Yanis Varoufakis is a Greek economist who currently heads the Department of Economic Policy at the University of Athens. From 2004 to 2007 he served as an economic advisor to former Greek Prime Minister George Papandreou. Yanis writes a popular blog which can be found here. His treatise on economic theory ‘Modern Political Economics: Making Sense of the Post-2008 World’, co written with Nicholas Theocrakis and Joseph Haveli is available from Amazon.
Interview conducted by Philip Pilkington.
Philip Pilkington: Without getting into too much technical detail what is it that you refer to in your book Modern Political Economics: Making Sense of the Post-2008 World the ‘inherent error’ in all economic theories and models?
Yanis Varoufakis: The essence of the economists’ inherent error is that they erred into thinking it is possible to tell a credible story about how values and prices are formed in complex (multi-sector) economies that grow through time. For decades economists struggled to produce such a narrative. But all their best laid plans for piecing it together crashed on the shoals of indeterminacy. Put simply, their mathematical models could not be solved. At that point economists did one of two things: Either they accepted that it could not be done, or they introduced hidden (and sometimes not to hidden) assumptions that ‘closed’ their model at the expense of credulity (e.g. an assumption that the economy comprises a lone Robinson Crusoe-like figure, or a single commodity, or that all exchanges occurred in a timeless universe and at a flash of a fleeting moment). The former scholars were forgotten by history, as their papers never saw the light of day. The latter built up careers, sometimes radiant ones. Alas, their economics were riddled with only thinly disguised ‘tricks’ the purpose of which was to disguise economics’ ‘inherent error’.
PP: One consequence of this is that you believe a meaningful theory of value cannot be established, right? In this estimation both the old Marxist/Ricardian labour theory of value is as meaningless as the supposedly sophisticated marginal theory of value. Can you give a gloss on why this is the case and what consequences it has for political economy?
YV: Allow me to be tactful about what our book refers to as the economists’ inherent error (including Ricardo’s and Marx’s): If a theory of value cannot be ‘closed’ properly (or embedded in a mathematically appropriate manner) within a theory of growth, then we have to cut corners to do it.
Neoclassical economists cut corners either by telling stories about how an economy creates relative prices in the absence of time, or by narrating the real-time growth of some single-sector or single-individual economy. Ricardo and Marx, in contrast, allow for different sectors to grow concurrently but introduce a single sector economy through the backdoor by implicitly assuming that the degree of capital utilisation in the same across all sectors (so as to ensure that profit rates are also equal across sectors). In the end, the same inherent error rears its ugly head and pushes economists into interminable debates of little or no value. The consequence of this, for political economics, is that we get absorbed in a self-referential, introverted mindset that allows us to lose sight of a really-existing capitalist reality which refuses stubbornly to fit into well-behaved models.
PP: You mention that economists seem to always want to ‘close’ their models. This appears to me to be a somewhat desperate desire to create a deterministic system. Yet, even in physics determinism is generally recognised to be incomplete and ‘open’. Why are the economists still stuck back in the 19th century in this regard?
YV: All model builders have a natural tendency to ‘close’ their models. When we see a system of equations we immediately wonder whether it can be solved. When (and if) we discover a solution –‘closure’ in other words – we feel a sense of pride and, even, joy. So, yes, if you are in the business of building models, it is natural to aspire to mathematical ‘closure’. This aspiration comes natural to all: mathematicians of course but also economists and physicists (the latter have been, after all, gunning for the Theory of Everything for a long, long while). The difference between economists and physicists lies not in their ambition to ‘close’ their models. It lies in the following:
Physics is blessed with an object of study – let’s call it Nature – that does, normally, not give a damn about the physicists’ theories of it (the Heisenberg Principle excluded). This means that Nature provides an impassionate assessor of the physicists’ models. If their models have been ‘closed’ in a manner that defies logic (i.e. by means of illegitimate hidden axioms), then Nature will expose them. It will defy their predictions in the laboratory (or thought the telescope) in a manner that forces the physicists back to the drawing board. So, if a particular model cannot be ‘closed’ (because it is not solvable), Nature will ensure that the physicists come clean and admit that this is so.
Economics on the other hand, by virtue of being a social science, is caught up in the famous infinite regress problem. What this means is that there exists no neat separation between (a) the economists’ object of study and (b) our theories about it. Indeed, our theories are part and parcel of the world of phenomena that we are trying to theorise about (unlike a theory of thermodynamics which is quite independent of the thermodynamic phenomena under study). Thus, on the one hand, our models can never be properly ‘closed’ (since they are, by construction, ‘open’ theories/beliefs about theories/beliefs etc.) while, on the other hand, when they are illegitimately ‘closed-shut’ by the economists (by means of hidden axioms whose purpose is to ‘close’ the damned models at all cost) there is no objective test that can either confirm or deny the validity of these models. Put differently, the Social Economy can never come to an objective verdict on our economic models (e.g. on some theory of bond markets) simply because these models are an intimate part of our Social Economy (i.e. of the larger game within which agents form beliefs about particular bonds and about the bond market in general).
This great difference between physics and economics means that, whereas physicists gain their discursive power in society from managing to explain and to predict the world we live in, the economists gain their discursive power exclusively from:
(i) convincing the rest (who are not sophisticated enough to be able to discern that they managed to ‘close’ their models utilising logically incoherent and well hidden axioms) that they managed successfully to ‘close’ their models, and
(ii) the great utility that these models offer to financiers who use similar models in order to pretend to value risky assets (e.g. CDOs)
(iii) the political utility offered by these ‘closed’ models to anyone who wants to argue that capitalism is a socio-economic system as natural as Nature itself, and thus amenable to the 19th century mechanistic approach which helped humanity conquer electromagnetic and other such natural phenomena.
PP: But some might say that the economists DO have a reality to contend with and that reality came screaming back in 2008. This certainly isn’t the first time that a crisis has occurred and yet economists remain evasive or they arrogantly assert that their models can in fact deal with it, when it is clear to any objective observer that they cannot. Does this not indicate that there is something fundamentally different happening in the economics profession?
YV: Economists may very well have a ‘reality’ content, to the extent that they are creatures of this really-existing world. The problem is that their own circumstances, within this reality, are improved significantly the less their own models of this reality has to do with actual… reality. In other words, the economics profession concentrates its rewards (tenured positions, large research grants etc.) onto the economists whose models subscribe to certain norms. The most important of these norms is that the truth of the offered economics be self contained within ‘closed’ models whose ‘solution sets’ are rather narrow. To meet this demand economists must ensure that their models leave no room for inconveniently open-ended phenomena like… crises.
This leads us to the question (that you pose): So, how do economists respond when some real crisis hits? How do their account for the fact that when their model had left no room for it? The fascinating answer is that they assume the crisis to have been the result of a random ‘disturbance’. Something like a meteor falling on an otherwise harmonious Earth; an event that is to be untheorised per se. Then, they employ all their mathematical prowess in order to ‘study’ how the ‘system’ (i.e. capitalism) absorbs this shock. As you may imagine, the result of this ‘recovery’ path is founded on assumptions which can only yield one plausible answer: The process of adjustment to the shock of the crisis is best left to the markets which, prior to the crisis, were assumed incapable of causing a crisis!
PP: That’s really twisted. How do you think students are responding to this chicanery after the crisis? And maybe you could say something about how students are imbued with this mindset in the first place.
YV : There have been some interesting studies that reveal how common decency and ‘other’-regarding norms are weeded out of students of economics very early in their undergraduate career. Take the example of the generalised prisoner’s dilemma below:
Young men and women are given some money (e.g. $10) and asked to contribute (each one separately from the rest, and in perfect anonymity) all or part of it to a common purse. Then, the contents of the purse are multiplied by the factor of, say, three and the contents redistributed among all of them independently of their contribution. Clearly, the best outcome for the group is that each contributes all of his or her windfall to the common purse and, that way, each gets a return three times as large. The problem here is that there is a temptation to let others contribute while you do not (since that way you get your share of three times of their contributions and, to boot, you have also kept your own money).
What we find in experimental studies involving real students, who play the above game with real money, is something quite startling: students of economics were significantly less willing to contribute to the common purse. Moreover they were more pessimistic about the prospects that others would contribute! So the question arose: Is it that economics attracts the less cooperative, more ruthless young persons? Or is it that exposure to economics makes them relatively more ruthless, pessimistic and aggressive?
To find out, the experiments were repeated separately for first semester first year undergraduates (before they were ‘contaminated’ with economics or other subjects) and for students who had just graduated. Guess what: Amongst the fresh(wo)men who played the game, the ones that had chosen to major in economics did not behave differently to the rest. They were equally willing to contribute. Therefore no evidence was found supporting the hypothesis that economics attracts misanthropes. On the other hand, amongst graduates those with an economics training stood out from the rest: they were much less likely to contribute, and more pessimistic about the others. The conclusion is inescapable: a training in economics significantly increases the probability that a person becomes less sociable, more aggressive, less cooperative; in short, miserable.
Why and how is this indoctrination taking place? The answer is simple: Economists are all about creating determinate models; ‘closed’ models that predict behaviour. To do this, they need to assume a particularly narrow minded form of rationality (which I call ‘instrumental rationality’): you are rational to the extent that you deploy your means efficiently in the pursuit of given objectives. When a young person is told repeatedly that to be rational means to be ruthlessly instrumental (i.e. to treat others as a means to one’s own ends), and that contributing in this game is for sissies (or, more ‘scientifically’, irrational), is it any wonder that a training in economics makes young persons more brutish and nastier?
The end result is that youngsters with a heightened sense of civic responsibility either drop out of economics, in a bid to retain it, or manage gradually to shed it; to adopt ‘instrumental rationality’ in their own daily life and mindset. Suddenly, the models begin to shape the modellers, rather than the other way round. It is a subtle process of change that turns economists into simulacra of themselves in a hopeless pursuit of ‘closed’ models that validate their own sad ‘conversion’. Seen from a different perspective, what we have here is a remarkable Darwinian process that guarantees the survival and dominance, within economics departments, of the anti-social, aka instrumentally rational, fools.
As for the Crisis and its effects on economics students and departments, I am afraid it has done nothing to ameliorate this Darwinian mechanism within existing economic departments. The norms of instrumental reasoning are so powerful that not even the earthquake of the Crisis has had the power to unsettle them. I have a hunch, however, that what the Crisis will do is speed up further the rate of decline in the number of youngsters interested in studying economics. This is the good news. The bad news is that they will not turn to other social studies but to pseudo-disciplines like marketing, business, advertising etc.
PP: That’s fascinating, let’s continue to run with this for a moment. If what you say is true – and I believe the evidence is unquestionable in this regard – then economics is not a science whatsoever. It more so resembles a school of morality or even a philosophical cult. The old Greek Stoics spring to mind. They were a school of philosophy that not only taught certain ideas but demanded that their followers live these ideas in their day-to-day lives. But in economics the students aren’t even told that they’re signing up for a moral vision, a sort of religion or belief system, they’re told that they’re being initiated into an objective science. Perhaps you could reflect a little in that direction and its implications?
YV: Quite so. It is a priesthood that truly believes it is not a priesthood but, rather, a community of scientists. How do they manage to maintain this delusion? The simple answer is because their incantations involve rather advanced mathematics and their rituals are steeped in statistical tests and projections.
Indeed, in aesthetic terms, the economists’ papers, models, presentations seem indistinguishable from those of physicists, bio-statisticians etc. The only difference is that, unlike the latter, economists generate nothing more than analytical propositions about economic variables which are, as Popper would have pointed out, profoundly non-falsifiable. And here is the rub. Once their non-falsifiable (and thus non-verifiable) propositions are expressed, the statistical tests that follow (usually referred to as econometrics) give economists a great excuse to imagine that their models have been tested. But tested they never are!
Let me explain this in more detail, as it goes to the heart of your question: The economist first builds a model – say: M – that seeks to explain one or more variables (e.g. wages and employment). Once that complex mathematical model is ‘solved’ (just like a system of two equations in two unknowns, y and x, can be solved by means of a function that links y to x; e.g. y = 3x+5), a so called ‘reduced form’ equation (or system of equations) is derived from that solution. Let’s call this R. What economists are good at doing is demonstrating that R corresponds to M (i.e. when the mathematical relationship R holds, this is consistent with the solution of model M). The virtue of R is that is can be checked statistically: data is collected and used to show that, indeed, there is no evidence that R does not hold in real life. At that point, the economist celebrates with yelps of joy: “My model M has been proven to be consistent with reality.” Alas, what the economist forgets to add is the crux of the matter. And what is that? Two crucial facts:
(a) There is a plethora of models, in addition to M, that are also consistent with R. Which means, naturally, that there has been no demonstration whatsoever that model M has been verified (since an infinity of alternatives could explain R just as competently). Now, of course this is also true in non-experimental sciences like astronomy. Yet, economics is unique. This is why:
(b) Model M, like all possible economic models, can only squeeze their ‘reduced form’ R out of their edifice if dodgy assumptions are made regarding time and/or complexity; i.e. only if they axiomatically dismiss what I call the economists’ ‘inherent error’. The practical importance of this is that the imposition of these assumptions may have succeeded in deriving R out of M but that success is bought at the price of having lost any capacity to predict what a complex economic phenomenon will generate (as outcomes) in the future (recall that if you account properly for both complexity and time in model M, no R is possible). It is in this sense that, as I claimed above, no test of M’s predictive capacity (regarding events unfolding in real time) is possible.
This is a most peculiar failure: The hapless economist uses the same tools as acclaimed physicists and astronomers. She has trained for years to speak precisely the same language as them, to understand the same advanced mathematics, to deploy most complex statistical methods which are an essential part of the scientific toolbox. It is, understandably, incredibly difficult to accept that her work is a form of higher order superstition; a religion couched in the language of mathematics and statistics. Tragically, this is precisely what it is. Come to think of it, what is it that separates science from mythology? The fact that scientific propositions are not self-referential. That, in science (unlike in mythology), when the facts clash with the theory it is too bad for the theory.
E.E. Evans-Pritchard (the famous anthropologist) once offered a brilliant insight into the social success of the priesthood within the Azande society. The question he asked is similar to yours (regarding economists): If they get it so wrong so often, how should we explain their continuing dominance? When the Azande priests and oracles failed to predict or avert disasters, why did people continue to believe them? His explanation of the Azande’s unshakeable belief in witchcraft, oracles and magic goes like this:
Azande see as well as we that the failure of their oracle to prophesy truly calls for explanation, but so entangled are they in mystical notions that they must make use of them to account for failure. The contradiction between experience and one mystical notion is explained by reference to other mystical notions. Evans-Pritchard in his Witchcraft, Oracles and Magic among the Azande, 1937
Economics, I submit to you, is not much different. Whenever it fails to predict properly some economic phenomenon (which is more often than not), that failure is accounted for by appealing to the same mystical economic notions which failed in the first place. Occasionally new notions are created in order to account for the failure of the earlier ones. For instance, the notion of natural unemployment was created in order to explain the failure of the market to engender full employment and of economics to explain that failure. More generally, unemployment and excess demand (or supply) is ‘proof’ of insufficient competition which is to be fought by the magic of deregulation. If deregulation does not work, more privatisation will do the trick. If this fails, it must have been the fault of the labour market which is not sufficiently liberated from the spell of unions and government social security benefits. And so on. The fact that these ex post rationalisations of theoretical failure are narrated in mathematically complex language, and accompanied by myriad statistical ‘tests’, adds to their social power to silence critics without and doubts within.