The dreadful suggestion that Britain really should have an industrial policy is raising its ugly head again. Which is an apposite time for a look at what industrial policies actually do to an economy:

The use of industrial policies to support a country’s steel sector has damaging effects on the export competitiveness of downstream manufacturing sectors that make use of steel. That is the central finding of research by Professor Bruce Blonigen, published in the September 2016 issue of the Economic Journal.

We don't, of course, have to look far to find people telling us that steel is one of those essential industries which a country just must have. And thus a sector which simply must be at the heart of any industrial policy.

The problem with this is that steel is an input into other processes - as any such basic industry we might try to protect will be. And thus that input into other industries becomes more expensive (obviously so, if it were already cheaper than foreign made then we wouldn't be trying to have a strategy to protect it, would we?) thus putting the boot into those other industries. That leads to:

One practical concern is that a layering of industrial policies often accumulates over time, leading to the presence of multiple policies at cross-purposes with each other.

Or as we might put that, government commits some other idiocy to try to adjust for the first. Instead of doing the sensible thing and just stopping doing the original idiot one.

Of course, the next stage of this argument is that no, this time around we'll really, and we mean this, study the effect of our industrial policy on all sectors of the economy. Before we intervene even! But that sadly runs into Hayek's objection, that we cannot use anything other than our market economy to calculate our market economy. Which means that we've got to use a market economy, sans intervention, to calculate our intervention - all of which means we'd probably better stay with the market economy in the first place, eh?