Monday, April 28, 2014

Post-Crash economics clashes with 'econ tribe'


Manchester University’s Post-Crash Economics Society (PCES) released a report recently into the way economics is currently taught at Manchester, which is relevant to economics teaching at universities worldwide.

Their well articulated advocacy for pluralism in economics teaching, where neoclassical approaches are just one of many analytic techniques in the curriculum, has tapped into the current media disquiet with economics.

The involvement of the Bank of England’s new Chief Economist, Andrew Haldane, certainly helped, and the report has now found media coverage at the Financial Times, the Wall Street Journal blog, The Independent, and Paul Krugman’s New York Times blog.

In this article I first provide an overview of the PCES report in the context of the modern history of economics and the dissenting voices that came before them. Then I examine the tribal defences raised by the mainstream in response to it, which only serve to reinforce the urgency of the task of transforming economics into a socially valuable endeavour.

What is it all about?
The PCES report provides a concise overview of the systematically narrow and outdated teaching approach in economics, which unfortunately simply reflects the mainstream practice of economics. It highlights how the economics profession, and economic teaching, has seen a ‘great narrowing’ over the past two decades, all but redefining the discipline away from the study of economic phenomena, to the study of a particular family of equilibrium marginal models.

At a practical level the report offers guidance for improving teaching, and a wholeheartedly agree with the points made, which I summarise as follows (emphasis is quoted text)
  1. Economics education should begin with the study of economic problems, where economic phenomena are outlined and the student is given a toolkit and must evaluate the strengths and weaknesses of how different theories explain different phenomena.
  2. Introduce pluralism so that students understand that different models and theories can be applied or are most useful in different situations. Students should be able to consider a variety of theories before forming judgements. This is important because economic theory is not universally applicable and much depends much on institutional, historical and social contexts.
  3. Include the study of institutional power structures and politics. In doing so, students should be aware of the ethics of being an economist and a consideration of the ethical consequences of economic theory.
  4. Ensure that the philosophy of economics, or the more generally the philosophy of science, forms a core part of the curriculum. Student should be able to understand which assumptions are justified in a scientific theory and how rigorous must the ability to falsify a theory must be.
  5. Finally, provide students an understanding of the historical development of a particular model or economic paradigm in order to contextualise the approach and provide insight into the problems it was designed to solve and how context influenced its formation
The PCES report, and its proposed reforms, is itself part of a recurring theme in the history of economics. In 1973 Joan Robinson tried to revolutionise economics teaching, writing a textbook reflecting most of the reforms advocated by the PCES. Unfortunately that book is all but forgotten in economics. I certainly wish Robinson had been successful, as it would have deterred the next generation of top economists from merely repeating the nonsense that went before them.

In a strange historical coincidence, William Baumol complained of poor teaching of economics as far back as the late 1930s. At the time he found the teaching ignored what was then the cutting edge work of Keynes and Joan Robinson herself.

In 1968 it was the Union for Radical Political Economists that set about disrupting the discipline. By 1972 this movement had agitated so successfully that the then President of the American Economics Association John K Galbraith, a supporter of the movement, noted in his 1972 Presidential address to the Association that "For a new and notably articulate generation of economists a reference to neoclassical economics has become markedly pejorative". He went on to say
Neoclassical and neo-Keynesian economics, though providing unlimited opportunity for the demanding niceties of refinement, has a decisive flaw. It offers no useful handle for grasping the economic problems that now beset the modern society.
But nothing changed.

In 1992 the Foundation for European Economic Development (FEED) funded work that produced a Plea for a Pluralistic and Rigorous Economics, which was published in the American Economic Review.  Again, nothing.

More recently the Post-Austistic Economics movement emerged in Paris in 2000 with a similar mission to reform economics, followed closely by a proposal to open up economics by a group of PhDs at Cambridge in 2001.

In 2003 Harvard economics students organised a group called Students for Humane and Responsible Economics (SHARE), which launched a petition calling for a new approach to economic teaching and offered an alternative introductory course that was ultimately rejected by the university. At Manchester, the PCES has had their proposed course on Alternative Theories of the Crisis rejected by the university just a month ago.

In 2011 we also saw the EC10 class walk out on N. Gregory Mankiw’s lecture at Harvard in support of the Occupy protests, which in retrospect seems completely appropriate given Mankiw’s prostitution of economic ideas to defend the wealthiest 1%. Published at one of the leading economics journals, Mankiw’s ‘Defending the 1%’ article provides a classic example of the unscientific nature of the discipline’s ‘respectable’ journals, and the perverse anti-scientific but ideological incentive system embedded in the culture of economics.

The PCES is now one of many student societies, including for example, the German Pluralist Economics Network, justifiably calling for change in economics. Again the time appears ripe for disruption of economics, and the PCES offers useful guidance for how to achieve that.

The tribe responds
First, let me say that ‘econ tribe’, the core of the economics profession, is truly a tribe - a clique, a social construct - rather than a scientific community. As senior econ tribe member once gave me the following advice.
...the group ‘elders’ have implicit theories in their heads as to how the world works, what forms of behaviour are appropriate, what beliefs are reasonable, what questions are truly yet unanswered, what articles of faith should not be doubted, and where they and their group ultimately make their money
Oh, and that’s just the start (added emphasis).
…many economists like to pretend that even if they give practical policy advice, they as economists make no value judgment on what people want, i.e. ‘preferences are taken to be stable and can differ over all individuals’.

Such statements are then implicitly ‘backed up’ by theoretical models of decision making arising from underlying preference maps (i.e. a ‘deeper layer’).

Now, as a statement of reality this is a completely absurd and purely religious position since no-one even comes close to having ever measured a preference map despite the feeble protestations of the revealed-preference crowd, but as a defensive shield against the criticisms of psychologists and mathematically-challenged economists it is very effective. It drags criticisers into a morass few will escape from and that no outsider would be able to follow anyway, making it sound like a good riposte.

Nearly all sub-fields of economics have similar defensive religious beliefs used to deter entry by unwanted outsiders
A give-away signal of a tribe member is that they will deny the religious nature of their tribe, and pledge loyalty wherever possible by defending the tribe against all challenges. This is the sell-out that most successful economists make at some point in their career, something Yanis Varoufakis has written extensively about.

We see evidence of overt religious zeal and loyalty pledges in the mainstream responses to the PCES report. Here are the links so you can read them for yourselves - Wren-Lewis, Yates, Farmer, Krugman. Expect more in the coming weeks.

The blueprint response was that economics teaching could be better, but economics is fine. I couldn’t disagree more. This is a mere loyalty pledge.

Poor teaching reflects the poor scholarship of the discipline more broadly. This is most clearly seen in Ferrero and Taylor’s survey of economists, where 78% of professional economists (including professors) incorrectly answered a multiple-choice question about the definition of opportunity cost. The profession subsequently kicked into gear to defend itself from embarrassment, with one article in a ‘respectable’ journal making the Alice in Wonderland response that any Humpty can use opportunity cost to mean just what they choose it to mean - “neither more nor less!”

But the big defensive play to combat the PCES threat was to assert that the ‘sciency-ness’ of the mainstream is the reason for its dominance, and the reason we should respect it. Again, deny the religious aspects of economics, be loyal to the tribe.

From Wren-Lewis.
…economics is a science. Its response to data and events may be slow compared to the normal sciences, for obvious reasons, but it is progressive
From Farmer
Economics is a science…the core approach to modern economics, both macro and micro, is the culmination of a process of intellectual argument in which ideas have been sifted, debated and compared with facts. Some have survived; others have not.
Farmer’s claim is not at all true. Most of the modern macro debates were already happening in the 1800s.

It is scary to think that these economists believe what they doing is science. There are parts of economists that do take those crucial scientific steps of prediction and experimentation, and the results usually contradict mainstream theory - conflict is real, costly punishment is real, social norms can evolve and change, preferences are not well behaved, markets boom and bust, and more. Yet for some reason the preacher’s models seem immune to evidence.

A comment from Yates’ post points to the non-scientific approach deemed acceptable in economics.
If economists wished to study the horse, they wouldn’t go and look at the horses. They’d sit in their studies and say to themselves, “What would I do if I were a horse?
Physicist-turned-sociologist Duncan Watts has warned about this approach
…social scientists, like everyone else, participate in social life and so feel as if they can understand why people do what they do simply by thinking about it. It is not surprising, therefore, that many social scientific explanations suffer from the same weaknesses—ex post facto assertions of rationality, representative individuals, special people, and correlation substituting for causation—that pervade our commonsense explanations as well. 
Krugman and Yates take the whole ‘sciency-ness’ one step further, pretending that economists do in fact know what’s going on in the economy, in banking, in financial crises, and in recessions. Or something.

Steve Keen fired off a response to Krugman, arguing that he contorted the failure of economics to foresee even the remote possibility of a debt-fuelled crisis, into a mere failure to keep an eye on shadow banking - as if better data on shadow banking would have somehow led to the whole econ tribe dropping their core money-less models of the economy in equilibrium to embrace theoretical approaches of outsiders.

A second defensive play, seen in Yates and Farmer’s responses, is that within the mainstream there is a model for everything, so no need to look outside. Again, no reference to the success of any of the models in predicting economic phenomena. It is as if the high priests response to evidence is to merely create a new god - “Ah yes, haven’t you seen, we have a Tornado God now. Solved that didn’t we” - and ignore reality.

In the past Krugman has actually made similar points. Mainstream economics ignored increasing returns to scale as a reason for trade, even though it had long been understood by ‘outsiders’. To address this he created the Increasing Returns God for the tribe.
I was, of course, only saying something that critics of conventional theory had been saying for decades. Yet my point was not part of the mainstream of international economics. Why? Because it had never been expressed in nice models. 
Which only serves to highlight the invalid non-scientific approach - ideas are not judged on the merit as predictive tools, but on whether they can be expressed in the pre-determined modelling methods of the tribe as nice totems of the modelling gods.

As I have said, the value of a scientific model is its ability to predict outcomes in new situations. In economics, the reverse is true. Most economists believe that models provide insights simply by their construction, whereas in reality we can never know if the model is useful unless it offers useful predictions.

It is telling that we see no arguments based on scientific principles in these responses. Nothing on the practical usefulness of mainstream approaches above others. No long lists of scientific battles won by the mainstream because of the useful predictive power of their approach, and plenty of silence on the battles they lost (eg. the capital debates).

That is to be expected from the tribe’s high priests. But it also shows a distinct ignorance of alternative approaches. If there was a wide understanding of alternatives a response comparing the merit of different approaches could be made. The reason it isn’t made is because once economists learn of heterodox approaches the usually leave the tribe. For example, I can’t imagine that any macroeconomist with a decent understanding of Godley and Lavoie’s integrated monetary models would really continue to believe that the mainstream ‘label the residual’ approach to a money-less equilibrium is more valid.

The future
I wish the PCES the best of luck. It is possible for economics to be a useful science, but that requires a much more humble discipline that is able to judge theoretical and methodological approaches based on proper evidence. It starts by acknowledging that in a scientific sense, the body of knowledge in economics is extremely limited, and much of what passes for economics is a collection of mere credible stories about certain phenomena, which are yet to be tested.

I hope indeed that the observation that Pilkington makes is the correct one, and that the time is ripe for transformation in economics.
…silently, behind the scenes, the heterodox have been winning the debate. Krugman would never admit this, of course — indeed, he seems to live in a bit of an echo chamber over at the NYT with all his fanboys and fangirls and he may not even be fully cognizant of it — but if you move in economics circles you’ll know this to be generally recognised.
But let me finish with more sober words from James K Galbraith, because I believe transforming economics is going to be one hell of a battle of tribal politics.
Leading active members of today’s economics profession... have formed themselves into a kind of Politburo for correct economic thinking.  
As a general rule—as one might generally expect from a gentleman’s club—this has placed them on the wrong side of every important policy issue, and not just recently but for decades.

They predict disaster where none occurs. They deny the possibility of events that then happen. ... They oppose the most basic, decent and sensible reforms, while offering placebos instead.

They are always surprised when something untoward (like a recession) actually occurs. And when finally they sense that some position cannot be sustained, they do not reexamine their ideas. They do not consider the possibility of a flaw in logic or theory. Rather, they simply change the subject.

No one loses face, in this club, for having been wrong. No one is dis-invited from presenting papers at later annual meetings. And still less is anyone from the outside invited in.

12 comments:

  1. Great post. A minor quibble... you left out some reform attempts, such as the 1992 Plea in the AER and the attempts by Rethinking Economics to change the UK subject benchmark.

    ReplyDelete
  2. Great! I will add your other campaign references to Wikipedia... thanks for those: http://en.m.wikipedia.org/wiki/Pluralism_in_economics

    ReplyDelete
  3. Wonderful economical report that's give me a clear concept of understanding. Thanks for nice summit . exceptional student education We support to not only parents and children dealing with special educational needs, but also support for schools and businesses for programs, assistance etc.

    ReplyDelete
  4. I wonder whose stupid and obnoxiously offensive idea it was to name a movement "post-autistic"? At least the "post-autistic economics review" has had the decency to abandon the name in favor of "real-world economics review."

    ReplyDelete
  5. Your next step should be to imagine how economics should work in order to be useful. The problem isn't modeling in and of itself. Attacking them is just classic "both sides do it". The problem is that there has been a lack of a mechanism for throwing out bad models and bringing in new ones. Chemistry couldn't have developed without the alchemists exploring new reactions, just as physics couldn't have developed without generating bogus models. (Some of those bogus models, like Newton's Theory of Gravitation are still in common use, and the much vaunted Standard Model is accepted as excellent, but demonstrably wrong.)

    One of the big puzzles to me is how poorly mainstream economic theory lines up with mainstream accounting. One might expect that for any economic model it would be possible to derive a set of accounting reports for each various entity that would give one insight into that entity's motivations and probably behavior. For example, if someone has no money, they will either have to borrow or not spend. In fact, most mainstream economic theories, as expounded in the press, Heritage Foundation reports, and in political speeches, do not allow this. I think this is why the more naive and less cynical among us tend to think of economists as autistic. Just as autistic people have trouble with handling social nuance or recognizing the danger of drowning or fire, economists seem to be strangely out of touch with everyday phenomena.

    When physicists finally manage to fix their broken Standard Model, they are still going to have to give us protons, neutrons and electrons so we can do chemistry. Economists seem to work under no similar constraint.

    ReplyDelete
    Replies
    1. Excellent comment. I agree the problem is with the filtering of bad models - the testing and prediction stages. I am drafting up a new post on how we could approach research using models in a more scientific manner.

      Delete
  6. 3. Heterodox tactics, and the self-reinforcing behaviour of the mainstream, both aside, there's an open question as to how good heterodox economics is as a replacement for mainstream economics. (If we're lumping "mainstream economics" into a category, I reserve the right to do so here for heterodoxy.) It's at least possible that the heterodox economics community has even poorer quality control and filtering processes than the mainstream.

    I have not personally invested lots of time and energy in exploring standard strains of heterodox thought, literature etc., and so it would be improper of me to be dismissive, but once struggle I've had is that I've encountered a distinct lack of reverse-engineerability in heterodox works. I should be able to read a paper in economics and pull it apart, so to speak, to see what makes it tick. I've tried a few times to read heterodox papers only to react at times going "where did THAT come from?"

    I don't think this is just me. Let's use Steve Keen as an example, since he's a major figure and you mention him.
    - I can't figure out Steve's work. It's likely that I am not clever enough, in which case I yield to;
    - JW Mason (heterodox-ish) and Nick Rowe (mainstream) who have both, as best I can tell, made good-faith efforts to understand Keen, and struggled.
    - David Stern and John Quiggin, neither of whom are right-wing ratbags, reviewed Debunking, and found serious analytical flaws. (Are they wrong?)

    (Again, not dismissing "heterodoxy", just posing the question of how and when we know it is superior? I'm sure some of it will be important into the future, but how does a pluralist approach deal with the comparisons?)


    So, summing up, I like and admire the principles set out above, but I don't like the tribal binary. The mainstream is a broader church than ever, and it isn't going to just die. And part of Krugman's push via his blog is based on a notion that the mainstream is effectively at war with itself; that a non-trivial part of the profession is politically tribal at its core, over and above any methodological tribalism. And that this part rejects (according to Krugman and others) the most basic element of scientific reasoning; changing your mind when the facts don't fit your argument. So there isn't one single Politburo of correct economic arguments or models.

    A side-note: the Political Economy group at Uni of Sydney has had decades as a self-standing department. Impressions from outside are that it has developed and delivered a very effective pluralist teaching program, with lots of successful grads going into lots of professional roles. It does not seem to have had parallel success developing an impactful research agenda that sees it ready to supplant the mainstream.

    (Part 2 of 2)

    ReplyDelete
    Replies
    1. Great points Michael. Let me respond with my views.

      1. You seem to be agreeing here that starting with observation is key (though there are exceptions). Economists usually do this when they motivate their writing with some empirical story or stylised fact, so clearly they realise that without some grounding in empirics, theorising gets you nowhere.

      2. I agree that the heterodox tribes mostly now define themselves as being outside the mainstream and have their own tribal issues. That certainly frustrates me no end. And it does nothing to aid communications between tribes. I've found that the mainstream is defined by their method, and hence they find it difficult to even absorb explanations of new approaches without adding in their implicit assumptions.

      Here's an example of a mainstream guy trying to get his head around agent-based models http://orderstatistic.wordpress.com/2014/08/12/are-agent-based-models-the-future-of-macroeconomics/

      In doing so he either 1. ignores the implicit assumption in his own methods, or 2. assumes that the same implicit assumptions are in ABMs.

      The heterodox guys should be trying to better communicate with the mainstream, but often the mainstream don't even know the assumptions they are bringing along with their models.

      Delete
    2. 3. Keen is a good example of this. His arguments about the inconsistency of perfect competition at the firm and market level are obviously true. And talking to some old hands in economics, it's been a debate for half a century - if all firms are price takers, who sets the price?

      Well read mainstream guys know it's bogus. So why then is there so much crap thrown around when Keen points this out. Sure, he might have not defined some of his maths in the identical way to others, but the basic point holds, and is well known.

      Quiggin's review is basically supportive from my read. Pointing out that many of Keen's critiques are well-known, and I think even here Quiggin does the mainstream thing of saying Keen is plain wrong on the welfare theorems, when (without reading the book again), it seems to be Quiggin who is bringing implied assumptions to the table here and I think it technically incorrect.

      My link to economists not even knowing what opportunity cost is really bring out the point that most mainstream guys don't know what implicit assumptions are in their models.

      Yes, the quality control is poor in some cases, but I've read a lot of rubbish in the AER as well, so I can't be certain here.

      To me the issues of heterodox v mainstream come down to language and communication. The mainstream, in my view, is not well-trained in the economics of the mathematics (if you get my point). So when an outsider questions the economics, the get a response in terms of the mathematics.

      As far as the mainstream being a broader church than ever I don't really see it. Sure, the big journals publish some interesting new modelling techniques and approaches from time to time, but that never seem to go anywhere.

      Here's another example the mainstream not really even knowing it's own models (hence the problem communicating with outsiders). I've submitted an article offering a new theory of the firm based on uncertainty and dynamic choices of capital. In essence I want to generalise the decision of firms so as to be consistent with capital-budgeting literature, the empirical record on economies of scale, and the nature of irreversibility and uncertainty.

      http://www.uq.edu.au/economics/abstract/497.pdf

      I sent draft of this to a bunch of professors, and now to a number of mainstream journals. I've had responses (including from editors) that range from "There is nothing new here, this is simply restating the profit maximising model" all the way to "This is completely wrong and arbitrary and doesn't make any sense"

      It can't be both.

      You can see why then I view the mainstream more as a social tribe than an alignment of social scientists interested in economics.

      Finally, yes of course there is argument in the mainstream. This merely goes to show what I have been saying - that they aren't united by the scientific endeavour, but by their belief in a method, regardless of whether or not they agree how to apply it in the real world.

      Delete
  7. Cameron, I've just realised I'm responding to a months-old post (I THOUGHT I'd heard about PCES stuff some time back!). But I'm glad to have the chance to.

    By the way, stick at it. I don't know if this'll make you feel better or not, and I forget the guy's name, but a prominent early contributor to behavioural economics (who I think worked with Richard Thaler at some stage) once described how he used to encounter referee reports that were either "That can't be right!" or "That's trivial, we all know that." I don't know whether in his case they were more sequential, or simultaneous like yours, but it's not an unknown problem, and pioneers in any field (assuming your paper is truly pioneering!) are not always met with generous acceptance (see Gans and Shepherd).

    But, while on behavioural economics, let's see how "the mainstream" has morphed. It's now far more empirical, because of both far more extensive data sets and far greater processing power. Levitt was someone who made (or helped make) interesting/fancy empirical analyses particularly acceptable at top places and in top journals (more on these later). There's behavioural economics using experimental methods. There's more use of random controlled trials (albeit with some pushback from the likes of Deaton). Although agent-based modelling is still troubling the macro crowd (although Peter Howitt's doing it now), it's been somewhat acceptable in environmental economics for some time. People are asking questions they didn't use to, and using methods they didn't use to, and there's a more catholic attitude to methods generally.

    That's, sort of, some of the good news coming out of the mainstream.

    There's another, grimmer, aspect of the mainstream that gets less attention than "saltwater/freshwater" -- it's the extent to which the Big Schools dominate the Big Journals and to which the Big Journals dominate the professional discourse. (Saltwater/freshwater sits alongside of all that.) In this way, the "Politburo" (assuming for argument's sake it's singular) not only decides what topics and methods are acceptable, but WHO matters and how. We saw this when Prescott was saying that only people "in the 3rd tier" institutions might have supported Obama's stimulus package. The "we're the properly smart ones" credentialling at the top tier places brings new meaning to the words "trickle down". ("We're the properly smart ones, doing the properly smart work, and you're either following what we do, or you're not in the game.")

    As a current example, there's a working paper circulating which is likely to come out in JE Lit or similar, prepared by two authors, one at MIT. It's on the "new exciting field" of "enviro-dev-economics", the intersection between the environment and development. Uncited and unacknowledged is any of the work done in the several decades of the existence of the journal Environment and Development Economics, which set out to tackle exactly that overlap, and has acted particularly to encourage submissions from developing countries.

    (Another example of faddishness with different implications: the attraction of doing RCTs in developing countries means that leading schools now want the fanciest experimentalists to join their faculty when a development job is open. This means the broader thinking, policy-focussed development specialists are crowded out in favour of those who can write novel RCT papers for top journals.)

    ReplyDelete
  8. (Part 2)


    Now, we know pretty much that top schools control top journals, and that chances of acceptance to Top Journal A is higher if you're top-tier-connected than if you're not. (Krugman can get a paper that name-checks Minsky into the QJE because he's Krugman.)

    So you have a Piketty-style endogenously growing discrepancy between what the Top Journals are setting as the standards of mainstream economic analysis and discourse, and what's happening out in the field journals and at the (ahem) third-tier institutions.

    I suspect you'll agree with much of the above, and it supports what you are saying pretty much, although to me it also underscores that what is "mainstream" kind of depends upon context.

    As for "of course there is argument in the mainstream", I would be concerned if there wasn't! All disciplines have disputes. That's how it ought to be. What I would say (and maybe what you're getting at) is that what matters is how these disputes get settled, and who (and what ideas) ends up prevailing. The things described above lead to the conclusion that the profession is set up in ways that enable a restricted set of methods and approaches to dominate the upper-level of the mainstream; and that novelty is only recognised when they discover something for themselves, regardless of what's been happening out in the further reaches of the profession/discipline.

    Anyway, it's been fun. I have more things to add but I've spent long enough at this and have a lecture to prepare!

    ReplyDelete