Poor people predominate as a subgroup of those who take part in healthy volunteer research. They are subjected to minimised but unknown risks and unpleasant burdens so that the safety of new medicines can be evaluated. This is prima facie unfair especially given that the poor are often unable to access expensive medicines. Although participants in this kind of research often do receive compensation for their time, these payments are usually capped at a very low level. This paper defends a version of a reimbursement model for the payment of research subjects. This model is practical, would benefit those without an income who take part in research, and would make it possible for those in regular work to take part in phase 1 research.