[debok]

Something that wasn't really addressed in the article is that smooth operation of the economy also saves lives.

To put in terms of the infinite trolley problem: The longer the trolley gets, the bigger the probability that one of the passengers is an EMT commuting to his shift, where being late could cost a life. You can add many such passengers, with various levels of complexity, where being late could end up costing a life or a livelihood.

So at some point, not hitting the brakes becomes worthwhile even just in terms of lives saved. The problem is that it is very hard to measure the exact impact of "inconveniencing" the 8 million passengers. I expect many lives will be lost if you stop a train with 8 million passengers for ~15 minutes. I would probably not stop the train with even less passengers, maybe a few hundred thousand.

Edit: added the last paragraph, to drive my point home.

[taneq]

Once upon a time I used much the same reasoning as your last point to calculate the expected total change in quality-adjusted life years for reducing the speed limit on our local freeway to various points. I don't remember the exact outcome (although I'm pretty sure that reducing it below something like 80km/h was a net loss) but it was an interesting thought experiment to begin with, and it got a lot more interesting once I started getting angry responses to my results. :P