Friday, 28 April 2017

Diesel fumes: Is the UK’s witch-hunt against diesel cars driven by zealotry and greed, not science?

(Neil's Note: I've now done an updated and much re-written version of this paper. It can be found here:

So I'm removing the old version.)


Emanuel Crisp said...

I'll admit to skimming as this is rather long, but it looks like you've calculated social cost based entirely on deaths. If the harm from pollution was more like stabbings (fine one minute; dead the next) then I'd be right with you. My "honest common sense" suspicion about air pollution is that there are many chronic lung and heart conditions costing the NHS oodles before the deaths occur. I can't see how that is accounted for in the model. Going further there are probably a very large number of non-life-threatening ailments such as mild asthma and occasional coughs etc. which have a social cost to be added on.

These statistical approaches to cost are therefore very problematic, but I accept that they may be the best tool we have for deciding how to allocate resources on a national scale.

Neil said...

Emanuel, many thanks for your comment.

What I tried to do was adopt the same approach to the numbers that the government did. The HPA, who did the original figures on PM pollution, gave both an excess deaths figure and a "life-years-lost" figure. These are not directly related to one another. Indeed, they gave the average life years lost per excess death as about 11.7. Of course, I can't be sure that figure (or, indeed, any of their figures) is actually reasonable; they could easily be systematically out by an order of magnitude or two, and no-one but a toxicologist would be any the wiser. But they're the best figures I have.

What I could, and did, do was cross-check their numbers against the NHS's figure for the value of a life-year, and DEFRA's figure of the social cost of a life lost because of pollution. The conclusion I came to in the essay was: "the contribution of the lethal part of the pollution effects is around 62% of the total social cost stated by DEFRA." Meaning, that the non life threatening ailments which you mention must account for about 38%. That doesn't seem unreasonable to me, although perhaps it may be a little on the low side.

Much more problematic, I think, is the identification of deaths and symptoms caused by NOx as significant. There's a point of view, expressed in one of the reports, that NOx may possibly not increase the number of deaths at all above those caused by PM. If that point of view were to prove correct, then PM is the significant factor and NOx is an irrelevance. And since the PM problem is fixed for diesel cars made since 2006, the whole scare would be just as much of a fraud as the original "CO2 causes castrophic global warming" humbug.