Why paying for blood donations reduced supply

Offering a fee to blood donors was expected to raise supply. In a Swedish field experiment it did the opposite among women: willingness to donate fell substantially once a payment was introduced, and recovered when the same payment could be redirected to charity. Nobody involved objected to money.

What everyone sees

The intuitive model is additive. People already have some reason to donate, a payment adds another, and more reason should produce more donation. When the number falls, the natural explanation is that the fee was too small to matter, and the recommended remedy is a larger fee.

What is actually happening

Motivation is not additive when the payment changes what the act means. An unpaid donation is a statement about the donor; a paid one is a transaction, and the statement is gone. Titmuss argued this before it was tested, and later crowding-out work found small payments performing worse than none at all in several settings, with the effect strongest where the original motive was reputational.

Why it stays hidden

The mechanism hides because the two motives are never visible at once. Once a payment exists, the unpaid version cannot be observed beside it, so the counterfactual is unavailable to the organisation making the decision. A scheme that underperforms is then diagnosed as underfunded, which produces a larger payment and a larger version of the same problem.