18 Signs Economists Haven’t the Foggiest

I’d like to thank Chris Auld for giving me a format for outlining the major reasons why economists can be completely out of touch with their public image, as well as how they should do “science”, and why their discipline is so ripe for criticism (most of which they are unaware of). So, here are 18 common failings I encounter time and time again in my discussions with mainstream economists:

1. They defer to the idea that “all models are simplifications” as if this somehow creates a fireguard against any criticism of methodology, internal inconsistency or empirical relevance.

2. They argue that the financial crisis is irrelevant to their discipline (bonus: also that predicting such events is impossible).

3. They think that behavioural, new institutional and even ‘Keynesian’ economics show the discipline is pluralistic, not neoclassical.

4. They think that the fact most economic papers are “empirical” shows economists are engaging in the scientific method.

5. They think ‘neoclassical economics‘ doesn’t exist and is just a swear word used by their opponents.

6. When pushed, they collapse their theories and assumptions into ridiculously weak, virtually unfalsifiable claims (such as revealed preference, the Efficient Markets Hypothesis, or rationality).

7. They dismiss ideas from the past or comprehensive study of previous thinkers and texts as “not science”.

8. They think positive and normative economics are 100% separable, and their discipline is “value free“.

9. They simply cannot think of any other approach to ‘economics’ than theirs.

10. They believe in an erroneous history that sits well with their pet theories, such as the myths of barter and free trade.

11. They think that microfoundations are a necessary and sufficient modelling technique for dealing with the Lucas Critique.

12. They think economics is separable from politics, and that the political role and application of economic ideas in the real world is irrelevant for academic discussion (examples: Friedman and Pinochet, central bank independence).

13. They think their discipline is going through a calm, fruitful period (based on their self-absorbed bubble).

14. They think that endorsing cap & trade or carbon taxes is “dealing with the environment”.

15. They think making an unrealistic model consistent with one or two observed phenomena makes it sound or worthwhile (DSGE and other models are characterised by this “frictions” approach).

16. They think their discipline is an adequate, even superior, method for analysing problems in other social sciences such as politics, history and sociology.

17. They think that the world behaves as if their assumptions are true (or close enough).

18. They think that their discipline’s use of mathematics shows that it is “rigorous” and scientific.

Every above link that is not written by an economist is recommended. Furthermore, here are some related recommendations: seven principles for arguing with economists; my FAQ for mainstream economists; I Could Be Arguing In My Spare Time (footnotes!); What’s Wrong With Economics? Also try both mine and Matthjus Krul’s posts on how not to criticise neoclassical economics. As I say to Auld in the comments, I actually agree with some of his points about the mistakes critics make. But I think these critics are still criticising economics for good reasons, and that economists need to improve on the above if they want anyone other than each other to continue taking them seriously.

PS If you think I haven’t backed up any of my claims about what economists say, try cross referencing, as some of the links fall into more than one trap. Also follow through to who I’m criticising in the links to my previous posts. And no, I don’t think all economists believe everything here. However, I do think many economists believe some combination of these things.

Addendum: I have received predictable complaints that my examples are straw men, or at least uncommon. Obviously I provided links for each specific claim – if you’d like to charge that said link is not relevant, please explain why, and if you want more, I’m happy to provide them. However, my general claim is simply that a given article trying to expound or defend mainstream economics will commit a handful of these errors, perhaps excluding the more specific ones such as history or carbon taxes. Here are some examples to show how pervasive this mindset is:

Auld’s original article commits 2, 3, 4, 5 & 12.

This recent, popular defense of economics as a science in the NYT commits 2, 4, 8 & 13 (NB: I forgot “makes annoying and inappropriate comparisons to other sciences”, although both sides do this).

Greg Mankiw’s response to the econ101 walkout commits 8, 9, 12 & 13.

This recent ‘critique‘ of Debunking Economics commits 9, 11, 15 (though, to its credit, it avoids 2).

Stephen Williamson manages 2, 6, 7, 8, 9, 11, 12, 13, 15 & 16 in his reviews of John Quiggin’s Zombie Economics (in fact, Williamson is a fantastic source of this stuff in general).

Paul Krugman committed 1, 7, 9 & 15 in his debate with Steve Keen

Dani Rodrik, who is probably the most reasonable mainstream economist in the world, committed 3, 4, 13 & 15 in his discussion of economics.

and so on…

(Note that, in the interest of fairness, I have left out the most ridiculous things I’ve seen since the crisis.)



  1. #1 by Ramanan on October 24, 2013 - 1:01 am

    Economists think that economics is the most trivial subject!

    ” …economics is the most trivial subject in the world …” – Milton Friedman. around 25:55

    and goes on to make mistakes. Hilarious Q&A starting 25:16

    • #2 by Cameron Murray (@Rumplestatskin) on October 24, 2013 - 9:29 am

      That’s very revealing about the ‘economic way of thinking’. Actually the evidence is that there is great stability of retail prices, like for shoes, and this is quite a challenge to economic theory.

      • #3 by Unlearningecon on October 24, 2013 - 9:04 pm


      • #4 by NicTheNZer on October 25, 2013 - 8:41 pm

        This is an amazing video. Was the audience here genuine lay-people? Or were some questions prompted by outside economists?

        One question points out that the inflation at the time was cost push, so this was clearly not a marginal theory. Another points out the problem with monetarist theory, endogenous money.

  2. #5 by First Financial Insights on October 24, 2013 - 1:38 am

    Where did it all begin? Why are they all wrong? What is wrong?

    Going back to the start of the industrial revolution in Britain, the seeds for the modern neo-classical economic thesis were planted by a Scottish philosopher and professor, named Adam Smith, in his historic inquiry and book, “The Wealth of Nations.” That’s where the problem started.

    First. economics as therein defined, concerns itself with allocating scarce resources; seeking to achieve an optimal production and consumption of goods and services, algebra. It is in the restricted context and definition of “production and consumption” where a quantum flaw occurs. It is a linguistic flaw too. (i.e. extraction is not production)

    What Smith and followers have advocated, was that extraction constructs or elements; particularly those relating to non-renewable resources, were equal or equivalent to production constructs. They clearly are not, as any geologist will confirm. Such elements are in fact produced by the planet over periods of millions of years, then extracted for human production, that coverts these elements into consumable items So to combine extraction and production constructs, creates a false view of reality’s physical flow of elements. Smith’s thesis, therefore, would have been more realistically accurate had he delineated this founding neo-classical economic thesis into three distinct constructs – separating extraction constructs from production and consumption constructs, in the overall model.

    Another major flaw in Smith’s thesis, is that the growth generated by free markets, will through its natural forces seek an equilibrium, thereby solving any immediate imbalances of an economy’s production and consumption algebra. Meaning, that the perpetual pursuit of growth is the single remedy needed to forever correct imbalances, and hence return an economy to its desired steady state. This is existentially, scientifically and logically absurd, because Smith ignores the finite constraints imposed by the extraction constructs.This flaw sets the stage for an inevitable collision between his abstract reality and existential reality – leading to a monumental collapse of Smith’s abstract version.

    While there are many other similar flaws in Smith’s thesis, these two stand out because they are the most damaging and pervasive to the human condition and its prospects to sustain any possible form of activity that aims to extend human longevity – thus avoiding premature extinction. Divorced from the hard constraint of reality, physics and mathematics; turns Smith’s thesis into a manifest of gibberish or likewise, utter poppycock. Yet to this day, all the leading economic textbooks, schools, professors and Noble Prize winners continue to premise their ideas, narrative and recommendations on these utter falsehoods that are disconnected clearly and logically from what is existential reality. Meanwhile, the end result of these flaws on human enterprise is sure to be catastrophic in some form.

    Why are we so disconnected? There are many reasons, some are distractions, but the ultimate causes must trace back to shortcomings of humanity’s neurological physics, affecting key social and psychological processes.We are neither individually nor collectively quite as smart as we think. From an evolutionary view, humanity’s cognitive functions did not keep pace with the responsibility entrusted and skills required for the extended management of the planet. The results and status of which are outlined clearly below in Brian McGavin’s featured abstract,” Sustainable Planet? The Silent Crisis”. Telling us, that we are in a heap of trouble, facing extinction or a serious transformation, if we do not get our act together soon to deal with this silent and invisible reality.

    To do so, economists and economics must first begin to embrace extraction as the third component of the overall model. And also, integrate their thesis with the laws and principles of objective mathematics and science – and most particularly; physics and exponential mathematics, and the hard constraints they impose on the extraction constructs that define human-activity’s limits.

    If we fail because of the neurological shortcomings, then the Darwinian outcome will reflect exactly how evolution destines our journey – joining the thousands of other species who had once inhabited this planet. But, if we fail because we continue to believe the dead wrong thinking of economists, lacking the will to correct their false and absurd thesis – no words can describe the magnitude of the tragedy, nor the loss of possible human opportunity. NO words.

    Moreover, there are no longer any excuses not to change the operative and logically-flawed economic thesis, as its structural shortcomings are self-evident and they are corroborated by a preponderance of irrefutable physical, scientific and mathematical evidence. There are no longer any excuses. There are no other sensible choices.

    Let us, just however hope. Let us, just however hope, that it is not too little – too late? That last question, is now for you to decide and – TO ACT UPON!

    For the full text click : http://drkinesa.blogspot.ca/2012/12/all-economists-are-wrong-dead-wrong.html?spref=bl

  3. #6 by Anonym on October 24, 2013 - 5:24 am

    A quick hover over your links reveals that almost all of them link to yourself, the post autistic economic review or other heterodox sources. If you want to try and critique the economic literature, maybe try and interact with, and read, some actual economics at some point.

    It seems a lot like you are talking within your (post) autistic bubble.

    • #7 by Unlearningecon on October 24, 2013 - 7:32 am

      A more detailed look, as well as following the advice at the bottom and looking at what was being criticised in said pieces, would reveal that I did link to things said by economists explicitly.

      And I am not associated with PAE.

  4. #8 by Luis Enrique on October 24, 2013 - 10:56 am

    oh you are a troll. This is a bit like the two step of terrific triviality. It could be that 0.1 per cent of mainstream economists are guilty of these things (those that are not just pure nonsense) which would make them no different with regard to their propensity to think silly things than climate scientists, biologists and post-Keynesian economists – of course you want to give the impression that some large proportion of mainstream economists are guilty of these things.

    1. It is a firewall against certain silly criticisms. It certainly is not guard against any of those criticisms, and that paper you link to does not claim that.

    2. countless conferences, journal special issues etc. say otherwise

    oh god I can’t be arsed.

    • #9 by Luis Enrique on October 24, 2013 - 1:22 pm

      you could have written “18 bad defences of economics”, which is how the generous Chris Dillow described this post, without making any claim that these defences are particularly widespread. But you’re doing more than that, you’re trying to claim the discipline has endemic problems, your post differs to Chris Auld, you have not written “here are 18 signs you are reading a bad defence of economics”

      • #10 by Roman P. on October 24, 2013 - 3:04 pm


        I think it is reasonable to note that all scientific disciplines have endemic problems, and some are probably more problematic than the others (political history has more problems than non-organic chemistry). It is hard to argue that economics is more problematic than a lot of sciences like geology or cryptography. Why be so exasperated with Unlearningecon then? The problems of economics are very real. A lot of good results remain in obscurity and bad stuff is still taught and published. SMD results deserved merely a vague reference in one of the best graduate-level textbooks, for example (Kreps).

    • #11 by newecon on October 24, 2013 - 4:38 pm

      Yes, saying something coherent is hard work, I’m not surprised you can’t be arsed. According to your comment, for example, 99.9% of economists would disagree with these statements. Really? Would they disagree that “their discipline is an adequate, even superior, method for analysing problems in other social sciences”? That will sure eliminate some grant applications. And 99.9% of economists would say that financial crashes are predictable? Does that include Eugene Fama?

      • #12 by Luis Enrique on October 24, 2013 - 4:44 pm

        learn to read.

    • #13 by Luis Enrique on October 24, 2013 - 4:43 pm


      one second thought, you are of course responding in kind – sort of – to Auld trolling people of your stripe.

      • #14 by newecon on October 24, 2013 - 5:43 pm

        Luis, you are responding to these points in the same dismissive and petulant manner as Chris Auld. You claim that only “0.1 per cent of mainstream economists are guilty of these things”. I’m afraid the problem is a little bigger than that.

      • #15 by Unlearningecon on October 24, 2013 - 8:14 pm

        Indeed, it is a counter-troll.

        I could have named it bad defences, true, but as I say in my new addendum, I am actually making a general claim.

  5. #16 by Mike Hall on October 24, 2013 - 12:21 pm

    I’m not a formally trained economist, just a pissed off citizen with time on his hands (yeah, for the obvious reason…)

    Just left a comment over at ChrisAuld’s that will doubtless be ‘moderated’ – ie not published. So, perhaps you’ll let it stand here, as a comment on behalf of the poor masses who have to suffer what people like Chris do?

    Here it is:-

    “I think Chris takes smugness to a whole new level.

    For the most part people using the terms above [in ChrisAuld’s blog] are usually talking in a political economy sense & terms like ‘neoliberal’ are completely valid.

    I think for any, yes, ‘mainstream’ macro economist to imply that their profession has not f**d up in an unprecedented fashion & writes such a pathetic attempt to deflect criticism is either callously self serving, blind or both.

    And it is patently obvious that heterodox schools like the post keynesians have been, and continue to be, shut out of mainstream discourse wherever possible. What passes for much mainstream ‘economics’, not just in the well funded ‘think tanks’, but academe as well, is completely politically infested in – you guessed it – the interests of established wealth.

    For example, the Euro shared currency was criticised heavily by heterodox economists before it was introduced & is clearly failing in precisely the way they said it would. (Failing for the masses of course, not the top few or those running the ‘financialised’ economy.)

    Five years of mass unemployment later & is the mainstream even acknowledging the flaws in their thinking? No, except perhaps in barely whispered tones, while, yes, the ‘neoliberal’ mantra of ‘austerity’, ‘structural reform’ & all the other coded terms for screw the poor & reduce public services continue to cause devastation & increasing social unrest, including the rise of vicious far right political parties.

    But, of course, we’re not all in this together are we Chris? You and your mates in the top few percent are doing just fine & even enjoying the asset ‘firesales’.

    So, pardon me, for my ‘bad critiques’, but I do now know what a ‘fiat’ free floating currency is, and that twats like you flatly refuse +any+ serious academic or other engagement with people who advocate how it could be used to counter the waste & deprivation of chronic unemployment & falling real wages for ordinary citizens.”

    • #17 by Sara on October 25, 2013 - 2:42 am

      It’s just what mainstream economics needs – another “filter” to avoid criticism. Alternatively, they could just put their hands over their ears and hum loudly.

      • #18 by Unlearningecon on October 25, 2013 - 2:21 pm

        I’m actually quite happy for economists to stick their fingers in their ears. The more they do it, the more everyone else will realise the discipline is not responding to the demands placed on it by its own failings, and will disregard it. That society linked to by gastro george demonstrates this point nicely.

  6. #19 by newecon on October 24, 2013 - 2:39 pm

    Chris Auld is the self-appointed defender of neoclassical economics:

    • #20 by Luis Enrique on October 24, 2013 - 2:48 pm

      “self-appointed”? as in Unlearning Economics is a self-appointed critic of neoclassical economics?

      is there a hiring committee somewhere I don’t know about?

      • #21 by B.L. Zebub on October 24, 2013 - 6:58 pm

        “self-appointed”? as in Unlearning Economics is a self-appointed critic of neoclassical economics?

        No, self-appointed as in “Luis Enrique is the self-appointed Chris Auld suck-up” 🙂

    • #22 by anon on October 26, 2013 - 5:23 am

      have you read the things Auld was talking about in those cases? if you really want to make some kind of intelligent criticism, you have to draw a line between intelligent and non-intelligent criticism, and if you defend David “externality means ‘we don’t give a shit'” Suzuki, you’ve drawn the line in the wrong place.

      • #23 by Unlearningecon on October 26, 2013 - 8:39 am

        As I said, I agree with some of his points. However, Auld uses these things as a way to ignore criticism, whereas suzuki’s bigger point (that the economy should be placed within the environment) is a good one. As I say, all too often it feels like economists think they can calculate the externality, slap a carbon tax on and say they’ve “dealt with” the environment.

  7. #24 by arijitbanik on October 24, 2013 - 7:53 pm

    Reblogged this on Arijit Banik and commented:
    From Unlearning Economics, the iconoclast writing in the shadows…

  8. #25 by Sara on October 24, 2013 - 11:30 pm

    The funniest part of Chris Auld’s original post is the way he compares the critics of neoclassical economics with climate change deniers – as if only “cranks” would question its self-evident truths …

    • #26 by Roman P. on October 25, 2013 - 6:35 am

      It is interesting to compare neoclassical economics with climatology, or more exactly with theories of anthropogenic climate change. Both are mainstream, both are widely believed by the general population, both are tied to ideology and political movements. They even have religious elements, insofar as economics promises riches to the worthy and the climate science speaks of imminent suffering and ruin in the future.

      It is curious, though, that the discourse of denying the anthropogenic climate change is virtually confined to the far right (I imagine that if somebody like Unlearningecon declared that he even doubted the climate science, he would have committed a social suicide) and the heterodox economics is mostly a thing of the far left (at least in the West). So, the Far Right, in contrast to their anxiety and fear of the different people, believes in Heaven but not Hell, mainstream believes in both suffering and salvation, and for the Far Left, it seems, there are only the sorrows of Hell.

      • #27 by Sara on October 25, 2013 - 2:16 pm

        Good point … what I object to in Auld’s post (or for that matter climatology) is the use of the “denier” label.

      • #28 by Unlearningecon on October 25, 2013 - 2:20 pm

        Being on the far left is all about confining yourself to the reality that meaningful change will never occur, at least not in your own life time, and also that when you do finally croak there is nothing on the other side.

        It’s what gets me out of bed in the morning.

  9. #29 by gastro george on October 25, 2013 - 11:19 am

    • #30 by Unlearningecon on October 25, 2013 - 1:12 pm

      On patrol in the comments 😉

    • #31 by Jim Rose on October 26, 2013 - 1:54 am

      more fool them for cutting themsivles off from a major body of knowledge.

      Many of the key issues about what modern macroeconomics has to say on financial crises are discussed at http://www.minneapolisfed.org/publications_papers/pub_display.cfm?id=4526 Interview with Thomas Sargent where he says:
      1. It is just wrong to say that this financial crisis caught modern macroeconomists by surprise: Allen and Gale’s 2007 book Understanding Financial Crises collects many of the dynamic models of the causes of financial crises and government policies that can arrest them or ignite them.

      2. Stern and Feldman’s Too Big to Fail doesn’t have an equation in it, but wisely uses insights gleaned from the formal literature to frame warnings in 2004 about the time bomb for a financial crisis set by regulations and government promises.

      3. Two polar models of bank crises and what government lender-of-last-resort and deposit insurance do to arrest them or promote them. In the Diamond-Dybvig and Bryant model, deposit insurance is purely a good thing, while in the Kareken and Wallace model, it is purely bad.

      4. Bryant-Diamond-Dybvig model has been very influential generally, and in particular that it was very influential in 2008 among policymakers. Many policy authorities correctly noticed that a Bryant-Diamond-Dybvig bank is not just something that has “B A N K” written on its front door. It’s any institution that executes liquidity transformation and maturity transformation.

      5. Policy makers saw Bryant-Diamond-Dybvig bank runs all over the place. The logic of the Bryant-Diamond-Dybvig model persuaded them that if they could arrest runs by convincing creditors that their loans to these “banks” were insured, that could be done at little or no eventual cost to the taxpayers.

      6. The Diamond-Dybvig and Bryant model makes you very sensitive to runs and very optimistic about the ability of insurance to cure them.

      7. The Kareken and Wallace model’s prediction is that if a government sets up deposit insurance and doesn’t regulate bank portfolios to prevent them from taking too much risk, the government is setting the stage for a financial crisis.
      8. The Kareken-Wallace model makes you very cautious about lender-of-last-resort facilities and very sensitive to the risk-taking activities of banks.

      p.s.Friedman predicted in 2001 that the Euo would not survive its first major recession. did anyone suggest that euroland was an optimal currency zone? The British stayed out for that reason.

      • #32 by Unlearningecon on October 26, 2013 - 12:36 pm

        They are not “cutting themselves off”; they simply want pluralism, including neoclassical theories and others. Economists are the ones who cut themselves off.

        A few examples of mainstream economists who were expecting/understood financial crises does not prove the crisis didn’t catch people by surprise. If you recall, people like Bob Lucas and Alan Greenspan were giving speeches about how depression prevention had been solved, and when the crash came the initial atmosphere was one of panic, not “we knew this would happen, let’s take steps xyz”. If you want to say Lucas & Greenspan are charlatans, be my guest, but they were reflecting a more general consensus. And to be honest, while I applaud a lot of the work you reference (as well as Romer & Akerlof’s work on “looting”), the ideas that (a) we should instill confidence in the creditors of failing institutions and (b) insurance will make banks take more risks, are hardly groundbreaking.

  10. #33 by Luis Enrique on October 25, 2013 - 11:30 am

    [first, I didn’t know who Auld was and didn’t read his post before reading yours and writing first two comments here.]

    okay so you are making a non-trivial claim that these things are widespread.

    but what to make of 2.? there’s no way that view is widespread and anybody who claims otherwise has undermined their own credibility on the topic. I will assume you’d qualify or retract on that one.

    as for the rest, well now I’d have to get into the thicket of trying to argue you have misinterpreted their arguments or have bad reasons for disagreeing with them and so forth. That would take far too long. So I am just going to discuss one small point [actually I would also agree with somewhat modified versions of some of your points, just as I’d take issue with some of Auld’s]

    take Williamson for example, who I don’t know terribly well but from what I’ve seen might be on my list of “bad” economists (alongside, regrettably, Mankiw, much of whose scholarly work I admire)

    I’d be surprised if any other economists think the EMH is merely the “assumption of rationality”. But what about his claim that rationality (and DSGE) don’t predict anything?

    let’s take rationality. Now there is one circular, or at least empty, thing to do with that, which is that because you can always think of an objective function that would rationalize observed behaviour, you can always claim behaviour is rational. Fine. But seen from the other direction, of course rationality doesn’t predict anything [*] because before you can say what people are going to do, in addition to having some idea about how they take decisions, you need to know what they want/like and what their constraints are. This point also applies to “irrationality” or psychologically realistic heuristic decision making or whatever, which also “do not predict anything” in the sense Williamson means. You can’t predict if somebody will purchase a chocolate bar unless you know they like chocolate, whether you think they are rational or irrational or whatever. [an aside, obviously you can observe them then predict they will repeat themselves, but that is not a theory of decisions, which is what we are talking about when we debate the assumption of rationality against the alternatives, such as psychologically realistic whatnots]

    Is pointing out that rationality does’t predict anything in this sense “a good defence of economics”? I would say only if you encounter a critic who appears not to understand that the assumption of rationality is merely one component of a model and by itself cannot predict things (other than things like [*]).

    similarly with DSGE. that doesn’t predict anything because it is method, or perhaps subset of “model space” defined by certain practices, and methods don’t predict things, theories do. Some critics (Rob Waldmann) claim you can write down a DSGE model that predicts anything. If you want to make predictions, you have to start discriminating between different DSGE models and choose one with what you think are the right ingredients.

    I might take issue with Williamson here and say that DSGE might not be capable of “predicting anything”, because it rules out certain phenomena and hence certain predictions. Hence you can potentially claim DSGE will be fundamentally misleading on certain questions no matter what you do with it. I am not sure about this, I can just see the potential. If Waldmann is right, and I have understood him correctly, this argument is wrong.

    again, when is “DSGE doesn’t predict anything” a good defence? When you encounter a critic who says things like “DSGE is wrong because it predicts this” A more nuanced criticism would be that commonly used DSGE models with the usual ingredients get *this* wrong and so aren’t suitable for *this* purpose. You could defend that model as saying it’s a work in progress that you realise gets things wrong, or that it is not intended for that purpose, but otherwise I think nearly all economists realise the certain models are highly unsatisfactory, and the view that standard DSGE is unsatisfactory is widespread.

    [*] actually it does, it predicts things like if A preferred to B and B to C then … and so forth.

    • #34 by Unlearningecon on October 25, 2013 - 1:12 pm

      As for 2 – stupid it may be, but it is all too common, sadly. Cochrane, Saint-Paul & Fama have argued something to the effect that it was a largely irrelevant ‘black swan’.

      I too would not consider Williamson representative of the profession; I just included him because I was astonished at the sheer number he managed to accrue ((16) in a review of Zombie Economics? Seriously?) and I do consider his characterisation of the EMH and DSGE quite bizarre. Nevertheless, he does illustrate (6) quite well, and there are plenty of other times I’ve seen economists do this: have you ever seen Scott Sumner defend the EMH? It’s like stabbing goldfish. I also consider the WARP quite a weak fallback for economists who are faced with criticisms of utility, as do a couple of commenters on Aziz’ post on the matter.

      A more nuanced criticism would be that commonly used DSGE models with the usual ingredients get *this* wrong and so aren’t suitable for *this* purpose. You could defend that model as saying it’s a work in progress that you realise gets things wrong, or that it is not intended for that purpose, but otherwise I think nearly all economists realise the certain models are highly unsatisfactory, and the view that standard DSGE is unsatisfactory is widespread.

      I know, and this is the criticism I try to make. Or that DSGE doesn’t seem to be approaching anything unified; there’s just this and that model to deal with this and that problem, but which one is actually a model of the economy as a whole? It’s weird: it seems like many economists will quietly admit flaws, but when an outside critic comes along they go nuts. Maybe it’s just an example of this?

      • #35 by Luis Enrique on October 25, 2013 - 1:26 pm

        I like that cartoon, yes maybe an example of that.

        I both defends and criticises DSGE, and find outside critics sometimes make exaggerated, misinformed or unfair criticisms, which is where the going “nuts” comes in.

  11. #36 by Jim Rose on October 25, 2013 - 12:39 pm

    I wonder if critics of the efficient markets hypothesis follow its advice for their own retirement savings.

    • #37 by Unlearningecon on October 25, 2013 - 2:17 pm

      This would only occur if the critic also happened to be an excellent investor. The obvious examples of this are buffet and soros.

      • #38 by Jim Rose on October 26, 2013 - 12:00 am

        Would you advise your children to invest their savings with Buffett or with a passive, diversified fund such as vanguard?

        John Bogle founded vanguard in about 1975 because he was inspired by Paul Samuelson’s 1974 essay in the Journal of Portfolio Management demanded “brute evidence” that active money managers could beat the market index. Such evidence has yet to be produced In Bogle’s view.

        Fama works for a wholesale fund that uses enhanced indexing e.g. investing in small cap companies.

      • #39 by Unlearningecon on October 26, 2013 - 12:00 pm

        Would you advise your children to invest their savings with Buffett or with a passive, diversified fund such as vanguard?

        Low risk, low yield over high risk, high yields, yes. Doesn’t mean Buffet doesn’t consistently beat the market though.

        And as somebody immediately noted, the notion that the market is “hard to beat” is in no way exclusive to the EMH, and is actually verging on a trivial prediction. The EMH’s other predictions (such as that the market will only really move with new information) are simply not true.

    • #40 by Sara on October 25, 2013 - 2:45 pm

      Last I heard, hedge funds don’t invest according to the efficient markets hypothesis. Otherwise they would just stay home.

      • #41 by NicTheNZer on October 25, 2013 - 5:09 pm

        There are also alternatives to the efficient markets hypothesis (such as the fractal markets hypothesis) which while plausible and without some of the implausible implications (such as there can be no bubbles in the market, leading to Fama’s post crash bubble denial) this theory also implies that beating the market will be difficult/extremely difficult. We should not assume that just because the theory is invalid that its opposite (e.g its trivially easy to make money in the financial markets) is true.

        Nobody does this, except for the purpose of defending their ridiculous efficient markets hypothesis (maybe ‘fantasy’ would have been a better phrase).

  12. #42 by Tom Walker on October 25, 2013 - 5:32 pm

    “1. They defer to the idea that “all models are simplifications”, as if this somehow creates a fireguard against any criticism of methodology, internal inconsistency or empirical relevance.”

    Except when they loath the specific simplification or its policy analytical consequences, in which case the model in question is denounced as being based on a “fallacy”.

  13. #44 by NicTheNZer on October 25, 2013 - 5:42 pm

    I quite liked point 19 actually.

    “19. Some kind of comparison is made between one’s pet theory and Copernicus on one hand and regular economics and Ptolemy.

    This one’s more general actually. There should be some kind of Godwin’s Law style injunction against invoking Copernicus/Ptolemy in any kind of internet discussion.”

    Amazing how many of these are just to deal with arguments Steve Keen has made. His criticism burns, especially as he actually got the crisis broadly right. Some of the many arguments I have seen rather underplay the Minsky point that Keen made in 1995, that “The chaotic dynamics explored in this paper should warn us against accepting a period of relative tranquillity in a capitalist economy as anything other than a lull before the storm.” though that is to be expected. Note Keen writes this in 1995, but he was warning about an impending crisis in 2005 due to observing a period of tranquillity (called the great moderation) and accelerating debt.

    Also 12,13,14 Mr Auld should not complain that lay people misconstrue technical jargon which is clearly inserted into much economics literature (maybe that should be the superficial text book literature) with the express purpose of being misconstrued.

    • #45 by Unlearningecon on October 28, 2013 - 12:04 pm

      A qualitative, general framework that accords with the “stylised facts” seems clearly to superior to first being largely unaware there was a problem, then doing some ad-hoc curve fitting once the crisis happened. I’d say the same for work like Ha-Joon Chang’s over various mathematical development models or the Solow growth model, or studies of firms over the basic marginalist theories. If only more time had been put into these, they would be as comprehensive as the neoclassical approach.

  14. #46 by NicTheNZer on October 25, 2013 - 6:34 pm

    I thought that the David Friedman vs Robert Murphy debate was interesting regarding 17. Of course Friedman spent the whole time spanking Murphy over methodological empiricism because Murphy can’t deal with this as an Austrian (also Murphy doesn’t have the foggiest idea about non-euclidean geometry so he just looks an idiot).

    On the other hand his father wrote the infamous as-if defence. If this is the defence of neo-classical assumptions then I can’t understand any way that a neo-classical model is more empirical than an Austrian one, they are both guesses where the actual behaviour of the agents is not captured in the model. As I see it if you use the as-if defence you are absolutely subject to the same criticism David Friedman uses to bash Murphy, even if your model fits behaviour well, you still don’t know why that behaviour fits.

    In fact this criticism appears to apply quite strongly to the monetarist model Milton Friedman hypothesised. It (purportedly) fit the available data quite well, but the internal mechanics were wrong (banks don’t lend reserves) and so it was a statistical accident that this fits over the period Friedman analyses data over, and the stable relationship collapsed when the theory was applied in practise. This is called over-fitting in statistics. The collapse of the theory might also have something to do with the changes in the monetary system due to the US closing the gold window.

    • #47 by Jim Rose on October 26, 2013 - 10:20 pm

      The disappearance of the correlation was predicted by monatarist ALAN WALTERS. When there is a stable money supply, the correlation disappears because the correlation of a constant with anything is zero.

      Kochin wrote a paper in 2006 on this.

      • #48 by NicTheNZer on October 26, 2013 - 10:34 pm

        The correlation of a constant with another constant is not zero actually.

        Is Walters explanation the same as saying that the money supply is an endogenous variable? Why do monetarists still not accept this simple fact?

  15. #49 by Noah Smith on October 25, 2013 - 6:46 pm

    I agree wholeheartedly with points 1,2,7,9,11,13,14,15,16,and 18.

    I disagree somewhat with points 3,4,5,6,8,10,12, and 17. Briefly:

    3. “neoclassical” has come to just mean “mainstream”, and has lost all value as a word. See http://noahpinionblog.blogspot.com/2013/06/what-is-neoclassical-economics.html

    4. Lots of economists don’t confuse statistics with controlled experiments. But sometimes experiments are impossible and statistics is the best you can do.

    5. Actually, this is correct. “Neoclassical” used to mean something but now it’s just a dog-whistle for people who feel their ideas are excluded unfairly from the mainstream.

    6. Sometimes those things don’t make sense; sometimes they do, though.

    8. Not gonna get into this, because it’ll mean I have to argue with Tim Johnson, and that is a task that daunts even me.

    10. I don’t think very many economists actually believe those things ever existed.

    12. Many economists do make this mistake, but there is also value in ignoring politics for some research.

    17. That’s just how any discipline works, scientific or no. Look closely and you’ll notice that biologists do this PLENTY. Physics is the only possible exception, and only with an asterisk.


    – N

    • #50 by Unlearningecon on October 25, 2013 - 7:14 pm

      3. OK but surely that just moves the problem – “they think mainstream economics is pluralistic based on behavioural etc.”?

      10. In David Graeber’s book, he gives quite a few examples of textbooks speaking about the myth of barter. People like Dani Rodrik have also pointed out economists’ widespread belief in free trade, theoretically and historically. However, maybe you are right that it’s more libertarians (who often claim the mantle of economics) than economists.

      17. This is true to an extent. But surely, once we can open up the black box – which we can with so many economic phenomena – our previous ‘as ifs’ become redundant?

      • #51 by Noah Smith on October 25, 2013 - 7:30 pm

        3. Mainstream economics IS pluralistic, and those things you cited – behavioral, etc. – DO represent a diversity of approaches. To ignore that fact is pure academic-politicking. Sure, mainstream econ doesn’t include ALL approaches, but that doesn’t mean it’s not pluralistic.

        10. Yes, real economists, including people like Mankiw who use those parables in their textbooks, do not believe there was ever a time when these parables described reality.

        17. “Black box” is just an epithet for “modeling at a higher level”. Sometimes higher-level modeling works well; we call those situations “emergence”. Sometimes it doesn’t work well and we have to go back to microfoundations (“reductionism”). The key is that we need to evaluate what level to model stuff on. We don’t want to model the digestive system as a bunch of colliding particles. But we do want to model a hurricane that way.

      • #52 by NicTheNZer on October 25, 2013 - 8:20 pm

        I disagree about this statement on “black boxes”. Micro-foundations can not provide a fall-back option. If you can’t aggregate the micro-effects (say due to SMD) and you can’t find some other model, then you simply don’t have a model. Just assuming that the aggregate behaves as a single agent of that aggregate is no better than any other model of the aggregate a-priori.

        Nobody models hurricanes at anywhere near a colliding particle level.

      • #53 by Noah Smith on October 25, 2013 - 8:39 pm

        SMD doesn’t stop you from aggregating micro effects, it stops you from using a representative agent.

        People model hurricanes based on collisions of smaller elements of atmosphere, not fundamental particles.

      • #54 by NicTheNZer on October 25, 2013 - 9:02 pm

        Quite right, my first statement should have been ‘if you can’t use a representative agent (say due to SMD)…’ I still think this point is valid however.

        Thanks for the explanation of weather models. I work in the weather forecasting industry, btw…

        Nobody should think that the geo-stationary pieces modelled by weather models work anything like particle collisions.

      • #55 by Noah Smith on October 25, 2013 - 9:22 pm

        The point about modeling different phenomena at different levels of complexity stands.

      • #56 by NicTheNZer on October 25, 2013 - 10:11 pm

        You certainly can, and sciences do, model different phenomena at different levels of complexity (actually aggregation might be a better term to use here). However I can’t think of any situation where higher level modelling fails here, but the lower level succeeds (can you suggest one)?

        This is based on the idea that if the lower level modelling works then you could reach the higher level model by aggregating up. The other case is when there are emergent properties and the lower level doesn’t simply aggregate up to exhibit these properties. So you must investigate some phenomena at a higher level.

      • #57 by NicTheNZer on October 26, 2013 - 11:48 am

        Maybe I should rephrase what I am asking Noah about. I genuinely don’t understand this usage of micro-foundations.

        Noah said: “Sometimes higher-level modeling works well; we call those situations “emergence”. Sometimes it doesn’t work well and we have to go back to microfoundations (“reductionism”).”

        But I would say no,

        Sometimes lower-level modelling works well and you can aggregate up to get a higher level model. We call these situations “strong reductionism”. Sometimes it doesn’t work well and you need to develop a model of higher level phenomena. We call these phenomena “emergent properties”.

        Richard Feynman explains an emergent property in physics here,

        If you are modelling an emergent phenomena but simply have not figured it out then falling back on your micro foundations is surely no better than any other conjecture on the matter?

    • #58 by roger gathman on October 26, 2013 - 8:39 pm

      On 3. To settle the dispute about the use of neoclassical, you don’t just present your impression – you actually go to places where usage would be accredited – economic textbooks, journals, etc. This isn’t that hard, given say Jstor and ebsco. I get 28 thousand some results for neo-classical economics in Ebsco business, which certainly means it is a current phrase. I get 113 references to the phrase in the American economic review. In his citation as a distinguished fellow, the term neo-classical seems to be used as a convenient classification for a certain economic methodology: :”These works criti-
      cized neoclassical economics for its assumption of costlessly enforced prop-
      erty rights and helped revolutionize the study of institutions, including political
      institutions, and their effects on the economy.”
      Just because neo-classical economics is used more by those who see themselves as working more outside the paradigm than those who simply assume the paradigm doesn’t invalidate the term. Making neo-classical a proxy for mainstream, however, is an invidious instance of the kind of economics imperialism promoted by Becker. By the way, Becker, not me, uses the phrase economic imperialism.
      Which gets us to pluralism. In the social sciences, of which economics is one, pluralism most often means the acceptance of both qualitative and quantitative approaches, and in that respect, I suppose, institutional and behavioral economics could count as evidence of pluralism. Unfortunately, I think the problem is that all adhere narrowly to the idea that equilibrium holds the key to economics. As Lucas once said,, equilibrium is the “condition of intelligibility” for economic analysis. And it is here that economics is not pluralistic. The equilibrium model goes back to thinking economics is social physics, which is, I think, a silliness that economists now a days deny they really believe, even as they also adhere closely to Lucas’ dictum. The idea that economic analysis should take equilibrium as its assumption, however, seems to me to make economics much closer to humoral medicine than physics, where the balance of humors was supposed to explain the workings of the human body. It was science, it had theories, it propounded cures – for instance, the humoral doctors proved conclusively that fruits could not cure scurvy, since scurvy derived from a disproportion of the bilious humors – and it was wrong. I’m reminded of those royal physicians getting sea captains to throw away their lemons and oranges every time I read an economist bigwig recommend austerity for getting out of the slump.

      • #59 by roger gathman on October 26, 2013 - 8:40 pm

        The distinguished fellow is Douglass North. I don’t know why that part of my sentence was skipped.

  16. #60 by Jim Rose on October 25, 2013 - 11:53 pm

    Stigler contended that economists exert a minor and scarcely detectable influence on the societies in which they live. As is well know, Stigler in the 1970s toasted Milton Friedman at a dinner in his honour by saying: “Milton, if you hadn’t been born, it wouldn’t have made any difference.”

    Stigler argued that if Richard Cobden had spoken only Yiddish, and with a stammer, and Robert Peel had been a narrow, stupid man, England would have repealed the corn laws to allow free trade in grain as its agricultural classes declined and its manufacturing and commercial classes grew in the 1840s,

    As Stigler frequently noted,
    • the ideas behind reform had been around for a long time.
    • To affect public policy, ideas must find a market among the groups influencing change.
    • when their day comes, economists seem to be leaders of public opinion.
    • But when the views of economists are not so congenial to the current requirements of interest groups and median voter, these economists are left to be the writers of letters to the editor in provincial newspapers

    These days they would run an angry blog.

    Can you think of any significant public expenditure, regulation, subsidy and welfare benefit that is not supported by the median voter? Interest groups fight for scraps from the median voters’ table in comparisons.

    Were the monetary and fiscal policies behind the great deviation in macroeconomic policies from 2001 to 2007 unpopular with the median voter?

  17. #64 by Lee Adler (@Lee_Adler) on October 26, 2013 - 2:11 am

    Economics is religion. Priests and followers. Bishops and churches. Gurus. Dogma. Faith, visions, and superstition.

  18. #65 by Steve on October 26, 2013 - 10:59 pm

    One sign you are dealing with an idiot:

    He extrapolates from a handful of anecdotes to a general principle and doesn’t test his theory.

    Pick some random nytimes op-ed under Economics View and see how many apply. Then eliminate the dumb ones (like 8, 9, 12,13 at least)

    • #66 by Unlearningecon on October 26, 2013 - 11:21 pm

      Ah, you got me – all of my examples just don’t count, especially the one that this article was a direct reply to. The definition was articles that are defending or expounding economics as a science, and the most recent one I could find on EV was the Raj Chetty one that I’ve already discussed. However, if you think you have an article that fits the criteria but doesn’t commit any of these, I’d be happy to read it.

      As for the “dumb ones”, you’ll have to do better than that. Explain why they are wrong, please. I’m especially intrigued as to why you think economics is value free, and why it is separable from politics?

      • #67 by Steve on October 26, 2013 - 11:31 pm

        You are making a claim about economics an then picked specific examples that stood out in your memory as bad for making these kinds if mistakes…. Then the reader is asked to extrapolate about economists on general based on these examples.

        If I said Harvard kids are dumb and then picked the ones that I knew that stood out from being dumb would that convince you that most are?

        Economics can be done value free just like physics, history, ethics or art. Here is an example: “I think monopolies probably try to maximize profit” Here is an ethics example, ” if people are trying to maximize life expectancy they would kill 1 old man to save 5 kids.”

        Of course it’s easy to do normative economics. Most, almost all, Econ is normative but it doesn’t have to be any more than literature or art does.

      • #68 by Unlearningecon on October 26, 2013 - 11:38 pm

        The examples I gave were simply major debates about the state of economics over the past couple of years, involving many prominent economists. It’s more like I picked the highest flying students at Harvard and pointed out they were dumb.

        And if you tell me you think monopolies are trying to maximise profits, you are saying that how monopolies behave is important. So it is at least partially normative.

        PS please do not follow up with childish snipes, they will be deleted and you will be blocked if you persist. I can assure you I’m more amused by economists than angry about them.

      • #69 by Steve on October 27, 2013 - 12:12 am

        I didn’t say I think monopoly behavior is important. I can see how you think that is likely but I don’t see how you think it is implied. Is it logically impossible so me to study something I know is not important? I used to memorize statistics on the back of baseball cards but I don’t think Ken Griffey’s XX home run in 1997 are important. (I left out the number because it isn’t important!)

        It could be evidence that I think staring at them is a good use of my time but then we’d be using the standard “neoclassical” theory of consumer choice… Revealed preference… *catastrophic music, world explodes*

      • #70 by Unlearningecon on October 27, 2013 - 12:15 am

        Well, “worthwhile”, “important”
        or any similar term. In any case the fact that you’re not studying the effect of recessions on snail deaths implies a value judgment of what’s important.

        An economist who releases a paper on monopoly pricing behaviour is surely implying that is important, no? Or at least someone with more power than them thinks so and has made them do it 🙂

        It just seems to me that all science is driven by what people find interesting or promising, and hence involves value judgments of one kind or another. Furthermore, in economics you’re dealing with people and policy, which makes it even more intertwined with ethics (that’s what the politics comes in).

  19. #71 by Luis Enrique on October 27, 2013 - 9:30 am

    “just seems to me that all science is driven by what people find interesting or promising, and hence involves value judgments of one kind or another”

    Ok, but isn’t that rather empty? I mean an economist, or any other scientist (or researcher, if you wish to claim social sciences are not sciences), says “I am just studying this question without making any value judgements, doing “positive” research” and you retort “aha, but the mere fact you are studying it is itself a value judgement”, doesn’t that strike you as a useless smart Alec point to make? The more important point is that, okay given the value judgement that the question is deemed worth asking, the researcher is simply seeking an answer without making any further value judgement. That seems to me using the word “positive” in a useful way, not a sign the economist hasn’t the foggiest. If you define things so that positive research is by definition impossible, you just need to invent a new word to distinguish between really value laden analysis, and simple question answering. And really, asking questions like “does increasing government spending during a recession reduce unemployment” is such an obviously useful thing to know, that turning round and claiming economics is value laden because it deems such a question worthy of interest, just makes the critic look like a fool. Of course we are interested in questions that matter to people.

    • #72 by Unlearningecon on October 27, 2013 - 12:36 pm

      I can understand what you’re saying, but I actually think the decision to say “look, I’m just a dispassionate observer, weighing up the facts” without considering other implications is in itself normative. I mean, I actually find more acknowledgment of the importance of “people” always being in the background from natural scientists. Consider Einstein’s quote about getting lost in your maths and so forth and compare it with Cowen and Mankiw’s pieces which I link to. In fact, when economists emphasise that their science is “value free”, it seems like they’re often just trying to evade criticism and convince themselves as much as anyone else.

      In any case, remember my initial point was the “economics” (as in, mainstream economics) is not value-free. I mean, surely you’d agree that you’d have a hard time leaving an economics degree without thinking efficiency was ‘good’, (too much) unemployment and inflation were bad, competition was good, ‘making someone worse off’ (pareto) was bad, etc?

      EDIT: Just saw this. Maybe you might just see it as a practical problem but it seems to me the way we present and interpret facts is impossibly intertwined with our biases.

      • #73 by Anonymouse on October 28, 2013 - 6:39 am

        This, and your comments above, reveal a lot.
        Just because you are driven by political and normative beliefs doesn’t mean everyone is (I find this is a common fallacy amongst the heterodox, both left and right).

        Is it really so hard to comprehend that some people enjoy positive theorizing, and that they actually don’t worry nor think about the normative at all? There is a lot you can do, and a lot of fun you can have, in a modelling world where the only normative assumption is that pareto improvements are good (this really should be uncontroversial!).

        When a theorist starts with assumptions A and draws conclusions X, most of the time all they are doing is saying that X logically follows from A. They don’t need to place any special meaning on A, or even like A. It is simply a fun intellectual exercise.

        Ariel Rubinstein is a good example of this.

      • #74 by Unlearningecon on October 28, 2013 - 11:19 am

        You berate me for thinking that normative assumptions are like, just my opinion, man, and then…go on to introduce a normative assumption (pareto) and claim that it “should be uncontroversial!” I don’t see how you could have made my point better for me.

        And I reiterate that if we are studying A, there’s something we’ve seen in A that is important. Not to mention how Leamer showed that, based on a few key – largely normative – decisions, even econometric results can vary wildly. Yes, I know, econometrics has moved on somewhat, and I’ve previously applauded some of this work. But Leamer’s point still remains, in theory and in empirics: the assumptions we use, the data, where we draw the line etc etc affects our results, even if we don’t know it. Chetty’s piece demonstrates this nicely.

        Even “fun” is normative anyway!

      • #75 by Anonymouse on October 28, 2013 - 5:26 pm

        “And I reiterate that if we are studying A, there’s something we’ve seen in A that is important.” No, this is not necessarily true. That is the point. Again, see Rubinstein. Or ask an algebraic number theorist. Sometimes an application is found for number theory, but the theorists who do the original work couldn’t care less – they are in it for the fun!

        And, we could just simply study Pareto efficiency because it is fun, not because we think it is relevant or important. But, do you really want to argue against using Pareto? It is so incredibly weak. Note carefully that the argument invoked when using Pareto is NOT “Pareto efficient points are good.” The argument is “non-Pareto points are bad,” and that these two statements are not converse of each other.

      • #76 by Unlearningecon on October 28, 2013 - 6:05 pm

        Again, I’m not arguing for or against anything. I’m saying pareto contains a value judgement – that “making people worse (better) off” is bad (good).

        And you think Rubinstein’s perspective has nothing to do with values?

        I believe that intellectual thinking – philosophy or logic or game theory – is very useful in the cultural sense. It’s part of the culture, it’s a part of our perpetual attempt to understand ourselves better and understand the way that we think. What I’m opposing is the approach that says, in a practical situation, “OK, there are some very clever game theoreticians in the world, let’s ask them what to do.” I have not seen, in all my life, a single example where a game theorist could give advice, based on the theory, which was more useful than that of the layman.

        He also goes on to imply that highly abstract thinking can eventually lead to something practical.

  20. #77 by Nick Rowe on October 27, 2013 - 12:49 pm

    Unlearning: On 17. first link is to something I said. You don’t explain why you disagree with what I said.

    • #78 by Unlearningecon on October 27, 2013 - 1:09 pm

      Hey Nick,

      Sorry if you feel that way – I wrote the post in a hurry and scrambled around to find the example that came into my head first. However, I think I discuss this satisfactorily here (?):


      As I noted on Scott’s blog, I don’t think we can analogise this to an oligopolistic market model, where aiming for p or q can have basically the same result. We tried aiming for q in the 1980s and it didn’t work. A commenter once helpfully compared the endogenous view of the banking system to that of an electricity supplier, who can control the price but not the quantity used. The central bank can raise or lower interest rates, but if there is enough endogenous demand then banks will just keep creating those loans. Obviously interest rates could go incredibly high, but that would just utterly devastate the whole system.

  21. #79 by Andres Carbacho-Burgos on October 27, 2013 - 11:38 pm

    I’m coming in late to this debate; while I have no sympathy for Auld, Mankiw, and some of the commenters here who back up their approach, I’d like to make the case that UE, and Steve Keen and others of their ilk go somewhat too far in bending the stick in the other direction. First through the rhetorical sleight-of-hand of conflating “economics” with neoclassical economics (that neoclassical economics dominates the curriculum and textbooks is no excuse to even momentarily deny to the layman the existence of other methods of economic analysis), and second by not emphasizing that neoclassical economics is just that–a generalized method of analyzing economic problems–and that neoclassical economics is therefore not a school of thought nor a political predisposition that an economist self-consciously identifies with; hence the myopic denial of most economists that they are neoclassical. Granted, this generalized method of analyzing economic problems creates serious pathologies that make it untenable as a long-term view of human relations, but the point is to change economics, not just to interpret or criticize it.

    Personally, I think that the description of Arnsperger and Varoufakis is not narrow or far-reaching enough. My reading of neoclassical economics is that it consists of the following five methodological tenets:

    1. Permanent foundations. That is, n.e. is characterized by rationalist epistemology: human behavior is grounded on certain permanent and unchangeable axiomatic principles reminiscent of Euclid’s elements (transitivity of preferences, declining marginal utility of consumption, etc) from which everything else is deduced. More importantly, these principles are _not_ based on observed human behavior, because observed human behavior is so changeable as noted in the vagaries of empirical studies or the proliferation of experimental studies in behavioral economics. This also implies that these axiomatic principles often have to be defended against the charge that they are unrealistic. Milton Friedman’s essay on positive economics gave us a very nice or an awful defense of unrealistic assumptions, depending on one’s point of view.

    The use of rationalist epistemology is emphatically not an assertion that economists don’t do empirical work or ignore the results of such work. Rather, empirical results that seem to contradict the axiomatic operating principles of homo economicus are instead co-opted by adding in complications (i.e., epicycles) from the outside world (asymmetric information, altruistic utility functions, etc). This is not quite the same as denying that neoclassical economics is unswayed by contradictory empirical results, as UE seems to assert.

    2. Atoms. That is, neoclassical economics operates through methodological individualism, as pointed out by Arnsperger and Varoufakis. So for example, macroeconomic aggregates are nothing more than the passive sums of individual decisions, and don’t influence those decisions as such. Hence the constant stress on microfoundations, and the denial or even consideration that there might be a feedback from macro aggregates to individual actions.

    3. Calculation. That is, the conception of rationality is one of either strict mathematical optimization through differential calculus or matrix algorithms, or something that closely approximates such a procedure, Samuelson’s theory of revealed preference the best-known example. Arnsperger and Varoufakis call this methodological instrumentalism, but I think they err defining this tenet as assuming that human beings are preference-driven. All of us have preferences around which we base our choices, the question is _how_. Mathematical optimization is one possibility, but imo it is not realistic and in many cases yields predictions that have been empirically falsified as long ago as Allais and Ellsberg.

    4. Gravitation. That is, most economic outcomes are either in equilibrium or are in the process of shifting from one stable equilibrium to another (A. and V. call this methodological equilibration). Economists have the mathematical tools to handle non-equilibrium and unstable equilibrium situations, but such results are too often not congenial to the view of economic outcomes as being self-equilibrating. I’ve seen much more prominent use of multiple stable and unstable equilibria, limit cycles, and saddlepoint dynamics by heterodox as opposed to neoclassical economists.

    5. A knowable, quantifiable future. That is, the future is simply an extrapolation of the past, so it can be modeled as a known range of outcomes with a known set of probability distributions. The technical term for this is ergodicity, and Davidson and others claim it is the prime shortcoming of n.e. Rational expectations and the EMH are logical outcomes of combining ergodicity with methodological individualism and calculation. Similarly, the life-cycle income hypothesis and its mathematical extensions (Ramsey model, OLG model), implicitly assume an ergodic future.

    Any sort of mathematical model that attempts to derive expectations or forecast future outcomes assumes ergodicity, and is therefore a highly hazardous undertaking. Econometric forecasts are extensions of the known past, but it is the unknown past (e.g., in the form of low-quality RMBS assets) that is waiting around the corner for us, armed with a big baseball bat.

    Whew. There are some implications of the above five tenets. First, self-consciousness as to method indicates that neoclassicism as a coherent approach to economics was _not_ founded in the 19th century. Jevons, Walras, and Menger, and before them Cournot, von Thunen, Gossen, etc, were the first unconscious practitioners of neoclassical economics, but the self-conscious derivation of these tenets had to wait another 70 years or so. Friedman, Samuelson, Popper, and even Lucas are in that sense the true founders of neoclassical economics.

    Second, these five tenets encompass a range of different schools. At the risk of grossly simplifying, I would say that Austrian economists are characterized by the emphatic acceptance of tenets 1. and 2., whereas the other three are not so much denied as simply discarded as irrelevant. Similarly, the older mainstream schools of monetarism and neoclassical synthesis Keynesianism are characterized by accepting tenets 1-4, while tenet 5. is mostly not considered. It is the full acceptance of tenets 1-5 that characterizes the mainstream, that is the new classical, RBC, and new Keynesian economists.

    Third and most importantly, these five methodological tenets say little about the political orientation of a neoclassical economist, or even about the technical policy recommendations that they would favor. Granted, a heavy emphasis on tenets 1. and 2. would predispose an analyst to be a libertarian free-market corporate apologist, but there are lots of ways of avoiding such an evil fate while still remaining in the mainstream. That is, introduce more institutional complications: rigid wages, staggered contracts, and asymmetric zero-lower bound effects to justify Keynesian macro-policy conclusions, non-excludable property problems to justify technocratic environmental policies such as Pigovian taxes and cap-and-trade permits, economies of scale to justify natural monopoly regulation, etc.

    In other words, it is perfectly possible to be a “good” neoclassical economist, Paul Krugman being the epitome. While I can sympathize with many of the heterodox disagreements with Krugman, who has some glaring theoretical blind spots, my normal reaction is to be highly irritated with such attacks, which ignore the practical need to put together some sort of broad coalition that can permanently modify not just mainstream policy but mainstream principles for the better.

    Take a lesson from Churchill guys, and start making favorable mentions of the Devil in the House of Commons. Mainstream liberal economists like Krugman, DeLong and Thoma, not to mention mainstream liberal bloggers like Klein, Rowe, and Yglesias may be theoretically faulty neoclassicals, but better them than the likes of Lucas, Cochrane and Fama. Whenever I read the sniping of Krugman and company that goes on in this blog or in Naked Keynesianism or Real-World Economics Review, the first question that comes to my head, regardless of how well-founded the criticism is, is “how does this help us rewrite the mainstream economics curriculum?”. Too often, the answer is that it doesn’t.

    Now that said, I do agree that history is in the process of giving left-of-center neoclassical economists a thoroughly deserved smackdown (and yes, the smug Brad DeLong is usually the most deserving). The most glaring blind-spot is the complete lack of any non-individualist theory of political economy, which results in left-of-center neoclassical economists being constantly frustrated because their favored policy solutions (fiscal expansion, cap-and-trade, single-payer healthcare, etc) are not only not enacted, but are continually libeled in the dominant political discourse. If they read even small bits of heterodox political economy, such as Kalecki’s “Political Aspects of Full Employment” or Hirschman’s “The Rhetoric of Reaction”, they would have a better understanding of why they are voices in the wilderness.

    In the long-term, I think, left-of-center neoclassical economists are a dying breed; the dominance of conservative economists in the political discourse will make them less and less visible and less influential, much in the same way that moderate Republicans today are being marginalized. But that’s _not_ going to be a good outcome for left-wing heterodox economists, who need to work with more center-respectable voices if the dominant economics discourse is ever to become something other than sophistry in favor of corporate interests.

    • #80 by Unlearningecon on October 28, 2013 - 12:21 pm

      Great comment, thanks.

      I’d like to make the case that UE, and Steve Keen and others of their ilk go somewhat too far in bending the stick in the other direction.

      Quite possible!

      Its is, however, worth noting that I have never identified the majority of economists as ‘free marketeers’, and have actually called attention to this as a mistake in the past.

      As for cooperating with Krugman – well, I’m not sure. In his debate with Steve Keen, which I link to above, he just behaved like the conservatives he usually decries: misrepresenting, name-calling (“mystics”). I’d be happy to cooperate with him, but when you attack a centrist or liberal from the left, well, they turn into a conservative very, very quickly. I for one suspect that he and people like Delong do think there are things fundamentally wrong with neoclassical economics, and they sometimes speak about this. However, when pushed, they simply fall back to what they know, so I’m not sure that they can be a positive catalyst for change.

      However, you are right about the Krugman fetish, and the overemphasis on repeating certain criticisms. It’s boring and it doesn’t get us anywhere.

      • #81 by Andres Carbacho-Burgos on October 29, 2013 - 10:20 pm

        True enough, though I was thinking not of cooperation with Krugman, DeLong, et al, more like a non-aggression pact. Heterodox economists have far more important things to do with their time than to play Ahab to the IS-LM model’s white whale.

        It’s bad enough to have free market (read: large corporate) apologetics polluting political discourse, but too much of the economics curriculum, especially at the undergraduate level, is polluted by such apologetics in a more sophisticated form: Ricardian equivalence, full Treasury-view crowding out, policy ineffectiveness under rational expectation, RBC theory, the strong EMH, and so on. The economics textbooks of the sixties and early seventies, plus their modern versions such as Krugman/Wells, for all their problems, would actually be an improvement compared to what students get fed today.

        And, for that matter, Krugman, Thoma, DeLong and all are slowly being dragged into an abandonment of full-fledged neoclassicism. Krugman’s recent assertion of the importance of _macrofoundations_ might yet become a milestone in his thinking if he follows it up properly, i.e. by rejecting not just microfoundations but the idea of any sort of permanent foundations for economic behavior.

        The point being that even adherents of neoclassical economics can make positive contributions provided they are willing to let their intuition override the methodological mechanics. Since I’m fairly new to this blog, I have to stick out my neck and ask: when are you going to rename it “Unlearning Mainstream Economics”? Honesty in advertising and all that…

      • #82 by Unlearningecon on November 2, 2013 - 5:49 pm

        Well, you are certainly more optimistic about Krugman et. al than I am, but I hope you’re right 🙂

        When are you going to rename it “Unlearning Mainstream Economics”? Honesty in advertising and all that…

        I have previously said this blog might be better titled ‘unlearning marginalism’. But it’s simply so much less catchy that I can’t give up the moniker 😦

    • #83 by Luis Enrique on October 28, 2013 - 2:23 pm

      some quick responses,

      1. assuming permanent and unchangeable axioms regarding human beings sounds bad. Thinking it might be useful to model human decision makers as if they have transitive preferences less so. The difference is a matter of emphasis.

      2. a better claim might be that economics gives insufficient weight to aggregate objects – there are plenty of models that do include a (limited?) role aggregates. A growth model might have private individual decisions both being affected by and feeding back into some aggregate feature.

      3. ok, it’s a (sensible?) way of getting results, not a simulacrum of reality. If more psychologically realistic models appear that are useful and better. fine with me, but it would be madness to constrain economists by insisting they never use any maths than is harder than the average person could perform whilst trailing round a supermarket with a child in tow.

      4. I am forever reading that mainstream economics ought to adopt non-linear systems with chaotic or cyclical dynamics (or whatever) [*] – my (poorly informed) understanding is that economics did flirt with these models in the seventies (hog cycle models) and also catastrophe models, but in the end decided they weren’t as interesting or as useful as people seem to think. If you are, say, trying to model how the labour market would respond to some policy intervention, or whatever, it’s really not obvious such dynamics make sense.

      5. this point is overdone too, imho. suppose I write down a stochastic process larded with regime changes and rare events, and say my agents know the distribution, that’s ergodic too if you take a million-year perspective, but so what? My agents still haven’t got much of a clue what’s going to hit them over the next decade. It’s one thing to point out the limitations of assuming people face uncertainty which is as predictable as a coin-flop, but going to full hog and trying to model non-ergodic, Knightian uncertainty etc. well it’s not clear where that’s going to get you. Although I am talking out of my rear end a bit here. and should probably read and think about this


      [*] actually plenty of people claim economics needs to adopt non-linear models full stop, without apparently being aware of the extent to which non linear models are already used. i.e. the friggin Solow model is non linear.

      • #84 by Andres Carbacho-Burgos on October 29, 2013 - 11:20 pm

        I grant all of the above, but am unpersuaded. Or, more charitably, I think you underestimate the importance of these methodological tenets to the mainstream economics edifice.

        To start, 1. seems to be innocent enough provided you stick to consumption hedonics such as transitive preferences or dimishing marginal utility of consuming the same good, but there are two problems with this. First, many of the axiomatic foundations are essential, but cannot be easily written down in a water-tight manner, and are full of contradictions.

        For example, tenet 3 is actually one such axiomatic foundation; and the problem is not just bounded rationality, but that any mathematical portrayal of individual optimization gives you a view of human decision-making that is too granular in both time and space and can therefore lead to fallacies that are similar to Zeno’s paradoxes. We may think we are optimizing based on constraints, but those contraints are created by other human beings making simultaneous decisions, and will therefore shift over time, possibly slowly, possibly with blinding rapidity depending on the circumstances. The liquidity-of-wealth paradox is a case in point: when individuals believe asset prices will fall, they cannot simply carry out an individual optimization exercise to decide whether to buy, hold or sell: they have to take into account the _unknowable_ reactions of fellow financial investors.

        Second, no matter how self-evident an axiom may seem, it usually has to relay on some sort of static assumption (“ceteris paribus” the most often invoked) to be useful. Transitivity of preferences, for example, can only be watertight if you look at the consumer at a single point in time. But preferences, and the choices being considered by preferences, change over time, often in response to feedback mechanisms. In the real world, there is no such thing as an exogenous variable.

        Anyway, the point of the above is not to throw out mathematics and models of decision-making, but simply to realize how limited, path-dependent, knowledge-constrained are human capacities; mathematical models should display such constraints rather than sweeping them under the rug.

        Brief comments on your reactions to the other four method tenets:

        2. The point is not just that models need to incorporate the feedback from aggregates to human decision-making, but that on occasion decisions that seem rational from the point of view of static, given aggregate quantities can lead to serious instability once they are aggregated. Not so much in growth models (though we’ve only just started factoring in ecological effects in such models), but especially in financial markets.

        3. Not only are there mathematical models of human decision-making that do not rely on optimization, they generally tend to be mathematically simpler than the machinery needed to find and confirm critical points. Kalecki’s model for industrial firm pricing is one such example. Similarly, it is possible to model the effects of peer behavior and persistence (“it was good last time, so I’ll buy it again”) on consumer behavior in mathematical models that don’t relay on optimization; note also that such models rely substantially on _past_ outcomes and therefore create the autocorrelation that we see in so many econometric studies.

        4. Heterodox models of business cycles rely substantially on unstable equilibrium dynamics to the present day. Not so much on multiple equilibria, which is harder to model, but on cases of cyclical or saddlepath instability. In many models, heterodox economists start right off with difference or differential equations without first building any sort of static model. See chapter 9 of Lance Taylor’s “Reconstructing Macroeconomics” for a good work-through.

        5. Well, that’s precisely the point. Though there could be exceptions that I’m not aware of, heterodox economists as a rule _don’t_ create mathematical models of optimization based on future outcomes (i.e. no Ramsey, OLG, etc models), precisely because any such model would involve ergodicity. Which is emphatically not to deny that individuals have expectations about the future; individuals always try to extrapolate from the known past. Rather, the point is that sooner or later the unknown past, aided and abetted by crooked accountants, will affect future outcomes in ways that will make at least some individual expectations suboptimal. The implication in financial markets is that there is always the possibility of individuals making systematic errors over at least a short-run time period, with serious and destabilizing results. But though financial markets have been getting all the attention recently, I fear that environmental considerations may teach us this lesson in a much sharper manner in the coming years.

  22. #85 by Jim Rose on October 27, 2013 - 11:59 pm

    Best evidence of bad criticisms and bias is going on about the methodology of economics. Who does that for sociology?

    • #86 by Andres Carbacho-Burgos on October 28, 2013 - 2:42 am

      I’ve not heard of the President having a Council of Sociologic Advisors, or of Bloomberg or CNBC interviewing sociologists, so I think the stakes are mite higher for economics. Even so, sociologists do have debates on methodology; every field of human knowledge needs to have occasional debates on methodology/epistemology/ontology in in order to have a clear idea of wtf it is doing. Or do you think Einstein debated Bohr about God playing dice out of a concern for government policy?

  23. #87 by Luis Enrique on October 28, 2013 - 11:28 am

    “a normative assumption (pareto) and claim that it “should be uncontroversial!” I don’t see how you could have made my point better for me.”

    um, doesn’t Pareto improvement mean “making some people better off whilst making nobody worse off” – you don’t need to be an economist to think that’s a good thing. Who doesn’t?

    Not to mention how Leamer showed that, based on a few key – largely normative – decisions, even econometric results can vary wildly”.

    WHOA THERE! Leamer showed econometric results can very wildly according to model specification (variable inclusion decisions) – that has nothing whatsoever to do with “normative” – now you’ve made me think you have no idea what the word means.

    • #88 by Unlearningecon on October 28, 2013 - 11:37 am

      I’m not making a judgment on pareto improvement; I’m just saying it is a judgment.

      And I’m pretty sure Leamer linked the chosen model specifications with peoples’ priors. I mean, they wouldn’t have to be, but I think his point was that based on what you suspected/believed, your results would differ. After all, he explicitly referred to political positions eg “a right winger expects the punishment variables to have an effect”.

      • #89 by Luis Enrique on October 28, 2013 - 12:38 pm

        yes but that’s not what normative means! Normative models are about what we OUGHT to do or judgements about what is good or bad. Example

        1. foreign aid does this (empirical, positive)
        2. this explains what foreign aid does (theoretical, positive)
        3. we ought to give this much foreign aid (normative)
        4. the loss of democratic sovereignty involved in donor conditionality outweighs the benefits of better development outcomes (normative)

        you may have priors, you may select your model in accordance with your political beliefs, and that may effect your results, but that could be true in all four examples above. Letting your beliefs influence your research is a different category from normative/positive

      • #90 by Unlearningecon on October 28, 2013 - 12:59 pm

        Hmmm, OK, but I think the right winger example is a good example of a normative belief (punishments should increase), even if it isn’t explicit.

      • #91 by Steve on October 28, 2013 - 1:24 pm

        So your argument is that everything is normative because you CAN infer normative beliefs associated with the positive statements …. In other words because you put words in people’s mouths. Sounds reasonable .

      • #92 by NicTheNZer on October 28, 2013 - 1:02 pm

        “um, doesn’t Pareto improvement mean “making some people better off whilst making nobody worse off” – you don’t need to be an economist to think that’s a good thing. Who doesn’t?”

        From this we should conclude that Pareto improvement is a good thing right?

        Lets say we conclude that some market change provides a Pareto improvement. If this is a non-normative statement then it provides no ethical judgement on if this is good or bad or we ought to implement this change. How did the good thing get in there then without it being an ethical statement?

        Also I don’t think real Pareto improvements actually exist, they always have externalities.

      • #93 by Steve on October 28, 2013 - 1:19 pm

        The are different definitions of Pareto improvemt. Normally better off is short for prefers the new situation. Sometimes people prefer things that make them sad or unhappy, esp if they are mentally ill.

        If you define a Pareto improvement as something that is good or neutral to everyone it is a tautology to say it is a good thing, basically.

    • #94 by NicTheNZer on October 28, 2013 - 12:45 pm

      The normative assumptions to ‘pareto improvement’ are in the externalities. Pareto improvement is great if there are no externalities, otherwise it depends what you don’t consider, there is a trade off between the improvement and externalities. Does anybody have a case of pareto improvement to show which has no externalities?

  24. #95 by Luis Enrique on October 28, 2013 - 1:17 pm


    yes it’s normative, it’s just not very controversial. Like saying enslaving people and poking sticks in their eyes is bad. It’s also not controversial to say pareto improvements don’t often exist (usually people talk about potential pareto gains that would necessitate redistribution)

    • #96 by NicTheNZer on October 28, 2013 - 1:34 pm

      Can you suggest even one Pareto improvement with no externalities? Maybe even one with very few would be a good starting point?

      Actually you can find plenty of evidence that people believed enslaving people and poking sticks in their eyes was not bad, circa when people were doing exactly these things. You can’t really get away from normative judgements here.

      Your Pareto improvement would be controversial depending on the externalities and also what you are judging to be good/bad. Does everybody really agree that that is a good/bad thing?

      • #97 by Steve on October 28, 2013 - 1:46 pm

        I don’t think you understand the pareto part. It has to make EVERYONE better off. If there are negative externalities that make other people prefer it didn’t happen but it didn’t make EVERYONE better off.

        The only real normative part is that sometimes we ask kids if everyone is better off when, for instance, everyone (as in, all people for all time) can catch more fish. But the fish that die are worse off (I guess?) as someone noted on a problem set. Usually we exclude the fish implicitly.

      • #98 by Luis Enrique on October 28, 2013 - 1:52 pm

        ” You can’t really get away from normative judgements here.”

        yes right. but there’s a big difference between “economics is riddled with questionable normative judgements” and “economics reflects some normative judgements that have been uncontroversial since enslaving people and poking them in the eye went out of fashion”

      • #99 by Unlearningecon on October 28, 2013 - 7:29 pm

        yes right. but there’s a big difference between “economics is riddled with questionable normative judgements” and “economics reflects some normative judgements that have been uncontroversial since enslaving people and poking them in the eye went out of fashion”

        This is sort of whiggish, but in any case it’s a matter of degree. Sure, the criterion of pareto isn’t controversial alone, but what about when people use it to avoid any real world distribution because it will make rich people “worse off”? You seriously think that doesn’t happen? Same with efficiency, growth etc. Sure, these things are ‘good’ taken in isolation, but they almost always have to be weighed up against other things, something that has been acknowledged in my courses at least.

      • #100 by NicTheNZer on October 28, 2013 - 2:09 pm

        So to take your Fishing example, well you stated (correctly) that its not necessarily good for the Fish. So its normative (we ignore the fish, that’s fine).

        There are other externalities though, like in the real world this might require that we limit the population of the world, or those who can consume fish (rationing), or the number of fish that people can consume out of fairness (a different form of rationing). Maybe in another case some people have a religious prescription against eating fish (say fish are sacred) and think this is immoral (similar religious prescriptions exist). Lots of kids don’t like the taste of fish (this is not irrelevant, as their choices will be limited by the abundance of fish). If somebody owns any of the current fish as their property then they are worse off (due to increased abundance, the value of their property goes down).

        If this is a simple way to say that Pareto improvements don’t exist in the real world then I agree.

      • #101 by Steve on October 28, 2013 - 2:35 pm

        if that is your definition of Pareto improvement then no economist thinks they exist.

        It is not an accident that you used a lot of economic modeling (trade-offs, interconnected markets, tastes and preferences) to come to this conclusion. It is the conventional wisdom in Econ.

      • #102 by NicTheNZer on October 28, 2013 - 2:36 pm

        @Luis Enrique, Well it is a value judgement.

        I think it closely resembles the value judgement that students of Greg Mankiw made when they walked out of his class in protest . Its quite clear that they didn’t say Greg Mankiw was advocating enslaving people and poking out their eyes with sticks. But he was making controversial normative judgements in his course (which they protested). The question was does economics make questionable value judgements (which it does, as you admit) not are these the moral equivalent of slavery and extermination programs, which they are not. This is far from isolated as an example.

      • #103 by NicTheNZer on October 28, 2013 - 2:48 pm

        I didn’t use any economic modelling to come to that conclusion, actually. I think you need to look back up the thread, the point was that Luis Enrique makes “a normative assumption (pareto) and claim that it “should be uncontroversial!” I don’t see how you could have made my point better for me.”. Clearly non-normative pareto assumptions don’t exist which you just said. There is always some level of controversy.

      • #104 by Steve on October 28, 2013 - 2:57 pm

        “I didn’t use any economic modelling to come to that conclusion…”

        Re: “If somebody owns any of the current fish as their property then they are worse off (due to increased abundance, the value of their property goes down).”

        So you independently developed the law of demand AND it doesn’t count as an economic model? So much for needing to stand on the shoulders of giants.

      • #105 by Luis Enrique on October 28, 2013 - 2:49 pm

        I have lost track of your point. I have been arguing that whilst regarding a Pareto improvement as desirable may be normative, it is not a “questionable value judgement”. At to anybody with their head screwed on

      • #106 by NicTheNZer on October 28, 2013 - 3:01 pm

        No, I have been arguing that the only non-normative content of a pareto improvement doesn’t exist in the real world. Any other content is all normative in some way, though the ethical framework may not be clear or defined. Economics makes normative judgements, also many of them are controversial. You can not claim its value free, or that its main-stream values are un-controversial.

      • #107 by NicTheNZer on October 28, 2013 - 3:18 pm

        @Steve, Oh you got me! I would never have realised that was possible without the economic model I keep on my desk for just these situations!

      • #108 by NicTheNZer on October 29, 2013 - 1:31 pm


        Seems to be a good example of senior economists engaging in “questionable value judgement”s, and also unscientific methods. The main basis for this appears to be the search for market efficiency.

      • #109 by Unlearningecon on October 29, 2013 - 4:38 pm

        I have to admit, Black’s self righteousness and meandering prose was too much for me.

      • #110 by jrkrideau on October 31, 2013 - 2:28 am

        Oh course slavery (West African – Americas) was a Pareto imrprovement: It gave those benighted savages a change to learn about Christan values and the enlightenment of European civilization. Heck a bit of torture, rape etc. was worth it.
        Sorry I have just been reading John Darwin’s “Unfinished Empire”.

    • #111 by Andres Carbacho-Burgos on October 29, 2013 - 11:42 pm

      I think it was Marx who make the distinction between “false” analysis (Smith, Ricardo, Mill), which comes to wrong conclusions because it starts with the wrong entry points or deduces from these entry points incorrectly, and “base” analysis, in which empirical data is either ignored or twisted in order to fit the economist’s (paid-for?) ideological bias (Malthus, Senior, etc).

      Even that fall short. I would add that it is more important to react properly to empirical data that contradicts one’s analysis/model.

      The problem with the positive vs. normative dichotomy is that it’s not even wrong, but doesn’t adress the real purpose of analysis. “Positive” economics assumes that economists undertake empirical analysis in a vacuum; that they collect data and interrogate it just because it hit them over the head. In the real world, economists undertake empirical analyses based on normative questions. Good economists do their best not to be biased in how they interpret empirical results. But to avoid bias in the nature of question that is posed, and that you want empirical data to answer, is not just impossible, but foolish as well.

      Secondly, the real acid test is how one reacts to empirical data that contradicts one’s analysis/model. The easier option, though not necessarily wrong, is to make the model more complicated by adding in secondary institutional detail. The harder option, not necessarily right but definitely more productive when it is, is to question one’s own preconceptions when reality smashes them to shards. One of the reasons I hugely admire Keynes (though he was undeniable a jerk and a snob in many ways), is because his economics changed in major ways in response to outside events; very few economists can make that claim.

  25. #112 by Luis Enrique on October 28, 2013 - 9:12 pm

    Further to #78 #80, I am glad I admitted incompetence, because I muddled, bafflingly, ergodicity and uncertainty. I’ll try again.

    One can have an ergodic model in which initial conditions matter for a long time, even if everybody ends up the same eventually. Or we may have a non-ergodic model that says poor countries will be poor forever. In both cases, we needn’t pay too much attention to the models very long run predictions. Non ergodicity is interesting, but it’s not a reason “economics is wrong”

    Similarly the knightian uncertainty criticism is a lot easier to make than it is to say so what next. You can make a know probability distribution very hard to predict, and if you have learning rather than ratex, it’s not obvious how important the differences between that and a model that somehow incorporates Knightin in uncertainty would be.

    • #113 by Unlearningecon on October 29, 2013 - 4:44 pm

      Similarly the knightian uncertainty criticism is a lot easier to make than it is to say so what next. You can make a know probability distribution very hard to predict, and if you have learning rather than ratex, it’s not obvious how important the differences between that and a model that somehow incorporates Knightin in uncertainty would be.

      Forgive me if I’ve misconstrued you, but here it seems you are almost implying that we want to be able to predict stock prices accurately in a way that is commensurate with their ‘value’. However, all we are trying to do is understand how the financial markets work and how investors price. To this end, we don’t need to – and cannot – ‘deal with’ Knightian uncertainty in a model, due to its nature: it is what we can’t know. Instead we can use the decision making processes used by investors to understand how the market behaves, for better or for worse.

      • #114 by Luis Enrique on October 29, 2013 - 5:25 pm

        er, yes I rather think so. Where did you get “implying that we want to be able to predict stock prices accurately in a way that is commensurate with their ‘value’” from? I have not written anything about any of that.

        I was responding to the idea that one of the things wrong with neoclassical economics is that it treats uncertainty as a probability distribution, and models decisions taken in the face of that (either by assuming probabilities are known or may be learnt about by experience), As opposed to modelling uncertainty in some fashion that cannot even be approximated by supposing people act as if they face a probability distribution, which is something I often read critics calling for (and actually see some economists working on).

        And I was suggesting that a model in which people face a sufficiently unpredictable probability distribution might do a good job of capturing decision making where people have much less of a clue of what to expect, so that this criticism loses its power.

      • #115 by Luis Enrique on October 29, 2013 - 5:31 pm

        [although not everybody would put models with learning under the umbrella of neoclassical. I think you would though, because it’s still within the optimizing framework]

  26. #116 by Luis Enrique on October 30, 2013 - 10:36 am


    thanks for long and constructive response. I am trying to argue something like: many heterodox critics of mainstream methodology are better seen as pointing towards areas for future research within the mainstream, rather than “mainstream economics is wrong because X”, and indeed there are lots of mainstream economists working in these areas. Also some of your points strike me as asking for models that are simply beyond our abilities, at least for now.

    Some responses to your responses.

    many of the axiomatic foundations are essential, but cannot be easily written down in a water-tight manner, and are full of contradictions.

    that leaves me none the wiser.

    a view of human decision-making that is too granular in both time and space and can therefore lead to fallacies that are similar to Zeno’s paradoxes.

    how important is that, in most applications? Don’t let the perfect be the enemy of the good.

    contraints … therefore shift over time,

    that strikes me as obvious, and already catered for in dynamic economic models.

    they cannot simply carry out an individual optimization exercise to decide whether to buy, hold or sell: they have to take into account the _unknowable_ reactions of fellow financial investors

    again, the importance of strategic interaction is well known and pervasive. There are various ways of dealing with it, that you may or may not regard as very satisfactory. Taken too far this sort of point just amounts to throwing up your hands and saying we cannot say anything.

    In the real world, there is no such thing as an exogenous variable.

    no but there are many applications in which it makes sense to treat various things as exogenous. Models that endogenize everything needn’t be better that those that leave some things exogeneous – everything is then determined by initial conditions (I think) – in many cases it’s more interesting to have a model in which you can ask what happens if *this* changes.

    decisions that seem rational from the point of view of static, given aggregate quantities can lead to serious instability once they are aggregated.

    what context do you have in mind? there are rational bubble models etc. in which individually rational decisions can leap to booms and collapses. But it’s not obvious that models of, say, how the labour market would respond to different forms of unemployment insurance would benefit from having “serious instability” added to the model. It all depends on what you are doing – and I agree in some contexts, the possibility of individually rational behaviour leading to instability is very important, and you don’t have to be “heterodox” to think that.

    the effects of peer behavior and persistence (“it was good last time, so I’ll buy it again”) on consumer behavior in mathematical models that don’t relay on optimization; note also that such models rely substantially on _past_ outcomes and therefore create the autocorrelation that we see in so many econometric studies.

    fine. obvious a more complete model of individual choice might include experimentation, learning and peer influence. This does not make me want to throw away more simple models, merely to appreciate their limitations.

    Heterodox models of business cycles rely substantially on unstable equilibrium dynamics to the present day. Not so much on multiple equilibria, which is harder to model, but on cases of cyclical or saddlepath instability.

    Right. And is that a good thing? A potential addition to Aulds list would imho somebody who just boldly claims that models with chaotic dynamics are superior to models with equilibria but which are continually perturbed. I mean I am open to this stuff when and where it makes sense, but it really isn’t obvious that it always does. I like Lance Taylor, though.

    In many models, heterodox economists start right off with difference or differential equations without first building any sort of static model. See chapter 9 of Lance Taylor’s “Reconstructing Macroeconomics” for a good work-through.

    the possibility of individuals making systematic errors over at least a short-run time period,

    I agree … but hasn’t mainstream economics been heading in that direction for a long time? I have a book called “Rethinking Expectations” edited by Phelps and Frydman. It has contributions from Woodford, Aghion, Evans and has praise on the back from Mortensen and Pissarides. These guys are the mainstream.

    • #117 by NicTheNZer on October 30, 2013 - 12:40 pm

      “A potential addition to Aulds list would imho somebody who just boldly claims that models with chaotic dynamics are superior to models with equilibria but which are continually perturbed.”

      The chaotic dynamics models are a super set of equilibria models in this regard. This means that if you were to build the same economic system as two specific kinds of model then we end up with a superior dynamic model. If the system is chaotic and dynamic then you can model this using a dynamic model, and if the system is not chaotic and tends to equilibrium then the dynamic model handles this too. On the other hand if the economic system tends to equilibrium then the equilibrium model can handle it, but won’t tell you much useful about the dynamic case. You might try to handle this using perturbation, but this doesn’t capture the model dynamics, and it seems unlikely this is going to produce a very good idea of what is going on in that economic system.

      “In many models, heterodox economists start right off with difference or differential equations without first building any sort of static model.”

      Because dynamic models encapsulate equilibrium behaviour if it is what falls out (you get a model which tends to some equilibrium stability level in this case). Why should anybody limit their model to static behaviour rather than going directly for the more generally applicable model?

      • #118 by Luis Enrique on October 30, 2013 - 1:39 pm

        ok, but this still hasn’t convinced me this mean economists should be using such models. What is the economic interpretation of whatever parameters or functional forms or whatever else determines the boundary between chaotic dynamics and dynamics that tend to an equilibrium? I can see that in some applications it might be meaningful and interesting to explore when this boundary is crossed, but that’s a long way from the “bad” criticism: mainstream economics is wrong because: chaotic dynamics.

        that other passage you quote was mistakenly left over from Andres, although you took at as a criticism when you thought I wrote it, I don’t think he meant it that way.

      • #119 by Unlearningecon on October 30, 2013 - 2:40 pm

        Brief reply: why do we have to start off with a static model?

      • #120 by Luis Enrique on October 30, 2013 - 2:47 pm


        I didn’t write that bit, it was a mistakenly repeated excerpt from Andres. I don’t really understand it, plenty of mainstream models start off dynamic. Although I guess people tend to tackle the easier problem first, and static is usually easier.

        here is a good example of some (in my opinion) very interesting ideas being introduced in a static model, presumably because the authors haven’t figured the dynamic problem out


      • #121 by NicTheNZer on October 30, 2013 - 7:20 pm

        Sorry, I didn’t notice this. The main reason I assumed you wrote this is that there is little difference between a perturbed equilibria and equilibria model.

        “What is the economic interpretation of whatever parameters or functional forms or whatever else determines the boundary between chaotic dynamics and dynamics that tend to an equilibrium?”

        There is not really a hard boundary here, but the economic interpretation is maybe similar to a pot boiling over (to make an analogy), you don’t want the pot boiling over in this analogy but you might want to heat the system up a bit. If its an equilibrium type situation then you heat it up (looking at some change in the economy, maybe a policy change) and the system contains feedbacks which stop it boiling over, or at least it takes a whole lot of effort to make it boil over. This system is inherently quite stable then, it tends back to the same value.

        On the other hand, it could be that you heat it up just a little and there is a kind of chain reaction. If you are in this situation then small changes can quite easily make the pot boil over. This system is not very stable and you want to understand the risks in this case.

        Its important in this regard to realise that a linear equilibrium model can not ever, because of its structure, produce an economic crisis like event. It needs to be at or near some equilibrium (they have two states, either they converge or they diverge and produce infinite solutions). We should also be pretty sceptical of the equilibria reached here, because if you are modelling a real dynamic thing with multiple equilibria then there must be an error in your modelling to convert multiple equilibria to a single one. Anyway if its a static model it can’t help to identify the instability in the economy. You need to impose exogenous shocks on the model to get any such thing out. This makes it obvious that its not modelling many of the dynamics in the economy, as the crisis came out of some process in the economy and was clearly not imposed (e.g it wasn’t a natural disaster which triggered it).

      • #122 by Luis Enrique on October 30, 2013 - 8:47 pm


        I can think of some economic applications where keeping something turned up high, analogous to the heat applied to a pot of water (say, interest rates being kept low) – – might cause the pot to boil over (say, inflation to spiral) but I think mainstream econ can easily handle that already. Remember mainstream economics does use non-linear dynamics, just not in the form that allows for chaos. I cannot immediately think of an application in where a certain level of something sets off chaotic dynamics. But that’s because it’s not obvious to me what in the economy does ever experience chaotic dynamics. GDP? Wages? Prices?

      • #123 by NicTheNZer on October 30, 2013 - 10:02 pm

        “I cannot immediately think of an application in where a certain level of something sets off chaotic dynamics.”

        I think Steve Keen is arguing that a high level of private debt can trigger a crisis, followed by debt deflation (actually Fishers hypothesis). There is actually a bit more to his argument than that though.

        In the contrary position, Rogoff and Reinhardt argued government debt does something to the economy, causing recessions (though I think their work was thoroughly discredited).

        It could be all kinds of areas though, some fairly mainstream arguments go that a too low level of inflation probably causes rising debt and crisis, and many MMT people say there is a clear pattern of too high government surplus followed by recession.

        Godley and Lavoie come to some interesting conclusions about the relationship between interest rates and inflation. Mostly this points to high interest rates boosting demand driven inflation (if the government issues all the money at least). Still if the CB is going to try to control bubbles this way they might want a better picture of if their actions are likely to lead to a downward spiral (though I think the correct conclusion is that interest rates are a weak lever anyway).

        The dynamics of a model are likely to lead to effects across the whole economy, not be isolated to one variable of such a model. So suddenly Krugman finds the economy in a liquidity trap but doesn’t know how it happened. Its like some of the old Friedman videos where he is explaining that there is a bank run, and it seems to be because people just worked out how banking works (and then proceeds to explain they can be fooled by slight of hand to settle it down again). Hilarious.

      • #124 by Luis Enrique on October 31, 2013 - 12:13 pm


        I should be clear that’s I’m not claiming models with chaotic dynamics are never going to be useful. In fact I am a big fan of agent based models, which are quite different from a bunch of differential equations, but still do interesting things.

        however, I am claiming that economists have generally not made use of models that consist of ODEs with chaotic dynamics, not because they can’t do the maths but because they have found them not to make much sense / be useful, in context.

        You mention models in which crises are triggered. This is an obviously useful thing, but it’s hardly a new or strange idea – check out standard models of currency crises, or public finance crises, for example.

      • #125 by NicTheNZer on October 31, 2013 - 6:20 pm

        “however, I am claiming that economists have generally not made use of models that consist of ODEs with chaotic dynamics, not because they can’t do the maths but because they have found them not to make much sense / be useful, in context.”

        The reason that economists have ignored ODE models are irrelevant to their usefulness. You could however provide evidence for this statement by showing the conclusion by economists that they tried ODE’s to solve some problem, but it could not do so (e.g not that it was too mathematically complicated, but that it didn’t work though the complexity was manageable).

      • #126 by Steve on October 31, 2013 - 6:41 pm

        Odes are all over economics, esp macro. See Ramsey, Solow, Romer, RBC, New Keynsuan models for instance just for the ones even non-macro economists all know.

        For micro, all the standard results in price theory are PDEs but they are pretty intuitive for comparative statics.

      • #127 by NicTheNZer on October 31, 2013 - 9:09 pm

        Yes, so all we need to do is convince them that there is a whole class of ODE models where the gradients don’t have to be solved for equilibrium conditions.

      • #128 by Unlearningecon on November 1, 2013 - 4:40 pm

        Having taken modules on ODEs from both the maths department and the economics department, it is quite apparent that economists only use them in a highly stylised way, looking for maximums and equilibriums. It’s really very different to the dynamic way a bouncing ball is modelled.

      • #129 by Luis Enrique on November 1, 2013 - 5:15 pm

        yes. the question is whether the form of ODE chosen when studying bouncing balls is useful when studying economics.

        some critics presume it is, but I really don’t think that’s very clear. If you want to have endogenous business cycles, the best model isn’t necessarily one that tries to model crisis, recovery, good times, crisis and repeat ad infinitum

      • #130 by NicTheNZer on November 1, 2013 - 5:55 pm

        “If you want to have endogenous business cycles, the best model isn’t necessarily one that tries to model crisis, recovery, good times, crisis and repeat ad infinitum”

        It is the best model if that is what happens (e.g if its accurate), on what basis could anybody possibly claim its not the best model? On the other hand (since there is no reasonable answer this this question), you are probably saying that you don’t want to create a model which perpetuates cycles (nobody wants to create a model which perpetuates cycles), but in order to attempt to limit cycles (or their impact) you need some idea of where they come from.

        The best model is the one which predicts the most real economic phenomena, so the question is do crisis occur endogenously (the existence of endogenous phenomena is purely empirical, having nothing to do with the structure of models). Anybody with basic scientific sense will be highly sceptical of a model of endogenous behaviour, which doesn’t function in an endogenous way, as this implies it doesn’t model the behaviour of the system and so we don’t understand from this model what is happening.

      • #131 by Luis Enrique on November 7, 2013 - 11:14 am


        I had been meaning to reply, saying something along the lines that I am not terribly familiar with all the endogenous financial crisis papers that have been written recently, and that maybe they do entail the sort of behaviour, and use the maths, you seek.

        this morning I came across this, which looks exactly like what you are talking about:


        I still maintain that different models serve different purposes, so not every model needs to try to incorporate crash, recovery, boom, crash (repeat) – some of them might do a better job just concentrating on one part. Plus lots of other interesting things happen in the economy other than crash boom repeat.

  27. #132 by Jamie on October 30, 2013 - 4:22 pm

    A typically combative post! Most of the comments have come from economists, so I thought it would be worth giving a non-economist perspective from someone who has helped a wide range of businesses and government departments to solve operational problems.

    Right at the heart of economics is the principle of division of labour. This principle works effectively only when specialists recognise that they are providing a service to their customers. That means there has to be agreement between supplier and customer on the scope and nature of the service; on its quality, timeliness and cost; and, crucially, on feedback loops from the customer to the supplier on the fitness for purpose of the service provided. When competition operates effectively in the private sector, the default feedback loop is that dissatisfied customers take their business elsewhere, although many firms provide other mechanisms to avoid this type of loss of business. In politics, the default feedback loop is that dissatisfied customers vote politicians out of office. Again, politicians try to avoid this eventuality by using other feedback mechanisms such as opinion polls.

    In the case of academic economics, I would argue that there appears to be insufficient agreement on the service that economists provide to the rest of society and on expected quality. Most importantly, there are no effective feedback loops.

    For example, there appears to be an ever increasing number of university students who complain about the quality of their education in economics. They complain because they are embarrassed that they are not being equipped to answer the questions that their friends and families ask them. What is the mainstream profession’s response to these complaints? I read a wide range of economics blogs. However, there are almost no posts discussing how academic economists could improve the quality of their courses or increase student satisfaction.

    Many (not all) academic economists appear to have little or no appreciation that they are providing a service to their students. They often come across as though they are doing the students a favour by teaching them at all.

    The same is true regarding communication between economists and the general public. The public looks to economists to help answer difficult economic questions mostly in the area of macro economics. In order to provide a useful service, economists need to meet certain quality criteria:

    • They must answer the question being asked, not the question they want to answer
    • They must express the answer in plain English. If a lay person needs a degree in economics in order to understand the answer then there would be no need to ask the question in the first place
    • The answer needs to be plausible. Economists can challenge the lay person’s prior beliefs but plausibility is essential. For example, when economists model the economy based on a representative agent, it feels as plausible as modelling an earthquake based on the behaviour of a representative piece of rock, or an avalanche based on the behaviour of a representative snowflake
    • The answer needs to be relatively consistent independent of which economist is speaking. When economists give totally inconsistent answers the onus falls back on the lay person to decide between the alternative answers. Again, without a degree in economics, that is impossible
    • When the answers appear to correspond mostly to economists’ own prior political beliefs then the answers are useless. Lay people are perfectly capable of forming answers for themselves based on their own political priors. Economists must provide objective answers based on evidence from the real world in order to add any value
    • If the answer is ‘I don’t know’ then that is the answer that should be given. Lay people respect seismologists who admit they can’t forecast earthquakes. The problem comes when economists claim to have knowledge and skills which they don’t actually possess in order to gain influence in areas such as policy advice.

    In my experience, most academic economists fail to meet most of these criteria most of the time.

    So what is my view of the debate between Chris Auld and yourself?

    Based on his post, Chris Auld appears to have little understanding of customer service. As a result, he is useless to people like me. (That doesn’t mean that he doesn’t know a lot about many areas of economics – that’s not the point). When he condemns lay people for making lame criticisms of economics then he doesn’t understand the required role of an educator.

    Does that mean that I am on your side? Up to a point. You understand that economists are providing a service. However, I am unclear whether you are arguing from the perspective of a dissatisfied customer or an alternative service provider. You appear to switch between the two roles. If you are a dissatisfied customer then you can’t answer my questions any more than Chris Auld. If you are an alternative service provider then I don’t really care what you think about Chris Auld.

    How do you think that post-Keynesian economists could improve their communications with the general public? How do the different strands of post-Keynesian thought fit together? What is your advice to the University of Manchester students who want a more pluralist education? Why is the general level of economics knowledge amongst politicians, media and the general public so low? How can economists claim that David Cameron has a poor understanding of economics when he has a first class degree in PPE from Oxford University? If we were discussing any other subject we would question the competence of the teachers. Should there be a mandatory economics course in schools? What topics should such a course include to help raise the general level of public debate? What would an effective introductory school-level post-Keynesian course look like? I’m not expecting answers to these questions in comments. I’m just posting the beginnings of an agenda of interesting questions beyond the endless bickering.

    • #133 by Unlearningecon on November 2, 2013 - 6:26 pm

      Thanks for an interesting comment. You are right that I come off as both a dissatisfied customer (which, as a student I am) and a purveyor of alternatives. I’m shifting more towards the latter recently, though, as it’s more constructive. Also, frankly, everybody knows the problems with mainstream theory, including economists (even though they would often hesitate to admit them). However, this response to Auld’s list was sort of “look, I can do it too!” That’s not to say I don’t believe everything on the list – the format gave me an opportunity to get it all down in condensed form.

      How do you think that post-Keynesian economists could improve their communications with the general public? How do the different strands of post-Keynesian thought fit together?

      Ha-Joon Chang once made me check my priors when he pointed out that, if any other ‘school’ of economic were dominant, it would probably be just as intolerant of other schools. So while I like a lot of post-Keynesian economics, I’m really for a ‘reality first’ method of teaching, followed by numerous relevant and plausible interpretations. For example with the business cycle you would have a marxist, austrian, monetarist, Keynesian and perhaps circuitist/Schumpeterian approach, but you would learn about the events themselves before these were introduced. (This also answers your next question).

      How can economists claim that David Cameron has a poor understanding of economics when he has a first class degree in PPE from Oxford University? If we were discussing any other subject we would question the competence of the teachers. Should there be a mandatory economics course in schools?

      Economics does indeed seem to have an unusual level of obfuscation surrounding it. From what I know of PPE (though things may well be different at Oxford) Cameron will have learned the basic Keynesian analysis and not much more, so his understanding of “economics” as currently taught is at best limited. In my experience few economists endorse Ricardian Equivalence.

      So do I blame economists for austerity? Not really. Ask an economist – even a relatively centrist or centre-right one like Tim Harford – about austerity in a slump, and they will say it’s a bad policy. In fact, taken as a whole, they’ve been one of the main groups of people arguing against austerity.

  28. #134 by Blue Aurora on November 2, 2013 - 5:27 am

    Although I’m obviously late to this discussion, I do have one thing to say about this part:

    6. When pushed, they collapse their theories and assumptions into ridiculously weak, virtually unfalsifiable claims (such as revealed preference, the Efficient Markets Hypothesis, or rationality).

    Still not quite getting there…spell it!





    • #135 by Unlearningecon on November 2, 2013 - 5:37 pm

      Although I’m also not keen on SEU, I’m interested as to where you’ve seen somebody “collapse it into a virtually unfalsifiable claim” ?

      • #136 by Blue Aurora on November 4, 2013 - 4:31 am

        Well, I failed to properly digest the first part of that sentence. Technically, S.E.U. decision theory is an extremely narrow and limited definition of rationality as it is used in the context of academic economics. So that’s why I named it here.

        As for S.E.U. decision theory being made by someone into a “virtually unfalsifiable claim”…well, does the following link from a while ago count?


      • #137 by Unlearningecon on November 4, 2013 - 1:07 pm

        Yes, he seems to do precisely that although commenters – even ivansml! – point out where he’s going wrong.

        Perhaps it’s time for me to look a bit more into alternatives to SEU.

      • #138 by Blue Aurora on November 4, 2013 - 3:35 pm

        Want me to help you with a literature review?

      • #139 by Unlearningecon on November 4, 2013 - 3:48 pm

        Go on, then – send me an email 🙂

      • #140 by Blue Aurora on November 6, 2013 - 4:13 am

        I sent it yesterday.

  29. #141 by Ramanan on November 4, 2013 - 6:10 am

    Scene during Diwali fireworks:

    Economist 1: Why do we see lights before we hear the sounds?
    Economist 2: Because our eyes are in front and ears behind.

    • #142 by Unlearningecon on November 4, 2013 - 1:02 pm

      That’s certainly a useful approximation, a benchmark if you will.

  1. Signs that let you know economists haven’t got the foggiest | LARS P. SYLL
  2. Thursday Morning Links | timiacono.com
  3. Advocates of Reason: 24 October 2013 | Economic Thought
  4. The macro foundations of micro (crossposted at Crooked Timber) | OzHouse
  5. 10 Friday AM Reads | The Big Picture
  6. The macro foundations of micro (crossposted at Crooked Timber) | PNCAU
  7. Assorted part 2 | Pearltrees
  8. Somewhere else, part 85 | Freakonometrics
  9. Six signs you're reading good criticism of economics - Evolving Economics
  10. 18 Signs Economists Haven’t the Foggiest | Enjoying The Moment