Economics and Its Discontents

May 29, 2007 4:58 AM

Heterodoxy is by definition not widely popular, and so it goes in the field of economics. Lately, though, the orthodoxy has been on the defensive against a faction that's named itself post-autistic economics. Arising in France, it has spread around the world, capitalizing on the widespread feeling that all's not well in the field of economics. [more inside]
posted by Kattullus (57 comments total)

For the record, I find the term 'post-autistic' rather tasteless, but apparently it has different connotations in French. Anyway, here's the original manifesto of the French students who started the post-autistic movement. Their main website is autisme-economie.org which is mostly in French, but has a selection of English texts. The leading light of the movement is Bernard Guerrien, whose short essay Is There Anything Worth Keeping in Standard Microeconomics? can be read on the site. The main English-language magazine of post-autistic economics is Post-Autistic Economics Review which runs the website Post-Autistic Economics Network. Cambridge University is the hub of the anglophone branch of the movement, and they have their own manifesto proposal.

Finally, if someone can find a link to an article that's critical of post-autistic and heterodox economics, I'd be much obliged. I couldn't locate any.
posted by Kattullus at 5:02 AM on May 29, 2007

I don't know if it can be called heterodox if Jeffrey Sachs, mr. UN, endorses pluralism in his best seller, "The End of Poverty."

Ecology had its "stochastic revolution" in the 1970's, after core assumptions and practices had been taken apart by biogeographers. that discipline was 50-odd years old. That economics has avoided a similar crisis for hundreds of years is a bit more sinister, and a reflection of its importance to the political debates of the last century.
posted by eustatic at 5:22 AM on May 29, 2007

By June, their outrage had coalesced into a petition signed by hundreds of students demanding reform within economics teaching, which they said had become enthralled with complex mathematical models that only operate in conditions that don’t exist. “We wish to escape from imaginary worlds!” they declared. Networking through the internet and reaching the media through powerful family connections, they made their case.

Heh. Whiners.
posted by psmealey at 5:24 AM on May 29, 2007

Maybe it should be "neurotypical economics" instead.
posted by TheophileEscargot at 5:29 AM on May 29, 2007 [1 favorite]

Seems to me that every time throughout history a movement gathers to suggest that current economic models are obsolete, a major economic "correction" occurs.
posted by psmealey at 5:34 AM on May 29, 2007

Meh. Economists will change to a different understanding if and when a different understanding can be constructed that better benefits the wealthy and powerful.

*goes back to advocating for Economics to be reclassified under Religious Studies*
posted by Pope Guilty at 5:38 AM on May 29, 2007 [2 favorites]

After the living wage ‘victory’, Harvard activists from Students for a Humane and Responsible Economics (SHARE) decided to stage an intervention. This time, they went after the source, leafleting Ec 10 classes with alternative readings. For a lecture on corporations, they handed out articles on corporate fraud. For a free trade lecture, they dispensed critiques of the WTO and IMF.

So actually this is just anti-globo tripe instead of corporate tripe? The article yammers on and on about how these privileged baby millionaires are oh-so-dissatisfied, but it sure as hell doesn't tell me anything of substance about this "movement." Reading the manifestoes and the other articles doesn't help either. Homo oeconomicus is discredited? You don't say. I'm reading Hirschman's books from the 70s, probably the most mainstream thing you can imagine, and he makes the same points.

What may be learned from this is that a) the French love their mock revolutions, and b) the Americans love to crawl after them, panting for a piece of the action.
posted by nasreddin at 5:39 AM on May 29, 2007 [1 favorite]

Yeah who knows. Economics has some serious problems, as do the WTO and IMF. You don't need to anti-globalizer to see that. Economics isn't called the "dismal" science for nothing.
posted by delmoi at 5:45 AM on May 29, 2007

The authors of all of these pieces seem to be treating Economics as a single body of work with one (or few) preferred methodologies of studying the world. My limited experience has been pretty much diametrically opposite. If you ask three economists, of the same basic school yet, a future-facing question you'll likely get a half-dozen answers. The most fascinating work (to me) in Economics is being done with fMRI machines and addresses exactly the issue of rational consumers. I see no meat here.
posted by Skorgu at 5:49 AM on May 29, 2007

The Standup Economist Translates Manikiw's Ten Principles.
posted by Pope Guilty at 5:57 AM on May 29, 2007 [1 favorite]

That economics has avoided a similar crisis for hundreds of years is a bit more sinister, and a reflection of its importance to the political debates of the last century.

It is difficult, if not impossible, to separate economics from politics. Turgot argued that "human societies pass through cycles of barbarism and civilization, the former attended by superstition, the latter the fruits of reason." Everyone likes to think their good ideas arise from reason, but as David Garrioch observed in The Making of Revolutionary Paris) Turgot (and Quesnay) were probably motivated more by political bias than economic reason.
posted by three blind mice at 5:58 AM on May 29, 2007 [1 favorite]

Homo oeconomicus is discredited? You don't say.

Well, orthodox microeconomics is still built on the assumption 'rational economic man', so it's not considered discredited by microeconomists, which is why things like PAE exist — they're not just tilting at windmills. In modern micro, it's assumed that each agent has a set of complete, transitive, continuous, reflexive preferences and acts rationally to maximise his utility based on these preferences.

privileged baby millionaires are oh-so-dissatisfied

Who do you mean by 'privileged baby millionaires'?

Skorgu, neuroeconomics is indeed interesting, but it's still debatable how mainstream it is. Orthodox economics has a tendency to see empirical contradictions of its results as interesting curiosities rather than the substantial criticisms they really are. (Or they just redefine their theory as normative, as Savage did with Expected Utility Theory when he fell for the Allais paradox).
posted by Aloysius Bear at 5:58 AM on May 29, 2007

Orthodox economics has a tendency to see empirical contradictions of its results as interesting curiosities rather than the substantial criticisms they really are.

Just imagine if physicists, biologists, chemists, or any other sort of scientists tried that. Why do economists get a free pass to ignore reality?
posted by Pope Guilty at 6:15 AM on May 29, 2007 [1 favorite]

Cute:

Harvard President Lawrence Summers illustrates the kind of thinking that emerges from neoclassical economics. Summers is the same former chief economist of the World Bank who sparked international outrage after his infamous memo advocating pollution trading was leaked in the early 1990s. “Just between you and me, shouldn’t the World Bank be encouraging MORE migration of the dirty industries to the LDCs [Less Developed Countries]?” the memo inquired. “I think the economic logic behind dumping a load of toxic waste in the lowest wage country is impeccable and we should face up to that . . . I’ve always thought that under-populated countries in Africa are vastly UNDER-polluted . . . ”

Brazil’s then-Secretary of the Environment, Jose Lutzenburger, replied: “Your reasoning is perfectly logical but totally insane . . . Your thoughts [provide] a concrete example of the unbelievable alienation, reductionist thinking, social ruthlessness and the arrogant ignorance of many conventional ‘economists’ concerning the nature of the world we live in.”

posted by delmoi at 6:19 AM on May 29, 2007 [5 favorites]

There was an interesting discussion on the existence of the neoclassical economics mafia
on Dani Rodrik's blog.

My impression is that it's easy to overstate the differences between neoclassical and "heterodox" economics like neo-Keynsians. They're not working from fundamentally different paradigms, like say behaviourist psychologists and Freudian psychologists.

The differences in their models tend to be about which effects are most significant, which can be ignored, and what the timescales are. Traditionally neoclassical economists have used simpler models and more-ideal markets, which ignore things like asymmetrical information, irrational actors and so on. These tend to favour market forces over government intervention.

When heterodox economists factor in these things, they often get outcomes suggesting more intervention and more regulation. However, they could be accused of tweaking the models to get the outputs they happen to believe in.

The other difference is what timescale you give weight to. Both neoclassical and heterodox economists might agree that liberalizing a certain market will cause short-term pain and long-term benefit. The heterodox might decide the short-term gain is most important, the neoclassical that the long-term benefit is most inportant. For instance, in his book Happiness Richard Layard points out that people derive much more happiness from security than a strict financial analysis would justify. Therefore, to promote security governments ought to avoid short-term rises in unemployment even if the eventual benefits are huge.

The trend seems to be that neoclassical economists are incorporating more and more market imperfections into their models. It seems to me more likely that there will be a new synthesis, than that neoclassical economics will be overthrown.
posted by TheophileEscargot at 6:30 AM on May 29, 2007

Just imagine if physicists, biologists, chemists, or any other sort of scientists tried that. Why do economists get a free pass to ignore reality?

C'mon Pope Guilty. You've seen Sesame Street. Physics, biology, chemistry, economics. One of these things is not like the others.
posted by three blind mice at 6:32 AM on May 29, 2007 [2 favorites]

What are the political implications of lack of empiricism in economics? This all seems more like a criticism of method rather than values. There's no reason why empirical economics (if such a thing exists) would make the world a better place.
posted by footnote at 6:55 AM on May 29, 2007

Also: Isn't the point of the Summers memo on polluting industries that third world countries should be able to benefit economically from pollution, just as first world countries have been able to do?
posted by footnote at 6:56 AM on May 29, 2007

It's unreasonable to take the statement "the economic logic behind dumping a load of toxic waste in the lowest wage country is impeccable" at face value, and ignore its ironic, black humour. Of course Summers (or whoever actually wrote the memo — Summers wasn't the author) was using an exaggerated tone for rhetorical effect. When the memo facetiously refers to "dumping" waste on the poor, it doesn't literally mean dumping waste on LDCs whether they like it or not — as one can tell from the larger excerpt, they are referring to an entirely voluntary transaction.

And anyway, to some extent at least, doesn't he have a point?
posted by Aloysius Bear at 7:09 AM on May 29, 2007

... If you are a non-economist, chances are, what you know as economics is neoclassical. This is the world as populated by homo economicus: the “economic man” who roams the world “rationally” “maximizing” his “utility,” generally through consuming goods. He acts, independently and using perfect information, to calculate and achieves maximum happiness as defined by his preferences, which must have other nice properties in order to be “rational.” Given his impressive detachment from social context and his ability to perform dauntingly complex utility calculations in his head, making use of a semi-infinite data, it would not be out of the question to diagnose homo economicus with autism, the medical condition. ...

That's always bothered me--usually we don't act rationally, and we don't do things that are good for us. And we certainly rarely maximise our utility at all. It's seems like economics is just cover for those with money and power to do whatever they want all over the world.

On most topics, and especially in terms of things that hurt real people, most economists speak of how good those things are, because they're rational and more cost-effective. Isn't slavery more cost-effective? Indentured servitude? No minimum wage? No benefits? No workplace safety or labor laws? Child labor? ...
posted by amberglow at 7:18 AM on May 29, 2007 [1 favorite]

I'm not sure about this latest "revolution" yet, but the current science (heh) of economics as it is commonly conceived sounds as if it was dreamed up by tipsy Republican students one night.

"Wouldn't it be cool if we could prove that everything we want is actually scientifically true!? That every man for himself is the only sane policy, that everyone should grab all he can, that we should maximize profits before everything else, that taxes are bad, that charity is stupid and counterproductive, that no one should be able to tell businessmen what to do..."

"Say! My dad practically owns Harvard!"

"And mine practically owns Yale!"

"Let's put on a show!"

"What?"

"I mean, let's put on a """theory""""

If the post-autistics swing things back the other way a bit, it will be a fine thing.
posted by pracowity at 7:21 AM on May 29, 2007 [4 favorites]

Beat the Press at Prospect always has good things on how economics and data and statistics are always misrepresented in the press, or just inherently flawed and worthless except as a scare tactic to get whatever political aim certain people and groups want (privatizing SS, using slumps in CD prices to justify greater legal crackdowns on downloading, healthcare, immigration, etc).
posted by amberglow at 7:26 AM on May 29, 2007

My impression is that it's easy to overstate the differences between neoclassical and "heterodox" economics like neo-Keynsians. They're not working from fundamentally different paradigms, like say behaviourist psychologists and Freudian psychologists.

I think this is a slightly distorted view. The point of PAE is that it is a fundamentally different paradigm. Orthodox microeconomics is based on rational agents with preferences which satisfy the axioms of choice, and PAE disputes that. The prevailing schools of thought in modern macro are very much based on this, via "microfoundations", to a greater (Rational Expectations) or slightly lesser (New Keynesian) extent.
posted by Aloysius Bear at 7:29 AM on May 29, 2007

This is really typical of autistic economics, i think, and extraordinarily common: Is Dangerous Food Good for the Economy?
That's what policy experts say, according to the Washington Post. The context is the proposals that have been put forward to ensure that food products that we import from China and other countries are safe for us and our pets. The point that the article makes is that improved regulation will hurt many importers and probably lead to higher prices.
While that part of the story is accurate, it is ridiculous to conclude that increased regulation will therefore "harm the economy." The economy is harmed when people and their pets get sick or die from eating contaminated food. This harm is real and measurable. People are willing to pay more money for food that will not harm them. ...
posted by amberglow at 7:29 AM on May 29, 2007

Fascinating stuff—thanks for the post.
posted by languagehat at 7:45 AM on May 29, 2007

If the post-autistics swing things back the other way a bit, it will be a fine thing.

Networking through the internet and reaching the media through powerful family connections, they made their case.

The more things change, the more they stay the same.
posted by public at 7:46 AM on May 29, 2007

This sounds like a restatement of the old criticism that economics contains lots of graphs, but no numbers.

The differences in their models tend to be about which effects are most significant, which can be ignored, and what the timescales are. Traditionally neoclassical economists have used simpler models and more-ideal markets, which ignore things like asymmetrical information, irrational actors and so on. These tend to favour market forces over government intervention.

I think this understates the case. What things like behavioral economics teach isn't that we should take the neo-classical model and add irrational actors, its that all actors are somewhat irrational.

When heterodox economists factor in these things, they often get outcomes suggesting more intervention and more regulation. However, they could be accused of tweaking the models to get the outputs they happen to believe in.

He was too early, and too popular, to be considered heterodox, but I believe Galbraith was accused of this.

The first practical example of this disconnect between neo-classical models and reality that pops to my mind is the application of the Coase Theorem to legal liability standards (which we talked about in my law and economics class). The Coase Theorem (from Ronald Coase of the University of Chicago, the neo-classical capital) basically says that in any problem affecting multiple actors, the overall efficient outcome can be reached by assigning clear property rights and letting the actors bargain their way to the correct result (assuming that bargaining itself is cheap), no matter how the property rights are initially assigned. When applied to the question of what legal liability standards a society should use (should the victim of an accident always be liable for all damages, or the perpetrator, or the victim as long as the perpetrator took reasonable precautions, etc), it teaches that it totally doesn't matter what negligence standard you use- as long as people are rational, they'll always take the efficient amount of precaution, and anyone without a direct incentive from the legal system will be "paid" in some way to do the right thing by those actually liable. This always struck me as ridiculous, and so far removed from reality, as to be not worth discussing (like most potential applications of the Coase Theorem), but apparently there's been a lot of scholarship on this issue.
posted by gsteff at 7:47 AM on May 29, 2007

I'm not sure about this latest "revolution" yet, but the current science (heh) of economics as it is commonly conceived sounds as if it was dreamed up by tipsy Republican students one night.

Hey, if Adam Smith was good enough for Karl Marx, he's good enough for me.
posted by PeterMcDermott at 7:55 AM on May 29, 2007 [1 favorite]

People like more rather than less. Except when they don't.
posted by MarshallPoe at 8:12 AM on May 29, 2007

gsteff, the Coase theorem is an interesting case. I get the impression that the 'Law and economics' group take the theorem more seriously than economists, and in particular Coase himself, do. Coase is a very special case, since it relies on a particularly strong assumption: that individuals have quasilinear preferences. This means their utility function looks something like: u(x1, x2) = √x1 + x2, or perhaps ln(x1) + x2. This is not generally regarded as a very realistic assumption, most of the time. If this condition holds (along with a couple of other less important assumptions), then the Coase theorem is valid, and every assignation of property rights (and hence initial endowments) will result in the same efficient amount of the externality.

I think that even Coase thought that his theorem had been misused. In this case, it's not necessarily the economists that are to blame.
posted by Aloysius Bear at 8:15 AM on May 29, 2007

I think that even Coase thought that his theorem had been misused. In this case, it's not necessarily the economists that are to blame.
posted by Aloysius Bear at 11:15 AM on May 29 [+]

[!]

Yeah, and would you believe that the only economics most law students get is the Coase Theorem? It's pretty sad. We had one indoctrination class that was a mishmash of Coase, some basic finance concepts like present/future values and insurance pools, and anti-regulatory missives showing that the only justifiable regulation EVAR has been seatbelt laws. Kip Viscusi, I shake my fist at you...
posted by footnote at 8:28 AM on May 29, 2007

Most fundamentally, economics ignores politics.

For example, the Coase Theorem gsteff talks about above doesn't work at all if the actors have different abilities to influence the rules by which the bargaining takes place (not to mention different existing resources which influence how long they are able to bargain and the opportunity costs of not reaching "consensus.")

Great post. I'm not sure I agree with everything being promoted by the post-autistes, but I'm glad to know someone is engaging in this debate.
posted by RandlePatrickMcMurphy at 8:31 AM on May 29, 2007

Thank you Kattullus. This post is great.
posted by halonine at 8:35 AM on May 29, 2007

There's something wrong with the study of Economics? I thought the whole point of the study of Economics was to keep economists employed and fairly well-paid, and their employers rich. It seems to me the current models do that fairly well.

I kid. Kinda.
posted by moonbiter at 8:51 AM on May 29, 2007

The problem with economics, and the reason why neo-classical economics is so prevalent, is the simple fact that in order to get published it looks better to have advanced mathematics in a paper. The problem with mathematics is that it simulates a perfect model. In physics this is okay, because the world has yet to change the rules on us. When dealing with governments, climate and a host of other chaotic behavior it is impossible to model things in all but the simplest terms mathematically.

This is nothing really new. Practitioners will tell you that if Efficient Market Hypothesis was really true then banks wouldn't spend millions and millions on R&D. Or if man were truly a rational animal we wouldn't see bubbles and collapses.

I enjoy Hayek who is one of the few, if only, Nobel winning economists who are not totally full of shit. I will admit that Merton Scholes and the rest are great mathematicians, and their models are incredibly rational and beautiful. It is too bad for them the world is not the same way.

In chaotic, dynamic systems like a market, or our brain, or a poker game, mathematical modeling is not a good representation of reality -- at least not right now. Of course it is hard to get published without it and that's while you still see cohorts of students learning CAPM or Value-At-Risk or Modern Portfolio Theory. Of course all this is based on Louis Bachelier's paper that was "discovered" around the 1960s, when we had the computational ability to do his equations on his fly. The fact that the great Poincare (arguably one of the founders of chaos theory with his 3-body problem) stated that Bachelier was a little too remote from what was taught in the day. It is interesting to note that Bachelier faced the academic blackballing that many people face today trying to get his ideas rejected.

I do think that Bachelier had some contribution, but as the saying goes, science is a progression of funerals. I think we'll see finance reluctantly turn to more of a craft, similar to surgery or law. At least if current trends are any indication, some influential papers are being published without crazy partial derivatives from physics toolboxes.
posted by geoff. at 9:19 AM on May 29, 2007

one of the funniest lines i ever saw in scientific american, one scientist telling another "orthodoxy is my doxy, heterodoxy is your doxy."
posted by bruce at 9:20 AM on May 29, 2007

I dropped out of a PhD program in Economics at a ranked school 10 years ago basically over this subject.

Economics always has been, and always will be, a social science.

For a long time, but at least the last few decades, there has been a push to make Economics more of a "natural science" where everything can be reduced down to a sufficiently complex equation. If we could just get the math right, these people say, then all our questions would be answered.

In my opinion, Economics is much more closely related to Psychology than it is to Physics. I am what is called an "institutionalist", and I believe strongly that the study of economics cannot be divorced from politics or sociology, and to attempt to do so is grand folly.

In fact, Economics is the last of the truly Renaissance pursuits... to study it well, one must be well versed in Political Science, Psychology, Sociology, Business, Mathematics, and Finance. At least, it used to be.

Now, all that is needed to be an exemplary Economist is a PhD in Mathematics. In fact, at the program I left, every faculty member under 50 held at least a Masters in Mathematics.

In a rather aggressive argument over this with one faculty member who was an advanced math wizard, one of those "scary smart" kinda people, he fiercely argued that if we could just use sufficiently sophisticated mathematical models, we could totally explain all economic behavior, I asked him what the model would look like for Beanie Babies (this was the mid 90's).

There is of course no answer, and THERE NEVER WILL BE. You will never be able to model behavior that is steadfastly rational and efficient, except when it is unpredictably, randomly, inconveniently, and stubbornly sub-optimal. These two things are at odds with one another, and it cannot be explained mathematically.

However, add in a little Psychology and an understanding of the prevailing social atmosphere at the time, and the answer becomes immediately available, obvious even.

In other words, any idiot can understand how the Beanie Baby craze happened, except classically trained PhD economists.

A pox on them all. It was truly a low-point in my life. And it seems that slowly the pendulum MAY swing back, but by then I will be too old to participate.
posted by Ynoxas at 9:23 AM on May 29, 2007 [7 favorites]

However, add in a little Psychology and an understanding of the prevailing social atmosphere at the time, and the answer becomes immediately available, obvious even. In other words, any idiot can understand how the Beanie Baby craze happened, except classically trained PhD economists.

Any idiot can construct a just-so story for why Beanie Babies became a fad. But can any idiot, or any psychologist or sociologist, construct an empiricially-verifiable theory which actually explains why Beanie Babies succeeded where other toys failed? Or, if the selection of toys that catch on is random (a very unsatisfactory result): of the set of successful toys, how can the magnitude of Beanie Babies' success be explained? The test for the theory is to go back in time to mid-1996 when they launched, and ask: given the hundreds of toys launched that year, which (if any) will catch on? A worthwhile theory should at least be able to come up with a convincing argument for why a particular toy will succeed, and preferably pick the Beanie Baby as the one that will catch on. If the theory can't do that, or at least make a half-decent attempt it, it's worthless. I'm open to being proved wrong, but I'm not convinced that sociological/psychological explanations of the Beanie Baby craze are significantly different from guesses, or have a higher predictive value. Of course neo-classical economics can't explain why Beanie Babies caught on, but at least it's coming at the problem of people's choices and preferences from the right direction: that is, building a theoretical model, rather than constructing ad hoc after-the-fact just so stories.
posted by Aloysius Bear at 10:24 AM on May 29, 2007

Of course neo-classical economics can't explain why Beanie Babies caught on, but at least it's coming at the problem of people's choices and preferences from the right direction: that is, building a theoretical model, rather than constructing ad hoc after-the-fact just so stories.

Why don't you believe it is just random? There is nothing inherent, empirically verifiable that can predict the success of the beanie babies a year before the caught on, or even the year they did catch on. I don't think it is any more predictable than someone in early Sept. 2001 predicting 9/11. The lack of predictability states that it is random, chaotic behavior. A behavior whose very properties preclude it from being predicted.
posted by geoff. at 10:30 AM on May 29, 2007

Economics is much more closely related to Psychology than it is to Physics.

I only made it through my second quarter of Micro back in my college days so this is possibly incredibly naïve to ask, but how is Economics anything other than the study of mass pyschology as it pertains to consumption/hoarding of resources? Are there serious economists that believe that, given adequate brain and computing power, every possible human and group economic interaction can be modeled?

Talk about your fool's errands.
posted by psmealey at 11:20 AM on May 29, 2007

Why don't you believe it is just random? There is nothing inherent, empirically verifiable that can predict the success of the beanie babies a year before the caught on, or even the year they did catch on. I don't think it is any more predictable than someone in early Sept. 2001 predicting 9/11. The lack of predictability states that it is random, chaotic behavior. A behavior whose very properties preclude it from being predicted.

That's proof by assertion, and it also begs the question: it's unpredictable because it's random! No one in this thread (assuming that no one here has spent significant time developing a model for beanie baby popularity) has earned the right to a priori declare that it was random. Give us some strong reasons why.

Indeed, I think the Ty marketing and product development people would disagree quite a bit.
posted by nasreddin at 11:20 AM on May 29, 2007

Why don't you believe it is just random? There is nothing inherent, empirically verifiable that can predict the success of the beanie babies a year before the caught on, or even the year they did catch on.

There's random and then there's random. There's a difference between saying: "dorkowidgets are gonna be huge this year!" and figuring out what made beanie babies objects people were willing to pay significant amounts of money for. There are many ways of approaching this conundrum. For instance, I don't see why a course of research exploring whether any certain economic conditions characterized eras where beanie baby-like fads took off couldn't lead to interesting and incisive results. It might not answer every question, but it could give us clues to the human condition. Have you ever looked at a tulip and tried to see it with the eyes of 17th Century Dutchman? I mean, it's just a flower, right? No point in mortgaging your house to get one. Right? It's a very alien mindset to me, and economics might not be able to tell me how people can come to value tulips, beanie babie or dorkowidgets at ridiculous prices, but it could at least tell me what kind of society allows it to happen.

hot tip from me, invest all your money in dorkowidgets, they're gonna be huge this year... huge!
posted by Kattullus at 11:24 AM on May 29, 2007

Personally, I refuse to accept Boyle's law for the expansion of gases. No physicist can predict the precise path an air molecule will take after even a few thousand collisions. Therefore, the motion of air is purely random.
posted by TheophileEscargot at 11:32 AM on May 29, 2007 [1 favorite]

TPM is getting some economists to weigh in on the matter. They promise blood will be drawn.
posted by caddis at 11:41 AM on May 29, 2007

Of course neo-classical economics can't explain why Beanie Babies caught on, but at least it's coming at the problem of people's choices and preferences from the right direction:

You make my point for me.

Except it is not coming from the right direction, it is coming from the only direction it knows anymore.

When all you have is a hammer, everything looks like a nail.

Economics has lost (I would say purged) everything except mathematical acrobatics. Therefore, every problem to solve, every question to answer, looks like a mathematical puzzle. Sometimes it is. Sometimes it is not. Sometimes we don't know what kind of problem it is.

But your insistence that the only "right" way to answer this is through mathematical rigour... that belies everything that is wrong with modern economics, in my opinion.

The answer lies in a blend of disciplines. There is room for the math, and the psychology, and the randomness.

But current economists and their "you can have my math when you pry it from my cold, dead hands" approach have become farcical.

The problem is, consumers' preferences and utilities cannot even be known, and certainly cannot be "measured" in any sensible way. They can't. At all. Ever. That's the problem facing mathematical economic analysis. That is its Achilles Heel.

In both micro and macro senses, it is unknowable. So a basic fundamental building block of the entire body of work has to be made based on assumption, and at that point, is indeed no better than "guessing".

In my opinion, of course. I am a frustrated student (disciple, really) of economics, not a Nobel Prize winner. So you're probably better off listening to them.

kattullus: that's my point... there are lots of interesting things that one could discuss regarding the tulip bulb mania, but few of those interesting things would have anything to do with high level mathematics. It was the social and psychological aspects that made it worth discussing at all!
posted by Ynoxas at 11:43 AM on May 29, 2007

The motion of a paticular air molecule might be random, but the aggregate is not. But I would argue that things like Beanie Babies are not stochastic processes. It is reasoning like that, that physics is similar to economics, that is the reason why we're int he place we are in.

No one in this thread (assuming that no one here has spent significant time developing a model for beanie baby popularity) has earned the right to a priori declare that it was random. Give us some strong reasons why.

Sure you can come up with a model that works mathematically and incorporates all sorts of quantitative data, and then you can back test it on say, "Look the Beanie Babies fit to the model!" As you can also create a model in which it shows certain comets and planets, in alignment has certain predictive power. The problem is that it lacks classic Popper falsifiability. I think no study would be worth itself in salt unless you study all the toys and fads that came along at the same time that did not explode in popularity.

I would say a much better way to describe such a phenomena would be to show that it shows the same properties of a many-body systems. We can never be sure new evidence will show up in the future that will completely invalidate any model anyone constructs off a social science. Your Beanie Baby popularity model can only be assured for its paticular place and time, vis a vis the problem of induction.

Economics is not a science and creating models and treating it like it is achieves nothing.
posted by geoff. at 11:55 AM on May 29, 2007

I should refine my point: I am not entirely against models. I am not a radical skeptical empiricist like Sextus, I just think that studying the result (that is the popularity of the Beanie Babies) and constructing a model purely based on math and induction is going to miss the generator. The result is not as important as the process which imparts. I don't believe we can ever really know the process which generates things like Beanie Babies unless we had computers modeling every single brain of every single consumer perfectly, that is of course impossible.
posted by geoff. at 12:00 PM on May 29, 2007

Ynoxas, I wasn't defending the assumptions of neoclassicalism, I was arguing that microeconomics has the correct basic methodology, for want of a better word. They're looking at the problem from the right angle: avoiding ad hoc "just-so stories" to 'explain' the past, and instead trying to build a generalised model from the ground up. Whether the assumptions are flawed is a separate issue.

But your insistence that the only "right" way to answer this is through mathematical rigour

I didn't mention maths. When I referred to rigour, I meant rigour in the general sense of intellectual or logical rigour.

geoff: Sure you can come up with a model that works mathematically and incorporates all sorts of quantitative data, and then you can back test it on say, "Look the Beanie Babies fit to the model!"

This is basically a mathematical equivalent of what I was criticising sociologists for earlier in the thread. I think the fundamental approach of microeconomics avoids this. Starting with preferences, the most basic thing, and building upwards from there is the right approach. You can argue that the particular assumptions made by microeconomics are not falsifiable, but that's a different issue.
posted by Aloysius Bear at 1:23 PM on May 29, 2007

Fascinating stuff. I haven't read all the links, but I have a pile of grading on my desk, and definitely will peek back in at this thread when the grading is done tomorrowish. Thanks for posting.
posted by HighTechUnderpants at 1:35 PM on May 29, 2007

I second caddis, the TPM Cafe discussion is very interesting. So far there are articles, by Galbraith the Younger, Tyler Cowen, Thomas Palley and the author of the Hip Heterodoxy article (the first link in the fpp) Chris Hayes.
posted by Kattullus at 1:39 PM on May 29, 2007

You can argue that the particular assumptions made by microeconomics are not falsifiable, but that's a different issue.

Yes, the way I see economics today is much like, as I interpret it, Popper saw political science in his time. Remember people looked at Marx and Freud as scientists and what they were doing as science. Marx predicted that the revolution would come from an industrialized nation like Germany or England from the proletariat class. Instead it came from what can probably be described as the most backwater of European nations at the time, Russia. Of course Marxists revised this and said that Marx simply did not see someone as influential as Lenin, etc. If you approached it as a science, as many contemporaries would have, you would have at least have said that it was a bad scientific theory of history as it became falsifiable.

Similarly I see too often someone creating a great narrative, using beautiful mathematics to show when the next Beanie Baby will come out (the amount of marketing R&D, balance sheet ratios, technical analysis of the market, etc.). If and when the next fad comes out, they will simply say their models were right, it is the world that is wrong. That they didn't predict the "Lenin" in this case.

I think that is perhaps the best analogy of the case I am trying to make. I first came across this when reading the multitude of case studies they give you in business school. They showed which companies succeeded and why, as if it were a formula. They had nice fairy tales to go along with it, with a lot of great buzzwords. The problem is by selecting only the winners, you delude yourself in thinking you know the generating process. I would wager to bet that while there may be conditions for a paticular case (and let us use case to mean a company achieving Beanie Baby like growth) and we may deem those conditions as necessary, it does not mean we'd be able to predict which Beanie Bay toy will explode in growth. Serendipity plays a large, large role in such things.
posted by geoff. at 1:54 PM on May 29, 2007

Eustatic, what was ecology's "stochastic revolution"? Google fails me (or vice versa). Given the numbers and mathy bent of economics grad. students, the field seems to me pretty catholic in its adaptation of mathematical tools.
posted by ~ at 2:12 PM on May 29, 2007

I specifically avoided claiming that micro can even come close to predicting things like Beanie Babies in my comment to Ynoxas. I don't claim that it can, and it's probably only business-school types who think that it does. But the basic idea of starting with the building-blocks, of things like preferences and choices and then attempting to build a consistent model upwards from there is surely the right way to go.

Modern micro assumes that these preferences are complete, reflexive, transitive and continuous: you can then debate whether or not these are valid assumptions. Of course, deciding whether or not they are valid assumptions is not quite as simple as finding a single real-world example where they are violated — by analogy, the assumptions of Newtonian physics have been proven wrong by relativism, but that doesn't mean there's anything particularly wrong with using Newtonian physics to analyse a car rolling down a hill. Micro assumptions are by necessity simplifying assumptions: if the model were every bit as complex as the real world, there wouldn't be any point in having the model in the first place. There's a trade-off between simplicity and tractability. Again, I'm not necessarily defending micro assumptions, just saying that it's a more complex issue than finding some contradictory evidence and immediately claiming the assumptions must be dropped.
posted by Aloysius Bear at 2:23 PM on May 29, 2007

I am in agreement with you.
posted by geoff. at 3:41 PM on May 29, 2007

Phoo-wee! The TPM discussion is getting bloody. From Galbraith's new entry:

Professor Cowen writes with the supreme confidence of the practitioner of intellectual eugenics, passing blithe judgment over who is fit to live and who is not. As a matter of compassion, he avers, he “would like to see heterodox economics survive, rather than perish.” But give it food and shelter? Allow the heterodox to earn a living? Give them access to students?? Heavens no.

Other new articles by Nathan Newman and Max Sawicky. Also of interest is an unrelated post about a bizarre new argument for not raising the income tax on the highest income bracket (it'll keep kids from going to college, apparently).
posted by Kattullus at 5:38 PM on May 29, 2007

In fairness, Cowen's argument was a bit tautological. As near as I can tell, he argued that some ideas that challenge the neoclassical model have already been assimilated by the mainstream, and the bad ones (by definition) haven't. Galbraith is one of the most caustic econ writers out there (see this dialogue with Paul Krugman, of all people); you don't bring a knife to a gunfight.
posted by gsteff at 6:08 PM on May 29, 2007

Regarding Beanie Babies, the whole theory behind having a free market economy (in at least some sectors) is precisely that no individual, committee, model or government department can possibly anticipate the wants of all the people.

The reason we need a free market for consumer goods is precisely because things like the Beanie Baby fad are unpredictable.

So I'm confused as to why that refutes neoclassical economics. If you believed things like that were predictable, you'd be a Marx-Leninist economist.

If you want to judge whether a field is scientific, you ought to look at the predictions it actually makes, and whether they are validated by observation or experiment.

You can't just say "I think it ought to be able to predict X, and otherwise I'm going to reject it". X might be chaotic, too complicated to analyse, there might not be enough data, or it might just be outside the scope of the theory.
posted by TheophileEscargot at 10:56 PM on May 29, 2007