The Economist: Free Exchange Blog | R.A. | Infrastructure investment: The unbuiltUmmmmm. Yes. But when you increase the numerator of a cost/benefit ratio you lower the value, even if the benefits remain constant, and high.
To follow up briefly on a story from yesterday, New Jersey Governor Chris Christie has indeed pulled the plug on the tunnel between his state and Manhattan. Mr Christie was concerned about cost overruns at the project, but you don't judge a project by its cost; you judge it by its cost relative to its benefit, and the benefit to the tunnel investment would have been substantial.
I have no idea if the cost overruns were enough to push the tunnel into "not worth it" territory. But neither could anyone else based on an argument of "but benefits are big!"
Keep in mind the tunnel currently in use was constructed a century ago. Tot up the value to the state of New Jersey over that period, and you get some pretty substantial numbers.This is exactly the error R.A. is accusing Christie of! He's only looking at one side of the equation. Yes, you accrue benefits over a long period of time. But you also must add up the costs over a long period of time. How long will NJ being paying back the bonds for this tunnel? How long will New Jerseyites be forgoing the opportunities they miss out on by spending money on a tunnel?
This rail tunnel actually sounds like a pretty good idea from the limited amount I've heard about it. But don't support it by making lopsided arguments and then criticize the guy who canceled it because you falsely claim his reasoning was lopsided!