Dr. Alexander Douglas specialises in the history of philosophy and the philosophy of economics. He is a faculty member at the University of St. Andrews in the School of Philosophical, Anthropological and Film Studies. In this series, we will discuss the the philosophy of economics.
Scott Douglas Jacobsen: Is there a lack of consistency in the terminologies used by economists?
Dr. Alexander Douglas: There’s a question about whether economists use terms consistently. But there’s another pressing issue, which is the gap between the language academic economists use and the language of public discourse.
I wonder if the retreat of economics into higher- and higher-level mathematics has done damage to democracy. Although there was a near-consensus among macro-economists in Britain that first austerity and then Brexit were bad policies, the government received popular support for both. The problem was that the macro-economists could say what they believed, but they couldn’t really explain why they believed it. The official argument rested on some of the most complex mathematics in the world, and there was no convincing ‘entry-level’ version.
Effectively, macro-economists have to ask the public to trust their expertise, even though we can’t see into their black boxes. It was easy for the media to portray the economic experts as elites with hidden agendas and vested interests. Normally the way to fend off that sort of ad hominem argument is to say, “Never mind me or my motives, just look at my argument”. But you can’t do that when the simplest compelling version of your argument consists of hundreds of differential equations.
I think this is a major problem. There is no bridge between the concepts of academic economics and the concepts we use to think about our day-to-day lives. Politics happens in the domain of the everyday concepts.
Jacobsen: What do you think of neuroeconomics?
Douglas: Neuroeconomics is very interesting and something I know little about. Philosophically, it raises more ‘conceptual bridge’ puzzles, this time between the scientific study of brain-events causing behaviour and the ordinary explanations we give for human actions. Some philosophers call this “folk psychology”. There are a range of opinions on this. The most extreme , “eliminative materialism”, suggests that our ordinary explanations, e.g. “Jane crossed the road because she prefers to walk in the sun”, are simply wrong and will one day be entirely replaced by explanations at the physiological/neurological level: Jane’s body moved in such-and-such a way because such-and-such events occurred in her brain. Standard choice theory in economics is, in my view, a regimented version of “folk psychology”. So one interesting question is whether the end game for neuro – economics is to entirely replace standard economics or whether it can somehow be fitted into the existing paradigm.
Jacobsen: What is the healthy perspective – the accurate view – on human economic decisions? What drives us?
Douglas: I’m not convinced that the individual economic agent is the right starting point. You can start instead at the sub-personal level, as the eliminative materialists propose. You can also start with institutions, which have their own ways of behaving that sometimes seem independent of the agents composing them. J.K. Galbraith’s entertaining book, The New Industrial State, is full of plausible-sounding claims about how committees, boards, and so on have their own strange ways of making decisions, which differ from the ways that individual people make decisions. His book on the 1929 stock market crash contains equally plausible descriptions of crowd behaviour, which can be very unlike the behaviour of individuals on their own.
Academic economists are beginning to study institutions in more formal and rigorous ways. The ‘New Institutionalists’ build models to explain why (rational) individuals might submit to the authority of an institution in order to avoid the transaction costs that accompany free exchange in the market. Economists like Herbert Gintis use models from evolutionary biology and game theory to model social norms and other emergent properties of social systems (properties that can’t be explained in terms of facts about the individual agents).
I’m sometimes tempted towards a much more radical view. There is philosophical literature that emerged from the work of the later Wittgenstein, concerning the nature of rule-following behaviour. One central claim is that rules can’t exist for an individual on her own; they can only exist for a whole community. Another is that the relation between a rule and the behaviour it governs can’t be captured by any causal relation – it is not the case, for instance, that knowledge of a rule causes behaviour in accordance with that rule. Rather, the relation is more akin to a logical connection: the rule and the behaviour stand in a similar relation to that of the premise and conclusion in an argument. I believe that preferences are effectively rules: a preference for A over B is a rule: choose A over B. This theory of preferences-as-rules, combined with the Wittgensteinian ideas about rules, suggests to me that both methodological individualism and the search for causalexplanations of choice-guided behaviour might be mistakes. If so, much of modern economics would rest upon a mistake.
Jacobsen: Can you imagine a future with ubiquitous artificial intelligence where mathematical models and algorithms could accurately predict all human behaviour?
Douglas: To the extent that the physical world is determinate then there should in principle be a system of equations that could accurately predict all human behaviour. Of course, the physical world might not be determinate. And even if it is, the finding of the relevant equations might be beyond not only our cognitive capacities but those of any cognitive system capable of existing.
Moreover, there is no reason to expect that any workable model will look anything like the choice theory used by economists. The perfect explanation of human behaviour might make no reference to choices at all; again, it might just track the motion of particles around the human brain and body, or it might track patterns at the institutional level. We don’t know what sorts of causes the perfect model would quantify over. Thus you don’t have to believe that there’s a perfect mathematical model of individual choice, even if you think there’s guaranteed to be a perfect causal model that explains and predicts all observable human behaviour.
Original publication in Conatus News.