Abstract

We examine welfare outcomes for non-market allocation mechanisms of public housing. We make use of a quasi-natural experiment in Amsterdam where public houses are allocated through a lottery to households already on waiting lists. First, we show that a lottery generates substantial windfalls to winners: it reduces their waiting time by 8.5 years, generating lifetime benefits valued at about €20,000. We continue to show that lotteries result in substantially different matches between households and public housing, as lottery winners tend to be much younger. Using information about the choice set of houses offered to households on waiting lists and lotteries, we demonstrate that lotteries induce inefficient matching: the monthly welfare loss of using lotteries rather than waiting list is estimated to be €113 per house, about 13% of the resource costs.