I was asked by widgetfox for my views on The Ten Pillars of Economic Wisdom by David Henderson.
I broadly agree with them. Not as the absolute Truth of Economics, but as a handy guide for students. More anon.
A conversation developed on twitter between widgetfox,
andrewducker and myself which we all agreed would not suit twitters 140 character format. I offered to set up a conversation on LJ. Here is it. I’m starting by recapping the position so far. In the interests of cohesion I’ve expanded my own original points a little. If I’ve misrepresented any of the contributions from WidgetFox or Andrew Ducker, then I’d welcome their correction.
To my broad agreement with the Ten Pillars I offered my opinion that one should remember that macro-economics sits on a foundation of human psychology and that you should be able to build up to macro-economic principles from the psychology of individuals and you should be able to work down from macro-economic principles to the psychology of individuals. I’d go further, if you can’t root your macro or micro-economics in human psychology then you’d better start work on the economics again. I think my favourite example of this is Modigliani and Millers Nobel Prize winning statement of the bleeding obvious in relation to the Weighted Average Cost of Capital.
WidgetFox disagreed (1). She suggested that 1) (a) humans are distinct from each other and (b) are inconsistent in their behaviours so are hard (impossible) to model. 2) people interact in complex systemic ways with each other and this makes their behaviour even harder to model.
Andrew Ducker said he disagreed with these statements from WidgetFox.
We touched on time preference and risk preference.
At this point we all decided that having this conversation on twitter wasn’t going to work.
I’m not sure whether I agree with WidgetFox or not. I certainly agree that humans are inconsistent in their behaviour and that they interact in complex systemic ways. I’m not so sure that, in macro-economic terms, they are distinct. In any event, this might make their individual behaviour difficult to model. I’m not sure, that it makes it impossible to model. I think I’m a gnostic on this point.
What I think we’re dealing with in economics is not the way individuals behave in a particular situation, but the way they behave on average and in aggregate and the way they make individual decisions. I think that macro-economics is the aggregation of micro-economic behaviour and that micro-economic behaviour rests on heuristics and bounded rationality (amongst other things, but heuristics and bounded rationality is most germane to this discussion.)
Whilst there is a difference between All People are X and On Average People are X I think in macro-economics the difference is cancelled out by the sheer volume of transactions and actors. For most, if not all, macro-economic situations I think it is unlikely that there is a significant difference between All People Being X or the Aggregate Position Being X (for an individual who is facing whatever curve macro-economics is offering.) A large part of the practical business of macro-economics is about trading positions. If there are evenly balanced outliers in a set of economic actors then the curve takes the same shape as if all people agreed that it should have a certain shape. (2)
On a micro-economic scale it’s not necessary to understand how particular individuals will behave. What you need to understand is whether heuristics tells you that whatever black box the human mind is, either single minds or complex systemic groups, sufficient people will be nudged into your shop to make opening that shop worthwhile. There might be sufficient people who will buy very low quality at a very high price to keep you in business (3) but on average economics and experience says not.
The fact remains that individuals and firms manage to anticipate the individual and group behaviour of others sufficiently well in order to engage in economic transactions with them.
(As a side note I wonder at the effect on macro-economics that many of the large volume actors are professionals.)
WidgetFox will undoubtedly have made and be about to make a more subtle and nuanced argument than I understood. I look forward to it, knowing that the fault for not understanding what she says lies with me.
I’m looking forward to Mr Ducker expanding on why he disagrees with WidgetFox on the modelability of humans.
(1) She didn’t call me Lord Copper this time. I’m not sure this strengthens my position.
(2) I’m quite interested in the ability of betting markets to predict outcomes. So far it looks like they no worse than any other method.
(3) and they are called tourists