The ludicrous 20-year timescale for HS2 is reason enough to abandon the whole thing

If I stand on the forecourt of Euston station tomorrow morning, I will be able to get to Manchester by high-speed train in 20 years, one hour and eight minutes. That’s only 19 years, 364 days and 23¾ hours longer than it took me last month.

But at least we know that 17 June 2033, the day earmarked for the opening of the London to Manchester High Speed Rail service, will be a nice, sunny day. As the inaugural train pulls out of Euston, it will travel under clear blue skies until the train reaches Birmingham (scattered clouds: chance of precipitation 20 per cent).

We know this, because, of course, we can forecast the weather 20 years in advance. Well at least we should be able to do this, since economic forecasters can clearly predict the demand for travel 20 years ahead. This same infallible skill also tells us that the economic effects of HS2 will spread wealth from London to the north — rather than in the other direction. Because, we know, don’t we?

We know nothing of the kind. Which is why the ludicrous 20-year timescale for HS2 should cause us to abandon the whole thing. Not only out of shame (Brunel built a railway to Bristol in five years, when the principal tunnelling technology was a navvy with a pick) but for another reason too.

The distinction here is between ‘optimality’ and ‘optionality’. Investment in a rail line may be ‘optimal’ — in terms of speed, comfort or convenience, given some startlingly confident assumptions about the future, but these predictions become increasingly unreliable the further into the future you peer. This uncertainty matters less if the investment has a range of uses, but HS2 doesn’t: it offers almost no ‘optionality’ at all.