Measuring the effects of borrower-based policies on new housing loans

The paper evaluates the outcomes of the borrower-based macroprudential policy measures that were introduced in Estonia in 2015. The core of the analysis is the response of the credit market in 2016-2021 to the upper limit on the debt service-to-income ratio for new housing loans. The paper employs a novel data-driven approach based on the distribution of loan-level data, which does not need to account for the changes in the macroeconomic environment, and which allows the analysis to be made under assumptions that are more flexible than those usually used in the literature. The average number of borrowers affected, meaning those who were rejected or got a smaller loan than they wanted, is estimated to be on average around 11 per cent of the total number of new loans, which translates into roughly 10 per cent of the volume of new loans. The effect is fairly stable over the whole period despite the volume of housing loans growing in recent years.