Many cities around the world have introduced paid parking but implicitly subsidize parking for example by providing residential parking permits for street parking. We study the welfare effects of residential parking subsidies through changes in car ownership for Amsterdam. We employ a boundary-discontinuity design that exploits spatial variation in the length of waiting lists for permits and therefore in the size of the parking subsidy. In the city center, the waiting time for a permit is up to four years. Our results indicate that one additional year of waiting for a parking permit reduces car ownership with 2 percentage points corresponding to a price elasticity of car demand of −0.8. We demonstrate that subsidizing residential parking induces a substantial welfare loss. On average, a parking permit induces an annual deadweight loss of €270. Furthermore, we show that the provision of parking permits is an income-regressive policy: rich households are five times more likely than poor households to receive these (implicit) parking subsidies.