I’ve spent the day at a workshop on benefit-cost analysis where a lot of discussion is on valuing policies that reduce risks to life of various kinds. US policy, for better or worse, is focused on the idea of Value of a Statistical Life. Typically a policy that reduces risks of death will be approved if the cost per life saved is below $5million, and not otherwise. (There are similar numbers applied to publicly funded health care services, prescription drugs and so on, usually per year of life saved).A striking thing I found out is that anti-terrorism policies of the Department of Homeland Security are subject to the same benefit-cost requirements as EPA and Transport. But Homeland Security is only one way the US government spends money with the aim of protecting Americans against attacks from terrorists and other enemies. Defense spending is far bigger and not subject to BCA, even though money spent on defense is money that can’t be spent on reducing terrorism risk through DHS or more reliably on reductions in environmental, health and transport riskThe numbers are quite striking. The ‘peacetime’ defense budget is around $500 billion a year, and the various wars of choice have cost around $250 billion a year for the last decade (very round numbers here). Allocated to domestic risk reduction, that money would save 150 000 American lives a year.So, since 9/11, US defense spending has been chosen in preference to measures that would have saved 1.5 million American lives. That’s not a hypothetical number – it’s 1.5 million people who are now dead but who could have been saved. I think its fair to say that those people were killed by the Defense Department, or, more precisely, by the allocation of scarce life-saving resources to that Department.
What can be said that might suggest that the defense budget didn’t really cost all those lives.?
Some objections can be dismissed fast.First, a lot of people are uncomfortable with notions like valuing statistical lives. But there’s no need to believe in this notion as far as the argument here is concerned. What matters is that there are lots of policy options that would save lives at a cost of around $5 million each, and those options are not being taken up.
Second, there are various responses that amount to the claim that refusing to do things that would reduce death risks is, in some important moral sense, different from doing things that increase death risks.  Maybe that’s arguable for individuals, as shown by the endless debates over trolley problems. But I can’t see that the distinction can be made meaningful for governments. Choosing one policy rather than another will raise the risk of death for some group of people, and perhaps lower it for another. This is true whatever choices governments make.More seriously, it’s not really plausible to think of eliminating defense spending altogether. But if the US spent 2 per cent of GDP like other rich countries (around $250 billion a year) and didn’t engage in wars of choice, it could have saved a million US lives over the past decade.A still more serious objection is that money saved on defense wouldn’t be used to save lives anyway. A couple of responses to this. First, even if the money was just handed back in tax cuts, around 15 per cent would probably be allocated to health care and more to things like education that are positively correlated with health status. Rounding to 20 per cent, that would still have saved something like 100 000 lives over a decade.Second, saving American lives is much more expensive than saving lives in poor countries. US military interventions are usually presented as being, at least in part, a kind of foreign aid. But civilian foreign aid can save lives at a much lower cost, perhaps 100 times lower. After deducting various forms of quasi military aid the US currently spends around $10 billion a year on development aid. Diverting 2 per cent of regular defense spending would allow that to be doubled, and could save something like a million lives a year.
fn1. As Clough put it “Thou shalt not kill, but needst not strive, officiously to keep alive”.
Update A lot of commentators have expressed doubt that there are policy options that could save 100 000 lives a year at a cost close to $5 million per life. There’s a valid point here, but not nearly as strong as many seem to think. Suppose the cost per life saved rose linearly from $5 million to $10 million as the amount spent increased. The average cost would be $7.5 million, which would allow for saving more than 65000 lives a year. And, in fact, there are many options that would save lives at lower cost than this. The Obama administration has adopted or proposed a range of air quality improvement measures that are estimated to save between 20 000 and 50 000 lives a year, with a best estimate of the cost that is substantially less then the official VSL of $5-6 million. These measures could have been adopted a decade ago, and saved between 200 000 and 500 000 lives. At the same time, Obama rejected ozone rules proposed by the EPA that could have saved thousands more lives per year, at a cost above the official VSL but below $10 million per life. That’s just one policy area, and a relatively small one compared to health care.
I’ll mention one other example that came up at the workshop. About 300 people a year, mostly children under 5, die in driveway “backover” accidents. The Department of Transport looked at a requirement to fit cameras that would provide rear visibility to drivers. This proposal would normally have been rejected because the cost came in at $7.5 million, but the accidents are so horrific in their consequences for families that they found a way to bend the rules. I’m pointing to this one, not because the numbers are large, but to point up the kinds of tragedies that are the opportunity cost of all those military interventions. The money it costs to keep 30 US soldiers in Afghanistan for a year ($1 million apiece) could save the lives of four American children, or many times that number in a poor African country.
fn2. More precisely, premature deaths avoided. For various political and theoretical reasons, the US standards treat all premature deaths saved as equally good, although they sometimes bend the rules a bit to place a higher value on children – this would happen automatically if the metric were life-years saved.