Having lots of shiny new infrastructure sounds like a really cool idea. Put the unemployed to work and get some asset that will enrich us all for the next 50 years or so. However, whatever it is that we do we do have to do a cost benefit analysis. How much is it going to cost us to do something and what is the benefit we get from having done so? And it is here that the case for HS 2 fails I'm afraid.

Such a cost benefit analysis was done for the train line. As is normal in these things it is the saving of time that contributes the greatest benefit. The assumption is that if you're locked up in a train carriage you cannot be doing productive things like meeting people, working, talking. Thus the time spent on a train is valued as a cost, a cost equal to the value of your time when you are working. This has the obvious effect of making the time of some senior businessman sitting in first class worth more than the kiddies on awayday tickets in cattle: but so be it, that's just the way the sums are done.

It is on this basis that the case for HS 2 rests. No, really, it does, the great girt chunk of the benefits to set against the costs is this calculation that getting people there faster means they spend less time unproductive and more time productive: less time in carriages unable to work and more time outside them able to do so.

Unfortunately for this case technology has changed [3]:

On-train wireless internet connectivity is growing fast in Europe - but even faster in the UK, which now has more than 2,000 Wi-Fi equipped carriages.

If people are productive while in a train then the benefit of getting them there faster disappears.

That's bad news for High Speed Rail though, as the justification for HS2 (the £17bn high-speed London/Birmingham connection) assumes all travellers are entirely unproductive during transit and thus the 30 minute reduction in travelling time benefits the economy.

At minimum it is necessary to do the sums again so as to take account of that change in the world and the value of time. Then we need to change the basic method of calculation used on all such projects: time that can be spent online is not unproductive these days thus we have to change the way that we do all such sums.

Amusingly this will affect the relative merits of train and car transport. Time spent stuck in a car really is unproductive: on a train not. So we have greater benefits from making car transport faster than we used to, fewer benefits from making train transport faster than we used to. But we also have greater benefits from having more trains (not faster ones) as this pulls people out of that unproductive time in the car and into productive time on the train.

I have a feeling that doing these calculations properly will lead to something of a change in how we think about rail transport. It could well be that this all makes more local, regional, commuter, lines vible while reducing the case for high speed long distance passenger lines.