Does there need to be an economic case for high-speed rail?
I like high-speed trains. I really do. Every time I travel on a French or German one I wish we had more of them in the UK.
Quite a few economists I know feel the same way. But none that I have spoken to thinks the economic case for HS2 is particularly strong.
"We should think of high-speed rail the same way we think about the Olympics," one of them said to me. "It's might just be a good way to make us all feel better about ourselves as a country. But we shouldn't kid ourselves that it's the most economically sensible way to spend £30bn."
And yet, as the National Audit Office (NAO) points out in today's report on HS2, the government's arguments for the project have tended to focus on the economic benefits to customers of increasing capacity and cutting journey times on the busy West Coast line.
Ministers have also stressed the broader economic and social benefits of building tighter connections between the North and the Midlands and the South-East. They talk about it being a "fast track to bridging the north-south divide".
There are also purported environmental benefits, with the government claiming it will help the UK meet its targets for cutting carbon emissions, though the NAO report suggests these will be rather small, and will depend on the way the electricity used to run the trains is generated.
Are the economic benefits worth the £30-35bn cost, in current prices, of building the network? The Department of Transport says they are.
Its own cost-benefit analysis suggests that every pound spent on the first stage to Birmingham would produce about £1.50 in benefits (depending on what you throw in). And they reckon the benefits of the second stage, further north, would be even greater - well over £2 for every £1 spent.
Critics have taken issue with the details of the analysis, particularly the assumption that time spent in trains by business travellers - counted as lost business time - is unproductive.
But for most economists, the scepticism about the economic case for HS2 goes well beyond the details of that elaborate calculation.
The point, for them, is not so much whether the benefits outweigh the costs - every government department and local council in the land could give you a list of projects that pass that test. The point is whether this is really the best one to be going ahead with, at a time when public capital spending generally is being cut.
Henry Overman, an LSE economist who has advised the Department of Transport on some of these issues, has written a useful summary of all this. It was written before the Department revised its case but the basic arguments still seem relevant.
At the time of the Eddington Review in 2006, he says the benefit-cost ratio for HS2 put it in the bottom fifth of investment projects that the Department for Transport had on its books. (And that is when the benefits of the first part of the project were said to be much higher than they are now).
That may all be true, supporters of HS2 would say, but what these narrow calculations are forgetting are the broader economic and social benefits that would come from this project, precisely because it is so big. It will so change the economic shape of the country - who knows what the ultimate results will be?
Who knows, indeed. Economists are not very good at capturing the broader dynamic benefits which might eventually stem from a project like this one. It's tricky to put a monetary value on the increase in national self-esteem that might come from the knowledge that we now have shiny fast trains too.
But, there is quite a lot of economic research into what helps to "bridge the North-South divide", and that does not suggest that high-speed rail is a particularly good way to do it.
That's because that high-speed rail line will, literally, go both ways. It might help northern companies compete more effectively with the South-East. But it might well also make it that much easier for the South to suck business and talent from the rest of the country.
This was brought home to me when I produced my report about the economy of London earlier this year.
By itself, the second stage of the project, connecting Birmingham to cities further north, might help boost economic connections between all those cities. But as long as it's all about giving these places an even faster link to London, the evidence from other countries suggests the majority of the direct and indirect economic benefits will go to the London and the South-East.
You might still say all this confirms how mean-minded economists are. Their cost-benefit analyses "discount" the future benefits - a pound of benefit for citizens in 50 years' time is worth a lot less than a pound today.
That is the standard way to do these kinds of calculations, since future generations are also likely to be a lot richer than we are. We can't invest infinite amounts now to help them. But you could argue that it systematically biases the outcome against projects that will deliver benefits to people for a really long time.
Looking back in history, it's hard to find many government "grand projets" that economists have liked (this book has a caustic summary of some of them).
The Millennium Dome is the famous recent example. But if ministers had only been concerned with the economics, we would not have had the Channel Tunnel. And we probably wouldn't have the Jubilee line extension or Crossrail either.
Note that many of those big projects involve trains. You would not have the same problem convincing the average economist of the case for expanding airport capacity in the South-East.
Governments like trains a lot more than we like airports, and so, apparently, do many voters. As I said at the start, plenty of economists like high-speed trains too. They just can't come up with a lot of economics to back that up.