Cambridge economics has died out

16 Jan, 2022 at 13:52 | Posted in Economics | 9 Comments

A couple of weeks ago yours truly had a review of Diane Coyle’s Cogs and Monsters in WEA Commentaires. As I wrote, there’s a lot in the book to like, but unfortunately also some things very hard to swallow. James Galbraith seems to argue along the same lines in his Project Syndicate review:

Cogs and Monsters: What Economics Is, and What It Should Be: Coyle, Diane:  9780691210599: BooksCoyle subscribes to the grand illusion that price adjustment is the economy’s prime mover. But as the Cambridge Keynesian economist Nicholas Kaldor noted in his slim 1985 book, Economics without Equilibrium, “the intuitive belief that prices are the key to everything” is simply wrong. The foundation on which Coyle places modern mainstream economics is a myth …

When I attended the University of Cambridge in 1974-75, I read Keynes, met Piero Sraffa, listened to Joan Robinson, and studied with Kaldor, Luigi Pasinetti, Richard Goodwin, Ajit Singh, Wynne Godley, Robin Marris, and Adrian Wood. Back then, it was understood at Cambridge that markets do nothing like what Coyle claims they do. Just as Einstein had erased Euclid’s axiom of parallels, Keynes’s General Theory had long since obliterated the supply curves for labor and saving, thereby eliminating the supposed markets for labor and capital.

It followed that the prices of production were set by costs (mostly labor costs and interest rates), while quantities were determined by effective demand. Markets were not treated as if they were magical. It was obvious that most resources and components did not move under the influence of an invisible hand. Rather, they moved according to contracts between companies on terms set by negotiation, as had been the case for more than a hundred years …

But the Cambridge school of economics that understood these things has died out. It was targeted in the great intellectual purge of the Thatcher era, and it was pried from its footholds in North America by early-stage McCarthyism, Reaganism, the MIT self-proclaimed Keynesians, and the Chicago School. Only a few scattered survivors remain today …

Coyle concludes that “economics needs to change.” She is surely right about that. But it is impossible for economics to advance as long as it remains anchored to the mainstream bedrock on which Coyle’s own training was based …

Cambridge has forgotten Cambridge, and it is poorer for it.


  1. Galbraith writes:
    “The foundation on which Coyle places modern mainstream economics is a myth …”
    I think it is worth remembering that Keynes himself said that classical theory came into its own when describing an economy at full employment. At this point, changes in relative prices, effect changes in resource useage and allocation. However, they do not have an impact on the overall level of output. And changes in the level of spending cannot change the level of real output.
    At below full employment, changes in the level of spending will effect changes in real output.

  2. “The great virtue of neoclassical theory is that it attempts to be systematic at the foundational level.

    I am not sure this a virtue, let alone a great one. Basically you have to put down the facts as we know it. Even if those facts contradict each other they need to be stated. If they cannot be reconciled, so be it. Just be aware of them and say so. That is the state of our knowledge.

    Really this is the nature of the humanities, where much of the study of capitalism ultimately belongs even if MIT doesn’t like it.

    It’s also a bit like psychology. Would you send your child who might have mental health issues to someone who ‘has a model’ or ‘a consistent theory’ over someone who does not but appreciates the complexities and contradictions of human behaviour and who is widely and deeply informed?

    • Well, I can certainly understand that your position simplifies life. Just the facts. And what are those facts? What makes them facts and not just claims? Presumably some evidence that you might have to address and sort out (though that comes dangerous close to having a theory of why some “facts” are more reliable than others, which is a rather important issue in this political climate, don’t you think?). How would you choose between cancer treatments, I wonder? And these, by the way, are backed up by theories of how the body works, and practices based on evidence. As for psychology, there you would have to choose among competing claims backed by evidence on how different pathways that affect the mind and behavior. These, are theories, believe it or not.
      I am an economist because I believe it is important to understand the effects of policies and practices. And of course there is debate, which is a good thing. My book is a fact too, and it is full of facts by the way. Still, I must say I admire the efficiency of theoretical position.

      • The facts will be determined by primary and secondary documentation. for example if you are looking for the cause of an interest rate rise in a certain place or certain time it might be minutes of a meeting in a bank or central bank.But the point is you do you have to caste your net widely: you need to put together a lot of this material from a lot of sources. You do not draw on a theory as evidence. Cancer treatments are based on evidence, as you say. The theories of causation are determined after a lot of analysis of cases. This is not the way sticky price rational expectations models for example are created. Economics, a major subject created to understand capitalism is an oddity in its use of fiction and simplifying assumptions. I don’t know any other subject that works this way unless they are laboratory subjects which can justifiably do so. Capitalism is complicated. But you must confront that complexity, like an historian does. If not major questions will be left unanswered and there will be arbitrariness. For sure it requires books like Das Capital or the General Theory. These books do not try and escape issues for the dubious aims of formalistic tractability. Another important issue is the one of universal laws. Keynes was a sceptic-you cannot remove time and place from analysis which is what classical and neo-classical theory tries to do. In the end recourse to neo-classical theory is inefficient: its fiction distracts you going to the facts (the evidence) and putting the pieces together. In linguistic philosophy, this is a clear case of the subject determining the analysis; not the object.

        • Funny that you should mention Marx and Keynes, who are central sources of my work. And btw, they strongly believe in theory, as you know. In any case, since you have dismissed my work without having even tried to look at it, I don’t know what more either of us can say. Best wishes.

          • I only questioned the supposed virtue of neo-classical theory that you raise.

            Your work is not dismissed: from what I have seen is important and of great value. Believe me the work you and others like you are doing is very much appreciated.

            I also appreciate the remarks I made need a lot of qualification: sure the General Theory is a general theory as the title makes clear. But his work also came with a lot of warnings about ‘pretty and polite techniques’; capitalism is ultimately unstable and continually changing and changeable with a crisis “always lurking beneath the surface”.

            There are a role for theories. I just believe that a particular theory must not determine analysis. For those reasons I think they way others work, for example historians, is more appropriate in the study of capitalism.

            For example there are theories of conflict, the Structuralist theories in international relations, for example. But to understand how a particular conflict happened you put those theories aside. You understand causation, why and how something happened or happens by following the facts (events, pieces of evidence) and putting them to together.

            If you wish to say it confirms or rejects a theory (Marxian, neo-classical or whatever) that observation is made after the analysis is done. As in history, hypothesis testing is usually not the appropriate way to approach the task.

            • Thank you for your kind words, and for your clarifying comments. I certainly agree with you that neoclassical (NC) theory is not a good way to go. But its claim of generality derives from its procedure of moving from some basic principles to more concrete patterns of micro, macro and general equilibrium. Given this, it is tempting to go the opposite way and overemphasize the particular and the conjunctural.

              I believe strongly that basic principles are fundamental, and following the Classicals and Marx, my task has been to show that the basic principles continue to operate in a powerful manner when we introduce more concrete factors, and that this procedure explains concrete patterns in a better way than any other framework. It is the purpose of my (damn) book to show this procedure of moving from the abstract to the concrete to the empirical can be done in a unified manner, but obviously in a manner fundamentally different from NC theory and post-Keynesian (PK) theory. I try to do this for commodity prices, stock and bond prices, exchange rates and patterns of trade, unemployment, inflation, growth, etc. In each case, the theory is compared to NC and PK arguments and to the empirical evidence.

              There are of course historical interventions that override basic forces and patterns, as in wars: Nazi Germany and the Roosevelt war-economies explicitly suppressed wage, price and interest rate increases despite of huge increases in production and in (deficit-funded) aggregate demand. I point out that this can explain the price, profit and wage patterns in the war era and the subsequent stagflation period in the post-war era after the suppressions were ended. It is not a case of ignoring history, but rather of taking into account the fundamental economic forces that drive the moving stage on which history plays out.
              Such tasks are never finished, but that does not mean that other patterns negate basic principles. Let me try to illustrate this with some more recent applications. Econo-physicists (most notably Victor Yakovenko) have recently found that within developed countries the empirical distribution of wages is always near-exponential, and that the distribution of profit rates is always close to a power law. What can account for this, given the well documented political and institutional differences over time and across developed countries? We know that the Classicals and Marx studied capitalism in great empirical detail. Still, at an abstract level they all began from the notion of equalized profit rates and corresponding prices of production. At the same time, further on in the very same works they all emphasize that at a more concrete level it is obvious that this equalization is a non-ending turbulent process that never settles down to equality. Marx explicitly notes that each sector will therefore go through cycles of “fat-and-lean years” whose patterns are industry-specific and time-specific. Thus at the empirical level, we have to see whether profit rates are indeed equalized in this up-and-down manner, and I discuss existing empirical studies and provide quite a lot of evidence showing that this is indeed the case.

              But then how to we reconcile turbulent equalization with more recently discovered evidence specific persistent probability distributions of wages (exponential) and profit rates (Pareto) in every year? Here my co-authors and I were able to recently show that when we treat the equalization ups-and-down as random shocks, we can derive exactly the observed profit and wage rate distributions. They are specific at the individual industry level but statistical at the level of the whole economy. It turns out that the distribution of wages within gender and race cohorts also follow the same patterns. So the fundamental equalization forces can be shown to arrive at observed concrete forms. Note that many disciplines, from physical to biology do the same. We can describe the exchange of energy of collision of two particles from fundamental principles, but when we consider millions of particles in a contained gas in which the particles all collide again and again, we can show from basic principles that the resulting distribution of energies in the simplest cases are exponential (thermal) or Pareto (super thermal).This is a consistent move from “the abstract to the concrete”, to quote you-know-who. History is of course relevant, but how we incorporate history into theory also depends on the theory.
              My apologies for this long response and thank you for your patience and kindness. If you are interested in the econo-physics discussion, I summarize the previous findings and explanations, and provide my own argument based on fundamental classical economic principles in these papers appearing on my home page (

              Click to access Economic%20Arbitrage%20and%20the%20Econophysics%20of%20Income%20Inequality.pdf

              Click to access Race%20gender%20econophysics%20physica%20A%202014.pdf

              • Thank you for your response. And I wish you the best with this work and I look forward to your findings. The end of wartime command economies is a challenging topic and I am aware of the huge debate about how much the post-WWII system was a legacy of one and for all changes made during the wartime period.

  3. Couldn’t agree more with Jaime Galbraith. The notion that orthodoxy is more rigorous because it is more mathematical is nonsensical. Sraffa’s work published in 1960 was highly mathematical, as was the subsequent debate on both sides (Samuelson vs. Pasinetti, for instance). But that is not the point. It is the embedded vision of capitalism that is the key, and here neoclassical economics has the great political advantage of portraying capitalism as perfect in theory and essentially perfect in practice. Moreover, it is not a question of realism, since neoclassical theory is riddled with what it itself calls “paradoxes”: the Leontief paradox, Gibson’s paradox, the equity premium puzzle, the stock market excess volatility puzzle which was Shiller’s focus, the bond vs dividend yield puzzle, the purchasing power parity puzzles of exchange rates, the so-called puzzling behavior of individuals, and so on. These are paradoxes and puzzles because they contradict the very fundamentals of the theory.

    Having said that, I have always argued against complaining about these many puzzles and paradoxes. The great virtue of neoclassical theory is that it attempts to be systematic at the foundational level. My own work, developed over several decades and published as Capitalism, Conflict, Crises (OUP, 2016) is an attempt to show that there is an alternative systematic theoretical and empirical framework developed from (one might say excavated from) key points in Smith, Ricardo, Marx and Keynes. It covers micro and macro economics, international trade, the financial markets, unemployment , inflation, etc. Fully half of the book is empirical. Jaime has been most supportive of this project.

    Most of all, I call on those of us who are critical of neoclassical theory to leave behind the notion that theory is perfect and reality is imperfect, since imperfection is merely the dual of perfection.

Sorry, the comment form is closed at this time.

Blog at
Entries and Comments feeds.