Mainstream economics — sacrificing realism at the altar of mathematical purity

3 Jun, 2022 at 17:44 | Posted in Economics | 25 Comments

e0f5b445c333de539ad33c6a63606b56This critique goes beyond the narrowly technical — that the workhorse neoclassical model of the economy was found to be lame when it came to running a real crisis race. The deeper critique is that these models, and the technical language that accompanies them, have played a role in policy and in society that has been disproportionate in two senses. First, disproportionate relative to our state of knowledge. Existing economic frameworks have shouldered a policy weight that is simply too great for them to bear, given the degree of uncertainty and fragility that surrounds them. Second, disproportionate because these frameworks placed an excessive degree of policy power in the hands of the technocrats wielding them …

Mainstream economic models have sacrificed too much realism at the altar of mathematical purity. Their various simplifying assumptions have served aesthetic rather than practical ends. As a profession, economics has become too much of a methodological monoculture. And that lack of intellectual diversity cost the profession dear when the single crop failed spectacularly during the crisis. This monoculture, it is argued, has also narrowed the economics curriculum in universities. This has generated an ever greater focus on the mathematical gymnastics of optimising models and too little focus on the everyday aerobics of how the economy functions. Accompanying this has been a neglect of disciplines that abut and illuminate economics: economic history, moral philosophy, money and banking, radical uncertainty, non-rational expectations. In short, neglect of the very things that make economics interesting and economies important.

Andrew Haldane (former chief economist at BoE)

25 Comments

  1. Nanikore said (in a post I agreed with up till then):
    .
    《take demand theory, if prices rise, often that will lead to reduced demand. (That such a relationship often exists is not controversial and has been known well before Anglo Saxon Classical Economic Theory was around).》
    .
    Why does demand for financial assets go up with price? If the stock market is at least twice the size of GDP, why ignore it in this “not controversial” model?
    .
    Does Fischer Black offer a much better story in “Noise”: prices are powerfully arbitrary?

    Comment by rsm— 19 Jun, 2022 #

    • The adverb was intentional – often. For sure in speculative markets asset prices can go up if prices increase. (Yet another example of how insisting on rational choice based theory distracts us from pursuing a proper understanding.) Many in the financial sector have pointed out that with surplus liquidity in markets asset prices have risen because QE has enriched asset holders, and more of these assets fall into their hands as only they can afford them, which further enriches them and pushes their demand up further.

      Comment by Nanikore— 20 Jun, 2022 #

      • If prices in asset markets (the self-proclaimed crown jewel of neoclassical theory) are that arbitrary (not exclusively signals of physical supply and demand imbalances), why not print and inflation-proof a basic income?

        Comment by rsm— 20 Jun, 2022 #

        • Like dealing with issues of inequality you are up against power, vested interests and politics. It’s like trying to get pension, taxation reform or containing the excesses of the financial sector.

          More on some of the things discussed here by the blogger and commentators:

          Comment by Nanikore— 20 Jun, 2022 #

          • And on some frustrations with economics teaching that would be familiar to more than a few:

            Comment by Nanikore— 20 Jun, 2022 #

            • @rsm a good documentary on how quantitative easing led to rising asset prices which interviews market players, was the BBC’s “The decade the rich won”. If you can access it, highly recommended.

              Comment by Nanikore— 20 Jun, 2022 #

    • Nanikore, the “gary’s economics” video underscores my leading question: bond prices rose as supply surged, because demand grew even faster; where does that leave the “often” “not controversial” model dictum that “if prices rise, often that will lead to reduced demand”? Are you really comfortable (like the professor gary interviewed) leaving out one of the largest and most liquid markets in the world, when you say “often”?
      .
      In other words, if one concedes basic concepts such as “price increases often destroy demand”, is one in effect letting the rich get away with very lucrative (for them) violations of that law, while imposing hardships on the poor through austerity, rate increases, etc. as one implements policies based on that fickle law?

      Comment by rsm— 21 Jun, 2022 #

  2. Is it ironic that the Fed (and BoE) are currently affirming their unwavering policy belief in the Quantity Theory of Money and other tales about having to destroy demand because inflation unequivocally means scarcity?

    Comment by rsm— 4 Jun, 2022 #

  3. Thank you, Andrew! Those of us who have gone against the grain have been long ignored by the orthodoxy. But we continue nonetheless. I agree that the key problem is the sterile framework of the orthodoxy, which math only disguises. Math is the Latin of this High Church, but one should not blame Latin for the doctrines of the Church.

    Anwar

    Comment by shaikh945— 3 Jun, 2022 #

    • Relinquete omnem spem, o Intrantes

      Comment by Henry Rech— 4 Jun, 2022 #

      • Not at all. Inferno is a pretty cool place, so to speak.

        Comment by shaikh945— 4 Jun, 2022 #

    • Shaikh I agree that maths disguises the problem, very true what you say, but I would argue that maths not only disguises the problem, but is also very much at the heart of the problem.

      When we are dealing with issues that really relate to matters of moral philosophy, human psychology and societal issues mathematics is not an appropriate tool. It “fences off” the really interesting, important questions:

      https://larspsyll.wordpress.com/2022/05/09/enlightenment-and-mathematics/

      Economics is in a 1950s Idealist/Modernist/Rationalist time warp from which I think its excessive faith in mathematical methodology and ‘technological’ solutions is derived. For some reason unlike other social sciences it never got out of it. The consequences of an insistence on formalisation at all costs has been a barrier to real understanding, and what has been understood has been forgotten or relegated to the margins and the heterodoxy. As Haldane alludes to, the consequences for our democracies are becoming very serious.

      Critical Theorists such as Adorno have argued that the development of the technocracy of which Haldane describes is very much part of the Capitalist Dialectic. Whether they are right or wrong, the thing is that they ask the important questions. And that is what is missing in economics.

      Comment by Nanikore— 4 Jun, 2022 #

      • “A just machine to make big decisions
        Programmed by fellows with compassion and vision
        We’ll be clean when their work is done
        We’ll be eternally free, yes, and eternally young, ooh”

        Comment by Nanikore— 4 Jun, 2022 #

    • I would say there is nothing wrong with going against the grain, but I would also say there is nothing (entirely) wrong with the grain itself.
      .
      It seems to me Lars wants to throw the baby out with the bathwater – Lars makes out as if there can be only one winner.
      .
      Lars argues that mathematized and highly formalistic neoclassical theory based on abstract models is not relevant to reality. It seems a fair enough argument to make. So chuck it out. He’s argued elsewhere that statistical models are only representative of reality (to a certain degree) in a given moment of time. He argues there can be no expectation that a statistical model will reliably apply in some other time frame. This seems to be a valid criticism. So chuck them out. Even anecdotal historical economic explanations, relevant in one time frame, cannot necessarily be expected to reliably apply in another time frame.
      .
      So what’s left in macroeconomics – not much, if anything. Macroeconomists may as well pack their bags and become astrophysicists or street sweepers.
      .
      I would argue that the orthodox (neoclassical general equilibrium theory) and the heterodox (let’s call it Keynesian theory) work together to explain the functioning of the macroeconomy.
      .
      Neoclassical general equilibrium theory is essentially about the resource allocative process based on relative price setting. It assumes full employment of resources. There can be no optimal setting of allocation without a constraint and that constraint is the full employment of resources.

      So it assumes away the basic macroeconomic problem, i.e. how is the level at which an economy functions determined?
      .
      Keynes developed a theory to answer this question. He developed a theory explaining how the level of output is determined. He demonstrated how an economy can be in macroequilibrium at any level other than full employment.
      .
      However, having been nurtured on the Theory of Value, Keynes understood that an economy at full employment functions as explained by what he called the Classical Theory. That is the Theory of Value would come into its own at full employment. He said as much in his chapter dealing with the classical theory of the interest rate and in the dying pages of the GT.
      .
      So, I would argue that Keynesian theory explains the level at which the macroeconomy functions and neoclassical general equilibrium theory explains how resources are allocated and priced at the existing level of output of the macroeconomy.
      .
      Put another way, Keynesian macroeconomics explains where the production possibility frontier resides and neoclassical general equilibrium theory explains how resources are allocated along this production possibility frontier.
      .
      Both are required to make sense of the functioning of the macroeconomy.

      Comment by Henry Rech— 5 Jun, 2022 #

      • No finance? How does the model explain a quadrillion in dollar assets versus a tenth of that as measured noisily by GDP, etc.?
        .
        Is the current price of gas due to scarcity or willful cartel policies and options market momentum traders buying out-the-money calls to gamma squeeze prices up?
        .
        How can you dismiss the quantitatively dominant financial sector?

        Comment by rsm— 5 Jun, 2022 #

        • RSM,
          .
          Finance is a funny little (big) game played on the side by otherwise useless people playing with the surplus of eons of accumulated wealth. (I sort of was one of these useless people).
          .
          The strategies gunslinger traders employ to play their game can only work because there is scarcity. It is the scarcity that runs the show, not the other way round.

          Comment by Henry Rech— 5 Jun, 2022 #

      • Haldane says

        “This has generated an ever greater focus on the mathematical gymnastics of optimising models and too little focus on the everyday aerobics of how the economy functions. ”

        Does the production possibilities frontier actually exist, or is it a geometric gadget that allows economists to think in terms of constrained optimisation and go off on their modelling?

        ‘Keynes understood that an economy at full employment functions as explained by what he called the Classical Theory.”

        Can you tell me when we had full employment such that classical/neo-classical conditions operated in the real world? Perhaps the early years of the Bretton Woods period? Perhaps those conditions were historically very exceptional and obviously not stable? And even then, were classical/neo-classical conditions really operating?

        I could look at the pre-war Japanese economy and says we had a lot of price flexibility, and it superficially looked ‘classical’. And many people have done this ( a predictable favourite is to show that random walk properties were evident in prices and interest rates). And I think this is really the point that Haldane is making. You can look at something and see Dear Model. But is it really telling us what we need to understand about capitalism and how to deal with its problems? I would say, as Haldane does, that requires thinking beyond things like constrained optimisation. It ventures into territory such as geopolitics, moral philosophy and psychology.

        Comment by Nanikore— 5 Jun, 2022 #

      • Nanikore,
        .
        “Can you tell me when we had full employment such that classical/neo-classical conditions operated in the real world? ”
        .
        Of course in reality, it is difficult to define full employment. I would say a basic test is where inflation begins an acceleration. (I believe in the Phillips Curve, muted as it has been in recent decades by globalization). At some point resource useage approaches completion.
        .
        “And even then, were classical/neo-classical conditions really operating?”
        .
        This is a facile black/white point of view. Reality, as you very well know, is a good deal more complex than the notion that there is full employment or not.
        .
        “But is it really telling us what we need to understand about capitalism and how to deal with its problems?”
        .
        I am not making this argument. We need a basis, a framework against which facts about economic behaviour can be understood. (Sometimes we can’t even agree about the facts, adding to the complexity.) These understandings will be imperfect given the nature of models. However, this is not an argument, as far as I am concerned, for jettisoning the whole kit and caboodle.
        .
        “…. that requires thinking beyond things like constrained optimisation. It ventures into territory such as geopolitics, moral philosophy and psychology.”
        .
        Yes, I agree with you. But it also requires some kind of foundation in economic principles to which the above factors can be applied. With baby gone, the bath water is not of much use either.

        Comment by Henry Rech— 5 Jun, 2022 #

      • Nanikore,
        .
        “Does the production possibilities frontier actually exist, or is it a geometric gadget that allows economists to think in terms of constrained optimisation and go off on their modelling?”
        .
        Yes it is that in the first instance. And then you ask how relevant is this to reality. It seems to me that the constrained optimization model explains much economic behaviour. But it is not the sole explanatory factor.
        .
        You (and Lars) want to turf the lot out. What are you left with? Historical, sociological and political facts (if you can agree on them) that have no co-ordinating foundation. Then there is the question, even if you can agree on these facts, what is their use in understanding and applying them to circumstances in other spaces and time? How can their relevance be adjudged?

        Comment by Henry Rech— 5 Jun, 2022 #

        • Henry, may I submit as evidence of arbitrary pricing two articles from today’s zerohedge:
          .
          https://www.zerohedge.com/markets/hedge-fund-elliott-sues-lme-456-million-over-losses-nickel-trading-halt
          .
          《prices more than doubled to over $100,000 a tonne in a matter of hours on March 8 […] a 250 per cent surge in the price of nickel to a record $100,000 a tonne that was triggered by a short squeeze […] The fund’s founder accused the LME of “reversing trades to save your favored cronies and robbing your non-crony customers”.》
          .
          What part does physical supply and demand play in those price movements?
          .
          Second article: https://www.zerohedge.com/markets/crude-hits-3-month-high-after-saudis-raise-oil-prices-more-expected-easing-chinese
          .
          《In another sign of improving sentiment toward oil, Bloomberg’s Sungwoo Park notes that net bullish bets by hedge funds are recovering, back near the levels seen before the invasion of Ukraine. What’s more, oil is effectively shrugging off some bearish news flows over the weekend around Iran and Venezuela supplies. This bullishness has more life in it, especially if those hedge funds continue to return.》
          .
          Are the hedge funds trading on psychological expectations and momentum, greatly exaggerating price movements in oil way beyond what physical supply and demand would warrant?
          .
          Bonus article: https://www.zerohedge.com/economics/absurd-market-hypothesis
          .
          《the Absurd Market Hypothesis: 25 Instances That Prove Samuelson and Fama Wrong》

          Comment by rsm— 7 Jun, 2022 #

        • Robert,
          .
          I’m sorry but I don’t think much of your linked articles. They don’t do anything for me.
          .
          And oil is not a useful market to look at. It is driven by politics and run by a bunch of oligopolists who can swing it around any way they want, mostly.
          .
          Trading/speculation does increase the volatility of markets but in the end the fundamentals have to rule.

          Comment by Henry Rech— 7 Jun, 2022 #

      • Henry,
        .
        Why is the dollar rising against international currencies, even as the Consumer Price Index rises? Does your theory fail to explain why more dollars makes dollars more valuable, due to the vast worldwide dollar demand completely left out of your account?
        .
        See https://fred.stlouisfed.org/graph/?g=QbRO for a graph of CPI and dollar index: why has the world dollar maintained value even as inflation increases?
        .
        Are you creating a strawman “throw out the baby” argument, because we can adopt a money view where arbitrary financial decisions made by select individuals for deeply personal reasons affect prices more than physical supply or demand?
        .
        In other words, why not view scarcity as created? If the Fed signals it wants tightening because Dear Model, will financing for real production (housing?) collapse, thus creating a real shortage by imposing a money scarcity?

        Comment by rsm— 5 Jun, 2022 #

      • Robert,
        .
        “Why is the dollar rising against international currencies,”
        .
        If I knew for sure I’d be a very wealthy person. (At the moment, it’s probably got something to do with Ukraine.)
        .
        “…where arbitrary financial decisions made by select individuals for deeply personal reasons affect prices more than physical supply or demand?”
        .
        Please show me hard evidence for this assertion.
        .
        “…thus creating a real shortage by imposing a money scarcity”
        .
        So what? That’s what the CB thinks it’s there to do.

        Comment by Henry Rech— 5 Jun, 2022 #

        • “Why is the dollar rising against international currencies”

          Because it is a safe-haven currency. Why is that? Because since 1945 the United States has been the hegemonic power that underwrites the peace and prosperity of the new world order.

          It’s to do with power. Political power. Geopolitical arrangements.

          Power. Crucial. This is no less a baby in the bathwater than rational choice (utlility) theory. In fact as far as relevance goes, it probably is very much more so.

          It requires a multi-disciplinary understanding and good historical knowledge. As an analogy think of case studies and how lawyers work.

          As Chomsky said in our previous discussion on these issues, what is important is that you know what are the right questions.

          However if you were familiar with events that saw the collapse of the Gold Standard and Sterling as reserve currency, you would likely to get a step closer to some of the real answers. This was related to the dying days of the British hegemony.

          And the more historical case studies you are aware of, the more likely you are to get closer to understanding such issues when they arise in contemporary space. (And the more cautious you will also be – you will find you will be less prone to jumping to conclusions than if you start with Model or ‘choose models”) You will be very much aware that each historical episode is likely to have its own uniqueness and the similarities are often only at a very superficial level.

          As Bartrand Russell said on a link posted on this blog, what matters in the end are facts – even if those facts cannot be universally applied to all time and space.

          Some international relations theory (particularly Realist and Neo-realist theory which also serves as good critiques and counter-arguments to Classical Liberal and Neo-Liberal theory which like Classical and Neo-classical economics are underpinned by a foundational philosophy of Utilitarianism and Rationalism will also be relevant. As will a lot of theories from other disciplines. As will Neo-Marxian theory. Do not spend 90 per cent of your time learning Neo-classical theory. Being familiar with it is fair enough. But it is not the baby in the bath water. It is one of many elements in the bathwater that may help you get the right leads. The more elements you are aware of, the better.

          But the point is ultimately you are going to have to trace the causation from the evidence, ground up each time. And it will be a lot of trial and error. Post modernist theory and particularly Deconstructivism and other areas influenced by Linguistic Theory is another exxellent form of training. “If it is not this, what could it be.” “Where, when, how and why did this idea come about and where, when, how and why did this idea become the dominant form of thinking?

          Basically ask questions.

          Likewise inflation, is usually not just about things like production frontiers, and hyper inflation almost never just about that. A lot of it is to do with factors that are psychological which rational utility choice theory distracts us from understanding, not help us. Rational choice theory might give us an answer but is it the right answer? This is what Adorno means by formal models “fencing off” where we really need to be doing our thinking. Periods of high inflation are sometimes connected with supply issues, but you will find many examples through history where you cannot really explain them that way: instead supply issues if it is part of them is only one factor in a complex set of interrelated factors.

          Often they are too do with weakening of states which debases their currencies and a lack of faith in powers to (implicitly or explicitly) guarantee things like trade routes and trade finances. Currencies and financing are basically based on the trust, authority and integrity of a system. They also happen when systems start to break down, but I do not see much use in incorporating these issues which might have some relation to supply into geometric gadgets like production frontiers. Again it fences off deeper understanding.

          But either way when you look at all these issues, you must look as much evidence as you can. If you leave out material, and you have made it clear that you have done this and justified it, you must be sure that even if it is a significant factor, your line of argument that establishing a route of causation is not impacted in way that makes your own identified route of causation weak or questionable.

          You cannot just resort to theory as a way out.

          On classical theory, OK take demand theory, if prices rise, often that will lead to reduced demand. (That such a relationship often exists is not controversial and has been known well before Anglo Saxon Classical Economic Theory was around). But I do not think we need geometric gadgets with questionable behavioural theory on top of saying all that to get down to work. But you will need to trace the causation. Do consumer organisations, businesses etc identify price rises in certain markets as causing reduced demand in those markets, if not what have their members identified as being the cause? What factors are their own members potentially missing? How do I go about finding (quantitative and non-quantifiable) material on that?

          You will not get perfect answers, and you must identify where they may not be, but you will get much closer than you will by arbritarily creating a whole of fiction so that you get tautological geometric and algebraic gimmicks.

          Perhaps this quotation from Keynes (1937) is right: classical theory can tell us about how markets can work; but when we are dealing with humanity and society we have to ask much deeper questions and put the abstraction aside:

          “Perhaps the reader feels that this general, philosophical disquisition on the behaviour of mankind is somewhat remote from the economic theory under discussion.

          But I think not…

          “Tho this is how we behave in the marketplace, the theory we devise in the study of how we behave in the market place should not itself submit to market-place idols. I accuse the classical economic theory of being itself one of these pretty, polite techniques ….

          I dare say that a classical economist would readily admit this. But, even so, I think he has overlooked the precise nature of the difference which his abstraction makes between theory and practice, and the character of the fallacies into which he is likely to be led.””

          Comment by Nanikore— 19 Jun, 2022 #

          • Nanikore,
            .
            Models are a beginning, perhaps a poor beginning, but a necessary beginning on which can be built a thorough analysis once real world factors are studied, as you propose.
            .
            I would not suggest throwing out the baby, but the bath water certainly does need a change after each situation analyzed.

            Comment by Henry Rech— 19 Jun, 2022 #


Sorry, the comment form is closed at this time.

Blog at WordPress.com.
Entries and Comments feeds.