Skip to content


April 9, 2012

Steve Keen is currently constructing a complex model of the economy based on Hyman Minsky’s financial instability hypothesis. The following animation (via Andrew Lainton) makes a stab at showing how it works. It’s interesting, but it makes a bit of caricature of orthodox neoclassical economics.



True, supply and demand curves don’t really exist, but used as rough rules of thumb for how some markets work they’re not such a bad way of explaining things. I remember my microeconomics students being fairly receptive to concepts like elasticity, and how shops weren’t necessarily in charge of the prices they set. Marshall was a bright chap. Some conventional economists have since developed fairly nuanced ways of looking at the economy. And complexity economics has been around for a while.

What’s more, mainstream economists like Krugman would counter-argue that they don’t actually set all that much store by the lines in a diagram. They’re just a temporary oversimplification which is subsequently relaxed. (This is methodologically dubious, however, as I noted in my previous post). If there is any sort of economics spring, it’s important to be aware of how neoclassical economists defend themselves, otherwise heterodox economists will continue to be ignored as they have been for decades.

Where I agree that one-dimensional lines are “so 19th century”, as the film asserts, is when economists do start imagining that supply and demand curves aren’t just a rough rule of thumb; that they exist in reality and are continuous and smooth.

In reality they might be spiky, vertical, backward-bending or have gaps in them. Empirical observation confirms this. Even Marshall purposefully left out the maths from the main part of his book and only put what was necessary in an annex.

It’s also extremely destructive to make supply and demand the totems that they are — to start from a basic position which is so patently biased toward competition, individualism and a rather perverted version of utilitarianism. These starting assumptions fatally shape any subsequent discussions.

Economies and markets don’t necessarily tend toward equilibrium, as some of the most interesting economists, like Schumpeter and Marx, have shown. It’s not good enough to point to something like the Dornbusch overshooting model, which introduces sticky prices to show theoretically that an exchange-rate market doesn’t always land smack-bang on its equilibrium and can spend some time away from it. As people like Keen has shown, sticky prices aren’t enough to make a model Keynesian. The much-fêted model uses lots of ridiculous assumptions. My other question about the overshooting model is: so what if you can retrospectively show an economic outcome mathematically (I emphasise the word outcome, because the process certainly isn’t true)? Isn’t that just academic grandstanding? Why not just look at what markets actually do and proceed from there?

I’ll be very interested to see Keen’s model, especially to see how it copes with randomness, the existence of black swans (unknowable events with a big impact) and reflexivity (people changing their behaviour in response to events that become predictable). Presumably this is what complexity theory is supposed to be good at, because attempts have certainly been made in the past to construct full working models of the economy, such as Oscar Lange’s socialist planning supercomputer, which was much criticised by Hayekians amongst others for being unable to cope with the subjective nature of the tacit information held by the ‘man on the spot’. The Philips machine was another dubious effort at trying to model the economy within a closed system. I’m sure Keen’s model will be far more sophisticated.

No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s