Is neoclassical economics a Mafia?

This controversial article from The Nation says yes, heterodox and non-neoclassical approaches are unduly ignored.  I’m likely writing more about this elsewhere (you’ll get the link), so I’ll save my thoughts for now.


I'm likely writing more about this, so I'll save my thoughts for now.

But you will violate the code of Omerta!

Strange that the article didn't mention Vernon Smith and Experimental Economics, Austrian Economics, or New Institutional Economics, that is, that there are more or less heterodox approaches that comfirm the efficacy of markets and free trade.

Social influences on peoples utility functions has already been analyzed--see SOCIAL ECONOMICS, by Becker and Murphy. More generally, work is being conducted on extending the Neo Classical paradigm to explain previously believed non-rational behavior, etc. While the paradigm is a crass qualitative description of human behavior, it does often turn out to be reliable quantitatively--often 90-95 per cent accurate. Sometimes utility-maximizing behavior just needs to be set up appropriately in the empirical framework.

Yes, the discipline has become more empirical and accepting of heterodox utility and production functions in models, as comes with better empirical tools.

Nonetheless, economists would benefit from a better understanding of economics history and more circumspection about contemporary mathematical models. For example, after Solow invented the first growth model, there were tons of thereotical replicas of it that made minor tweaks here and there. Allegedly, Solow was not impressed, nor should he have been. Developing economists need to know what a genuine contribution to economics knowledge is and what is not. There a lot of interesting questions waiting to be researched--ask them and answer them!

I'm sure the author prefers an *irrational* government to correct for *irrational* actors.

I would just add that the article continually goes back and forth between defining the "mainstream" or "neoclassical" as:

1. a methodology
2. a set of theories
3. a set of policy conclusions

Among the heterodox-left crowd, this same slight-of-hand is often seen, having been the lone Austrian in their midst at a conference a few years ago.

One problem is that there are plenty of neoclassical/mainstream economists who don't believe in the "free market" policy conclusions. The other is the Austrians, who reject good hunks of 1 and 2, but come to the free market policy conclusions.

If you're going to talk about the mainstream as mafia, you better get a clear handle on what defines them.

Looking forward to the expose on bias against heterodox climate scientists.

Your posting speakes about utility maximisation and production function-all neoclassical concepts.
Newinstitutional economics is application of the same old neoclassical theory to institutions.It is not an example of heterodox economics.

I consider Post Keynesianism to be COMPLEMENTARY vis a vis the other paradigms. Many Post Keynesian theoretical models have been earning some empirical credence recently (Pressman & Holt's "Empirical P-K Economics").

Remember, when the capital controversy arose last century (Cambridge U.K. vs. Cambridge U.S.), neoclassicism conceded to Robinson & Sraffa et al. The only thing that kept neoclassicism from going belly up was the weight of Samuelson's authority and the fact that neither theory could be proven empirically. Literally, faith-based. Consequently, as an intermediate micro instructor, I do my best to be well-rounded...

I love the argument strategy: assume the conclusion that mainstream economics is wrong, and then argue from conspiracy that it must be true. You could insert "intelligent design" for "heterodox economics" and have your typical Discovery Institute article.

Also, calling Marx a heterodox economist when he was really just an uber-Ricardian doesn't really inspire confidence in the author.

Then again, most of the Ricardians today would certainly fall into the heterodox category, would they not?

I think what the last one or two decades of economics have shown is that the "mafia" is not so much about resulty, but about method. If standard methods (econometrics or formal theory) are used, you can get into the top journals, regardless what the political implications of your research are. You have to be fairly blind not to see a whole lot of papers with paternalist, anti-free-market implications in the top journals. But they use formal theory. Or up-to-date econometrics. And that is what the bulk of the heterodox camp won't produce.

So, the discussion in The Nation is a bit misleading, I think. If it were about: "What sort of science do we want, what is the adequate methodology?" then it would be relevant. But whining about too little left-wing results in top journals is plain silly.

I think neoclassical economics is persistent primarily because it's easier to learn, comprehend, and apply than most of the alternatives

Brian asks:

Is the number of Austrians larger than the number of so-called heterodox economists in American academia? That's a genuine question, not something rhetorical.

I can tell you that the Society for the Development of Austrian Economics has around 100 members and the Mises Institute's mailing list dwarfs that. The SDAE list probably has a much higher percentage of economists with university or policy affiliations, but the number of such folks willing to identify as "Austrian" or sympathetic to it overtly is certainly in triple digits in the US.

Not sure how that compares to "heterodox" folks.

I would add that, in my view, Austrians *should be* considered heterodox as well. I have an essay in a collection on heterodox economics coming out this year that makes that argument.

It's easy to say that "people aren't rational," because it's easy to choose a very high standard of rationality, and then give umpteen examples of how people fall short. It's very, very, very hard, though, to say that people do not respond to incentives -- which is essentially all I (and I think most economists) mean when I use the term "rational." The divide between orthodox and heterodox really doesn't have much to do with the debate over whether people are rational. Of course they are, with only very, very rare exceptions. A few behavioral economists have leapt onto the "irrational" bandwagon, but they don't seem to realize that the use of heuristics, i.e. mental shortcuts for decision-making, is itself rational behavior -- a way to reduce information costs. For the most part, behavioral economists are, as Hayes notes in the article, part of the orthodoxy.

Statler makes the correct point. Neoclassical economics, or, better, the broader field of today's mainstream economics, is about methodology, not about conclusions and policy implications.

Economics should be about learning how people behave in economic decision-making, about being able to describe the implications of these decisions, and draw inference for what-if scenarios, so that we are able to recommend policies that are in line with what we want to achieve. And the only way to make such inference is, as Becker puts it, to "tie our hands" with methodology.

Heterodox economists (at least most of the "left-heterodox") are completely wrong at least in two ways. First, they confuse this methodology with free-market argument. You feel that market imperfections exist? Build them into the model, verify it with data, and then you'll see there may be space for governmental action. The same about limited rationality of agents, informational advantage etc. The fact that the models often find the space for governmental action limited is not because this is some mafia conspiracy, but rather that the data simply do not support the idea of a beneficial governmental action, or find the cost of such an action as too high.

Second, the argumentation of these heterodox economists very much lacks any methodology at all. When you read the articles in the Post-autistic economics review, they are full of argumentation on how different social norms, different perceptions of value, or different histories make it impossible to transfer policies from one place to another, invalidate the free-trade argument, and strenghten the case for social action. But they never say HOW different these norms, perceptions, or histories are, and how important this actually is. They use this just as an excuse for not being able to set up a coherent argument that could be use for valid analysis. Although I very much favor the importance of things like institutions in policy analysis, you cannot do it without a framework. Otherwise it is just useless blabbery.

Heterodox economists advocate policymaking on a case-by-case basis. But that is a step from serious research and analysis toward disprovable arguments, very much in style of intelligent design. For research is not about proving arguments (which is impossible by definition), it's about disproving them. Neoclassical economics (with its current wide extensions) has shown to be sufficiently robust to extensive analysis, and flexible enough to be able to incorporate a lot of possible scenarios. Using this framework, you can argue which factors do not play a role, and then omit them as a reasonable approximation. Using the argumentation of heterogenous economics, you can never dismiss any argument, since there is no framework based on which you can dismiss it.

We should be thankful that we have a framework which actually allows us to show which things do NOT work, and which can be invalidated by research. If we embraced the approach of (most of) heterodox economists, we would never be able to do so. And that would be the end of economics.


Which Becker? Probably the one that argued that the law of demand is the most fundamental proposition in economics. In one of his articles, he made a brilliant argument that even if people acted ramdomly, the law of demand would hold (he argue this simply on the basis of the income constraint, avoiding utility entirely).


Becker's article was the perfect response to the challenges (I think in the late 1960s or early 1970s) being made by other social scientists about the absurdity of the assumption of rationality. His argument hugely advanced confidence in what he called the most robust concept in economics: the law of demand.

But it seems that this article has not been read by many of the economists newly minted in the last 15 or 20 years, because there have been a small number of publication with theories in economic journals that postulate upward sloping demand curves. No empirical support for the upward slopes, just hypotheses and some in top journals too. I wish more people shared your awareness of Becker's article.

I could comment at length on much of what has been said here so far, but will stick
to just two points.

1) In connnection with Steve Horwitz's remarks, there is an organization that attempts
to unify or act together among the various heterodox schools of economic thought, the
ICAPE, (sorry, I forget what exactly this acronym stands for). Austrians are definitely
included in that organization and I believe active in it, although it is certainly true
that more leftist heterodoxies probably dominate the group. In that regard, of course
The Nation is a leftist outlet and it is not surprising that Hayes focused more on the
identifiably leftist heterodox economists.

2) Ric Holt, mentioned above, has a book with me and David Colander entitled,
_The Changing Face of Economics: Conversations with Cutting Edge Economists_, 2004,
University of Michigan Press, pbk. In that book's first chapter (and a related article
of that year in the Review of Political Economy), we distinguished between "orthodox,"
"mainstream," and "heterodox." We argued that the first category is strictly intellectual,
an established set of ideas widely accepted and established and (more or less) internally
consistent and coherent. The second is sociological, the group that is in charge of the
main journals and funding sources, and the top departments. They may have no ideas at all,
just pure opportunism, but they run the show, and they are not all "orthodox" (or "neoclassical")
in their thinking. Indeed, the "non-orthodox mainstream" was very much a topic of our book.

However, "heterodox" is both an intellectual and sociological category; they are both
anti-orthodox intellectually while also being sociologically alienated from the power
centers of the profession, which are in some sense self-feeding and reinforcing. I hope
that this is useful to various readers.

"A few behavioral economists have leapt onto the "irrational" bandwagon, but they don't seem to realize that the use of heuristics, i.e. mental shortcuts for decision-making, is itself rational behavior -- a way to reduce information costs."

FWIW, I think the thing to do is to tie it back to the part of the brain doing the "thinking."

Becker (1962, JPE) Irrational Behavior and Economic Theory

Regarding the business of upward sloping demand curves, part of the problem
is defining exactly what is to be held constant when one assumes "ceteris paribus"
to draw the curve. Usually it is stated that "everything else is held constant,"
except for the price and the quantity of the good itself.

The problem arises when we allow for a change in the price of the good itself to
change some of these things that are being held constant, in particular expectations.
Thus, in the real world with speculation we see lots of situations and times where
an increase in the price of something draws forth an increase in the quantity demanded,
or at least an increase in people trying to buy the thing. The standard way to
view this is to say that the demand curve shifted with the change in expectations,
but the change was induced by the change in the price of the good itself. The
situation really is rather murky.

I actually ran into this in a weird sort of way quite some time ago. I had submitted
a paper to a journal about speculative dynamics. The paper was rejected because a
referee (whose identity I figured out and is actually a very prominent economist)
said that there can be no such thing as speculation pushing prices and quantities
purchased up. This referee said that this would constitute an upward-sloping demand
curve, and then proceeded to cite George Stigler on the claim that we have never
observed any upward-demand curves. Sic transit gloria, gustibus non disputandum est.

The Econ mafia is part methodological and part elite based. Especially now, when there is no such thing as anonymous submissions, your department really affects your submission reviews. And of course mainly the top departments recruit from the top departments, so there is not a lot of room to break in if you start out at a lesser program, even if you overcome the lack of resources.grad students, etc.

But the other part is methodology, that the heterodox articles look like sociology or polic sci to most of us. It might be good or not, but I can't judge it. I can critique your identification strategy but not your rhetoric.

The problem with this is that it doesn't provide room for economists with a non-mathematical approach like Galbraith. Galbraith would have to be a political scientist now. That's a problem with the discipline that it has narrowed down and doesn't have a way to judge and award the next Galbraith.

Ya know, I read a working paper version of a paper that was published in the prestigious and uber-mainstream Quarterly Journal of Economics, which showed that systemic use of a rule of thumb by stock analysts created large mispricing in the stock market over decades.

If you can actually use evidence to demonstrate instances in which people deviate from the neoclassical model of behavior, especially when the costs of deviation are high, you do quite well in the profession.

So a lot of this "heterodox" complaining just looks like kvetching.

I think the Austrians are actually getting a lot more respect, mainly because their notions of "radical ignorance" rather than just probabalistic uncertainty make a lot of sense given what economics is learning from cognitive psychology. People's cognitive shortcuts and biases do create radical ignorance, and competition does then act as a discovery mechanism.

In some sense, the social darwinist sociologists of the 19th century (Spencer, Sumner) did have deeper insights about economics than a lot of modern neoclassical economists.

Barkley: The demand curve a) is supposed to represent a set of mutually exclusive possibilities that occur over a given planning horizon for b) goods where purchase motives are dominated by use rather than resale.

Condition a) means thqt movements along the curve are entirely notional. (Oskar Morgenstern got this point wrong in one of his papers, where he argued for a complex scheme of accounting for purchases already taken off the market when prices change.) For many markedts, the demand curve is pretty stable from one period to the next, in which case aggregating data across periods will show real "shifts along the curve" that approximate this notional exercise. In general, though, there is no guarantee that time-aggregated data capture the demand curve if it's shifting around.

Condition b) means that for assets that can be resold, it's pretty hard to tell demand curves from supply curves. In the case of financial markets, I recall Steve Ross pointing out that in finance all "supply" or "demand" curves are either vertical or horizontal, with all the action occuring as these curves shift around rapidly under the impact of new information. Ross argued that this was exactly what separated finance from standard microeconomics, which seems like a pretty good demaarcation to me.

So on the basis of these conceptual points, the impact of today's prices on tomorrow's valuation of assets should definitely be thought of as shifting the "demand" curve rather than moving along it.

Then there was that Battalio lead AER article in 1990 that developed a Giffen Good for rats.

As for the asset price demand argument, it is standard 101 that expectations of future price are curve shifters. But maybe I'm just being the neoclassical mob enforcer here. Of course, earlier I mentioned that a lot of old-school social darwinist sociologists had some deeper insights than many modern neoclassical economists, so I may be due for a whackin'.


Sorry about being less than clear; I should have reported the authors names as opposed to just listing the journal, volume, issue number, and page ranges. It is much easier to use JSTOR with the author name, so, again, sorry (I just got lazy in my post).

Regarding the argument that my colleagues (Coelho and Klein) had with Pesendorfer on the topic of fashion cycles, the references are:

Regarding the Becker article on restaurants, I have an unpublished working paper that I will be presenting at the Western Econ Association Meetings this Summer; if you attend, I would be pleased if you would consider coming to my session.

Ah, so indiana jim is McClure. Sorry, won't make the WEA.
Used to belong before they jacked up the price (my own demand
curves are downwardly sloping, thank you).

Ah, but does not your institution have a constrained amount that they give you
for that? Mine does.

Neoclassical economic theory according to me is persistent because it's way more simpler and easier to learn, understand, and implement than other theories.

Great post to read ...want to know more about these non neo-classical aproaches.

Comments for this post are closed