As currently practised, economics works rather poorly apart from relatively short periods because dynamic and living systems cannot be modelled. This is not calculating load factors for beams or the melting point of copper.

Dynamic systems are characterised by their non-linearity, whilst humans are noted for their tendency to rely on linear extrapolations. That’s one reason why economists are wrong 75% of the time when asked to predict the direction of interest rates 6 months hence — direction (up or down), not magnitude.

If you flip a coin for that question you’ll be wrong only 50% of the time. Any ‘discipline’ in which flipping a coin doubles its practitioners’ predictive success rate … is not particularly useful.

Dynamic systems do, however, include waves and cycles enabling a certain level of prediction. The brilliant economist, Peter Bernstein, in his seminal work ‘Against the Gods’ (a history of risk), points out that something like 80% of risk can be controlled by two factors: a) recognise that there is a cycle, and b) understand roughly where you are in it.

In very banal terms you don’t plant tomatoes in October … unless you live in the southern hemisphere.

Macro-economists tend to stuff it up because they’re not thinking macro enough, and consequently get caught out by natural non-linearity. In dynamic systems you can predict the what and the why reasonably well if you’re willing to accept considerable uncertainty regarding *when.*