Tuesday, November 17, 2015

Cost-Benefit Analysis on Accepting Syrian Refugees: A Back-of-the-Envelope Calculation

Disclaimer: All numbers below are estimates. Some or all are likely wrong. Please don’t get mad at me for using some estimate other than the one you would have used. Feel free to do your own Fermi estimates, if you like. I repeatedly disclose the fuzziness of my numbers below (as well as in the very title of this post!), so please, dear reader, don’t think I’m trying to pull one over on you.

220,000 people have been killed in the Syrian civil war, out of a population of about 22 million. Assume that an average Syrian has about a 1% chance of dying from the civil war. This is huge. Americans find this kind of risk intolerable. Most of us would flee from it, too, if we had the chance.

There are about 200,000 terrorists among the world’s 1 billion Muslims (and this is probably an inflated estimate). So a random person from the Muslim world has about a 0.02% chance of being a terrorist. Assume (crudely) that this is the proportion of Syrian refugees who are terrorists.

Using these estimates, if we let 10,000 Syrian refugees into our country, we save 100 lives and let in 2 terrorists. Assuming (generously) that each terrorist manages to kill on average one person once here, the balance is still favorable to letting in refugees. There is a great deal of room here for making the assumptions more conservative before you reach the opposite conclusion. Let’s go ahead and increase the “chance of being a terrorist” by a factor of 10; the resulting balance still favors letting the refugees in. Presumably terrorists are trying extra-hard to sneak in; then again, presumably we’re trying extra-hard to screen them out. Presumably those people seeking asylum are at a *greater* risk than the average Syrian, so the above estimate of “lives saved” is probably conservative. Also, we failed to account for the benefit to those Syrians who wouldn’t have died anyway, but who still escape daily terror and enjoy a vastly improved quality of life. (Quick quiz, what’s the GDP per capita in Syria and what is it in the US? Wouldn’t it be nice to increase your annual income by such a large factor overnight?) The potential gains in quality-of-life are enormous.

Suppose you accept the estimates in the first sentence of the previous paragraph. (I won’t call them “My estimates”, because I could just as easily convince myself of some other set of estimates.) There’s one obvious way to square the circle and reach the “ban all Syrian refugees” conclusion. You might say, “Well, I value an American life at 50 times a Syrian life.” Some people, explicitly or implicitly, take this position. That’s fine. I’m not going to denounce you or spew venom at you for taking that position. (You need to be part of the discussion, too!) But I do consider this radical discounting of human life kind of ugly. I want that ugliness out in the open for everyone to see. I don’t want it buried under hollow slogans. I don’t want it shrouded in clever-but-imprecise rhetoric. By all means, value an American life more than a Syrian life. But by a factor of 50? Something in the 1-4 range might be reasonable, but at some point this differential starts to look monstrous. (Of course, a different back-of-the-envelope estimate might reach a more modest answer. Feel free to supply your own figures and assumptions.)


I think there’s a lot of low-hanging fruit here. We can probably select for those refugees who are almost certainly *not* terrorists. Families with young children, old women, people who can be credibly vouched for, etc. We can always be selective and exclude young single men or people with questionable histories. And we can always use the refugees themselves as intelligence assets. Who better to inform on ISIS terrorists than their former neighbors? And how great would it be to lure a known ISIS terrorist to our country and promptly arrest him? Crude as it is, I think my “100 lives saved per 2 lives sacrificed” is fairly conservative because we can do quite a lot to guard against the 2 lives lost.