Working Paper Series no. 877: Lost in Negative Territory? Search for Yield!

We study how negative interest rate policy (NIRP) affects banks’ loan pricing. Using contract-level data from France, we show that NIRP affects bank lending rates to firms through a portfolio rebalancing channel: banks holding a one standard deviation more of cash and central bank reserves offer a 8.6 basis points lower loan rate after NIRP is introduced. The impact concentrates on medium-term loans (with maturity comprised between three and six years) but not on loans to risky firms, indicating that banks conduct a search for yield focused on term spreads. These findings suggest that NIRP complements quantitative easing policies.

When a central bank sets its target interest rate in negative territory, the reserves that banks hold in excess of minimum requirements become costly to hold, and banks’ revenue may reduce. In an effort to offset the contraction in revenue, two mechanisms may determine how banks modify their loan pricing in response to NIRP. The first mechanism hinges on the downward rigidity of deposit rates: because they may not pass negative rates on depositors, banks react to NIRP by not fully transmitting the interest rate cut on borrowers. That is, banks relying more on deposit funding offer loans at relatively higher interest rates. The second mechanism works through portfolio rebalancing: since the opportunity cost of holding cash and reserves increases, banks aim to substitute away from those assets. Hence, banks holding more cash and reserves become more competitive in pricing loans.

We test which of these mechanisms explains banks’ reaction to NIRP by focusing on the introduction of this policy by the European Central Bank (ECB) on June 11, 2014. We exploit contract-level data on 121,519 loans lent by 77 banks to 84,048 firms from France. Considering French data is a distinctive element of our analysis: The French banking system is a priori amongst those most affected by the reduction in revenue caused by NIRP, being one of the main excess reserves holders in the euro area. Our approach employs difference-in-differences: we measure the difference in lending rates before and after the introduction of NIRP depending on the lending banks’ deposit ratio and cash and reserve ratio.

Our first finding is that the more a bank holds cash and reserves, the more it reduces lending rates when NIRP is implemented, in line with the portfolio-rebalancing channel. According to our preferred specification, a one standard deviation difference in the cash and reserve ratio leads a bank to offer a 8.6 basis points lower loan rate, that is a 3.6% lower loan rate relative to the sample median. Banks holding more cash and reserves are more competitive in pricing loans as a means to attract more corporate borrowers and compensate for otherwise larger drops in revenue.

NIRP incentivizes banks to substitute cash and reserves, which yield zero or negative remuneration, with corporate loans, which yield positive remuneration. That is, NIRP pushes banks to search for yield by taking more risk to increase return. To characterize the search for yield at play, we examine whether banks target borrowers at longer maturities–to earn the corresponding term spread–or borrowers with higher credit risk–to earn the related credit risk spread. We show that the impact peaks in magnitude primarily on medium-term loans. A one standard deviation difference in the cash and reserve ratio leads a bank to offer a 16 basis points lower loan rate after the implementation of NIRP if the maturity of a loan is comprised between three and six years, and a statistically insignificant 3 basis points lower loan rate if the loan maturity is up to one year. Conversely, we obtain no heterogeneous effect depending on firm credit risk (as measured by the firm credit rating). These results indicate that when NIRP is introduced, banks holding more cash and reserves target primarily borrowers at intermediate maturities with the purpose of earning term spreads. Those banks offer medium-term loans at relatively lower interest rates, suggesting that they associate such loans with a lower price of risk. Overall, these findings imply that NIRP flattens the middle of the corporate loan yield curve and can act as a complement to asset purchase programmes, which rather affect the longer end of the yield curve. 

Download the PDF version of this document

Working Paper Series no. 877: Lost in Negative Territory? Search for Yield!
  • Published on 06/28/2022
  • 40 pages
  • EN
  • PDF (1.1 MB)
Download (EN)

Updated on: 06/28/2022 09:55