Heterodoxy and the Fly-Bottle

(I have a review in the new Review of Keynesian Economics of a collection of essays on pluralist, or non-mainstream, economics teaching. You can the full review here. Since I doubt most readers of this blog are interested in the book, I’ve posted a shorter version of the review below – just the parts on the broader issues rather than my assessment of these particular essays.)


Wittgenstein famously described his aim in philosophy as “showing the fly the way out of the fly bottle.” The goal, he said, was not to resolve the questions posed by philosophers, but to escape them. As long as the fly is inside the bottle, understanding its contours is essential to getting it wherever it wants to go; but once the fly is outside, the shape of the bottle doesn’t matter at all.

Non-mainstream economists have a similar relationship to dominant theory. Because we’ve been inculcated for years that the best way to think about the economy is in terms of the exchange of goods by rational agents, criticisms of that framework are a necessary step on the way to thinking in other terms. But the logical and empirical shortcomings of thinking about economic life in terms of a perfectly rational representative agent optimizing utility over infinite future time don’t, in themselves, tell us how we should think instead.

The essentially negative character of economic heterodoxy is a special challenge for undergraduate teaching. You can’t teach criticisms of economic orthodoxy without first teaching the ideas to be criticized. Finding our way out of orthodoxy was, for many of us, central to our intellectual development. Naturally we want to reproduce that experience for our students. This leads to a style of teaching that amounts to putting the flies into the bottle so we can show them the way out. But how useful is it to our students to understand the defects of a logical system it would never have occurred to them to adopt in the first place? Having spent so much time looking for a way out, it sometimes seems we don’t know what do in the open air.

This dilemma is on full display in The Handbook of Pluralist Economics Education. In order to present a realistic model of the economy, Steve Keen writes in one of his two chapters, “an essential first is to demonstrate to students that the ostensibly well-developed and coherent traditional model is in fact an empty shell”. Many of the volume’s other contributors make similar claims. This is the spirit of Joan Robinson’s famous quip that the only reason to study economics is to avoid being fooled by economists. But if that is all we can offer, better to send our students to the departments of history, anthropology, engineering, or some other field that offers positive knowledge about social reality.

What then are we to do? Pluralism as such is not a useful guide; carried to an extreme it would, as Sheila Dow says here, amount to “anything goes,” which is not a viable basis for teaching a class (or for any other intellectual endeavor). This is a problem with pluralism as a positive value (and not only in economics teaching): Pluralism implies a number of distinct perspectives, but to be distinct they must be internally coherent, that is, unitary. Carried to an extreme, pluralism is self-undermining. To challenge the mainstream, at some point you must argue not just for the value of diversity in the abstract, but in favor of a particular alternative.

In practice, even economists who completely reject mainstream approaches in their own work often give them a large share of time in the classroom, in part because they feel obligated to prepare students for future academic work and in part, as Keen says, simply because of “the pressure to teach something”. Teaching is hard enough work even when you aren’t reconstructing the curriculum from the ground up. It’s much easier to teach a standard course and then add some critical material.

But pluralism in economics teaching doesn’t have to mean simply presenting orthodoxy and adding some criticisms of it. It could also mean approaching the material from a different angle that avoids — rather than attacks — the dominant formalisms in economics and gives students a useful set of tools for engaging with economic reality. For me, this means a focus on the definition and measurement of macroeconomic aggregates, and on the causal relationships between those aggregates. Concretely, it means reliance on flowcharts where the nodes are some observable variable, as opposed to the normal emphasis on diagrams representing functional relationships — ISLM, AS-AD, etc. — that can’t be directly observed.

A more specific problem in heterodox teaching — and heterodox economics in general — is the weight put on the financial crisis as an argument for alternatives to the mainstream. Many of the authors in this collection present the crisis of 2008 and its aftermath as a decisive refutation of economic orthodoxy. Edward Fullbrook declares that ‘no discipline has ever experienced systemic failure on the scale that economics has today.” David Wheat, less hyperbolically, argues that “the failure to foresee the financial epidemic in 2008” demonstrates a need to shift the focus of economics teaching away from long-run equilibrium. One might push back against this line of argument. It is true that several large financial institutions went bankrupt in 2008, and some financial assets fell steeply in value, to the dismay of their owners; but with the perspective of close to a decade, it’s less clear how much of a base these events offer for critique of either the economics profession or economic institutions. Singleminded focus on “the crisis” risks implying that the problem with our economic system is the rare occasions on which it fails to work well for owners of financial assets, while ignoring the ongoing problems of inequality, hierarchy and privilege; tedious and demeaning work; environmental degradation; and the fundamental disconnect between ever-increasing money wealth and unmet human needs – none of which has much to do with the failure of Lehman Brothers. As people used to say: capitalism is the crisis.

It is true, of course, that the economics profession failed to foresee or explain the 2008 crisis, but that’s nothing special. To make a list of phenomena unexplained by orthodox economics, just open the business pages of a newspaper. In any case, while it might have been reasonable at the time to expect some degree of self-criticism in the economics profession, and some increase in openness to alternatives, seven years later it is clear that there has not been. With a handful of exceptions – Naryana Kocherlakota is probably the most prominent in the US – mainstream economists have not revised their views in the light of the crisis; even those who were initially inclined to soul-searching have mostly decided that they were right all along. The case for heterodoxy must be made on other grounds.

19 thoughts on “Heterodoxy and the Fly-Bottle”

  1. One problem is that most heterodox economists have abandoned microeconomics, perhaps because they are convinced that microeconomics when combined with a representative agent cannot yield involuntary unemployment.

    I have demonstrated the exact oppposite. In a paper published last year I show that the right microfoundations must result in involuntary unemployment. The argument is too involved to go into here. But the paper can be read at http://www.paecon.net/PAEReview/issue80/George80.pdf (real-world economics review, June 2017)

  2. Is that Godley and Lavoie, “Monetary Economics An Integrated Approach to Credit, Money, Income, Production and Wealth” text book a suitable base for teaching economics from a heterodox perspective? Are there courses that do that?

    1. It really depends what you’re looking for. I believe that Godley and Lavoie is the main macroeconomics text in the Bard-Levy Institute MA program.

  3. Excellent post. It is easy to criticize but unless we offer alternative models we will not carry our students along. I’ve been struggling with this for a whole career.

    In upper level courses I have had good experiences using our Growth and Distribution text because it emphasizes alternative closures so students can see how different visions of economics lead to different causal ordering. The Keynesian and Classical traditions have well-articulated models for the very long run that students get. We (Duncan Foley and I) are in process of revising G&D (with the addition of Danielle Tavani), BTW.

    But that leaves how to teach basic intermediate macro. It would be nice if we had a workhorse model on the same level as the text 3-equation model, but with more attention to financial instability. As it is, I follow Carlin/Soskice and just paste that on.

    One thing I tried this term is adding hysteresis to the textbook model. Path dependence is perhaps the one thing that heterodox economists all agree upon? Adding it dissolves the ‘short-run keynesian, long-run neoclassical’ property of the text model–it becomes ‘long-run keynesian’ more or less. My best students definitely connect to this approach. (Your Roosevelt paper “What recovery” is very teachable here.)

    1. Thanks, Tom! I actually taught out of your book once, many years ago.

      I think Carlin & Soskice is pretty good if that’s the general approach you want. My own personal preference is to deemphasize formal models of the economy as a whole and to focus more on causal links between particular aggregates. That’s partly based on my judgement of what works for our students (who are less comfortable with math than yours, I would guess), partly because it lets me link the classroom material more closely to news stories and policy debates, and partly just because I prefer a more descriptive, historical approach to macro in general. But I can definitely imagine using Growth and Distribution in a different setting, maybe with our MA students.

      Anyway as you say the most important thing isn’t which specific approach we take but that we are giving our students a coherent set of tools to udnerstand the world and not just the standard stuff plus our complaints about it.

  4. I’ve thought about this a lot over the last two years and this has really helped make things clearer in my head. I think I agreed with your whole argument, and what seems to make most sense to me is for heterodox professors in this conundrum to just teach econometrics.

    I think it would solve nearly all the problems you mentioned.
    First, it would offer positive knowledge about social reality.
    Second, it would offer skills beneficial to the average student and to the few who will go on to study econ further in graduate school.
    Third, you could engage in pluralism by going over papers published in both mainstream and heterodox journals.
    And finally, and this might be the most important, you wouldn’t needlessly be putting students in the bottle. Mainstream undergrad econ is an absolute waste of time. When I took it, my heterodox advisor made sure to tell me “just to be safe, don’t believe any of it.” He was only slightly kidding and mostly right.

    We agree that traditional undergrad macro and micro offer no positive knowledge about social reality. So just don’t teach it. And if a student ever needs that knowledge in their future studies, then they are likely a particularly driven student that will pick up those concepts pretty quickly with no problem. Perhaps through an independent study or on Kahn academy.

    1. Mainstream undergrad econ is an absolute waste of time.

      One of the funny things about mainstream undergrad econ is that it is not even useful as preparation for mainstream graduate work. There really is this closed system of undergrad econ that exists only in the college classroom. (Tho to be fair there is some convergence of the undergrad curriculum with “real” economics, which I don’t like much better.) On the other hand, the good thing about it is that you can do something else without worrying that you are shortchanging the students who do want to go on to graduate school.

      1. Undergrad econ has an important function in social reproduction. Just look how appeals to econ 101 are used in politics.

      2. My intermediate macro made use of no calculus and my micro course made use of only basic calc 1. Meanwhile I was taking differential equations and intro to real analysis to prepare for graduate school. I had a feeling that both couldn’t be useful preparation

        1. My undergraduate (very) heterodox department didn’t have intermediate macro and micro for both these reasons (why put the flies in the bottle in the first place) and lack of student interest revealed in surveys. This has be thinking though, that they could just teach courses called intermediate macro and micro but whose content is all left-Keynesian, Marxist, Feminist, Developmental, or whatever heterodox school of choice the professor prefers. As you said, it wouldn’t even be short-changing the students who want to go to graduate school

          1. Wait, does heterodoxy just mean leftism?

            There are fields of economics outside “the right economics” today, like political economy, history of economics, transportation economics, institutionalism, spatial economics and economic history.

            Being heterodox doesn’t mean being leftist.

          2. Being heterodox doesn’t mean being leftist.

            I agree. But I also don’t think it just means being outside the mainstream in general. From where I am sitting, heterodoxy constitutes a distinct intellectual community. There is a set of journals, PhD programs, conferences, research institutes and social networks with sufficiently overlapping membership to constitute a distinct community. We can discuss what if anything the people in this community have in common intellectually or politically, but in a lot of ways it’s an ascriptive status like being in a club — you either are a memebr or you aren’t, independent of any other qualities you might have.

          3. as an empirical matter most heterodox economists are leftists and even the ones that are liberal are far to the left of the current democratic party. Their liberalism basically consists of lesser evilism and downplaying how right wing, say, the Obama administration was.

            calling “history of thought” or “political economy” or “economic history” et al “heterodox economics” is incorrect.

  5. Not an economist. One central issue with orthodox economics is that it uses large categories to talk about general welfare, but operationalises these in terms of money. Since welfare is about much more than money, and many aspects operate in ways that are antithetical to the processes and values embodied in monetary systems, this leads to a sort of contorted vagueness.

    An alternative might be to focus on monetary systems – what “moneyness” is, how it operates, what the key indicators and aggregates are, what are the key developments and variations historically (Merijn Knibbe does a good deal of this). So long as you kept reminding students that monetary processes are a key system, but not the only system, nor even the dominant system, they would see orthodox economics as just a peculiar way of treating the part as the whole?

  6. We have ended up in a very strange situation with economic education. JW put it well: “There really is this closed system of undergrad econ that exists only in the college classroom.” Econ 101 should be titled The Theology of Capitalism, because it’s the one college class dedicated to defending the status quo. I think the fact that economists teach this class explains the status the field has in society (or had before 2008).

    An odd consequence of this is heterodoxy ends up being “heterodoxy to undergrad econ”. I have a mainstream grad education in econ, and about half the time when I read a heterodox claim about what the mainstream believes, I think “…what? I have never believed that, or been told to believe that.” So I think JW has the right of it — ignore the undergraduate textbooks and concentrate on empirical questions.

  7. Hi — I just got around to this thread, so I might be talking only to myself…..

    Of course, figuring out how to teach undergrad econ with integrity (but not being a windbag) has been a preoccupation of mine since forever. In my current job I teach only intro econ, since the concept of prerequisites doesn’t exist.

    I wrote intro textbooks in part as a way out of my own particular dilemmas, but also to help myself think through what I was doing. The micro text was written first and is more jumbled: I had figured out one piece of the puzzle — present economics as a story about “them”, the tribe of economists — but not the practical side that Josh emphasizes. Students reading that book would not come away with useful knowledge they could apply to questions that interest them.

    The macro book was written later and is, for me at least, about 90% successful. (It lacks all the other stuff that adopters are looking for, like problem sets and ideas for workshops and projects, but I have all of that in my own files for my use.) I don’t think there really is any single model that sums up usable macro knowledge, but there’s a lot of very practical stuff about aggregates and how they tend to be related, again as Josh suggests. I focused on that area and emphasized actual history, including the history of policy. Obviously the book has not exactly caught fire, but it has done its job for my own students.

    The deep problem that still gnaws at me and lies beneath the conundrum Josh brings up is that many of us, myself among them, don’t see “economics” in the same functional role as the mainstream. We don’t think of the economy as the expression of a few primitive, timeless forces, so we can’t claim to explain everything in a model or two. Myself, I see the economy as an interlocking set of processes, sometimes reinforcing, sometimes pulling apart. Identifying them and seeing how they operate is certainly knowledge of an important sort, but it doesn’t permit the tidy exposition of conventional undergrad econ. I like to compare it to biology: we know a lot about, say, marine biology, but there’s no “general theory of fish”. The deepest question is how we reframe what economics is supposed to do when we are being asked to present and explain a pre-existing frame that doesn’t work and we don’t agree with. It’s not just the model but whole intellectual strategy.

    1. Not just talking to yourself – I read this with interest. As you know, I used your macro textbook for a couple of years, so I have a lot of time for your views on this stuff.

      I wrote intro textbooks in part as a way out of my own particular dilemmas, but also to help myself think through what I was doing.

      I think this is a common experience. I also think — maybe you agree — that a lot of people approach teaching, and writing textbooks and teaching materials, as a kind of service activity – as useful and maybe rewarding work but not a contribution to scholarship. WHereas I’ve come to feel that figuring out how to present material to students is actually an important part of coming to understand it yourself.

      We don’t think of the economy as the expression of a few primitive, timeless forces, so we can’t claim to explain everything in a model or two. Myself, I see the economy as an interlocking set of processes, sometimes reinforcing, sometimes pulling apart. Identifying them and seeing how they operate is certainly knowledge of an important sort, but it doesn’t permit the tidy exposition of conventional undergrad econ. I like to compare it to biology: we know a lot about, say, marine biology, but there’s no “general theory of fish”.

      I like this a lot. I will quote that fish line, for sure. But I think this doesn’t have to be such a problem at the undergraduate level since students aren’t necessarily expecting a general theory to begin with. A grab bag of heuristics and local causal relationships feels like a perfectly acceptable body of knowledge to them, usually, I think.

  8. I have a lot of sympathy for the Wittgensteinian perspective on economics. A lot of macroeconomics suffers from conceptual confusions that bias analysis. A superb example is highlighted in BIS working paper 525 by Borio and Disayat. They show completely convincingly how ignoring finance, and the analytical distinction between “saving” and “financing” in international macro theory, has led to a range of theoretical dead-ends and conundrums (such as the Lucas paradox and the Feldstein-horioka puzzle) that simply dissolve when the conceptual framework is richer.

    1. I like Borio and Disyatat a lot. As always, the hard part is to move from negative critique to substantive claims — it’s easier to say what is not true than what is.

Comments are closed.