Crowding out in blood donation: Was Titmuss right?
Mellstrom and Johannesson try to offer a very clean test of Titmuss' crowding out hypothesis (not called that at the time) offered in his book The gift relationship. In his book, Titmuss argues that the introduction of monetary rewards can reduce the supply of blood donors.
Mellstrom and Johannesson's paper presents a field experiment designed to test this. They use three treatments:
- Subjects are given the opportunity to donate blood without any special incentives.
- Subjects are given the opportunity to donate blood with a payment of 50 SEK (~ 7 USD).
- Subjects who choose to give blood were give the option of accepting 50 SEK or donating the same amount to charity.
The findings in the full sample showed no crowding out. However, when the authors split the sample into men and women, they found a statistically significant effect with women, but not with men. Women were less likely to donate blood when money was offered while they could not conclude that were. The charity option was associated with an increase in the rate of donations for women, but not for men.
There were several important limitations. For one, the authors advertisement for the study (due to ethics boards requirements) explained that the study was on attitudes to donate blood. Second, they do not actually measure blood donation but rather when the people recruited went to a physical exam which in Sweden is necessary to give blood. Apparently, most people who receive these exams will subsequently give blood (~70%).