This paper uses a survey-based approach to test alternative methods of channeling tax relief to donors - as a tax rebate for the donor or as a matched payment to the receiving charity. On accounting grounds these two are equivalent but, in line with earlier experimental studies, we find that gross donations are significantly more responsive to a match change than to a rebate change. We show that the difference can largely be explained by the fact that a majority of donors do not adjust their nominal donations in response to a change in subsidy. This evidence adds to the growing empirical literature suggesting that consumers may not react to tax changes. In the case of tax subsidies for donations, this has implications for policy design - we show for the UK that a match-based system is likely to be more effective at increasing money going to charities.