How do you make speculation endogenous to economic theory?
Further how do you make the full suite of potential profit making activities, speculation, hedging, arbitrage and investment endogenous?
By endogenous I mean a variable that is determined alongside other variables rather than outside the economic model.
The reasoning in this post comes was prompted in part from speculation by Steve Keen on to what extent the speculative drive is a necessary component of capitalism even though it is destabilising, partly from some dissatisfaction with the Minsky ‘Ponzi’ model of asset speculation, which has been too easily dismissed by neoclassicals as somehow individuals not behaving ‘rationally’. The model here is a generalisation of our earlier model of default risk in banking and insurance across all sectors.Decisions, Decisions, Decisions
The Cost (that is Price) of Speculation – Going Beyond The Minsky ‘Ponzi’ Model
Andrew Lainton
Interesting model, but it sounds to me like Minsky's hedge phase after which the Ponzi phase.
5 comments:
I think it's a mistake to over-mathematize the phenomenon of capitalist financial instability and build it into a model as though it were some kind of deterministic outcome. To me its seems better just to view it as the highly probable, but not determined, consequence of a pervasive tendency in human nature, not an iron-clad law.
It's like the "eyes were bigger than the stomach" phenomenon. People are imperfectly rational and have a tendency to excessively discount or excessively overestimate the future. They borrow and promise more than they should because the present enticements of immediate-term liquidity are seductive, and they lend more than they should, because the enticements of huge interest profits lead them to overlook credit risk.
The best of the models can do no more than say, "People are kinda dumb, and they're kinda crooked and avaricious, and as a result they cause all kinds of cock-ups if they don't sufficiently regulate their own dumb and crooked behavior."
Dan,
I think that depends on the purpose of the model. I'd agree that you can't expect to have a model that will tell you what sort of instability is likely to occur, much less when or how extensive it will be.
However, a model might help you understand in more detail what certain forms of instability might look like. Verbalising ideas about instability is fine, but it only takes you so far. Examining them within a consistent mathematical framework helps draw out greater insight into their dynamics and the conditions which drive them. What it won't do is tell you whether they will or will not happen.
OK, fair enough Nick. But the model should be stochastic, so that there are indefinitely many possible outcomes.
And then there are the unknown unknowns.
The financial industry was using very sophisticated risk models but they didn't account for control fraud even though they were sitting in the midst of it.
Post a Comment