The monkey is cheaper ... and a "counterfactual portfolio" where the researchers randomly added the money to an existing holding instead. The problem arises when they come to sell.