It’s important to make sure that your empirical claims are correct–especially if you’re using them to make arguments that directly affect people’s lives. Last year the World Health Organization published a document entitled “Toward Self-Sufficiency in Safe Blood and Blood Products based on Voluntary Non-Remunerated Donation”, which argued that the UN states should, well, become self-sufficient in blood and blood products based on voluntary non-remunerated donations. Leaving aside the oddity of the desire that states become self-sufficient in blood procurement, there are two remarkable things about this document. The first is its hostility to the idea of compensating people for donating blood and plasma. The second is its cavalier disregard for the facts. I suspect that these two aspects of this document might be related. To take just one example, the WHO re-iterates the hoary old Titmussian line that “Paying some people to donate blood while others donate on a voluntary basis has been shown to crowd out altruistic donors” (“Toward….” 10). As far as I know, this claim is simply false; paying people to give blood or plasma secures more blood, not less. So I was interested in the study that the WHO was citing to support this claim: Alena M. Buyx, “Blood Donation, Payment, and Non-Cash Incentives: Classical Questions Drawing Renewed Interest,” Transfusion, Medicine, and Hemotherapy 36, 5 (2009), 329 – 339. So I read it. The closest that Buyx comes to supporting the WHO’s claim here is to refer to an unpublished Swedish study that was designed to examine whether crowding out of blood donation occurred in a real-life donor situation. The data showed that a small crowding-out effect occurred among some donor groups, but that this could be countered by the introduction of non-financial incentives. As Buyx notes in conclusion, “Quite generally, a crowding out could of course also be potentially offset by a significant increase of the amount of money paid, since it is only present when relatively low sums of money are paid”. To be fair to the WHO, strictly speaking Buyx’s reference to the unpublished Swedish study does support the claim that crowding out “has been shown” to occur–especially if one is being both pedantic and disingenuous. But to take just one part of one unpublished study that shows that some crowding out can occur, ignore the rest of the very same study that shows that such crowding-out is not the final result of offering incentives, and also to ignore the conclusions that are explicitly drawn by the author of the paper you are citing…. Well, at best that’s incompetent, and at worst intellectually dishonest and deliberately misleading. I think there are two lessons to be learned here. First, those who claim that compensating blood and plasma donors will crowd out altruistic donation and lead to less blood and plasma being secured overall still need to provide support for their claims. (And this support won’t be forthcoming–the evidence is overwhelming that paying for blood and plasma secures more rather than less.) Second, also check references. This is especially important for a classical liberal. After all, given the wealth of empirical evidence that shows that markets work, getting the truth out and correcting misinformation is a consummation devoutly to be wished.