Original URL: http://www.theregister.co.uk/2013/05/16/hs2_plans_not_adding_up/

Mobile tech destroys the case for the HS2 £multi-beellion train set

And robot cars will mean no need for new roads, too

By Tim Worstall

Posted in Financial News, 16th May 2013 12:36 GMT

Comment Finally people seem to be waking up to the dog's breakfast which is the economic case for the proposed High Speed Two London-Birmingham rail link.

You know, this lovely train set that the politicians want to plonk down in the middle of England. I've never quite been sure why it is that politicians love such train sets: most of us get over this around the age of 14 when we discover girls.

The basic problem is that all those terribly clever people who did PPE at Oxford seem not to have noticed what y'all Register readers have been doing this past 15 years: connecting all the world and its Mum to the internet wherever they may be.

A little diversion into the economics of these things: when making a decision on whether to splurge £12bn or whatever on a project, you do what is called a cost benefit analysis. Simply, you add up all the costs - the money you've got to spend, the value of the countryside you'll trample, what else you could be doing with the money instead and so on.

You also add up all of the benefits: clearing that countryside of ugly trees, the extra economic growth that will come from whatever the project is, etc. With transport projects the largest of these benefits, by far, is the time saved by people travelling.

Which seems pretty obvious really: when we're talking about projects to transport people around we'd not be surprised to find that the major benefit is that, er, people can be transported around. The question, though, becomes: how are we to value that time that is saved by those people?

A rough rule of thumb is to look at what people would be paid if they weren't travelling. Not perfect, by any means, but it's a reasonable enough approach. Or rather, it used to be reasonable enough, but now it isn't; which is the problem with the HS2 calculations.

Take our first class passenger: no one else can afford the damn seats so we assume that it's business travel on expenses. When they're in the office doin' stuff they get paid according to the value of their output. (yes, yes, I know, but please, this is economics we're talking here)

When they're on the train they can't be doin' stuff. So, if their time in the office is worth £50 an hour (say) then their time on the train is a loss of £50 per hour. Halve the journey time and we've thus created a value of £25 per passenger per trip. This forms part of the “value added” benefits of such transport projects.

No, really. This is the way that the numbers are worked out, even though those aren't the correct numbers. And this is how HS2 comes to have a £19.8bn "value" ascribed to travellers. Of that, £12.6bn comes from those business travellers: the remainder from lowly commuters and leisure travellers. That's offset against all of the costs, along with the other benefits, and is the major reason why HS2's cost benefit analysis shows that the benefits are higher than the costs. So, let's go build the train set.

But, as the perceptive among you will note, people actually are doin' stuff while on the train these days. We've got this whole communications revolution that large numbers of you dear readers have been working on for the past 15 to 20 years. We can make phone calls from trains these days, imagine that! We can work on spreadsheets because we've got laptops. We can write emails firing people because we've got mobile connectivity on our laptops. We can actually work while spreading out in a first class seat.

Anyone who has actually been on a train recently (anyone who's been able to get a second mortgage to afford a first class seat, at least) will know that most of the people on those trains are indeed working. Thus, we can't say that the value of the time shaved off the journey is worth the pay rate of the people who save the time.

Valuing business' travellers time - a false economy

Anecdotally, I've heard many a time that people find working on a train to be more productive than being somewhere: you can just get on with it without having to deal with the usual office interruptions. Something which, if true, would make the value of a reduction in travel times a negative.

Of course, we don't need to go that far, but the problem of the valuation of business travellers' time is at the heart of the HS2 case, whether for or against. Today the National Audit Office has released a report on it. (PDF)

The fun stuff starts at para 2.10, on page 19. In para 2.20 we get the following:

The Department’s methodology uses a simplifying assumption that time spent travelling is unproductive and business travellers will use all the time saved from faster journeys to work. While this approach is used in appraisals in other countries, it has been challenged by opponents of the High Speed 2 programme on the basis that it is unrealistic for rail travel ...

Research commissioned by the Department suggests that business travellers do work on trains for at least part of their journeys, and a proportion of the time saved from faster journeys may be used for leisure purposes. Taken alone, these findings would have the effect of reducing assessed time-saving benefits. On the other hand, the Department argues that if business travellers work on trains then reducing crowding would allow them to do more work – a benefit that is not currently assessed – and, in the long run, business travellers will use time saved from faster journeys to work.

That's what's known as changing your argument (and your evidence) after you've been found out. And yes, the essential problem really is as I've described. The value of business passengers' time is based upon surveys done in 1998-2000 or so: that is, before the introduction of all this shiny-shiny that you IT guys produce really became fully usable in a mobile environment.

And yes, this really is £12bn-odd of the so called benefits of this scheme. Write that number down to zero, as arguably one should, and it's extraordinarily difficult to see that there are any net benefits in building this particular train set.

Sure, freight congestion might be removed, commuter times might decrease, lots more people might travel, but the benefits don't seem to be enough to cover the costs of the project. We move into the “do not build” space rather than the build one. The politicians will have to go and spend £20k with Messrs. Hornby to build a model train set instead of £12bn of our cash on the fullsize version.

While the NAO report is interesting, the real problem is that none of this is new information. Andrew Gilligan wrote, back in December:

Earlier this year, The Sunday Telegraph revealed a report, suppressed by Department for Transport officials, which undermined a major plank of HS2’s business case.

Ministers have argued that less time spent on trains would benefit the economy by increasing time available for productive work. But the report said such benefits were largely “illusory,” since Wi-Fi, laptops and smartphones have made long-distance train carriages an extension of the office.

Or George Monbiot fully three years ago:

Almost all of it is money deemed to have been saved by reducing travel times. Business customers, it says, will save £17.6bn by getting there faster; leisure customers £11.1bn. Nowhere in the documents are these figures explained or justified. I spent the whole of Monday pressing the Department for Transport, asking for an explanation of how it converted time into money. The department spent eight hours of frantic searching to discover, just before 5pm, that it did indeed have a model, which it described as “frightfully complicated”.

By then my copy deadline was almost up, so I cannot tell you whether or not its consultants accounted for the fact that business travellers can work on the train, sometimes as productively as they can in the office.

And of course there have been people like me screaming in various places about this for years now.

I have to admit that I use that Monbiot piece as my clinching argument in discussions at times. George may well be a National Living Treasure to some but no one's ever going to accuse him of a deep and detailed knowledge of economics. If even George, three years ago, could see through the economic case for HS2, then we can be certain that it really is a total dog's breakfast.

There is also a much wider point to be made. I think we all agree that Google's driverless cars are going to be on the roads soon enough. Or if not Google's, then someone else's driverless cars. At which point we'll face exactly the same point over whether to build new roads or not: both scenarios, train and car, will use the same justification.

Time spent behind a car wheel right now really is dead time. Building roads to get people from A to B faster is indeed an economic benefit, then, unlike the building of train sets. But when the robot is driving the car and the passenger is fully linked up with the vehicular mobile comms then that time is no longer dead, is it? Which leaves us with the quite delicious thought that robot cars will reduce the economic case for building more roads. That isn't how anyone is thinking as yet, but it just might be, not too long from now. ®