KEY:
I: Interviewer Dr. Constantinos Repapis
R: Respondent: Professor G.C. Harcourt

I: We're here with Professor Geoffrey Harcourt, Emeritus Reader in the History of Economic Theory at the University of Cambridge, and Emeritus Fellow at Jesus College, Cambridge, and also Professor Emeritus at the University of Adelaide.

Professor Harcourt, thank you for agreeing to participate in this interview series organised by the Independent Social Research Foundation, and Economics at Goldsmiths. I would like to start, if I may, with the following question. What does it mean to be a Keynesian economist? 

R: To be a Keynesian economist first, not a post-Keynesian; I'll go onto that.

I: Yes, a Keynesian economist.

R: A Keynesian economist means a number of things. Keynes, in particular, put aggregate demand alongside aggregate supply in producing a new theory of the determination of level of employment and activity. And he claimed that Thomas Robert Malthus, whom he called the first of the Cambridge economists, had this idea, but was defeated in his debates with Ricardo, and so the whole concept of aggregate demand vanished for 100 years, and Keynes brought it back when he was thinking about: "How do I explain these prolonged and terrible levels of unemployment?" Both in the 20s, and even more so in the 30s. And he developed his new theory around the interplay of aggregate demand and aggregate supply, and resurrected the term that Malthus, amongst others, used, effective demand, where effective demand was the point where aggregate demand and aggregate supply were equalised, in the short period. If you don't like the phase short period, use rest state. This implied a rest state where what business people thought people were going to spend on investment goods and consumption goods, and what business people were willing to supply in the existing conditions and at the existing price levels, and their expected profits, matched up.

When they matched there were two meanings to aggregate demand. One was in the minds of the entrepreneurs, the business people, concerning what people were going to spend. In the case of investment, a reasonable assumption, you could say, was that they knew exactly, as far as fixed assets were concerned, because they would have been placed as orders. What they didn't know exactly was whether they would achieve or not achieve their planned rates of inventory accumulation, which is also a very important part of investment, as we know. But they had to anticipate what the consumers would spend, out of whatever incomes they had, on consumption goods.

Then the all-seeing economist, macroeconomist, looking at the economy as a whole, would see a function for planned investment expenditure, and the aggregate consumption function, which showed, in the given conditions, at each level of personal disposable income, what consumers in aggregate would spend on consumption. And when all those three things matched up, you're at the point of effective demand, and that was a rest state unless something happened to change the level of planned investment expenditure, or the level or consumption expenditure out of a given level of personal disposable income.

I: But on some level the idea of demand was not new in Cambridge. Marshall worked on demand.

R: But not on aggregate demand.

I: Isn't it an obvious move from demand of individual firm to aggregate demand?

R: But it was only because he accepted, at least for the long period, the quantity theory of money, the level of activity was given at full capacity, full employment working. He implicitly assumed Say's law. So in the quantity theory, if we take the Fisherian version, you've got T, the total level of transactions occurring; that is the Say's law level, not the actual level, but the Say's law level in the long period. V is given institutionally and by the habits of the people in their use of money as what they need for transactions. V is assumed to be constant. So you've got V, T and it was assumed that the monetary authorities created M, the monetary supply, so there's one unknown, the general price level. Thus in the quantity theory equation, M and V and T determine P. Now Marshall, though he had a very subtle version of that, and distinguished between the short period and the long period, he accepted that, so there's not need to ask what determines total demand, because it's just a simple addition. And I mean, he was, what, eighth Wrangler or something like that, which means that adding up was a piece of cake.

I: As you say, the aggregate demand gives a completely different understanding of the economy.

R: Well, it's a new concept.

I: How does it change our understanding of demand, of what's happening at the micro level?

R: Well, it doesn't change it at all, because Keynes always argued that whatever market structures ruled, whether they be freely competitive, or imperfectly competitive, or oligopolistic, it made no qualitative difference to the total outcome. The systemic effects might vary because of different behaviour in market structures, but the overall picture of aggregate demand interplaying with aggregate supply was not affected. 

I: How is that possible? How is it possible that variations in the structure of the market would not affect it? 

R: Well, it affects it quantitatively, but not qualitatively.

I: This brings me to ask you the following, which you've said on a number of occasions.

R: I do repeat myself.

I: -Yes, with profit- and that is, you say that an important aspect of Keynes is that the sum is more than the parts.

R: The whole is greater than the sum of the parts.

I: That's more correctly put. What does that mean?

R: Well that means that you just can't add up micro-behaviour, firm and consumer behaviour, and get what the systemic outcome will be. The system has a life of its own. 

I: And you think that is something that economists generally agree with?

R: No, because they tossed it overboard when they brought in representative agent models, and Dynamic Stochastic General Equilibrium Models. Because in general equilibrium, the whole is the sum of the parts. That's why when you're doing micro-foundations, which I think is a fallacious approach, actually, the general equilibrium people say, or the extreme ones say, there's no such thing as macro-economics; it's just micro-economics added up to the economy as a whole.

I: Do all Keynesians agree with your understanding of Keynes in this aspect?

R: I've no idea. I would have thought that people like Victoria Chick and Sheila Dow would go along with it.

I: What about Paul Krugman and other modern macro-economists? 

R: No, but they're new Keynesian economists, and that's a misnomer. They see imperfections as the reason why you get Keynesian results of prolonged unemployment or prolonged inflation. But the Keynesians who go back to Keynes, through Joan Robinson, Richard Kahn, Lorie Tarshis, who all knew what Keynes was saying in his lectures, and understood what he was doing, and knew he was a Marshallian, not a Walrasian, they would certainly, I think, go along with this, and that's one reason why they didn't like IS-LM, because it was illegitimately bringing a simple Walrasian model into what was essentially a Marshallian setup and a Marshallian approach. And the persons who recognised this, after having first of all gone along with Patinkin and other general equilibrium versions of Keynes, were Clower and Leijonhufvud, who in the 70s said, "No, scrap what we've done. We're going back to do it in a Marshallian framework.” And then you have the recursive method, when you start off with one relationship and you move into the next one, and go right round the economy and come out with a true story at the end. Marshall says that's what he hoped to do.

I: The recursive method? 

R: Do you know my article in the Economic Record on a short-run model of employment and the distribution of income? 1 It was then that John Cornwell told me about the recursive method. I said, "What am I doing?" And he said, "Well, that's what you're doing, and that's the proper way to do it." Because he did that himself, over his lifetime.

I: You mentioned Leijonhufvud you mentioned other important theorists as well. If you have to explain briefly to a student the difference between the Keynesians, how would you explain it? There are a lot of economists who go by saying they're Keynesians.

R: Well, I tell them to have a thorough read of ‘A ‘Second Edition’ of The General Theory’, edited by Harcourt and Riach. 2 No, you'd have to point out that lying behind proper Keynesian analysis is the assumption that all important decision makers are doing it in an environment of fundamental uncertainty, so that expectations - both short-term expectations and long-term expectations - have a central role to play. It's in the light of people's expectations, and then the total outcome of what they do, we see whether expectations are realised or not. If they're not, then in Keynes's analysis there are a variety of different stories of how the decision makers react to the signals that are given out by the initial non-realisation of expectations. So you can tell a story of approaching the short period rest state as Keynes does.

In Keynes' case it was a price signal, because he assumed in the first case that you're operating in the Marshallian market period where there is a given stock, and the price that is eventually set clears the market, and then that reacts back on the short-period price, and so on. In more modern analysis, people said, "Well, let's suppose that you have market structures where prices are quite sticky or inflexible in the short term” and then the signal is given by unintended inventory changes. So if you've overestimated what demand is going to be, you'll find that you've over-invested in stocks. And if you've under-estimated, you'll find that you haven't achieved your desired rate of inventory accumulation. Now, so long as you assume that that doesn't feed back and change your planned inventory accumulation, then you get stabilising signals. Because as people either cut output or increase output according to the signals, they change output more than they change either increased demand or decreased demand, because of the leakage into saving, and so you hive in on the rest state. That is, you go towards the point where aggregate demand and aggregate supply, or aggregate investment and aggregate saving, in a planned sense, as opposed to an identity sense- are equal.

I: You started by a discussion about uncertainty, and you said that uncertainty is central -

R: Absolutely central

I: - in post-Keynesian economics.

R: Yes. Joan Robinson defined post-Keynesian economics as an approach which takes account of the fact that you can't escape uncertainty, so you have to make decisions about the future, which you don't know what it's going to be.

I: Do you agree with that definition?

R: It's a good start.

I: Is it a good finish?

R: No, well, you need to add more things on. But Joan was very pithy. She could sum up in a sentence what most people would take a whole essay to write.

I: You've discussed in your work -as have others- the difference between uncertainty and risk as being central in post-Keynesian theory.

R: Yes.

I: Could you explain why?

R: Well, risk can be handled by probability theory, and a definite number be put on it; uncertainty can't. Keynes has a famous description in his 1937 QJE article, 3 ending up: "We simply do not know." What is going to be the price of wheat 20 years on? When will the Third World War start? Well, we know the answer to that: if Trump becomes president. But leaving that aside, we don't know when the Third World War is going to start, or if it ever will.

I: That's the difference between uncertainty and risk, and as you say, a constitutive feature of post-Keynesian economics is uncertainty.

R: Yes, because all major decisions are made in an uncertain environment, and therefore depend on expectations.

I: But, Austrians also consider uncertainty.

R: Well, their approach is very similar. If you look at Lachmann's review of Hicks' Capital and Time, 4 it could have been written by Joan Robinson, except that their political and policy conclusions are diametrically opposed. But the understanding of the process is very, very similar.

I: What I was aiming at is the fact that they place the same central role on uncertainty. This means that this is not enough to describe the post-Keynesians, because other schools also consider uncertainty basic. What else would you use to describe post-Keynesian analysis? Why is it a school of thought other than the central role of uncertainty?

R: Well, because it also includes a very negative critique of mainstream economics, including the Austrians and their policy responses. But as far as positive attributes are concerned, the way of defining particular functions, like the consumption function, the investment function, putting in how you model exports, how you model the demand for imports, how you model the government's behaviour, they are all peculiarly post-Keynesian because they are based on observations rather than on axiomatic assumptions. 

I: So the relation with reality in some ways is central to the school. 

R: Very. It's summed up in Nicky Kaldor as ‘stylised facts’. You look at what the processes in the world are, and then you abstract to make up a set of relationships which have as inferences those stylised facts.

I: So do you think there are many ways to do post-Keynesian economics, or is there only one way?

R: Well, it depends whether you ask Paul Davidson or not.

I: Well, I'm asking you at the moment.

R: Well, I think we're quite a - what do you call it? - pluralist lot. Joan, for example, was a left Keynesian. Kahn was much more a liberal, I think. Austin Robinson was a pragmatist. Reddaway was, if anything, a liberal. But they all had a very similar approach. In the case of people like Austin and Brian Reddaway, they'd learned Marshall and they'd learned Keynes, and from that they developed everything else they ever did- very sensibly. 

I: What about Garegnani, for example, or Pasinetti. 

R: Well, they should not be lumped together. Garegnani was very influenced by Piero Sraffa, or his take on Piero Sraffa, and he had this view that the only refined formalistic theory you can do is about the long period. And it's very peculiar, because his school says you can only theorise about sustained and persistent forces. But there is no more sustained or persistent force than fundamental uncertainty, and that is ruled out in their long-period analysis, as you'll know from the work of Heinz Kurz and Neri Salvadori, in their great book on the long period. 5 And Ian Steedman to a certain extent, as well. 

I: Are they post-Keynesians?

R: Well, historically, yes. You cannot exclude Piero Sraffa from under the rubric of post-Keynesianism. Here I disagree with Peter Kriesler. He won't have a bar of including Sraffians as opposed to Sraffa, and of course Sraffa was very critical of parts of Keynes' theory because it was based on Marshall, and in turn that was based on subjective value, rather than the objective value theory of the classicals and his dear friend David Ricardo. "Please be kind to my dear friend David," or, "Don't be harsh on my dear friend David." But of course, Sraffa was a great admirer, most of all, of Marx, and he thought that Marx and his approach was the last word in how to understand capitalism, and that his own role was just to fill in some unfinished business, or correct some things that may have gone wrong in a minor sort of way. But the whole structure, and thought, and multi-dimensional system of Marx was his starting point. He thought, "That's it. I belong in that."

I: What about Pasinetti? Because you mentioned Pasinetti.

R: Well, Pasinetti is a unique figure, because he's a great admirer of the classics, and of Ricardo, and of Marx, and of Keynes. But he's also a very devout catholic, so that has put him aside from Garegnani and Sraffa, who loathe the church of Rome. In fact, Garegnani once told me, "Do not make jokes about religion, and especially the church of Rome. It's far too serious to joke about." Once, when Sraffa lost his temper with me when we were having a discussion about the review article that Vince Massaro and I wrote of his Production of Commodities, 6 he said the following when I said, "But Piero, last time we met, you agreed with what I just said." He said, screaming at me with his white eyebrows hitting the sky, "I am not the Pope! I am not infallible!" So there's a gulf there. I think that's why Pasinetti, who was much more positive in his contributions than Garegnani, was not chosen to be Sraffa's executor. But that's conjecture- that's what I think.

But Pasinetti, I've always argued, is probably the last of the great system builders in “our miserable subject’”, as Keynes called it. No one less miserable than Keynes would be hard to find. But Pasinetti has this view that you have to distinguish between fundamental relationships, which are independent of institutions, which exist prior to them. And then you do particular periodic analysis when particular institutions are dominating, and these principles are an a priori set. They're there underlying it, but they take a particular form, according to the historical period you're looking at. This is consistent with one of the greatest books ever written on the development of the subject, A K Dasgupta's Epochs of Economic Theory. 7  

I: So where would you put your analysis, your contributions, within this group of post-Keynesian economics? 

R: Well, as far as post-Keynesian economics is concerned, I've developed a structure based on Joan Robinson, Kalecki, Keynes, Marx, Smith and Ricardo, and Kahn, trying to understand the processes at work in a modern capitalist economy characterised by, on the whole, oligopolistic market structures, and the Keynesian and Kaleckian forces at work. And I've come increasingly to be influenced by the work of Richard Goodwin, which culminated in his growth cycle model, and putting together production interdependent models and aggregate cycle models, and late Kalecki, where he said, "The long-run trend is not an independent entity; it grows out of a succession of short-period happenings." So I think that's the best way that I know to understand what is going on in modern capitalism. And of course you have to then put in Marx's insight and follow it up by Keynesian insight as well, that finance capital and industrial capital, and commercial capital are all interwoven. And when finance capital is out of kilter with industrial capital and commercial capital, then you will get instability and often crisis. And I think that vision, as it were, makes far more sense of what's happened in the last 30 or 40 years in the modern world, particularly when it's allied with the rise of multinational oligopolistic industries, which dominate national governments, international trade, and borrowing and lending. 

I: Do you think your work has moved/changed from where it was in the past? If you had to periodise it in influences, how would you do it?

R: Well, I mean, in the 50s as a student I was bowled over by The General Theory. 

I: I understand, but in the 80s you worked a lot on Joan Robinson’s... 

R: Yes, but her structure of thought became more and more that of Kalecki. And that means going back to Marx's schemes or reproduction as a starting point, rather than the Marshallian tradition.

I: Some would argue that in An essay on Marxian Economics, 8 in the 40s, Joan Robinson read Marx in a Keynesian or Marshallian, way. 

R: Yes. Well, she never really completely understood Marx. If you read Jorge Araujo’s archival work on the exchanges between Shove, Dobb and Joan, whilst she's writing An essay on Marxian Economics, 9 shows that she just had a blind spot on what Dobb was telling her about the nature of what Marx was doing, and she never really fully got it. She was very hostile to the labour theory of value, even though I think in the end she was using it, but as an explanation of the origin and size of profits as surplus value and surplus labour in the sphere of production, she never agreed with it. However, the distinction between what is happening in the sphere of production, on the one hand, and what is happening in the sphere of distribution exchange on the other, is absolutely essential both to post-Keynesian economics and to Marxian economics. And, of course, that was never accepted by Sidney Weintraub or Paul Davidson. It was by Jan Kregel.

I: But it is a part of what you see post-Keynesian economics as being. 

R: Oh, very much so.

I: Do you think mainstream analysis is equally adept at dealing with questions of distribution?

R: No.

I: Why? What's the difference between how mainstream analysis does distribution with post-Keynesian distribution?

R: Well, the mainstream still has one form or other of marginal productivity theory. Not necessarily a crude aggregate production function, but nevertheless, lying behind it is the idea that rewards are related to variations at the margin, but there are of course overlays of monopolistic and institutional factor to explain- for example, why CEOs get such ridiculously high salaries, and so on.

I: Why do you think that's a problem when doing policy and real world analysis? Why can't we get good outcomes from that framework?

R: Well, for example, if you take Piketty's work, the formal structure is an aggregate production function, but that assumes marginal productivity rules in a rough way, and full employment. But if you put in the Keynesian and Pasinetti strands, as Lance Taylor's done, for example, then you're dealing with a much more realistic view of how the economy is working, and therefore what sorts of policies are likely to be effective. And also you're much more aware of the political constraints, because if you think that technical factors dominate the distribution of income, then it's much more easy to suggest common or garden incentives and disincentives as ways to affect the distribution.

I: Didn't Keynes say that economists should be humble as dentists? 

R: Yes. 

I: And therefore the type of technical analysis like that, just going in and doing a bit of estimation…

R: Well, technocratic is a terrible approach, because it forgets about power structures. What has become more and more obvious in the workings of the modern world is the rise of conflicting power structures that you have to take into account. People like Ken Galbraith understood that from very early on, and people like Duncan Foley and Tom Michl, their work on growth and distribution theory, and similarly Marglin's work with Bhaduri, they all recognise the fundamental structure that derives from Marx and the classicals feeding into the Keynesian views of the interrelationship of the financial, production, and expenditure parts of the economy. 

I: So you don't agree with the mainstream distinction between normative and positive economics? 

R: Not at all.

I: Why?

R: Because ideology and analysis are indissolubly mixed. The greatest book on this was by the wonderful Australian historian and polymath, Hugh Stretton. His book, The Political Sciences, published in 1969, 10 was on the interrelationship of ideology and analysis in the social sciences: economics, sociology, history, politics, law. Maurice Dobb had very similar views on the interrelationship of ideology and analysis, and so did Schumpeter.

I: So when an economist says, "I'm not responsible, I'm a technocrat."

R: Well, they're dodging.

I: Dodging?

R: Yes. Because it's like Chomsky's great essay in 1967, I think it was, in an otherwise not very distinguished collection called The Dissenting Academy. 11 He has an essay called ‘The responsibility of intellectuals’, and in that, he says it's no good making napalm and then saying, "Don't blame me for what it's been used for." You have a social responsibility to follow up what use is made of your findings. And I remember when talking to physics students at Adelaide, this didn't mean anything to them. They said, "We did physics, period." And I said to the professor of physics, who was a very old friend of mine, "Well, don't you ever give them lectures on the social responsibility of natural sciences?" He said no.

I: So to return to the question of: What is the job of an economist? As you say, he is not a technocrat...

R: Well, he's partly a technocrat, because particularly with modern inventions and econometrics and all that, it's become quite a difficult subject. Keynes said, "An easy subject at which very few excel," but that was probably true of the tools that were available when Keynes was writing. Now it's much more technical than it was, and we shouldn't be techniques Luddites. But the basic reason for doing economics, I think, is to end up with policy proposals to try and explain why the world is behaving or misbehaving, where the misbehaviour most falls, and then to define realistic policies which alleviate the impact of the misbehaviour of the economy on those least able to defend themselves. 

I: So what qualities do you think would make a good economist today?

R: Ah, well they're set out by Keynes in his essay on Marshall. 12 And in his essay on Malthus, 13 on the optimum path that he thought Malthus took, to ultimately become an economist. He had a sequel to his essay on Malthus in the Economic Journal, 14 I think, in 1933, where he's paying tribute to Malthus. It was a celebration of something or other in Malthus' life, and he sets out this wonderful paragraph of how Malthus followed the right path to become a very good economist. And if you ally that with the traits you have to have, which he said Marshall had, but not in the right proportion, by implication, and also that Keynes had, again by implication, in the right proportion- He never considered modesty a virtue. I'm being unfair. But put those together and you have the perfect outcome. I quoted them both in a symposium we had in Sydney a few months ago with Mary Morgan and a whiz kid at UNSW called Richard Holden, on the future of economics. I said that these characteristics still are the relevant ones that people who want to be good economists ought to strive to have.

I: What do you think the challenges are for the way economics is taught today? What are the problems? What could be fixed? 

R: Well, I don't go as far as Joan Robinson, who in one of her last papers said, "Scrap the lot and start again." It was originally called ‘Spring Cleaning’. 15 I gave a valedictory lecture at SOAS in May 2010 called ‘The Crisis in Mainstream Economics’, in which I went through what I thought- drawing on other people's work as well as my own- were the inadequacies and shortcomings of mainstream economics, and therefore pointing the way to a more appropriate approach to both understanding and designing policy. And that was published in Real-World Economics Review in June 2010, 16 and reprinted in a revised version in the Festschrift for Harald Hagermann. 17  

I: More specifically today, what would you say are the key issues? Would it be the way it's taught, or the material that is taught? Is it what is taught?

R: Yes, because a lot of the things that make sense of how the world works, like the fallacy of composition, for example, can't get into the formal models, because you've a representative agent.

I: What do you think about some movements that ask for pluralism?

R: I'm all for it, because as a liberal educator, you have to introduce students to all possible approaches, sympathetically critical or antagonistically critical according to your own view, provided you've told them where you stand. After I read Stretton, I've never started a course of lectures without telling the students at the beginning what my religious, political philosophies are, and what approach I take to economics. And then I say, "I don't expect you to agree with me, and I would rather read a first class attack on me than a third rate agreement, but I want you to know where I come from."

I: So pluralism would mean what to you? That we would use different models?

R: Well, it depends how you want to do it. I think you should always start with conceptual foundations before you model. I mean, I've always taught theory in a historical context, saying: who originated it, what were the institutions dominant in their own time, what was their own personal background, and why did they come at it like this? Then I would ask, "Well, okay, this is an outline of the structure that they provided. Is it or is it not now applicable to what's happening in the world today?" And very often it is, because economics doesn’t evolve like physics does, so that you can disregard what's gone before. Often the ideas of the greats of the past are still relevant, sometimes in a modified form, to be put into your structure of thought in the present day. 

I: You agree with Keynes when he said that, "The study of the history of opinion is a necessary preliminary to the emancipation of the mind"?

R: I would have thought so.

I: So history of economic thought is important in teaching? 

R: Well, I would prefer it to be history of economic theory, because there's an antiquarian tinge to history of economic thought.

I: Why not history of economic ideas?

R: Because Schumpeter said that. No, no- Well, yes, but by calling it theory you relate it not only to what was relevant for when it was first produced, but also to the modern day. So it's both highly interesting and useful, and it's fair, because it's giving credit to where the ideas came from. That's just elementary scholarship. 

I: A student today who's doing his/her undergraduate in economics, what would he/she gain from reading Adam Smith or Ricardo? The world is very different to what it was then.

R: Well, there are some constants that still linger on.

I: Like?

R: Well, wondering about how markets work, power relationships. Adam Smith on how the labour market works is a very modern explanation of how the labour market works. Smith was a very shrewd guy. I wrote for A-level students many years ago an essay, a short essay, called, "What Adam Smith really said," 18 and I put together key extracts from The Theory of Moral Sentiments and The Wealth of Nations to show that when you put them together, the vulgar Smithians didn't have a clue as to what his work was about. At least, there were very serious shortcomings in their view of the relevance of Smith to the present day.

I: You said that past texts in economics are still relevant and important today. That means that not the latest model is always the best. 

R: No, they're often inferior. Because we tend to train graduate students, now, as though economics began ten years ago with a moving peg, lots of modern economists rediscover, in an inferior way, wheels from the past.

I: Does this relate to questions of whether economics is a science or not?

R: Economics is not a science. 

I: Why not?

R: Because, as Hicks said, it's on the borderline of history and on the borderline of science. We call it a science because many economists suffer from physics envy. I greatly admire physics, but I do think that physicians who've strayed into becoming economists have done more harm than good. 

I: If it's not a science, how does that change the way we teach it?

R: Well, we just teach it as a social science. But it's a way of understanding the world. Marshall, for example, said, "It's mankind in the ordinary business of life," and there's a lot to be said for that. You are looking at the behaviour of key groups. He looked at individuals, though he understood about business people and he understood about consumers; he was a very great economist. But coming from Marx, you look more at classes and the key unit from which you start theorising, which is something Lorie Tarshis stressed, was the capitalist firm. Because that's where decision making starts to set the whole thing going. 

I: Why is it important to think of classes and not individuals? 

R: Because classes are still dominated by their different behaviour with regard to saving and consuming, and what their principal role is. Business people have to organise production and employment, set prices in modern market structures, and plan the capital accumulation.

I: Don't we live in a classless society?

R: Rubbish, classless society. You live in Britain. I mean, if the upper classes didn't exist you'd have to invent them. No, no, class structures are inbuilt to our social system. And think of a country like India with the caste system. That still has a huge hold over what people can and cannot do.

I: What about Australia?

R: Oh well, now, it’s completely classless... No, of course it isn't. It's a capitalist society, and with the blow-out of inequality in wealth and income distribution in the last 30 years or so, the domination of the very wealthy is getting greater and greater, especially when they're connected with multinational oligopolistic firms. 

I: Yes, but class analysis also suggests that they have some coherence as a structure, don't they?

R: Yes. Well, so they do. They're characterised by different marginal propensities to consume and save, different marginal propensities to spend. Of course, consuming has become more like investment in the post-war period, because as you extend credit to all the behaviour changes. So consumption is now not only a function of your income, which in Keynes' day it largely was, it's also a function of your net wealth, and so the short-period  consumption function, is not nearly as stable as it was when Keynes was writing. He was aware, very aware, of all these other influences which were later developed by Duesenberry, Friedman and Modigliani in particular, but he made a judgement that in his time overall personal disposable income was the key variable.

I: Do you think class consciousness is an element of class structure? Today very few people identify with a specific class.

R: Well, they all think they're middle class.

I: Exactly.

R: But actually, you can't say that after the Brexit vote, because these were very disaffected traditional Labour supporters in the neglected regions with terrible unemployment, and they are really getting onto being an underclass. With the exception of Heseltine, no government of either complexion has taken any systematic action to try and bring capital to the depressed areas. De-industrialisation, the reinforcement of that through the Thatcher years, and now globalisation and austerity– means that these people feel completely shut out of society, and they've tended to blame it on immigration and the EU, but really they're the victims of these long-term trends.

I: Why is it important for an economist to understand that these people are, as you say, shut out of society, and not simply the same as us but with lower income?

R: Because they feel shut out. If you're in a family that's had three generations of no one's had a job, I mean it's completely different to your experience and my experience.

I: And would that have an effect on how we conceptualise the economy?

R: It certainly would, because it's a source, if it's left alone, of great social anarchy. Marx's idea that the proletariat would arise and transform- it was a daydream, I think. There may be unpleasant right-wing risings in Europe, and so on, but they're so spread out and so different to one another, the most likely outcome is just anarchy.

I: So Marxism was the good scenario?

R: Well, Marx was a brilliant analyst of how capitalism worked, and there was no greater admirer than he of its production potential. But his views on how you got to socialism, what socialism would be about, were more unrealistic than Christian socialists, and that's saying something.

I: Isn't it interesting that Marshall was never a big admirer of Marx?

R: Well, that's not quite true.

I: It's not true?

R: No. He was influenced by Marx's analysis, and of course, they both went back to be very much influenced by Hegel.

I: Yes, but Marshall’s Principles of Economics...

R: Well no, because he was basically a liberal, and he wanted to work within the system; “Capitalism with a human face”, same as Keynes.

I: What I was getting at...

R: You're getting at me.

I: I am getting at you, but I'm doing it in a roundabout fashion. What I am getting at is that a lot of the followers of Keynes reverted more to Marx's work in their analysis than Marshall himself.

R: Well, Joan Robinson was an admirer of Marx, and learned a lot from him analytically, but she was an old-fashioned left Keynesian. She came from a family of dissent on from the word go, an upper class English family, but she was never a communist. She had a period where she fell in love with Maoism, but she grew out of that, and similarly with Stalinism.

I: Was this much more of a stance than a true belief, you think?

R: No. No. No. She always passionately sought after truth, but also, as Samuelson says somewhere, she was admirable- she had a noble aspiration, to create a more just and equitable society, which is what I think the best economists have always tried to do.

I: I'd like to turn to your personal history. You mentioned on a couple of occasions that you came to Cambridge in 1955 to do a PhD. As you note this was a very good year, or a very good vintage, given who was in that class. A lot of very famous economists were starting then their graduate study: Luigi Pasinetti, Amartya Sen, Garegnani...

R: Charles Feinstein.

I: Charles Feinstein, to name three or four.

R: John Whitaker, Hugh Hudson.

I: And many others. But I wanted to focus on, you, Pasinetti, Sen and Garegnani, because you all developed new systems of economic analysis. Or you developed new lines of thought outside established schools or modes of thought.

R: Well, I can't speak for the others, but like Tom Asimakopulos, we thought we were just being economists. You know, when he refused to be in the first edition of Dissenting Economists, 19 he said, "Why should I be? I come out of Marshall, I come out of Keynes, and I now come out of Kalecki." That's all mainstream economics.

I: Nevertheless, I didn't get to my question. My question was: what do you think were the elements of the intellectual atmosphere in Cambridge at the time that allowed you to develop new paths, even while building anew on old paths? I'm not saying you diverged from the tradition, but you all developed systematic schools of thought that became your own. And I think in some ways that is not typical in economics. First of all, it's atypical with any other cohort to have had such wealth of output.

R: Well, don't forget we coincided with the development of growth theory building on Harrod, and also on the critique of the conceptual foundations of neoclassical economics which became associated with the capital theory controversies. We were in an atmosphere of negative and positive ideas. That's how I came to see it.

I: So how did you feel at the time? What do you remember from back then?

R: I just thought economics was economics. I'd come to write a PhD on the implications of secure profits as well as maximum profits, for the trade cycle, the setting of prices. It grew out of my undergraduate dissertation on saying that if you had Rothschild oligopolists as your principal market structure, what is the systemic effect of this on Keynes' theory? And I wanted to investigate that. Well, when I came, Robin Marris was writing The Economic Theory of Managerial Capitalism. 20 So when he called us all together at the beginning of the year and said, "What are you going to work on?" and I told him this, he said, "You're the first cab off the rank," and I immediately had to write a paper and give it to the research students' seminar, and Joan Robinson came along to it.

I: When did you start thinking that the type of analysis that you do is of a more systematic nature? What I mean is, it starts to produce a body of work which gives an idea about society which is unified. It's not simply taking something that you've seen in a textbook and doing some straightforward extensions.

R: Oh, that would have come many years later.

I: I'm sure it wasn't at the time, but I just wondered your frame of mind then...

R: Well, having to read all the literature on capital theory on the one hand, and locking myself up for a term in 1956 with The Accumulation of Capital 21 on the other, and then reading the paper to the research student seminar over two consecutive meetings, and then Joan coming along… That formed my structure of thought from then on. Furthermore, although I'd been introduced to Kalecki as an undergraduate, it was only much later that I came to realise that Joan had absorbed Marx through Kalecki, and her own thought was moving more and more towards that, and finally towards late Kalecki's and Goodwin's cyclical growth models. But those ideas were formulated well into the set of lectures I'd been giving in Adelaide and Cambridge; much later, I would say, they became explicit in my mind.

I: An important part of your work is on the Cambridge capital theory controversy. This is a key focal point for post-Keynesian analysis more broadly. Could you explain, briefly, what the issue was then, and also whether it's still relevant?

R: Well, it was taken by the mainstream to be an aggregation problem. How do you aggregate capital to explain the distribution of income? And of course, some  said, "Well, you don't have to if you're a general equilibrium theorist." But really as Joan Robinson stressed more and more, and so did I, it was really what vision of the workings of the economies you have in your mind. Do you come out of the Fisherian view- well, Fisher's a good example - that the consumer queen is the guiding factor trying to maximise her expected utility over her lifetime by dividing her income between consumption and saving, and all the other institutions of society, firms, stock exchanges and so on - are means to her achieving or not achieving these ends? That's one vision, very much an individual, subjective value vision. The other comes out of Marx and Smith, and then Keynes, where the swashbuckling ruthless entrepreneur capitalist is the principal decision maker, and all the other aspects of society dance to their tune. And according to which of those visions you have, so you have very different views on what you're analysing and what your policies are going to be.

I: So it's what frame of mind you have when you're doing the analysis.

R: Well, you know, Schumpeter said, "You've got to have a vision before you start."

The other thing was a methodological point. Joan Robinson said that the great mistake of mainstream economics was to use comparisons or differences to explain processes or changes. She said you can't do that. And that's in her 53/54 article. 22 She said that's methodologically a terrible error. Now, the irony is that people like Samuelson and Solow, Bliss, would agree with her. They would say that comparisons of equilibrium situations are just useless for telling you about processes. But they would have the Fisherian view of how the system works. So they agreed with her on method, but had the different vision. People like Garegnani on the other hand had the Marxian, Sraffian view of who really ran the place, but had the methodology that Joan criticised: comparisons of long-period positions, the only way in which you could do rigorous theory.

I: Comparative statics is also in Keynes.

R: Well, that was his method, but he was really being dynamic.

I: How?

R: As Kregel explains in the three models you can see underlying his work at the time: the one where you assume that short-term expectations are always realised, long-term expectations are given and are independent of short-term expectations, so you go immediately to the point of effective demand, so you've proved existence. Then you drop the assumption that short-term expectations are always given, but not that long-term expectations are influenced by this, and then you tell the stabilising story of getting to the point of effective demand by the signals, either by unexpected inventory changes or unexpected prices being established. And then the third model- what he called the shifting equilibrium model where short-term expectations aren't given at the beginning and short-term expectations and their fulfilment or not fulfilment affect long-term expectations, which in turn affect the point of effective demand towards which the economy is tending to go. So that is the methodology which underlay the early beginnings of Cambridge growth theory in Nicky Kaldor, Kahn and Joan. But then, I think, Joan went further, and though she didn't have the formal analysis to do it, she was all of a one with late Kalecki. I mean, Kalecki started off with a trendless cycle and distinguished between trend and cycle, and said different factors were responsible for each. But Goodwin always thought they were indissolubly mixed, and moved towards the growth cycle, and that's implicit in Joan's later work.

I: I wonder then if you always must use the same overarching framework. You've mentioned in many of your writings you are a horses for courses man.

R: Yes, so you adjust according to the issue you're looking at.

I: Do you think that what you call mainstream analysis can provide you with answers to some types of question?

R: Well, of course. I mean, thinking of the paper I wrote with Kenyon on pricing the investment decision. 23 You can use ideas from mainstream micro in that, but you feed them into your particular structure.

Tom Asimakopulos' micro text with Oxford University Press, 24 which Joan did not approve of, but I approve of thoroughly. Because he goes through the mainstream and then ends up with non-mainstream price theory. But it's preceded by a thorough teaching to the students of what the mainstream pricing theory is, and then how it's modified when you take a post-Keynesian view. And that's good teaching. Tom was a brilliant teacher, so the students were introduced to all approaches, left to make up their own mind, but they weren't at all uncertain where Tom himself stood.

I: Do you think the mainstream is not posing the right questions, or is it giving the wrong answers to questions?

R: It's giving the wrong answers, and there's a ruthless hegemony. They're trying to drive out any alternative approaches. And it was crystallised in my mind many years ago when I was giving a seminar at Canberra and Steve Turnovsky was there. And he said, "Geoff, what's all this talk of schools of thought? You and I are just economic theorists, period." To which I replied, "Steve, the difference between you and me is that I failed university entrance mathematics and you failed university entrance history," both of which were true.

I: Do you think the crisis shows that we need to reconsider some of these issues?

R: Yes, that's what I argue for in that valedictory lecture I gave, and I said I'm somewhere in between Joan's Spring Cleaning, "scrap the lot and start again," and the apocryphal story of Sunny Jim Callaghan coming back from holiday at the end of the 70s and saying, "Crisis? What crisis?" Which he never did, but it's attributed to him. Hicks wrote a book, The Crisis in Keynesian Economics, 25 and as I say in my lecture, I pinched the title from him - with acknowledgement, of course.

I: The other thing is you've collaborated a lot in your work -

R: I have.

I: - and there's an article out about 50... 100 collaborators. I can't remember.

R: It's supposed to be going into a book, but I can't get the editors to get in touch with me.

I: Why are collaborations important in your work?

R: They happened. I would just be talking to people, or I was working on something, but I couldn't solve something, I'd go and see someone. For example, Geoff Whittington's first ever article, which was an article with me in the Economic Journal, 26 arose from me drawing on his expertise as a professional accountant as well as a very good applied economist to work out some detailed things about the British tax system. Or take my paper with Araujo, one of my PhD students. Mike Lawler had discovered in the Keynes papers, this correspondence between Dobb, Shove and Joan Robinson on whether it was possible to have a growing economy with only a normal rate of profits being received. And it was a three way exchange of letters, and then we had the idea of what models we should use to see whether you could reconcile the different points of view, and what was the actual answer to this question. We ended up using the apparatus from Paul Davidson's great book, Money in the Real World. 27 So Araujo was much better at mathematics than I was; and he did this excellent archival work, and then we put it all together and wrote it jointly.

I: You've collaborated a lot, but that's not the only interaction with the rest of the academic community. You've written a lot of book reviews and other review articles. Your survey articles on the Cambridge controversies were also constitutive of what the issues are, and how to read other people’s work as well. This brings me back to the original question on post-Keynesianism. You say it's a diverse field, many different...

R: A broad church.

I: Exactly.

R: In fact, the first survey article I ever wrote on it said there was only one unifying thing, they all hated the neo-classicals.

I: You've said that before. But that means that they may be a reaction to an existing school.

R: Yes, but they've taken different aspects of it.

I: Are they freestanding theories, or are they simply there to tear down the mainstream?

R: Well, there's the famous remark about Hicks going around the building and seeing the same thing from a different point of view. but I don't think that's quite right, because people like Paul Davidson won't have a bar of Marx, or Kalecki, or Sraffa, and people like Garegnani probably won't have much of a bar of Kalecki, even though Kalecki and Sraffa were very good friends, and Dobb thought highly of Kalecki.

I: But these are self-supporting systems. You don't need the mainstream to make them intelligible. Or they're not? What do you think?

R: No, they're independent of the mainstream. Or to put it better, they're the positive side of the critique.

I: Do we need to work towards a working definition of what is post-Keynesianism in order to be able to do that type of analysis?

R: Well, Peter and I tried to set that out in the Oxford Handbook 28 , in the first chapter.

I: Can we do it without having a definition? Can we identify pieces of post-Keynesian analysis without having a definition?

R: No, I think you have to have the definition, and then you see how particular people's work fits into it, and how they emphasise different aspects. But it's quite fluffy around the edges. What does Sheila say? The Babylonian mode of thought. Everyone in talking in their own languages, but there is understanding between them.

I: With this answer, you've given me a prompt to my next question. I was reading something by Isaiah Berlin. In one of his autobiographical essays he writes of an incident in 1979 when he received the Jerusalem prize. He was phoned up by the Israel Broadcasting Service, and the interviewer asked whether he is formed by three traditions, the Russian, the British and the Jewish. I wondered if you were to identify strands of your thought, what would they be?  How do they constitute your overall identity?

R: Well, as you know from some of my commissioned autobiographical essays, I'm a democratic Christian socialist coming from a Jewish background that was assimilationist agnostic right-wing. Because I became a believer a long period after I became a socialist. So in a sense, I have grown up as a carbon copy of my parents' thoughts, especially my father's.

Nevertheless, although I have unbounded admiration for my father as a person, we didn't agree on whether God existed or not, and we didn't agree on political matters, and we didn't agree on the need sometimes to take direct action, like in the anti-Vietnam war movement. But, he had all the old fashioned verities: a very, very good, admirable person.

I: You mentioned in another of your interviews the Vietnam war. It obviously had a big intellectual influence.

R: Yes, the student revolt in the Vietnam war was absolutely life-changing. I mean, the intellectual basis of it was Chomsky and Stretton, but the actual experience of taking direct action, and spending two and a half days a week for eight years in political activity -direct action- had a huge effect on my research, my teaching, my life.

I: How did it change your research? How did it influence the way you do economics?

R: It's much more explicitly value-ridden. When I started, I was very proud to overhear some students after one of my lectures on the economics of Keynes, which was the basis of my first book written with Peter Karmel and Bob Wallace. 29 They said, "Well, I can't tell whether Harcourt's Labor or Liberal from his lectures." And I wrote my PhD like it was a scientific report, with numbered paragraphs and writing in the third person. But by the time I was completely affected by political activities... I mean, I'd always been politically active, but through the usual channels. So I was in the Labor Party, I was in the Howard League for Penal Reform, I was an immigration reformer. But I acted through the usual, official channels. But I completely changed when the student revolt and direct action became an additional way of tackling important social issues, and that all fed into my work. If you read my papers, they're much more personal. There's much more of me in them as you go from the first papers out of my PhD thesis, for example, up to what I write now.

I: To finish off, I'd like to ask you the following. Has it ever happened to you that you were going to do some empirical or some policy work, and you come in with a frame of mind, and as you were there...

R: Did I change my mind?

I: Well, did it change your mind, yes?

R: In fact, most of the empirical work I did, I had a conjecture and the empirical work confirmed it. For example, the only really econometric paper I ever wrote was with Peter Praetz and Al Watson, 30 and we were criticising some work by Alan Powell and Fred Gruen on estimating the constant elasticity of transformation function for Australian agriculture. And the question we asked was: "Suppose you were God" - as you could be in the computer age - "and knew what the world was really like. Would the econometric methods that you were using give you unbiased estimates of what we know the true values of the parameters to be?" And my conjecture, as a result of an article I wrote in the Review of Economic Studies on biases in empirical estimates of CES production functions was that they would be chock-a-block full of biases. 31 And we did a Monte Carlo study which confirmed that; at least we thought it did. We were attacked by Ray Byron and, needless to say, the authors.

I: So thinking, of a bigger picture, were there, at any point, developments in the political sphere that challenged your broader view?

R: Oh, yes. I started off as a very idealistic democratic Christian socialist when I graduated, but by the end of the 70s, after Whitlam was chucked out and the Labor party was doing terribly, I decided that you really had to lower your sights, and be pragmatic. That is within the political constraints the society, consider a package deal of policies in order to move towards a more just and equitable society. But you always had to take into account the political reaction to that, or opposition to it, and so on, and that's reflected in a paper that I wrote with Prue Kerr, called The Mixed Economy, 32 which was subsequently the basis of discussion paper number six, the first draft of it, which I prepared for the ALP's - Australian Labor Party's - National Committee of Enquiry at the end of the 70s as to why they were doing so badly. And they re-thought all aspects of their policies, approaches, institutions. This was a paper called ‘Economic Policy and the Future of Australia’. My proud boast is that Bob Hawke, when he became prime minister of Australia in 1983, implemented what I suggested for a good half hour after he became prime minister.

I: So the 70s did not...

R: It lowered my sights.

I: It lowered your sights. But did it not question your belief in a managed capitalist economy?

R: No.

I: With labour unions’ power at the time.

R: No, no. Because in work I did with Eric Russell, Phil Bentley, and Barry Hughes, part of our package deal was an incomes policy which adjusted money incomes for overall productivity plus prices, and we argued that was the only just and equitable incomes policy that you can have if you were going to have sustainable employment. This work was started by Wilfred Salter, Eric Russell, and Nicky Kaldor.

I: You haven't, I think, in any of your interviews, discussed extensively living in Thatcherite Britain in the 80s.

R: Well, I was living in feudal Cambridge, so why should I discuss living in Thatcherite Britain? I thought it was a horrible society. The rise of the number of homeless in Cambridge, for example, over those years- utter disgrace for a developed, so-called civilised society. Mrs Thatcher had very few redeeming features, and her regime was appalling, and her advisor, Alan Walters, who I was fond of personally, allegedly said, ‘We must implement what Pinochet has done in Chile, and it's a great pity we can't go further, like he did, and either kill or lock up the trade union leaders’.

I: What does it mean to be an economist today? What advice would you give to someone starting a degree in economics today.

R: Ah. Well, I would not do as Joan Robinson did, who pleaded with her best student in the last course she ever gave at Williams in the US, "Whatever you do, don't become an economist." But I would say, no, there's a world to win- help carry on the flame. But I would say before you do that, you must read Heilbroner's The Worldly Philosophers, Partha Dasgupta's A Very Short Introduction to Economics, and in Australia, there's two very good collections of essays by our leading economic journalists, Ross Gittins and Jessica Irvine. If you read these and they don't turn you on, don't do economics. In this country I'd say if you don't like Larry Elliott and Bill Keegan, plus those two books I mentioned, it's not for you.

I: Thank you very much for an exciting discussion. 
 

(End of recording)