Steve Keen’s new edition of Debunking Economics is a great read. It’s not quite a philosophy of economics book, but it comes close, and its popularity might be a sign that philosophy of economics is a discipline people may one day be willing to take seriously.
Keen also, I’m happy to say, says something in defence of history of economics – history, that is, of economic theory, not economic history. Several times he has publicly winced at how Paul Krugman wrote, during their rancorous blog war, that ‘[t]his is economics, not Talmudic scholarship.’ For Keen this is absurd because it suggests that economists, like natural scientists, can get away with dealing only with the most cutting-edge theories, which are likely to be the most accurate. In fact, since modern economics is, in his view, almost all wrong, there is no reason to assume that the best ideas won’t be found in past authors rather than in the most up-to-date textbooks.
Accordingly, he begins his chapter on demand curves with a look at Bentham’s utilitarianism. But, contrary to his avowed scruples, he treats Bentham very unfairly. Bentham, he correctly points out, promoted a kind of methodological individualism, according to which one determines the level of social welfare by aggregating the pleasures and pains of all the individuals in a society and subtracting the latter from the former. Bentham was, in his view, the true originator of Margaret Thatcher’s (or Barry Goldwater’s) ‘no such thing as society’ claim.
Now Bentham’s fundamental idea – that the welfare of a society is the mere sum of the individual welfares of its members – can be defended on philosophical principles, and these seem pretty sturdy. An individual’s welfare (pleasure minus pain, for Bentham) is a brute psychological datum – its presence is a fact perfectly apparent at least to the person experiencing it. But (as A.N. Whitehead once put the point) nobody experiences society’s welfare. A society’s overall level of welfare is not a brute psychological datum. So what else can it be but a mere abstraction: the sum of the really existing individual welfares? Wouldn’t it be strange if it were anything other than this? That is the heart of Bentham’s ‘methodological individualism’ – it hardly seems the most controversial part of his philosophy.
So what does Keen think is wrong with it? For Keen, individualistic thinking lies at the origin of a thoroughly incorrect idea in economics, that the demand curve for a market is the mere sum of the individual demand curves of the participants in that market. This is wrong for any number of reasons (most involving the Sonnenschein-Mantel-Debreu theorem, if you’re interested). But what does it have to do with Bentham? Bentham never said anything about demand curves.
The link, as far as I can tell, is this. Keen claims that the fact that market demand curves aren’t just the sum of individual demand curves undermines the standard economic claim that (under perfect competition, etc.) a free market will maximize social welfare. Perhaps, then, he has confused Bentham’s claim, that social welfare is the sum of individual welfares, with another claim, that allowing each individual to pursue her own welfare will automatically maximize social welfare. One claim says that social welfare is just individual welfare aggregated; the other says that the way to get social welfare is just to pursue individual welfare in each case. They couldn’t be more different – one is an analytic claim about the meaning of a concept, the other is an empirical claim predicting a causal outcome.
In fact, having identified social welfare with the aggregate of individual welfares, Bentham was quite ambivalent about how best to pursue it. Very quickly there was a division into two camps – one promoting (to use Elie Halévy’s terms) ‘the artificial identification of interests’, the other promoting ‘the natural identification of interests’. The artificial identification of interests arises when a legislator or governor works out what system of constraints to place upon individuals so that their total welfare is maximized. The natural identification of interests arises when simply by pursuing their egoistic aims, individuals automatically and possibly unintentionally maximize total welfare, without any artificial interference from on high.
Often Bentham seemed to believe only in the artificial identification of interests. ‘Law alone’, he wrote ‘has accomplished what all the natural feelings were not able to do.’ For instance, he argued against the view of Montesquieu and others that the punishment for a crime ought to be equal to the damage done by the crime, which is what our natural feelings supposedly lead us to desire. Bentham reasoned that, on the contrary, the punishment ought to be only severe enough to deter crime until the pain visited on the punished criminals equals the pain suffered by the victims of crime. This is very much an artificial identification of interests: the artificial interference of the legislator brings the interests of potential criminals more into line with those of potential victims.
On the other hand, there is the possibility of the natural identification of interests. A great deal of the animus for this idea came, of course, from Adam Smith and his ‘invisible hand’. Smith, however, did not use that phrase to refer to the welfare-maximizing work of the market. The idea that there are natural laws governing the distribution of resources, independently of individual intentions, came from the French physiocrats, not from Smith. The idea that individual behaviour in competitive markets should lead naturally to an efficient distribution of resources really came from Ricardo, who took the idea of natural economic laws from the physiocrats and used it to defend the natural identification of interests.
But the idea of the natural identification of interests was used to defend many other completely different positions. For Edmund Burke it was the basis of an argument against the reform of existing institutions, on the grounds that these had evolved out of the natural identity of interests and were therefore likely to maximize total welfare as far as genuinely possible. For William Godwin it argued against the institution of private property – if people could use all goods freed from the artificial constraint of instituted property rights, the natural identity of interests ensures that their behaviour would lead towards the best results for all. By no means was it an idea used only by the early economists to defend laissez faire. But it certainly was used in that way – Bentham himself supported Ricardo’s argument for economic liberalism, and there is therefore an inevitable tension between the paternalism of his legal philosophy and the liberalism of his economic philosophy.
The natural identification of interests has one advantage over the artificial identification of interests. The latter leads to a puzzle. If egoistic psychology is right, why would you expect a legislator to act to maximize total welfare, instead of just her own welfare? Perhaps in a democracy, the legislator, answerable to the people, would act in the interests of the people (speaking very hypothetically now). But then that would just turn the whole instance into one of the natural identification of interests – a natural process of give and take between the legislator and the people, and not an artificial imposition of one upon the other.
Perhaps, then, in a very vague and indirect way, Bentham’s theory does lead to a defence of the natural identification of interests. If his egoistic theory is right, it is hard to see how social welfare could ever be maximized, except as some natural consequence of the pursuit of individual welfare by each member of society. To avoid the conclusion that his entire project was hopeless, Bentham might therefore have been inclined to endorse the view that a free market can maximize social welfare naturally and automatically. But this hardly seems enough reason to blame him for the dodgy tricks with demand curves modern economists have played in order to defend that view.
 John Quiggin’s review of the first edition of Keen’s book pointed out that this doesn’t seem right – the claim about efficient markets is grounded on Arrow and Debreu’s theory of general equilibrium, which doesn’t require market demand curves of any shape. But Keen has made the argument more subtle in the new edition.