We propose a model of collateral choice by banks to quantify how changes in the haircut policy of the central bank affect the collateral used by banks and the funding cost of banks. We estimate the model using data on assets pledged to the European Central Bank from 2009 to 2011. Our results suggest that an increase in the haircut on low rated collateral by 5 percentage points would have reduced the use of this collateral by 10% but would have increased the average funding cost spread between high yield and low yield countries by 5% over our sample period.

Should central banks lend against low quality collateral? We characterize efficient central bank collateral policy in a model where a bank borrows from the interbank market or the central bank. Collateral has favorable incentive effects but is costly to transfer to lenders who value the collateral less because of imperfect collateral quality. We show that a fall in the quantity or the quality of the bank’s collateral can increase interest rates in the economy even with a constant policy rate. A looser central bank collateral policy can reduce the spread, alleviate the credit crunch and increase output.

We model mortgage refinancing as a bargaining game involving the borrowing household, the incumbent lender, and outside banks. We show that bargaining can provide a competitive advantage to the incumbent bank. In equilibrium, the borrower’s ability to refinance depends on the incumbent bank’s cost (dis)advantage relative to locally present competing banks and on the average creditworthiness of borrowers in the relevant market. It is also driven by borrower impatience and switching costs. We find empirical support for the key predictions of our model in an administrative data set covering the universe of mortgages in Belgium.

Using security-level holdings for all euro-area investors, we study portfolio rebalancing during the quantitative easing program from March 2015 to December 2017. Foreign investors outside the euro area accommodated most of the Eurosystem’s purchases. Duration, government credit, and corporate credit risk did not get concentrated in particular regions or investor sectors. We estimate a demand system for government bonds by instrumental variables to relate portfolio rebalancing to yield changes. Government bond yields decreased by 65 basis points on average, and this estimate varies from 38 to 83 basis points across countries.

We use new and comprehensive data on the security holdings of euro-area investors to document facts about the ongoing quantitative easing program. The holdings of purchase-eligible government bonds have strong home bias not only for banks but also for insurance companies, pension funds, and mutual funds, especially in the vulnerable countries. In response to the program, foreign investors sold most of the purchase-eligible government bonds. Banks also sold purchase-eligible government bonds to a lesser extent, but insurance companies and pension funds bought them. Thus, quantitative easing may have reduced the duration mismatch for these institutions.