This paper analyzes the relation between the quality of the legal enforcement of loan contracts and the allocation of credit to households, both theoretically and empirically. We use a model of household credit market with secured debt contracts, where the judicial system affects the cost incurred by banks to actually repossess the collateral. The model shows that the working of the judicial system affects both the probability of being credit-constrained and the equilibrium amount of debt. In the empirical part, we test our predictions using data on Italian households and on the performance of Italian judicial districts. Controlling for household characteristics, unobserved heterogeneity at judicial district level and aggregate shocks, we document that an increment in the backlog of trials pending has a statistically and economically significant positive effect on the household probability of being turned down for credit. Furthermore, we show that moving a household from the high-cost judicial district (in southern Italy) to the low-cost judicial district would reduce his probability of being credit-constrained by 50% on average, other things being equal.