage, as I understand it there is disagreement about ability to measure effects, the data is marginal and the theory isn't solid.

Of course the analogy holds in the the trivial sense that everyone can agree that setting a minimum wage of $1000 USD/hour would be silly, in the same way they agree that setting the earths atmosphere to 99% CO2 would be silly. This has no bearing on useful analysis of either situation.

Was that Owen?

A Yale Divinity washout named John Noyes -- the community was in Oneida, New York, part of the Burned-Over District that produced so many 19th century radical religious movements.

James, I'm confused -- climate modeling is designed to make predictions; when those predictions don't work out, the assumptions are revisited and the models are tweaked. On the other hand, reductive Econ 101 prediction that a minimum wage increase would cause job lost didn't pan out, and... what, there are confounding factors so the model doesn't need to be revised? (We'll ignore the proposed cushy $6/hr gigs people are reluctant to give up in favor of better paying ones.)

(On preview, pwned by sz.)

Regarding arguing with people who have taken Econ 101, I think the best solution is to take Econ 101 yourself (or the equivalent) so you can recognize when they are wrong and point out why.

I don't think it is really the same as arguing with people who say invoke astrological principles which you can safely automatically dismiss.

Hrm. May we continue to point out when they're wrong about their predictions even in the absence of the required coursework? Or need we call in someone who's taken the class?

Wait, why is psychology not a science? It's a young science, but what elements of science are missing? People do experiments and do get predictible results and also observe behavior outside of the laboratory setting with what at least attempt to be the methods of historical science (is that the right term for things like paleontology?) and looks for theoretical explanations that can underpin both lab results and observed behavior. I don't get what component of science is missing from that.

Math, silly.

I don't mean to say that paleontology and psychology have the same methods, but just that there are standards for observation that attempt to be systematic.

There is math! There are lots and lot of statistics!

Psychology seems even more fragmented than economics. If all you knew about someone is that they had a psych PhD from an accredited university, you'd really have no idea what you wer getting. Some of it seems scientific and some not.

When I say that economics is not scientific the way physics is, I'm not saying that it should be, just that it shouldn't claim to be. Psych would be about the same.

222: The title is bad, and I should back off from the reference to science in it. The distinction I'm really trying to make is between the kind of science where you can usefully deduce things from first principles (a category which includes physics and not all that much else) and the kind of science where it's all about staying tightly tied to the data, and using math not to advance outside the data you have so much as to interpolate within that data.

It's not that psych isn't a science, it's that it's not the kind of science that's capable of providing a strong theoretical underpinning for a claim that "Large group of people X, under circumstance Y, will do Z". Physics can do that for chemistry sometimes ("Large group of subatomic particles X, under circumstance Y, will do Z") but psych can't do it for econ in the sort of mathematically rigorous way they want it to.

It's not a question of methodology, but subject matter and generally really hazy operational definitions. This is actually a hazard in any discipline that starts verging into the terriroty of studying epiphenomena; see popular misconceptions of the implications of uncertainty, emergence, cog sci, etc etc.

Which isn't a condemnation, I should point out, just a reflection of my personal bias. I think the entire hard science/soft science divide as silly. There are the sciences, and then there are disciplines that are using the scientific method on phenomena that we can't quantify well enough yet, which may or may not yield useful insights regardless.

Or, what LB said!

Okay. Lunar rockette said in this thread that psychology wasn't a science, and you agreed. If neither of you wants to defend that, that's fine. I just didn't understand what concept of science undergirded that assertion.

But I think there are probably at the very least a few elements of cognitive psychology that are at the deduction from first principles level you describe for physics. Maybe Tom should speak to that.

227:"The title is bad"

"The distinction I'm really trying to make is between the kind of science where you can usefully deduce things from first principles"

Like the Categorical Imperative? I really felt you were complimenting economics, but then I like 19thc philosophy.

That economics was born in politics and ethics is its greatest virtue. The two foundational works are Wealth of Nations and Theory of Moral Sentiments.

But wait, you can make a lot of accurate predictions about what large numbers of people under circumstance X will do; it's just that circumstance X is a lab where you've controlled for lots of other variables. If you put large numbers of people in small groups where everyone else says the bar is longer than it is, most of them will agree with the other people in the group that the bar is longer than it is. You can show that again and again. It doesn't mean you can predict exactly what will lead to an individual's conformity or rebellion in a real world small group where the question is more loaded than bar length, but you've successfully demonstrated the existence of an impulse to conform, even though you don't know precisely what drives it.

I definitely think that cog sci, AI, and physiological psych are sciences. There's a lot of other stuff in psych though.

What I especially disagree with is that the pecking order should be based on "Who is most scientific?" I think that linquistics, for example, is and always will be more scientific than history, because language is more susceptible to scientific investigation than history.

I do think that the hard/soft science division is useful, but not the way it's ordinarily used.

The physics standards is for prediction to be exact and without outliers. That's what "predictive" originally meant. Some degree of predictivity can be attained even by simple common sense.

229: This is what I deserve for being sloppy about what I'm agreeing with -- you're absolutely right about what I said, and I shouldn't have. I looked at that, and thought, no on the calling psychology not a science, but yes on the fact that the problem with with econ is that psych doesn't have the sort of physicsy rigor that econ wants as an underpinning, and agreed with it as a cheap shot against econ. But psych didn't deserve the 'non-science' characterization, and I shouldn't have said it without teasing it all out.

Took a cheap shot, and hit the wrong discipline. Sorry.

It's not a question of methodology, but subject matter and generally really hazy operational definitions. This is actually a hazard in any discipline that starts verging into the terriroty of studying epiphenomena; see popular misconceptions of the implications of uncertainty, emergence, cog sci, etc etc.

And not to be snippy at all, just because I'm interested in the question and would like to understand, I frequently am lost in comment threads when people gesture at things without saying them assuming I, jane reader, have the background to know what they're gesturing at. I usually don't.

231, 233: And for predictions to be useful even under most non-laboratory conditions, which is what we're talking about here.

Statistics is girly math. Real scientists use math to build models.

I think that linquistics, for example, is and always will be more scientific than history

Damn straight.

I once saw a math major deliver a fine, fine rant about how engineers don't love math. They just use her and leave her weeping in the gutter.

Regarding communal enterprises, Israeli Kibuttzim provide examples of various communal forms of social and economic organization. However over time they have tended to evolve towards more conventional forms.

Silly rabbit, segues are for kids.

Okay, it's obvious psychology doesn't aspire to this no outlier standard of predictivity. Not that you're saying it was, JE, but I don't think that was LB's original goal for the foundations of economics; I doubt there will ever be a totally deterministic science of human behavior (or maybe I just hope there won't. Or maybe I think nuclear winter will have forced the remaining 1/50 of the earth's population into subsistence farming with no resources for research before we get there.) And actually I think there are probably some cases in which even social psych can predict what people will do, just not to this no outlier standard. I think some social psych discoveries have been applied with some success but I don't have examples at my fingertips. Perhaps I should google.

But history is much more interesting than linguistics. I forgot the end of my sentence.

Hutterite colonies are also communal, and have been successful for 300-400 years on three continents, but you don't want to go there.

Read today that probably the two most influential economists of the last thirty years, Friedman & Greenspan, are the people who talked Nixon into ending the draft. Some of you know my feelings about that, I think it cut the legs from under the welfare state in America, but whatever, Greenspan & Friedman did not separate their politics from their economics.

Another leg was cut out by monetarism. Now the problem of accusing Greenspan of using monetary policy for ideological ends is that, in the language of economics, it is a simple ad hominem. Verboten.
I somehow have to find the math that proves inflation-targetting had/has political and distributional effects. And in any case, until fairly recently, economists didn't seem interested in the political and distributional issues.

I guarantee Greenspan was. It is kinda inconceivable to say some physicist was propoundng his variant string theory in order to enrich his friends. Economics is no science.

239: That reminds me of a joke a math major friend of mine told me.

An engineer, a physicist and a mathematician were all staying in the same hotel one night. In the middle of the night, when everyone was asleep, a fire started and quickly spread to each of their rooms. All three were awakened by the alarms and the smoke.

The engineer went to the sink, lit a match, put it under the running faucet, and saw that the water put it out. Then he grabbed the ice bucket and started pouring water on the fire until it was out.

The physicist also went to the sink and did the same test with the match, then looked at the fire and calculated exactly how much water it would take to put it out. He then measured exactly that much water and poured it on the fire.

The mathematician too went to the sink and did the test with the match. When he saw that the match went out he said "Aha! A solution exists!" and went back to bed.

But history is much more interesting than linguistics.

No argument here.

207: what doesn't call for mocking Wolfram? Since I read ANKOS, I can't stop.

239: Bitch was just asking for it.

I don't think there will ever be a totally deterministic science of anything, actually.

I'm trying to formulate a way to say what I want to say about psych/econ definitions vs physics definitions, but I can't without babbling uselessly about elasticity and droplet detachment procedures. Which helps no one.

And you know you could always scrap the minimum wage and replace it with a jacked-up earned income tax credit. That way, the increase in the cost of labor would be shared by all taxpayers, instead of just by the entrepeneurs who create the jobs.

I don't think that was LB's original goal for the foundations of economics;

My point wasn't that psych needs to attain that standard to keep economics from being useless crap; it was that psych would have to be the sort of thing that was capable of attaining that standard to make the Econ 101 arguments I'm fulminating against have any validity.

Advances in psych are likely to (and are now in the process of) feeding into advances in econ. But it's never going to be the case that "This simple algebraic rule I learned in the second week of class explains everything relevant".

Wait, shit, did disowning determinism put me in the same epistemological camp as Wolfram? Shit!

And you know you could always scrap the minimum wage and replace it with a jacked-up earned income tax credit.

Why print out all that paper for people to push? Just institute a guaranteed income.

Regarding worker owned enterprises, in my opinion the labor theory of value notwithstanding capital is a independent factor of production and efforts to deny this by artificially tying capital and labor together are likely to be unstable. Workers leaving an enterprise want to take their capital share with them and existing workers don't want to dilute their capital stake by giving new hires an equal share of earnings.

You see the effects in law and accounting firms organized on a partnership basis. Partner payouts will tend to reflect what amounts to ownership shares as well as current contributions to earnings.

That's a fair point, and IMO one of the things that makes anarchosyndicalism a fantasy. (And that leads to some godawful litigation in the partnership dissolution context. I swear it's nastier than matrimonial work.)

252:"Just institute a guaranteed income"

Or a negative income tax? Because it is a Friedman idea, he was a supergenius and I am a patzer, and I don't trust him. I do trust that Milton & Rose were smart enough that all their ideas are somehow connected and support each other.

Marx's state faded away because everybody became workers and concious of their role in creating the state. At that point public & private, people and gov't becomes indistinguishable. Somehow I don't Friedman had the same goal with his efficient small-state policies.

I suspect, as a libertarian, Friedman's negative income tax had an intended purpose of alienating people from the state and each other.

254: Indeed, "islands" of anarcho-syndicalism are unlikely to work well.

I would contend that "islands" of capitalism in a fully socialist world, or a state that for instance did not provide secure health insurance in a world where other economies had it covered, would also perhaps have some competitive problems.

257:Or better maybe distributional problems, but let's noy talk about taht. I'm outta here.

The physics standards is for prediction to be exact and without outliers.

But not for the folk rummaging around in the sub-atomic realm tho'. What they report happening depends on the probability of certain events happening, not on the fixed and predictable behavior of each "particle".

And yes, IMX many experimental psychologists have physics & math envy and cover up nonsense with chalk dust.

But, lastly:

Burley Trailers

Story about a worker-owned cooperative in Eugene, OR, successful for 25 years. via Mark Thoma

256--
""islands" of capitalism in a fully socialist world"

the joke *we* heard in Econ 101 was that after the USSR took over the world, it would leave Finland alone so that it could be the sole remaining capitalist country. Because without at least one market, somewhere, the central planners would not know what to price anything.

But, no, I don't think intro econ does any ideological indoctrination. Just rigorous science.

If psychology really wants to be considered a science (as oppose to a discipline with some scientists and some non-scientists) they'd better stop teaching about Freud in Psychology 101. It's hard to respect disciplines that think Freud is worth studying (with the exception of studying literature whose authors were influenced by Freud).

Okay, I've read the whole thread (some skimming), and I refuse to be sucked into this in a material way, but here are a few quick thoughts:

66: You really don't understand this hostility? It's because most "economic arguments" non-economists encounter are poorly-understood econ 101 bullshit. If we had an entire pundit class writing editorials based on flawed physics, that came to incorrect conclusions that hurt people's lives in real ways, all while dismissing anyone who disagreed with them as simply ignorant of "fundamental" principles of physics (since 101 is the fundamentals, right?), I think you'd see a similar hostility to physics. It's not hard to understand.

73: I think you're underestimating econ here. Econ 101 will give you "correct" policy answers in some contexts, exactly like physics 101 will teach you about firing the cannonball. The problem I think is that most real-world economic analysis is really hard -- there's a *lot* of variable moving around. In other words, makign a real-world economic policy prediction might be better analogized to launching a space shuttle than firing a cannonball. All those little things you ignore in physics 101, like friction and wind and temperature etc. etc., are going to make your rocketship explode if you don't take them into account. Perhaps one big problem with Econ 101 courses as they are currently taught is the examples they use: they teach undergrads how to model launching a spaceship with a physics 101 toolkit. This unfortunately leaves too many undergrads (who never go on to advanced econ) out in the real world thinking they have all the tools necessary to lauch a spaceship (read: tell you about the effects of a trade treaty). Perhaps Economics professors ought collectively to work a lot harder to find some cannonball examples to use instead in teaching their intro courses.

145: Sorry, but this is just wong. If you correctly understand the causation, then you can make exactly that prediction ("all else equal"). And I know your point is "hah! but all else is not equal!", but that doesn't invalidate the model. Returing to our cannonball, just because a meteorite fell from the sky and knocked down our cannonball far short of its projected trajectory, you can't then say "well your model was all wrong-- the cannonball didn't go nearly as far as you predicted." You have to remember that economists are firing cannonballs through an airspace that is heavily meteored. Just because once, twice, repeatedly the cannonball doesn't go as far as predicted doesn't mean the model is wrong (especially when they have very good reasons to explain what happened -- how "all else equal" was violated -- in most cases). Unless you have a plausible contrasting causal mechansism to suggest, or have some good evidence to dispute their theory (with proper controls, etc.), I think you ought to pay it a little more respect. And if you have either of those things, by the way, I think you'll find economists generally very willing to listen, and adjust their views accordingly.

Do the various subpopulations that have existed throughout time in various countries count as "islands of capitalism"? Kulaks, the first Italian moneylenders, etc?

263: In a sea of mercantilism, you mean? I dunno. It's an interesting question.

Hrm. Let me put it another way. Imagine if you will, that we were some alien life-form that breathed a dense, viscous, sludgy substance with a lot of turbulence in it -- stormy jello. And that scientists had discovered this neat science called 'ballistics' that predicted the movement of bodies moving freely in enclosed boxes that had had all of the jello pumped out of them. And they were trying to use 'ballistics' to predict the motion of projectiles in everyday life. One response would be 'It's a perfectly good theory -- the answers are just affected by all of the confounding factors'. But I think a more reasonable response, if it were important where the projectiles landed, would be to say 'Shut up with the ballistics, and if you want to do something useful start analysing the patterns of turbulence.'

And to make my analogy clear, economics proper includes turbulence analysis. It's the Econ 101 arguments that are what I'm calling ballistics. If the confounding factors can reliably be counted on to swamp your theory, stop using the theory to make predictions without accounting for confounding factors on the front end. And if you don't have the chops to deal with the confounding factors on the front end, accept that you can't do predictions.

Brock, it seems that the simple model economists teach isn't a very good one, for the reason you said. Ceteris paribus only is immediately useful in cases when ceteris are usually paribus, and if they aren't, you certainly shouldn't teach the model to freshmen.

As I understand, economics works pretty well as a description of urban produce markets.

265: to some extent, but even in our world where air is comparatively non-viscous and non-dense, aerodynamic drag has a huge influence on how things move, but a clear understanding of how things move independent of air resistance (Galileo dropping big and small iron spheres, as opposed to a sphere and a feather) was a necessary prerequisite to a good understanding of how tumbleweeds move.

268: Yep. And if artillery were more like tumbleweeds than like cannonballs, people wouldn't think of 'physics 101' as a useful means of determining the motion of a projectile, and anyone who suggested that it was would be made fun of. Artillery would be aimed by people who had spent an awful lot of time studying turbulence in fluids, rather than by people armed with a couple of freshman level equations. (not that wind velocity doesn't need to be accounted for even for a cannonball, but it's a small enough effect that the basic equations get you pretty far.)

I don't know what psych 101 classes you're talking about, pause. No one teaches Freud in cog psy or social science surveys that I know of. Freud is relevant to the history of clinical psychology, and for that matter, people are still working out operationalizing and testing psychodynamic theories of various sorts of behavior (they're trying a bunch of stuff at Penn State I gather), so it's not like it's a good idea to just sweep all of it under the rug just yet, and since the later theories are responses to Freud, it would be kind of awkward just to pretend all that never happened. In the experimental disciplines in psychology (as opposed to psychoanalytic therapy, which I guess you're talking about), no one (in my experience) is interested in propping up anyone's theory of anything in the absence of evidence.
Anyway, I think Freud is worth studying, as are psychoanalysis and later psychodymanic theories of therapy in general, though it's certainly not science. I'd probably roll my eyes right back at someone who waved their hand and declared it all of no use. Clinical transference is an awesome concept. You go, Sigmund.

oops, I meant to say my last comment was to 261

And I meant to say social psych, not social science.

There was some study done sometime by someone which demonstrated that economics majors behaved according to game theory predictions, when placed in these fuck-your-neighbor situations. All the other majors performed like regular people.

So okay fine none of the macro predictors work as well as we'd like, but as an econ major one can predict the behavior of other econ majors, and considering how many of us there are that's got to be worth something.

Also,

If we had an entire pundit class writing editorials based on flawed physics, that came to incorrect conclusions

Assholes will be assholes. The same people brought you social Darwinism in the name of biology.

221, you don't have to take the course, going through an Econ 101 textbook would suffice. If you are good at math it is all pretty easy.

Pointing out they are wrong about their predictions of the future is good but requires patience. Perhaps you mean pointing out their methods do not correctly predict the past. Also good but you have to be fair. A prediction of job losses is not necessarily wrong just because no losses can be observed, the losses may be hidden in noise. Some quantification is necessary.

Prior to 1970 there is not much evidence of global warming in global temperature trends. This did not mean the theory was wrong just that the signal was not yet large enough to stand out from the noise.

269: Ah, OK. But if you don't understand turbulence or ballistics, even in a situation where turbulence dominates motion, ballistics is sufficiently easier to understand that you are better off figuring that out first. Or at least it is plausible that this is the case. Or it's not that Econ 101 is useless, just that it doesn't completely (or very accurately) represent the real world, but then neither does particularly anything else.

The thing is "not 'yet' large enough to stand out from the noise" is a much more interesting statement with regard to CO2, in that its addition to the atmosphere is a cumulative process, than with respect to the minimum wage. Sure, there's some level at which a minimum wage would affect employment, but knowing that fact doesn't tell us anything useful about whether it's likely to affect employing within the range where we're talking about raising it.

And thanks for the tip about Econ 101 textbooks not being that difficult.

If you're good at math, LB.

217 it is fair use. Actually I think we largely agree. See for example this old usenet post of mine. So I not exactly sure what was so offensive in 196 as to trigger a rant.

Drat. Math is hard! Let's go shopping!

I feel that it's in the same vein as "while it's very hard to do better at picking stocks than picking them all via an index fund, it's very, very easy to do much much worse."

If the signal doesn't stand out from the noise, then don't teach it to freshmen.

Unless you're trying to indoctrinate them.

But that was unfairly snippy of me. I'm sure it was meant kindly.

279: I doubt 196 was offensive, just that it reminded her of a pet irritation.

LB! How many times have I told you!

Yes, Sensei.

Kill! Kill! Kill!

(279: And I like the usenet post, particularly the closing paragraph:

Simple models have the virtue that their flaws tend to be more obvious which discourages placing excessive weight on their results. For some reason people will often give greater weight to models that they do not understand.

Kill! Kill! Kill!

See, this is the sort of thing I was advocating in the Dodd thread.

LB, compare the following two scenarios. They're admittedly artificial (to preserve the same examples), but I think they get to the heart of my point as to what's wrong with 145:

(a) People: "Let's launch cannonballs!" Physicist: "Let me caution you, while that might be a lot of fun, your cannon is aimed at your village, and if you keep launching cannonballs you're going to destroy your houses." People: "Silly physicist, we've lauched cannonballs from this cannon in the past, and it's never been a problem." Physicist: "Well there are a lot of meteorites between here and there, and in the past you've gotten lucky, but if you keep doing that eventually you're going to get a bad result." People: "Your models are useless!"

(b) People: "Let's make unions stronger!" Economist: "Let me caution you, while that might raise wages for some, it may also increase the natural rate of unemployment." People: "Silly economist, we've had stronger unions and low unemployment in the past, and it's never been a problem." Economist: "Well there are a lot of confounding variables that could have masked the effects in the past, but if you keep doing that, eventually you're going to get a bad result." People: " Your models are useless!"

145 sounds to me like the people in scenario (b).

287 - It makes me think Nancy "Faster, Pusscat!" Pelosi, and that's not happy for any of us.

288: I'm running out of my office, but I'd argue that the confounding factors are strong enough that "Eventually, gonna be a problem" isn't a useful prediction.

Indoctrinate freshmen? I'm shocked that anyone would ever consider such a thing.

279 - Looks like we do largely agree. That is actually an old rant of mine, that lurks untriggered in my heart on most days. It was set off by your comment that the problems with modeling stem from balancing simplicity with predictive ability and being unable to test the results. I think the problems with modeling lie elsewhere, as I described in ranty detail.

288 is pnwed.

281: really? Very easy to do much, much worse? How?

Sure, you could burn up some return in transaction expenses -- it that all you meant?

Predicting which stocks will go down is no easier than predicting which stocks will go up.

Brock, do you my object to my point that, since the theoretical negative effects of raising the minimum wage do not seem empirically to rise above the noise, that this is a bad example to use when teaching freshmen? I understand the dynamic you are talking about -- it's not exactly rocket science -- but in real-world circumstances it doesn't seem to be a very powerful factor.

Part Two: Supposing that you do teach the negative effects of the minimum wage to freshman, can you at least agree that it's wrong to let them think that you have shown them that minimum wage laws are a bad thing, since in real-world terms you haven't?

293: Trying to pick stocks based on some model is one way. This can be as simple as "buying in industries after an article comes out in the popular financial press that said industry is the hot new thing." People smarter than you will figure out what you are doing and act in opposition to take your money. Another way is selling stocks after they go down and buying stocks after they go up. Which sounds like a clearly bad idea, but people do it ("Cisco is taking off, I should buy some!
Oh wait, it tanked, I should sell it!").

And also transaction expenses.

would be to say 'Shut up with the ballistics, and if you want to do something useful start analysing the patterns of turbulence.'

At which point the scientist says, "Please move over here in front of the cannon while I make a few adjustments. We need to understand the simplified case before we start with turbulence."

The problems with economic models are the politician's latching on whichever one seems to their benefit, and the economist's willingness to be bought by flattery and perks.

Damn you, Emerson. I said I didn't want to get sucked into this, and I'm about to be off to bed, but here goes:

I'm not sure I like your characterization of this in terms of "noise", but I can look past that. I'd agree that teaching freshman about simple supply and demand with respect to the minimum wage is dangerous. It shouldn't be done without: (a) stating very, very clearly that empirical work doesn't seem to back up the standard supply and demand story, and (b) using the empirical results to probe into what might be wrong with the model. How could the model be tweaked to better capture reality? What is the data telling us about our assumptions here? This is a fruitful discussion that has a lot of different possible answers, none of them easy, and it gets students used to thinking more deeply about these sorts of ecomomic arguments. This 's what I always did with my freshman when I taught 101.

And yes, I think it's wrong to let them leave your class with the impression that "economics says if you raise a minimum wage it must increase unemplyment." Wrong for lots of reasons, at least one of which is that they may later run across empirical data to the contrary and thereafter doubt everything you taught them.

And no careful economist would ever teach that minimum wage laws were "a bad thing." They'd teach that the standard model suggests they'll raise wages for those with minimum wage jobs, but also raise unemplyment. Whether that's "good" or "bad" is obviously a judgment call based on your evaluation of the trade-off. Whether the standard model is itself correct is a separate question, discussed above.

Really, Brock, economists may avoid the word "bad", but they're about as value-free as Savonarola. The economists' presentation game is really annoying.

And our point has been that the bad effects of minimum wage laws should not be the example you choose if you are teaching undegrads, but that a lot of econ professors do use this example, complete with the implied moral, without any embarassment at all.

Brock's right about what should happen, but John is unfortunately right about what does happen most of the time.

277, this is not qute correct. The Econ 101 job loss effect is always operating, it just may not be big enough to detect in employment statistics for modest raises in the minimum wage. This does not mean it is fair to ignore it when considering the costs and benefits of an increase in the minimum wage. Consider for example regulation of carcinogens. Regulations are imposed even when the number of cancer cases predicted by (often crude) models is too low to be detected in epidemiological studies. The failure to detect an increased cancer incidence in such studies is not considered proof that the models are wrong.

Btw while the Econ 101 effect is always operating it may not be dominant. It is possible to think up models for which a small increase in the minimum wage increases employment. For example after an increase pleasant $6/hr jobs become a smaller number of pleasant $8/hr jobs causing some job losses. These pleasant $8/hr jobs attract some previously unemployed workers causing some more of the old $6/hr workers to lose their jobs. However all of these newly unemployed $6/hr workers go fill previously unfilled unattractive $8/hr jobs. The result total employment rises. This may not be the most plausible result but I don't see that it is impossible. However as the minimum wage increases further I would expect the job loss effect to become dominant.

Regulations are imposed even when the number of cancer cases predicted by (often crude) models is too low to be detected in epidemiological studies.

It's controversial whether this should be done. And in general, the economists who worry about the invisible effects of minimum wage laws are exactly the same ones who tell us not to legislate against the invisible effects of pollution.

Regulations are imposed even when the number of cancer cases predicted by (often crude) models is too low to be detected in epidemiological studies.

Which leads right to silly If-Saves-Just-One-Life-It's-Worth-It legislation.

300: No, that isn't quite correct either, because it has in no way been shown that the econ101 model is correct but the effect is lost in the noise. That is one possibility, but not the only one. IIRC the `old standard' method (Marshall) of deriving this model doesn't even work mathematically, let alone `from first principles'.

Think of this sort of economics as an axiomatic system where sometimes there is argument about both the consistency, and the question of whether or not the the axioms match reality. People happily suggest theorems in the system anyway.

From way back, but to 151: has there ever been a private corporation where the workers were compensated purely through shares in that corporation (as in, all of the shares, collectively)? basically, merging the union with a board of directors type function?

Compensation isn't solely through company stock, but a lot of companies (including the one my dad works for) are wholly employee-owned. They usually operate through an ESOP (employee stock ownership plan) that lets them get around the shareholder cap on S Corporations. They still have to pay a salary, though, since most companies don't make nearly enough profit to cover their entire payroll, and basing it on revenue would sort of defeat the purpose of incorporation.

However, it's not really all that different from the typical setup (except for some tax implications) since there's nothing that says the stock has to be distributed equally.

General Equilibrium is Still Dead ...thanks to Bruce McF at Ezra's

...pdf by Frank Ackerman of Tufts. Mostly on the SMD theorem, implications, denial. Has a section on the history of economists directly using physics as a guide that is fun (statics, dynamics, entropy).

"In the aggregate, the hypothesis of rational behavior has in general no
implications."
-- Kenneth Arrow (1986)

You know, after having read that Ackerman paper, I find it hard to scroll back up without making rude noises. I will keep them to myself.

Isn't it a weird sort of serendipity that Milton Friedman dies and the DJIA hits a new high?

Ah, Milton Friedman, may he be in Hell a week before the Tenure Committee in Heaven hear's he's died.

SMD Theorem

Neat. This Wikipedia stub was written by "radek". I presume the same radek who was very active on the Crooked Timber thread.

I am probably being unfair. Give me a few hours, and I will defend Ackerman against all comers. Now what was Walras's Law again....

:)

Oh, one more thing, regarding Alan Greenspan: If he was so fucking smart, then why, during our last bout of "irrational exuberance" did he contiually screw around with the fed funds rate, rather than upping margin requirements, which would have had the effect of dampening the overheated stock market, without messing up the real estate market and otherwise ham-handedly affecting the economy? It's not that he didn't bring economic principles into political life, it's that he did just the opposite with absurd results.

309: Use of past tense is intended to indicate Greenspan's departure from the Fed rather than his departure from this mortal coil. Unfortunately.

contiually screw around with the fed funds rate, rather than upping margin requirements

It wasn't the stock market itself that concerned him, it was the overall rate of economic growth--he was trying to stave off inflation, one of his big fears. It seems that he was somewhat unnecessarily worried about this, but that's just my highly uniformed take on it.

My husband's an econ professor, as are a lot of our good friends from grad school. Really liberal folks, all of them. Maybe there was some self-selection there--and it was one of the more liberal departments to begin with--but I think the profession's getting kind of a bad rap.

What he says is: Econ 101 & makes you conservative. Econ 201-701 (or whatever #s the grad classes have) make you liberal again. (Granted, he went to one of the more liberal departments. I am sure that things are different at U Chicago).

There are assumptions that you make to simplify things that can lead in very useful directions--but they're only useful assumptions, not revealed truths. In many cases they're pretty obviously false (e.g.: a dollar is worth the same to everyone, so we just maximize total dollars). And so once you get past the basic level, you need to do the empirical work and add the complications back in.

Yeah, I think the problem LB's addressing here really doesn't have very much to do with academic economics. The people who make claims like this are mostly folks who took Econ 101 and believe everything they learned there is gospel truth. I think the ultimate problem is that so many undergrads take a little econ (because it's perceived as "practical") and never get beyond the level where the simplifying assumptions are massively unrealistic. I'm not sure how you could fix that, though, since you really do have to start out with those assumptions to build the theory before you bring in the complications, and at least some of those students in the giant 101 lectures are going to go on to become economists (and therefore need a thorough grounding in the theory). Maybe separate classes for majors and non-majors? In any case, econ profs should stop using examples like the minimum wage to illustrate the theories in 101.

Whoops, in 311 "uniformed" s/b "uninformed".

288: The problem with that gedankenexperiment is that, in the case of the cannonballs, we've actually replicated the freefall projectile scenario enough times to have an accurate model -- or at least accurate enough absent large-scale effects from friction, air-resistance, quantum uncertainty, etc. -- precisely because we can eliminate the extraneous variables to within a fairly fine tolerance. [To abuse your example past the breaking point, because we can stop the meteors at will. Or alternatively, because we can fire cannonballs where the meteors aren't falling.] When it's meteors all the way down -- abusing your metaphor like Cornyn on a box turtle -- as in your second example, the claim that one knows "eventually you're going to get a bad result" becomes a lot less convincing.

184 was sarcastic, of course. "No one cares about the unemployed more than management trying to screw workers."

I am reluctant to accept the self-reports of economists who say that they mere motivated to enter the field out of compassion for all mankind and a desire to do good for the poor.

Regarding anarcho-syndicalism, it has to be said that there's a pretty strong critique of anarcho-syndicalism/industrial unionism (to lump together historically distinct and sometimes competing political tendencies) that it merely leads to an unacknowledged government-like thing, with elites and big power differentials and the whole bit. Yet and still, anarcho-syndicalism has also proven one of the most enduringly popular forms of anarchist thought, particularly during anarchism's heyday in the 5 decades between Haymarket and the end of the Spanish Civil War. As this discussion shows, it's an ideology that's pretty intuitive for a lot of people in the current system -- transfer the good works of government to the unions, let workers control the unions in a directly democratic fashion (e.g. federal structure, immediate recall of delegates) and you have a system that can make a lot of sense, even in this age when man is wolf to man.

Shorter minneapolitan: When Adam delve and Eve span, where then was teh gentleman?

This is exhausting.

317: is that a response to me? Because it's not even close to what I said. (If not, sorry, but what is it a response to?)

Let me flog a dead horse here a little more:

300: 277, this is not qute correct. The Econ 101 job loss effect is always operating, it just may not be big enough to detect in employment statistics for modest raises in the minimum wage. This does not mean it is fair to ignore it when considering the costs and benefits of an increase in the minimum wage. Consider for example regulation of carcinogens. Regulations are imposed even when the number of cancer cases predicted by (often crude) models is too low to be detected in epidemiological studies. The failure to detect an increased cancer incidence in such studies is not considered proof that the models are wrong.

Btw while the Econ 101 effect is always operating it may not be dominant.

Look at this claim. Even in a range of changes in the minimum wage where we can't detect the 'Econ 101' effect, we nonetheless know that it's always operating.

How do we know that? We haven't seen real world data that demonstrates that it's 'always operating', have we? No, the claim has to rest on the belief that we can deduce the existence of the effect, even when it's too small to measure, from first economic principles.

What economic principles? That on average, people will always act as rational profit maximizers? I don't believe there's the sort of strong psychological support for this claim that you need to be able to deduce effects too small to measure -- it's not a first principle, it's a convenient rule of thumb, and that means you can't use it to deduce effects that don't show up in the actual data. That the relationship between changes in the minimum wage and employment are going to be linear, so that if we know that a very large change in the minimum wage will reduce employment, that tells us that a small change in the minimum wage must also reduce employment to a smaller extent? Again, it's a nice assumption, probably true over some range of data, but there's no empirical support for it that I'm aware of, which means that you can't use it to claim that 'the Econ 101 effect is always operating' in a range where you don't actually measure any such effect.

It was not necessarily a response to anything. Someone seemed to be on the point of saying that economists really do care, and I've heard them talking about how much they wanted to help people, so I thought I'd head them off at the pass.

i'm willing to grant that the formal effect is there, and that if the minimum wage is very high the effect might be quite significant, but the evidence seems to be that in the American context it hasn't been enough of a factor to make raising the minimum wage a bad idea on the net.

The main points remaining: for a long time the profession was saying something about policy that turned out not really to be true at the concrete policy level, and that even though the empirical facts became known about ten years ago, the profession has changed its mind only very slowly. And last, at the undergrad level students are still being taught an effectively-misleading version of this principle without the caveats which should be attached to it.

312, 313: I don't mean to give the entire profession a bad rap (again, there are lots of liberal economists, more advanced econ is valuable and interesting, all of those things) but I think the problem I'm snarling about goes beyond the fact that the oversimplifications of Econ 101 accidentally mislead people who don't go any further in the topic.

IME, there is a substantial class of people who will use Economics 101 arguments as if they had real-world validity in order to support their desired policy ends, despite what must be enough economic education to know that the arguments don't stand up. I ran into this in law school in arguments with professors. For example, in one discussion, my Contracts professor made a claim that it is incoherent and impossible to talk about reducing corporate profit margins, because it can be economically proven that true 'profits', rather than mere accounting profits which are fixed and unchangeable by the requirements of Teh Market, don't exist. And, you know, bullshit. Historically, profit margins vary from company to company, industry to industry, and time period to time period; no argument from economic first principles is going to convince me that a change in legal regime that affects profit margins is impossible (I'm afraid I don't remember the specifics of the issue being discussed.) But the conversation shut down completely on the point -- the professor wouldn't talk about the possibility of reducing profits, because it was impossible -- economics proved it.

That's not economic ignorance on his part, that's bad faith, and it's a common bad faith usage of Econ 101 arguments.

324, say hello to 274. I think you two could become friends.

Is this the "few bad apples" theory? The stuff LB describes seems quite prevalant in the profession, to the point that to many of us it seems characteristic.

Yeah, but it's not just that assholes will always be with us, it's that Econ 101 arguments are treated as respectable. Mocking Social Darwinism and race science didn't reduce the number of assholes, but it did mean that reasonable people didn't have to listen politely to discussions of how blonds represented a higher type.

The professor was a bully, but the 'Economics is a science in which important truths can be deduced from first principles without reference to the data' belief, which is common and respected, even if real academic economists don't believe it, is a powerful and pernicious tool for bullying, and one that I'm trying to do some damage to here.

you mean blonds don't represent a higher type?

I thought science had proved that.

The encomiums to Milton Friedman I've heard in the past day, from and appealing to this Econ 101 worldview you're referring to, show just how much work needs to be done.

Methodology of Positive Economics ...via T Cowen

Links to a Friedman essay that may show influence of Quine...really is a trip

"To be important, therefore, a hypothesis must be descriptively false in its assumptions;it takes account of, and accounts for, none of the many other attendant circumstances, since its very success shows them to be irrelevant for
the phenomena to be explained." ...MF

Well, LB's professor wasn't "in the profession", so to speak. Or, to the extent he was, then yes, he was just a bad apple. That's the point I think the economists here have been trying to stress: you don't have to be an asswipe to be an economist, and most economists are not, in fact, asswipes. The fact that so many asswipes use poorly-understood and incomplete economic arguments to justify their asswipery certainly gives the profession something of a bad name, but that's really a bastardization of the discipline. I think the comparison to social darwinism in 274 is quite apt: it was the "economics" of its day, and many "pubilc thinkers" presented "scientific" arguments for why any sort of charitable programs were horrible ideas, against the march of natural selection. But of course this was never really what darwinian biologists claimed (except perhaps, again, a few bad apples).

When social darwinism was finally discredited, the next generation of editorialists lached onto free market economics, and picked up right where the social darwinists left off. People who want to justify their greed and indifference will find some theory or another by which to spread their gospel.

330 superficially might appear to directly support the 2nd paragraph of 327; but I read Milty as recognizing the paradoxes, and, hell, what I understand of Quine Beyerstein carries under a fingernail.

331 to 326, although I think it touches 327 as well.

LB: To the extent that your project here is just cleansing social darwinism from the waters so that biological darwinism can thrive, I commend your efforts. (Some of your assualts seem aimed more at biological darwinism, which is what I'm defending.) To the extent that your project here is instead just to emphasize that I, personally, am a bad person (because your contracts professor and I took some of the same classes in college), I disagree with you. At least in this respect.

I read that Friedman quite awhile back. It struck me as gimmicky special pleading. Redman "Economics and the Philosophy of Science" discusses all this in great detail, as the economists went from Popper to Kuhn until finally Lakatos gave them the most convenient justification of what they were doing.

I think that the whole "Is X truly scientific?" / "Is X more scientific than Y?" debate is really usually fake. People are just trying to get a little extra juju for whatever it is they're doing, and they think that being very scientific will get it for them. It's clear that that was what Friedman was doing.

They usually end up having to redefine "science" to fit what they're doing, while dressing up what they're doing in the most scientific way they can.

I think that the big single mistake social sciences make is trying to present the study of historical data (a changing, developing, evolving reality) in analytic terms only suited to constant realities like physics. Someone going back 2500 years to study physics wouldn'thave to do anything different and wouldn't find any new, different facts, whereas an economist going back 2500 years would have no idea where to start. (And to a degree horizontally: economically, different societies behave differently, for path-dependent reasons.)

Hodgson's "How Economics Forgot History" is excellent on this topic.

I think the problem I'm snarling about goes beyond the fact that the oversimplifications of Econ 101 accidentally mislead people who don't go any further in the topic.

Oh, for sure. I phrased 313 to give these folks the benefit of the doubt, but I'm sure a lot of them (particularly in the pubic sphere) are just arguing in bad faith for ideological reasons.

"Yet the belief that a theory can be tested by the realism of its assumptions independently of the accuracy of its predictions is widespread and the source of much of the perennial criticism of economic theory as unrealistic. Such criticism is largely irrelevant, and, in consequence, most attempts to reform economic theory that it has stimulated have been unsuccessful." ...MF

I guess this goes directly to the Ackerman essay and the SMD Theorem. "General Equilibrium" may indeed be utter bullshit (mathematically refuted) as a premise or model and may even by contradicted by data and evidence but that doesn't matter if it provides accurate predictions. Fun stuff.

I think the comparison to social darwinism in 274 is quite apt: it was the "economics" of its day, and many "pubilc thinkers" presented "scientific" arguments for why any sort of charitable programs were horrible ideas, against the march of natural selection. But of course this was never really what darwinian biologists claimed (except perhaps, again, a few bad apples).

This is true, and I've read somewhere that it plays into the William Jennings Bryan involvement in the Scopes Trial. While we associate evangelical Christianity with the right, Bryan was much more of a left-wing populist. Apparently his opposition to teaching evolution wasn't rooted solely in biblical literalism, but partially in a political reaction against Social Darwinism, which he thought of as evil and untrue, and a mistaken connection between Social Darwinism and genuine evolutionary biology. Makes him a little more sympathetic, doesn't it. (All this is sourced to 'something I read somewhere'.)

I have argued to my husband that economists have an obligation not to let their field and arguments get misrepresented and appropriated for bad policies. (We actually got into a fight about this once on the T in Boston, though I think he was actually partly playing devil's advocate).

To the extent that your project here is instead just to emphasize that I, personally, am a bad person (because your contracts professor and I took some of the same classes in college), I disagree with you. At least in this respect.

No, no, I think you're a bad person for entirely different reasons.

337: Almost posted the last bit of Bryan's "Cross of Gold" speech here, as an example of how far we have let "positive economics" take us from "normative economics". I doubt that Bryan had a profound understanding of exchange rates and macro, and he didn't frigging care.

334: Hell, John, as a Marxist, if data and assumptions are irrelevant to theory, that sounds really useful. Ya know?

339- such as?

Crap, I'm such a wimp. I was considering pasting a '(just kidding! :-)!!!)' on the end of that, and decided not to, but now I have to.

No, I don't think you're a bad person at all, not in the slightest. I just saw a straight line and took it.

Oh, pooh. Surely you could have come up with something. (341 was in jest.)

Goddamnit, I am a wimp, I thought you were actually offended.

Alright, I think you're a bad person for being able to successfully con me with a 'mock offended' routine.

(I must now meditate on the teachings of Master Emerson. Concern about hurt feelings is the enemy.)

LB, the correct answer is, "You're a bad person because your nose is in the middle of your face".

The public face of economics is pretty bad. A lot of it may be people outside the field, though frankly I'm reluctant to let economists off the hook for "Law and economics". I think that the Chicago school and the right libertarians probably are overrepresented in the public sphere. (For me, the bland complacency and optimism of economists grates on me too, policy recommendations aside. It's pretty ironic that people so worshipful of the market should be mostly off-market in secure tenured positions made possible by a credentials monopoly).

It would be interesting to know what proportion of the profession voted for Bush in 2004, because only a very hard core rightwinger could do that, and not only that, they'd have to ignore Bush's fiscal policy too.

At least 368 economists, including 6 Nobelists, did support Bush.

At least 10 Nobelists supported Kerry. And Modligliani would have, so that would make MIT (my husband's grad school) 3 for 3. You will also note the MIT people at the top of this list.

Law and economics is law professors and judges who took econ 101. (Okay, some of them actually know how to run regressions etc., but not many. And it has some place in common law fields--torts, contracts, etc.--but Posner needs to stop applying it to Con law.)

321 The Econ 101 effect in question is that some employers are sensitive to the price of labor and will employ less of it as the price of labor goes up. Do you really doubt this? If there are one million people employed at a minimum wage of $6/hr and the minimum wage is raised to $7/hr, you don't think any of the employers will figure out a way to get by with fewer workers? Nobody will decide that maybe they can wait another year to repave their driveway or paint their house? Maybe their lawn only needs mowing every 10 days instead of every week? Maybe the employee bathrooms only need cleaning every other day? Maybe it no longer pays to stay open all night, or to midnight instead of 10pm? Etc, etc. What do you think, that there will be no job losses at all until the minimum wage reaches say $10/hr and then all of a sudden a whole bunch of employers will independently decide that is too much and cut 100,000 jobs? I don't think that is very plausible even if that is the point you start seeing the loss in employment statistics.

The Law and Economics folks really do make economics look bad.

Sure, but you're making an argument just on the basis of "Come on, it just makes sense that there'd be some effect." And it's a reasonable argument, and one that makes sense to me too -- there's a whole lot of people out there, and they make decisions for a whole bunch of reasons, and hiring fewer people because the cheapest rate you can hire them at goes up is a perfectly sensible thing to do, so probably someone does it for that reason. But saying that it's reasonable is relying on folk knowledge about people's motivations and reactions (often reliable, sometimes not), nothing much more than that.

But you don't 'know' anything here unless you can measure the effects. It's perfectly possible, and offhand I'd say it's not unlikely at all that there are actually effects too small to measure, instead of no effect at all, when the minimum wage is raised within a reasonable range. But you can't say that you 'know' that there are if you can't directly see the effect, and you don't have a solid basis to reason that the effect must exist, rather than that it seems really very likely, based on the kinds of things people do, that there would be.

(And, more importantly from a policy point of view, if it's too small to measure, it's not important enough to consider.)

348: That's mostly why I'm so cranky on this subject -- as a lawyer, L&E types are the face of economics that I see.

324 Regarding bad faith, people can sincerely believe all sorts of absurd things. Would you apply the same standard to religious beliefs?

That said there is a portion of the economics profession which seems to be rather blatantly sucking up to the rich and powerful. So you get absurd arguments that executive salaries are justified or that Enron's accounting was actually perfectly ok. Actually maybe these guys are finance professors or something rather than in the economics department, I am not too clear on how academia is organized. Anyway it seems to me that the economics profession could spend more time refuting these guys lest people believe all economists have been bought off.

Finally I am puzzled by the constant profit margin bit. According to Econ 101 it is return on capital which is constant, margins are expected to vary.

Apparently the problem is less than people claimed it was for several decades. Not nothing, but much less. No one bothered to check because opposition to minimum wage laws was a confortable idea to economists.

In many cases a dollar or two an hour for labor is just not a decider. It might be critical for a struggling company highly dependent on cheap labor, or someone thinking of buying a laboor-saving device. But most people hire the labor they need. It's not like they can usually switch to non-labor, the way you might switch from expensive apples to cheap oranges.

Fine, but even when you see a problem it still might possible to say that the tradeoff is worth it from a welfare standpoint. So in the frequent actual cases when a problem is not seen, then there's no significant tradeoff, and the wage increase, which is not invisible, is definitely worth it.

As I pointed out in the pollution example, the people who are most vigilant about invisible tradeoffs when it's the minimum wage are the same people who are most skeptical about invisible tradeoffs when it's pollution and regulation.

I repeat that this is not a good example to teach to beginning students, unless you're trying to inject a little right-wing propaganda into their heads. Introductory examples should be typical and unambiguous, shouldn't they?

Regarding bad faith, people can sincerely believe all sorts of absurd things. Would you apply the same standard to religious beliefs?

Yes. But no claims to religious tolerance are allowed for economists.

Economists seem so smug so much of the time.

324 Regarding bad faith, people can sincerely believe all sorts of absurd things. Would you apply the same standard to religious beliefs?

I manage to avoid getting into hostile conversations with people about the insupportability of their religious beliefs by assuming that they are supported by a subjective experience of the Divine. This may not be the case, but it keeps me polite on the subject.

On the econ front, when you repeatedly run into people making fallacious Econ 101 arguments, and you know they've been exposed to enough Econ 201, 301, and 401 that they should know better, bad faith becomes an attractive assumption. Believing that their arguments are supported by a direct revelation from Ayn Rand doesn't work as well to generate charity.

Finally I am puzzled by the constant profit margin bit. According to Econ 101 it is return on capital which is constant, margins are expected to vary.

Reasonably so, it was a decade ago and I've forgotten the details of the argument completely. I did, however, at the time, sit down with an Econ 101 textbook in the library to figure out what the hell he was talking about, and in the context of the discussion we were having it was completely irrelevant handwaving. Shut the conversation down nicely, though.

354- the puzzling thing about your story is: was no one in your class fluent enough in economics to call him on this irrelevent handwaving? Seems odd, unless you were all just timid.

I argued it for quite a while, but a 1-L classroom really isn't a venue for freewheeling discussion -- it's tightly controlled by the prof. The discussion ended with 'Econ 101 proves you wrong, next topic'; I don't think another student weighed in.

I really shouldn't have brought this up as an example without being able to remember the specifics of the argument better, and it's just not coming back.

I wasn't accusing you of lying or exaggerating, if that's how 355 came across. It's just that law students are usually godd about pressing points they believe are correct, and I would have thought *someone* in the class would have been comfortable enough with economics to challenge him. I guess not.

349: Yes, this is what I was trying to point out in 303. It isn't that there is no plausible argument there, it is just that people totally misrepresent the case that this is what actually happens.

331: I don't think anyone is arguing that someone must be an asswipe to be an economist. The argument is more that the economists are not accepting responsibility for abuse of their fields results, and in some cases they are fascilitating it. I'm certain plenty of economists are decent people.

349 I disagree in general but I particularly disagree with the last part:

"(And, more importantly from a policy point of view, if it's too small to measure, it's not important enough to consider.)"

One big problem with this is that whether you can detect something depends on how carefully you look. So this encourages people to game the system by defunding data collection. This is the Bush let's not count Iraqi casualties so we can more easily ignore them philosophy.

I assume that the law prof's argument was "economic profits are zero; Econ 101 proves it".

According to one literature survey I saw recently (on Kash's blog, maybe?) economists did look at minimum wage before, and the effect was not that small. But then they looked at it again using a more sophisticated empirical analysis, and the effect did turn out to be small. So it's not that they didn't look, it's just that economists have become more sophisticated statisticians. (Both statistical models and the best practices in applying them have improved steadily over the years.)

I just googled around a bit, trying to refresh my recollection, and this page of class notes from a UMass econ course covers what the professor was talking about: the difference between 'accounting profits' -- revenues minus explicit costs, and 'economic profits' -- revenues minus explicit costs and implict (opportunity) costs. And it includes the claim that "In long-run, economic profits in all industries will be zero." (Because if you're making a positive economic profit, people will notice and come compete with you, and if you're making a negative economic profit, you'll exit the industry.)

As I recall the argument, the professor was claiming that it was impossible to design a liability system that would place costs for something or other (it was contracts, so it must have been some category of breach) on a company -- any attempt to do so would simply result in passing the costs along to other customers. I asked how that would work if there were sufficient competition in the market to make a higher price non-competitive (and a sufficient profit margin that it was possible to absorb the additional cost of liability within the profit margin.) The prof's response was that real profits are always zero (this, of course, isn't even honest Econ 101 -- 'in the long run' is a far cry from 'always'), and so that any decrease in the profit margin would necessarily have the effect of forcing the company to exit the market. Econ 101 demonstrated that there were no circumstances under which imposing liability on a company would decrease its profit margins without forcing exit.

I was completely flummoxed in the classroom, largely because he kept on talking about how economics demonstrated that there were no profits, without explicitly drawing the distinction between accounting and economic profits. I'd say "But companies do have profits, there wouldn't be a word for it if it didn't exist," and he'd respond that if I knew more economics, I'd know that they didn't really. This left me with no real options between saying "What on earth are you babbling about, you stupid stupid man?" and giving up and allowing him to proceed to the next topic.

One big problem with this is that whether you can detect something depends on how carefully you look. So this encourages people to game the system by defunding data collection. This is the Bush let's not count Iraqi casualties so we can more easily ignore them philosophy.

This isn't a bad point at all -- perhaps a better way of saying what I wanted to is that an effect that remains de minimus, even after a sincere, serious effort to detect it, isn't worth shaping policy around.

361- Ah. Well, that answers 355. I don't want to get into it now, but as a general matter I agree with your prof. (Depending on what the specific example under discussion was, of course.) He may have had a mediocre grasp on the subject and been misusing ideas (not emphasing the long run or distinguishing economic profits), but I can at least understand why the economically literate didn't object.

Seriously, I'd be fascinated. That appears to me to be a claim that any increase in costs must inevitably be matched by an increase in prices, or the firm will exit the market. Which seems like an overly ambitious claim to make. (Not arguing that it couldn't happen, but must happen?)

359: I was saying was that we should ignore the effect until it's measurable, not that it shouldn't be measured or observed.

In reality, of course, what really happened was that the worst was assumed for decades, and shouted from the rooftops and taught to undergrads, before anyone bothered to try to measure it and found out things weren't really that bad at all.

And I'm confident that the Cato institute, the Restaurant Association, and all the others will be vigilant for evidence supporting their point of view. Until they bring something to the table, let's just raise the minimum wage.

This is the second inversion we've had -- you're worried that something might possibly happen that you don't like, when the corresponding thing has been going on forever favoring your point of view.

For example, picture a firm with a privileged position in a particular market -- say, possessors of a valuable trade secret relevant only to that market, and having the effect of lowering their costs of production. It seems possible to me both that such a firm might be garnering economic profits, and that entry into the market by other competitors, not possessing the same trade secret, would not have the effect of eliminating those profits. A firm in such a position would be able to absorb increased costs without raising prices or exiting the market, no?

Now, in the long run, all information probably becomes public, but is this really such an unlikely possibility as to be dismissible as theoretically impossible?

Were you talking about imposing costs on a particular firm or on a whole industry? Assuming the latter and trying to shorten as much as possible-- I'll expand later if there's a particular that bothers you: the basis of the claim is that in the long run the expected rate of return (risk-adjusted) on all capital should be equal, since capital is mobile (esp. long run) and driven by a search for profit. So if, for example, costs in some industry go up, but the industry for whatever reason can't pass those costs on to consumers, the capital base (and therefore the industry) will shrink until the risk-adjusted expected rate of return gets back up to levels seen across other industries. Keep in mind that the "long run" here could be 50+ years, especially if you are talking about heavy manufacturing or some similar industry. Keep also in mind that this is only really true with perfect competition, which is a pretty good assumption over a long run timeframe (even if not true for shorter timeframes), but could be violated even in long-run periods (particularly with legal barriers to entry, etc.).

So if you were talking, for example, about forcing the "very profitable" insurance companies to not raise their rates for people who get in auto accidents, you're going to see an across-the-board price increase in response. If you also leglislated that no price increase could occur, capital would abandon the industry (long run). Assuming of course, that the industry is not experiencing above-normal profits presently. If they are, then the legislated "cut in profits" would just trim away at this excess, and thereby lessen pressure for entry. Then the government policy is cutting profits much in the same way as new competitor, potentially just bringing the industry back closer to its LR equilibrium.

I feel like that was waaaaay more confusing that it needed to be. Sorry.

LB -- Jesus Christ, your professor sounds like a moron. Free entry and competition provides a theoretical basis for an invisible hand setting price equal to marginal cost (i.e., no economic profit). A casual look at REALITY, however, suggests that economic (not just accounting) profits do exist. There are all sorts of obvious reasons why this would be the case --barriers to entry, problems of information, "natural" or government-granted monopolies, the fact that competition has to take place OVER TIME, first-mover advantage, etc. How did your professor think that Bill Gates got rich???

The problem here is simple -- morons who have had some exposure to Econ 101 should not be smug bastards, and yet often they are.

(By the way after mulling it over for a couple of days I think I have some more nuanced thoughts on gender differences and math, maybe I'll shoot you an email later.)

Perfectly reasonable in the context of 'the long run' and 'the entire industry'. You'll have to take my word (based on a very minimal recollection of the specifics of the argument) that 'the long run' and 'the entire industry' were not the relevant context, and that when I found (essentially) that explanation in an Econ textbook later on, my reaction was not, "Oh, that's what he meant, now I see," but "Sure, that makes sense, but it can't reasonably be said to apply to what we were specifically talking about."

368: Excellent.

366- what policy example can you think of that would be relevant here? The answer is of course the firm with proprietary technology could have economic profits. And the "long run" isn't any fixed period of time -- it's literally however long it takes for this sort of thing to go away, and the playing field to even out.

Only "perfectly competitive" industries lack economic profit in the short run, and that's not a perfect model for any industry, and only a good model for some industries (commodities, mostly).

To follow on to 371, which may at this point be completely pwned -- this isn't advanced econ. These are points that are stressed in 101. So he was either terribly confused, arguing in very bad faith, or you're misremembering the details. I don't mean that harshly; I understand it was a long time ago. I can't remember last week.

So he was either terribly confused, arguing in very bad faith, or you're misremembering the details.

Those are the options. While I don't remember the initial topic of the argument well enough to be certain now that I'm recalling this fairly to him, at the time, after going to a textbook to figure out what he was talking about (well, actually after going to office hours to figure out what he was talking about, and then finding a textbook when that didn't help), my conclusion was bad faith -- he was arguing a policy position he thought was correct, and misusing Econ 101 as ammunition for it. But extreme confusion is also a possibility, I suppose.

But this is a ridiculous amount of rehashing an argument I had a decade ago and can't recall perfectly, with someone who isn't here to stand up for himself. I shouldn't have originally brought it up at all under the circumstances.

Let it be a lesson to you all, though -- the Irish can hold a grudge.

Then 355 remains mysterious. Oh well.

Note (probably irrelevently, since I'm not trying to defend it) that a staunch believer in the competitiveness of our economy could argue that economic profits are zero even in cases like this. The idea being that some incredibly talented person (or group of persons) at the firm must be creating that competitive advantage for the firm (through superior effort and skill), and could do so at any firm. An industrial superman. (Or woman, sure.) So the extra profit is really just a fair return on this guy's incredible talent. In other words, the firm has more capital than you can see with the naked eye -- it's got some incredible human capital. So even though its return on capital looks high on its accounting books, it's total real return on capital is not inflated. And no economic profits are being made.

Believing that something like that explains our economy as a whole requires a pretty serious lack of attention to the details of reality in corporate america, but that's not uncommon. And it requires neither bad faith nor theoretical misunderstanding. (It just requires you to believe we are a perfect meritocracy.) So I retract my statement in 372. Although I'm not sure your professor comes off any better for my having done so.

Back to 355, you're a lawyer, aren't you? or at least I thought you were. Maybe NYU 1L's were more cowed than the students at your school, but it's not surprising to me at all -- while there were arguments with professors, in the 1L classes they tended to be very much under the professor's control, with the students intimidated and receptive.

We were feisty.

Oh, we were soaking up the revealed wisdom like obedient puppydogs. The feisty ones were the ones with the courage to actively fawn on the professors, not so much disagree.

It got better in 2 and 3L.

LB- what about 375?

You mean this?

Believing that something like that explains our economy as a whole requires a pretty serious lack of attention to the details of reality in corporate america, but that's not uncommon. And it requires neither bad faith nor theoretical misunderstanding. (It just requires you to believe we are a perfect meritocracy.)

Boy. Believing something like that may not require theoretical misunderstanding, but I think it does require either real dimwittedness, of an unlikely sort, or a sort of willful blindness to reality that I have a hard time separating from bad faith.

368 - I remember reading someone (Warren Buffett's business partner Charlie Munger, maybe? Possibly addressing some business students.) who mentioned a contrived example of a toll bridge and long-term profit maximization. The classic business school argument is to figure out where you set your pricing so that if a competitor wanted to enter the market, she'd earn an insufficient return on capital (to build a new bridge or establish a ferry). Maybe-Munger said that he'd hire anyone who said that the real-world response would be to crank up the profit margins as high as possible in order to bribe the regulatory officials.

Or Rodney Dangerfield in Back To School: "then there are long-term costs such as waste disposal. I don't know if you're familiar with who runs that business, but I assure you it's not the Boy Scouts."

Not that human capital doesn't show up poorly on balance sheets - there are certainly people out there, especially in small companies - where the company would be better off having their headquarters burned to the ground than have this person quit coming to work. But this is about five layers of abstraction below discussions about "economic profits". Not that discussions of "economic profits" are useless - theory indicates that you don't expect to see them, so if you do see them and don't have a good explanation as to why, it's a good indicator that there is something fundamental that you are missing, as opposed to just the vagaries of existence.

362 Well what about for example radon . Models and extrapolations from high dose exposures suggest residential exposure is killing 21000 people a year. Epidemiological studies have not detected this but they have not included nearly enough people to be expected to detect it. What do you do? Is 21000 deaths a year really de minimis? I think you should use your best estimate when evaluating policy options.

Similarly for minimum wage increases. If there really is strong empirical evidence that the job losses from minimum wage increases have been small then a best estimate of the losses from a proposed increase will also be small and acknowledging them will not preclude an increase.

Except that the linear dose-response model is fairly well supported for toxins (oh, there's argument about it, but there's some basis for it), which lets you extrapolate from high doses to low doses, and therefore estimate how many deaths will be caused by low-dose exposure in the absence of epidemiological data. There is not, in so far an I am aware, the same basis for a claim that the relationship between an increase in the minimum wage and unemployment is linear. The 'it just makes sense' argument (which is a reasonable one, of course) suggests that there should be some effect, but it doesn't tell you a thing about how big it is.

Sorry, that was a response to the first paragraph -- I somehow didn't read the second paragraph, which I agree with.

I really didn't want to come out in favor of radon toxicity.

I don't think that there is a subtle delayed effect of minimum wage such that people could be become unemployed without our knowing it, or suddenly fall unemployed decades later when it's too late. It would seem that if the effect of a rise in the minimum wage is there, it will be visible, and if it's not visible, it must be small. With radon there's a long lag time between exposure and diagnosis, and radon toxicity can easily be misdiagnosed or its cause misattributed. I guess I don't think that the comparison is very germane.

I'm not even denying that the questions we're talking about are something to think about. I'm just saying that several decades of ranting about the minimum wage turns out to have been overstated, and that several decades of undergrads were basically misled, and that several decades of allowing the minimum wage to decline against inflation probably were not as good a thing as economists believe it was.

Well, right. We aren't talking about too-small epidemological studies -- unemployment rates are robustly measured in any area that changes its minimum wage (or doesn't).

388- but only rarely with a good control. In saying the empirical evidence doesn't support the standard econ. theory here, it should be noted that the empirical evidence is relatively thin, or, better, has only very recently become robust.

We always measure unemployment rates in areas that change their minimum wages-- sometimes they go way up, sometimes they go way down. As do the local unemployment rates of hundreds of different locales that didn't change their minimum wage. Sorting out all the causes and effects is not a trivial task.

Yeah, what I meant more is that there isn't a firm theoretical prediction of an effect that we know we haven't done a large enough study to see -- the radon analogy is off both in that the radon theory is more robust, and the understanding of what data would be necessary to check it but is still unavailable is clear in the radon case, foggy in the unemployment case.

The reason why the evidence has been thin is mostly that until about ten years ago, the harmfulness of minimum wages laws was just assumed on general principles, and broadcast everywhere as truth. As soon as a study was done the orthodox position came into question.

Standard supply and demand tells me, at least, that if we start importing things and at the same time a lot of immigrants come in, that there will be a negative impact on wages. But economists all want free trade and almost all want immigration, so they forget about basic economic principles and do empirical studies proving to their satisfaction that there wasn't much harm.

If theory works for you, argue theory, if the facts work, argue facts.....

There's an interesting overarching question here. Something like: What is the responsibility of people within a profession to police how the discipline is being publicly represented?

I usually hear two answers:
1. "But those guys aren't REAL [economists/doctors/etc.]!" (i.e., disclaim any responsibility)

2. "Yeah, those guys aren't REAL [e/d/etc.] and isn't the media stupid for putting them front and center. Here, let me tell you what economics/medicine/whatever really is about." (i.e., take responsibility only in a private/individual setting)

I think economists need to stand up and affirmatively say, "That is NOT economics. We don't believe that, and anyone who says we do is misrepresenting our field. Here's why."

Models for this already exist -- law is a good one. Lawyers are great policing their boundaries. Practicing law without a license is treated very, very seriously. That doesn't mean there aren't kooky lawyers with crazy beliefs, or that law isn't a big tent with competing belief systems. But it does make for an effective set of ground rules. With hundreds of years of agreement on how laws are enacted and modified, nobody can go on Nightline and argue that the law is whatever the President says it is. (Oops, bad example.)

But seriously. When I see purported economics used in political discussions, and economists are silent in response, I tend to think they agree. If not, wouldn't they be shrieking in outrage?

(Yes, I understand that the "ground rules" of economics may not be as agreed-upon as basic principles of law. Even so.)

Part of the problem there is the Ivory Tower thing: a lot of academics (in many disciplines) are more inclined to look at how their field is presented in the media, say "tsk, tsk, sure is a shame the way our work is misrepresented out in the world; good thing we don't have to worry about that here in the academy" and go right back to their work without doing anything about it. Why deal with the rough-and-tumble of the public eye when you could be safe in a comfortable office doing interesting research? Obviously not everyone does that, but it's pretty common in my experience.

I think that the real problem is that a lot of real economists have really obnoxious views, and that for all their pretentions (see Lazear) there isn't as much consensus in economics as there is in a real science.

What are you basing that on, John?

Gary Becker strikes me as a nut case, and to a lesser extent the Chicago school generally. A lot of the Libertarian economists too. And a lot of the law and economics people. The furthest-right people are very far right.

I think that the "lack of consensus" question could be answered by getting a list of proposed universal truths of economics and mailing it aound. There's really an enormous body of accepted physics, with scattered areas of controversy, but I think that consensus economics would be pretty thin. You could probably pump it up by translating it all into ceteris paribus hypothetical propositions, but they would often be pretty empty -- e.g. the assumption of rationality with rationality undefined.

Tyler Cowen has listed five different categories of economic rationality, and he doesn't mentioned Sen or Gintis's takes on rationality.

There are a lot of economists out there, though; they're not all like Becker. I don't really know where they fall politically. They may well be more conservative on average than most people, but they're definitely not all going to be out on the extreme right wing with the guys you mention.

As for the lack of consensus, well, the same could be said of many disciplines. Probably most of them. Physics is kind of unusual in that it has been studied for a very long time and is largely about things that can be studied fairly directly and easily. No social science has that kind of consensus; hell, even most natural sciences don't. And the humanities don't even come close.

I've gotten the general impression that you expect more from academia than it can reasonably provide.

No social science has that kind of consensus; hell, even most natural sciences don't.

I don't know about the social sciences, but aren't you overstating this about the natural sciences? Where do you find well respected academics in the same field who have a hard time agreeing on basic tenets of their discipline?

I probably am overstating it; I'm no scientist. I just think John's holding economics to an impossibly high standard.

People are always asking me to forget about Becker, forget about the Law and Economics people, and so on. Why? Becker is a Nobel Prize winner and teaches at one of the top schools, if not THE top school.

What I would actually hope for from economics would be a lot less arrogance, not more unanimity. Physicists have been hammering away at the "We're Real Scientists" thing, in explicit contrast to everyone else in social science and history, for fifty years or more now. Their claims have always been problematic, and I think that by now a lot of claims have to be scrapped.

For the claim to science to have been plausible, you really whould have had to have had more agreement about more things. There isn't that kind of disagreement in physics, or chemistry, or biology, and there's an enormous body of consensus.

My impossibly high standard is motivated by economics' unrealistic claims. I'm not asking them to reach the standard, but to drop the claims. There'd still be a lot left.

I've been working on this stuff for awhile, and there's stuff up at my URL.

I agree that economists often make overblown claims that economics is a "science." If that's all this is about, then fine, I'm with you. They should stop.

Personally, I care just as little about whether X is a "science" as about whether Y is "art." While claims to the first can obviously have greater (and worse) practical effects in the world, neither is really a very useful way to look at things.

Maybe we can explain it using faux-economic terms.

LB has identified a kind of bad behavior that includes things like irresponsibly throwing around economics terms in political discussions, citing so-called principles based on no factual evidence, and using those techniques to shut down debate (as her law professor did).

Disincentives for engaging in this behavior are limited. Nobody in "authority" is going to call you on it.

Incentives are strong (e.g., as a pundit, you can have recourse to a "scientific" discipline to back up your preferred policies).

Given all this, is it a surprise that we see so much of it?

I don't much care whether economists form a stronger cartel/create higher barriers to entry, or whether they just get more humble. Either would be an improvement.

"Models for this already exist -- law is a good one. Lawyers are great policing their boundaries. Practicing law without a license is treated very, very seriously. That doesn't mean there aren't kooky lawyers with crazy beliefs, or that law isn't a big tent with competing belief systems. But it does make for an effective set of ground rules. With hundreds of years of agreement on how laws are enacted and modified, nobody can go on Nightline and argue that the law is whatever the President says it is. (Oops, bad example.)

But seriously. When I see purported economics used in political discussions, and economists are silent in response, I tend to think they agree. If not, wouldn't they be shrieking in outrage?"

Making a really stupid economic argument != practicing law without a license. It's more like making a really stupid legal argument. Which happens, oh, I don't know, constantly. The National Review's reporting on the Supreme Court is every bit as dismayingly stupid and wrong as what Donald Luskin puts out--every bit. John Yoo not only has tenure at Berkeley, but I'm not sure that he couldn't get it again at some fancy school. He still gets invited to the cocktail parties, too. All sorts of legal arguments on torture, executive power, etc. that I think are a hell of lot worse than the minimum wage oversimplification are treated as credible and have been for five years now.

There are academics in law who are openly hostile to this sort of thing--your Ledermans, Balkins, etc. And among lawyers there are a bunch of nonacademic practitioners who get quoted in the newspapers, who will not pull their punches at all. Economists probably see themselves as more apolitical technocrats, so the divide is less apparent to the public--though there are people like Krugman and DeLong. But really--I think there's less room in econ for total charlatans than law, if anything.

404: Point taken.

I will say that I see a raging legal debate where I often don't see a raging econ debate. That is, (some) lawyers are willing to speak up and say that John Yoo is crazy, or what have you.

I don't go looking for the econ stuff, so maybe I'm missing it. But it seems like a higher number of economists are staying in the world of grant-funded, low-profile research rather than engaging in public debate.

Or maybe it's just that the stakes are lower.

For instance, take this interview with Ed Meese. With Gonzales and Ashcroft, that makes at least 3 out of the past 6 Attorneys General who I would say are agnostic at best about the Rule of Law.

405: No, I agree--lawyers are more used to arguing and saying nasty things about each other, much more at home with their profession involving taking sides, and a much lower % of the profession is academics. So you get more at both extremes. I mean, Bush's economic advisors' professional reputations suffered enough for being hacks that they actually got turned DOWN by several conservatives when they were looking for replacements. Whereas the supply of what one friend of mine calls "Federalist Society Robo-Hacks" is endless.

I don't know if it's preferable or not. Most people here take a side in this debate--but look, we're a bunch of liberals, we're overwhelmingly for a minimum wage increase too. I think a large % of the population just think "well, the experts disagree" and can't follow all the legalese.

In the interests of balance, Emerson --

http://www.msnbc.msn.com/id/15227667/

"More than 650 economists, including five winners of the Nobel Prize for economics, called Wednesday for an increase in the minimum wage"

I have been told that one reason for political reluctance to raise the minimum wage is that contracts for workers at higher ends of the pay scale are often indexed to the minimum wage, so a raise for the low end means a raise for (say) union machine operators or the like. Does anyone know if this is true?

Unionized machine operators.

409: seems hideously unlikely, or possibly far-fetched. The canonical argument about organized labor and the minimum wage is that increasing the minimum wage helps those who do the same jobs as those who's wages just got increased - which is frequently organized yet unskilled or semi-skilled labor - by making them more competitive.