Economists use models to make astrologers look good

So goes the not entirely unfair joke, that an economic model is less accurate than pondering the influence of Saturn on which horse is going to win the 3.30 at Wincanton. And yet, if this is true, economic models are beginning to look pretty good:

Why has Sweden done so much better than many predicted? Because others failed to see that society could respond voluntarily to the pandemic. For example, the influential Imperial College model estimates a higher reproduction rate of the disease in Sweden than in other countries, “not because the mortality trends are significantly different from any other country, but as an artefact of our model…because no full lockdown has been ordered.”

In other words, the model could only handle two scenarios: an enforced national lockdown or zero change in behavior. It had no way of computing Swedes who decided to socially distance voluntarily. But we did.

For near the entirety of an economic model is trying to puzzle through how people will respond to changed incentives. Exactly the thing that - so it is claimed - the Imperial epidemiological model did not even try to consider. The assumption there being that there are only two sets of behaviour, that without any incentive to change habits at all, that is no presence of an infectious disease, or only those changes enforced by the power of the state.

This, clearly and obviously, not being how we humans do respond to a change in the incentives around us. Thus the model isn’t, in the slightest, representative of the alternative outcomes possible. The alternatives that are, if we are to restrict ourselves to dealing with only two, being “What would everyone do knowing about the problem but without being forced?” and “What would everyone do if forced?”

The answers being that some to many would change their behaviour with the information and without the force and some - an increasing number over time - would not change said behaviour even with the insistence of that state force. The merits of a forced lockdown are therefore very much less than is being assumed - simply because the model being used is worse than any economic one would ever be.

We are of the opinion that the damage done is vastly worse than the benefit gained so we’re not unbiased here. But it does still come as a surprise to find that the powers over us weren’t even considering the correct question. Which is, to repeat, what is the effect of a forced lockdown given that humans are humans? Instead of the question actually asked, what is the effect compared to no change in behaviour in the presence of a pandemic?

Given that, you know, we’re trying to figure out what humans will do it does seem odd to have left humans out of the question.

But then Hayek wasn’t right enough, was he? The centre can never have the information necessary to make decisions. Then we find out that they’re not even asking the right questions? The validity of government planning is what again?