Monetary policy transmission may be impaired if banks rebalance their portfolios toward securities. We identify the bank lending and risk-taking channels of monetary policy by exploiting—Italy's unique—credit and security registers. In crisis times, with higher central bank liquidity, less capitalized banks react by increasing securities over credit supply, inducing worse firm-level real effects. However, they buy securities with lower yields and haircuts. Unlike in crisis times, in precrisis times, securities do not crowd out credit supply. The substitution from lending to securities in crisis times helps less capitalized banks repair their balance sheets and restart credit supply with a one-year lag.

Do banks extending government-guaranteed loans simultaneously reduce their risk exposure to firms? Using unique euro-area credit register data and the COVID-19 guarantee programs as a laboratory, we find that 1 euro of guaranteed lending was associated with a reduction of 28 cents in non-guaranteed credit, relative to other banks lending to the same firm. Substitution was highest for riskier and smaller firms in more affected sectors and for stronger banks. Nevertheless, banks offered cheaper credit and longer maturities to guaranteed loan recipients, especially more fragile ones. This improvement in lending terms is the flipside of credit substitution.

By providing liquidity to depositors and credit-line borrowers, banks can be exposed to double-runs on assets and liabilities. For identification, we exploit the 2007 freeze of the European interbank market and the Italian Credit Register. After the shock, there are sizeable, aggregate double-runs. In the cross-section, credit-line drawdowns are not larger for banks more exposed to the interbank market; however, they are larger when we condition on the same firms with multiple credit lines. We show that, ex-ante, more exposed banks actively manage their liquidity risk by granting fewer credit lines to firms that run more during crises.

We show that negative interest rate policy (NIRP) has expansionary effects on credit supply through a portfolio rebalancing channel. By shifting down and flattening the yield curve, NIRP differs from rate cuts just above the zero-lower-bound and has effects similar to QE. For identification, we exploit ECB’s NIRP and the Italian credit register and, for external validity, European and U.S. datasets. NIRP affects more banks with higher ex-ante liquid assets, including net interbank positions. More exposed banks reduce liquid assets, expand credit supply, especially to financially-constrained firms, and cut loan rates, inducing firms to increase investment and the wage bill.