The responsiveness of long-term household debt to the interest rate is a crucial parameter for assessing the effectiveness of public policies aimed at promoting specific types of saving. This paper estimates the effect of a reform of Credito Bonificado, a large program in Portugal that subsidized mortgage interest rates, on long-term household debt. The reform established a ceiling in the price of the house that could be financed through the program, and provides plausibly exogenous variation in incentives. Using a unique dataset of matched household survey data and administrative records of debt, we document a large decrease in the probability of signing a new loan after the removal of the subsidy.