Negative Monetary Policy Rates and Systemic Banks' Risk‐Taking: Evidence from the Euro Area Securities Register

• Author: JOHANNES BUBECK,ANGELA MADDALONI,JOSÉ‐LUIS PEYDRÓ
• Date: 01 October 2020
• DOI: http://doi.org/10.1111/jmcb.12740
• Published date: 01 October 2020

DOI: 10.1111/jmcb.12740

JOHANNES BUBECK

ANGELA MADDALONI

JOSÉ-LUIS PEYDRÓ

Negative Monetary Policy Rates and Systemic

Banks’ Risk-Taking: Evidence from the Euro Area

Securities Register

Weshow that negative monetary policy rates induce systemic banks to reach-

for-yield. For identication, we exploit the introduction of negative deposit

rates by the European Central Bank in June 2014 and a novel securities

register for the 26 largest euro area banking groups. Banks with more cus-

tomer deposits are negatively affectedby negative rates, as they do not pass

negative rates to retail customers, in turn investing more in securities,

especially in those yielding higher returns. Effects are stronger for less

capitalized banks, private-sector (nancial and nonnancial) securities and

dollar-denominated securities. Affected banks also take higher risk in loans.

JEL codes: E43, E52, E58, G01, G21

Keywords: negativerates, nonstandard monetary policy, reach-for-yield,

securities, banks

Wethank Raj Iyer for providing valuable suggestions in the early stage of this project. We also thank our

discussants Ugo Albertazzi, Thorsten Beck, Matteo Crosignani, Mathias Kronlund, and Skander van den

Heuvel, as well as an anonymous referee and the editor Robert DeYoungfor useful comments and sugges-

tions. We also thank the participants at EFA2020, FIRS 2019 in Savannah (Georgia), the CEMFI/Banco

de España Second Conference on Financial Stability in Madrid, the 50th Anniversary Conference of the

JMCB in Frankfurt, the Eight Italian Congress of Econometrics and Empirical Economics in Lecce, the

Second Conference of the ESCB Research Cluster 3 on Financial Stability, and seminars at the Univer-

sity of Mannheim, BIS, ECB, Deutsche Bundesbank, and the Federal Reserve Bank of New York.Vale-

rio Morelli and Niklas Grimm provided excellent research assistance. This project has received funding

from the European Research Council (ERC) under the European Union’s Horizon 2020 research and in-

novation programme (grant agreement No 648398). Peydró also acknowledges nancial support from the

PGC2018-102133-B-I00 (MCIU/AEI/FEDER, UE) grant and the Spanish Ministry of Economy and Com-

petitiveness, through the Severo Ochoa Programme for Centres of Excellence in R&D (SEV-2015-0563).

The views expressed do not necessarily reect those of the European Central Bank, Deutsche Bundesbank,

or the Eurosystem.

J Bis at Deutsche Bundesbank (E-mail: johannes.bubeck@bundesbank.de). A

Mis at European Central Bank (E-mail: angela.maddaloni@ecb.europa.eu). J-L P-

is at Imperial College London, ICREA-Universitat PompeuFabra, CREI, Barcelona GSE, and CEPR

(E-mail: jose.peydro@gmail.com).

Received January 24, 2019; and accepted in revised form February 11, 2020.

Journal of Money, Credit and Banking, Supplement to Vol. 52, No. S1 (October 2020)

© 2020 The Ohio State University

198 :MONEY,CREDIT AND BANKING

C     series of unconven-

tional monetary policies during the last decade. An important difference in the imple-

mentation of monetary policy between the euro area and the United States has been

the use of negative policy rates (NPRs). The Federal Reserve in particular (and also

the Bank of England) has been somewhat critical of the use of negative rates (see

Bernanke 2016). In the euro area, however, negative policy rates were introduced by

the European Central Bank (ECB) in June 2014, when the deposit rate for banks with

an account at the central bank was lowered to −10 basis points. Negative rates have

also been introduced by central banks in several other countries (Japan, Denmark,

Sweden, and Switzerland), and the value of securities (bonds) yielding negative rates

reached around 15 trillion U.S. dollars (USD) as of July 2019.1Therefore, negative

rates currently represent an important policy tool, which going forward may be con-

sidered part of the standard toolkit, especially given the secular low levels of rates

around the world. Thus, a key question is what are the effects of negativepolicy rates

for banks?

In this paper, we analyze the impact of negative monetary policy rates on reach-

for-yield behavior by banks. For identication, we exploit the introduction of nega-

tive deposit rates by the ECB in June 2014 and a novel securities register for the 26

largest euro area banking groups. Banks with more customer deposits tend to be more

negatively affected by negative rates, as it is more difcult to pass negative rates to

their retail customers.2In sum, our results show that, after the introduction of nega-

tive rates, banks with higher (as compared to lower) ex ante customer deposits invest

relatively more in securities, especially in those securities yielding higher returns.

Reach-for-yield effects through this deposit channel are stronger for less capitalized

banks. Moreover, we nd that more affected banks also take higher risk in loans.

The transmission of monetary policy at the “zero lower bound” and below has

become a topic of particular interest for researchers and policymakers since several

central banks have chosen this path in the last few years (Brunnermeier and Koby

2018, Eisenschmidt and Smets 2018). A very recent literature is assessing how nega-

tive policy rates are transmitted through the banking sector and howt heyaffect credit

supply to the economy and the equity valuations of banks (see, e.g., Ampudia and Van

den Heuvel 2018, Basten and Mariathasan 2018, Bottero et al. 2019, Heider, Saidi,

and Schepens 2019, Eggertsson et al. 2019). However, to the best of our knowledge,

ours is the rst paper to analyze how negative policy rates affect investment choices

in the securities portfolios of banks.

Analyzing banks’ reach-for-yield in securities is important for several reasons.

First, securities holdings represent an important share of the assets held by banks

around the world: For example, in Europe they account for around 20% of total

banking assets. In part because of lack of detailed micro data, banks’ securities

1. See, for example, https://www.bloomberg.com/news/articles/2019-08-06/negative-yielding-debt-

hits-record-15- trillion-on-trade- woes.

2. In this paper,we refer to customer deposits or retail deposits (or simply deposits) as the same concept

(see Section 1).

JOHANNES BUBECK, ANGELA MADDALONI, AND JOSÉ-LUIS PEYDRÓ :199

portfolios have not been analyzed as much as credit. While credit register data are

available for several countries (in Europe, Asia, the Americas, and even Africa),

the new data set at our disposal, which provides detailed information on securities

holdings at the bank and security (the International Securities Identication Number

[ISIN]) level is not generally available for other regions in the world. For each secu-

rity, we have information on yield, issuer, ratings, price, and remaining maturity. In

particular we observe, even within the same issuer (like a sovereign), all the different

securities with different yields and maturities held by each bank in a certain quarter.

The overall nominal holdings in our database are valued at around 3 trillion euros.

Second, banks take risk through their securities holdings in addition to their lend-

ing portfolios. In Europe, for example, during the sovereign debt crisis, the sound-

ness of banks in some countries was severely affected by the decrease in prices of

sovereign bonds held in their portfolios, which represented a strong link between the

stability of the banking sector and of the sovereigns issuing the securities—the so-

called sovereign-bank nexus (Acharya and Steffen 2015, Brunnermeier et al. 2016).

As banks may take signicant risk via securities holdings, policymakers have dis-

cussed how to regulate the investment of banks in securities, which is at the center

of an important policy debate. In Europe, the Liikanen and the Vickers Reports ad-

dressed these issues and in the United States the Volcker Rule in the Dodd–Frank

Act. Finally, securities are more liquid than loans and bank risk can be more easily

reallocated by changing the positions in liquid assets (Myers and Rajan 1998).3

How are negativerates transmitted through the banking sector and affect bank risk-

taking? A reduction in the policy rate transmits to short-term rates rst. Since banks

tend to have long-term assets and shorter-term liabilities on their balance sheets, a

rate cut should result in an increase in bank net worth via banks passing-through the

rate cut on their liability side. However, the pass-through of negative rates is some-

what different: banks do not generally pass on negativerates to their retail depositors,

differently from their wholesale deposits. As we discuss below, these funding struc-

tures tend to be persistent and high retail deposit banks have less wholesale deposits

than low deposits banks, which implies that they are overall less able to pass the neg-

ative rates to their liabilities. Evidence from our sample of banks, and in related work

(see Altavilla et al. 2019), also suggest that these high retail deposit banks are not

increasing fees more than other banks on average.

When interest rates are positive, the impact of a cut in policy rates on bank risk-

taking is theoretically ambiguous (see Dell’Ariccia, Laeven, and Marquez 2014,

Dell’Ariccia, Laeven, and Suarez 2017). There are at least two different forces at

work. First, a portfolio reallocation effect—safe assets become less attractive, which

induces banks to take more risk. Second, a risk-shifting effect—lower interest rates

increase franchise value, which would induce less risk-taking. The net effects of these

3. Myers and Rajan (1998) allude to the concept of transformation risk. They argue that “greater asset

liquidity reduces the rm’s ability to commit to a specic course of action” and hence “managers could

substitute risky assets for safe ones”—or the other way around. Some recent evidenceprovided by Timmer

(2018) shows indeed that banks react strongly to past returns in the securities holdings and adjust their

holdings accordingly, leading to procyclicality in investmentdynamics.