Who Borrows from the Lender of Last Resort?

• Author: ITAMAR DRECHSLER,DAVID MARQUES‐IBANEZ,THOMAS DRECHSEL,PHILIPP SCHNABL
• Date: 01 October 2016
• Published date: 01 October 2016
• DOI: http://doi.org/10.1111/jofi.12421

THE JOURNAL OF FINANCE •VOL. LXXI, NO. 5 •OCTOBER 2016

Who Borrows from the Lender of Last Resort?

ITAMAR DRECHSLER, THOMAS DRECHSEL, DAVID MARQUES-IBANEZ,

and PHILIPP SCHNABL∗

ABSTRACT

We analyze lender of last resort (LOLR) lending during the European sovereign debt

crisis. Using a novel data set on all central bank lending and collateral, we show that

weakly capitalized banks took out more LOLR loans and used riskier collateral than

strongly capitalized banks. We also find that weakly capitalized banks used LOLR

loans to buy risky assets such as distressed sovereign debt. This resulted in a reallo-

cation of risky assets from strongly to weakly capitalized banks. Our findings cannot

be explained by classical LOLR theory. Rather, they point to risk taking by banks,

both independently and with the encouragement of governments, and highlight the

benefit of unifying LOLR lending and bank supervision.

LENDER OF LAST RESORT (LOLR) actions represent one of the most dramatic in-

terventions by governments in financial markets, particularly during financial

crises. Such interventions have long been an important part of economic pol-

icy, even providing the motivation behind the establishment of central banks

such as the Federal Reserve. Given their central role in policy as well as their

∗Itamar Drechsler is with NYU Stern and NBER. Thomas Drechsel is with ECB and LSE.

David Marques-Ibanez is with ECB. Philipp Schnabl is with NYU Stern, NBER, and CEPR.

We thank Viral Acharya, Tobias Adrian, Jennie Bai, Markus Brunnermeier, Stijn Claessens,

Hans Degryse, Douglas Diamond, Mark Flannery, Xavier Freixas, Xavier Gabaix, Mark Gertler,

Simon Gilchrist, Pierre-Olivier Gourinchas, Anil Kashyap, Todd Keister, Anna Kovner, Arvind

Krishnamurthy, Ross Levine, Robert McDonald, Stijn van Nieuwerburgh, Gordon Phillips, Ben-

jamin Sahel, Tano Santos, Amit Seru, Amir Sufi, Alan M. Taylor, Stan Zin, and seminar partici-

pants at Bocconi University,Chicago Booth School of Business, Chicago Fed Structure Conference,

Columbia Business School, Cornell University,Darden International Finance Conference, Federal

Reserve Board, Kellogg School of Management, MIT Sloan, NBER Monetary Economics Meeting,

NBER Risks of Financial Institutions Workshop, NBER Capital Markets and Economy Group,

New York Federal Reserve Bank, New York University, SFS Finance Cavalcade, Tuck School of

Business, University of Connecticut, University of Illinois at Urbana–Champaign, University of

Lausanne, University of Maryland, University Pompeu Fabra, UC Berkeley, UC Davis Financial

Institutions Conference, UCLA Anderson, USC Marshall School of Business, Western Finance

Association Meeting, Wharton Business School, Wharton Conference on Liquidity and Financial

Crises, YaleUniversity, and the World Bank for helpful comments and suggestions. This paper was

written primarily while David Marques-Ibanez was at the Federal Reserve Board and the Federal

Reserve Bank of New York. The views expressed in this paper are solely the responsibility of the

authors and should not be interpreted as reflecting the views of the European Central Bank, Board

of Governors of the Federal Reserve System, or of any other person associated with the Federal

Reserve System or the European Central Bank. The authors declare that they have no relevant or

material financial interests related to the research in the paper. All errors are our own.

DOI: 10.1111/jofi.12421

1933

1934 The Journal of Finance R

magnitude, it is important to understand how these interventions impact the

financial system and, ultimately, the economy.

In this paper, we ask which banks borrow from the LOLR and why they do

so. The answers to these questions are important for understanding the costs

and benefits of LOLR interventions. If banks borrow from the LOLR to stop

a banking panic, then LOLR interventions will help alleviate a contraction in

bank lending (i.e., “credit crunch”) and reduce the impact of financial crises on

the real economy.This is the scenario envisaged by classical LOLR theory. How-

ever, if banks borrow from the LOLR for other reasons, for instance, to increase

risk-taking or in response to regulatory pressure, then LOLR interventions will

do little to mitigate financial crises and may harm financial stability.

We analyze the take-up of LOLR lending during the European sovereign

debt crisis. The European crisis is an almost ideal setting for our study because

LOLR lending was a central tool of the European Central Bank’s (ECB) strategy

for addressing the sovereign debt crisis, and the intervention was arguably the

largest and longest lasting LOLR program ever conducted. Moreover, the ECB’s

implementation largely followed the recommendations of LOLR theory. That

is, the ECB provided unlimited lending to a large set of banks, but only against

collateral that was good prior to the crisis.1

We use a novel, proprietary data set that contains weekly bank-level records

of all banks’ ECB borrowing for the period August 2007 to December 2011, as

well as precise, security-level information on the collateral each bank pledges.

We match these data to publicly available data on bank characteristics, bank

asset holdings, and security characteristics. The sample covers the universe of

banks in Europe.

Our analysis yields three main findings. First, following the onset of the

European sovereign debt crisis in May 2010, weakly capitalized banks took

out more LOLR loans and pledged riskier collateral against their loans than

did strongly capitalized banks. Using banks’ precrisis credit ratings as a mea-

sure of bank capital, we find that a one-standard-deviation decrease in a bank’s

credit rating leads to a 11.2 percentage point increase in the likelihood of LOLR

borrowing and a 14.6% increase in the amount of LOLR borrowing after May

2010.2It also leads to a reduction in a bank’s value-weighted average collat-

eral rating of 0.3 notches and a 36 bps increase in its pledging of distressed-

sovereign debt (Cyprus, Greece, Italy, Ireland, Malta, Portugal, and Spain)

relative to bank assets.3The results are economically significant, accounting

1Our focus is on LOLR lending undertaken as part of the ECB’s normal credit operations, which

represents the bulk of its lending during the sample period. The ECB has referred to this lending as

the “monetary approach” to LOLR (Praet (2016)). We do not analyze lending under the Emergency

Liquidity Assistance program, which focused on banks that had insufficient eligible collateral and

was administered by national central banks.

2We use precrisis credit ratings (August 2007) to avoid capturing the direct impact of the crisis

on ratings.

3Throughout the paper,we define distressed countries as those that were downgraded below AA

after the onset of the sovereign debt crisis, namely,Cyprus, Greece, Ireland, Italy, Malta, Portugal,

and Spain. We refer to their debt as distressed-sovereign debt.

Who Borrows from the Lender of Last Resort? 1935

for 23%, 18%, 22%, and 21% of a standard deviation of the respective outcome

measures.

Second, weakly capitalized banks borrowed at least in part to buy risky

assets such as distressed-sovereign debt. Using panel data from the European

bank stress tests, we analyze whether banks that pledged more distressed-

sovereign debt with the LOLR also actively purchased this debt. Indeed, we

find that a 10% increase in a bank’s pledging of distressed-sovereign debt is

associated with a 4.5% increase in its holdings. This relationship is driven

entirely by weakly capitalized banks and is stronger for distressed-sovereign

debt originated in the bank’s home country.

Third, we show that, in aggregate, risky assets pledged with the LOLR moved

from strongly to weakly capitalized banks. Following the start of the European

sovereign debt crisis, about a third of total distressed-sovereign debt pledged

with the LOLR moved from strongly to weakly capitalized banks. Similarly,we

find that for the pool of all distressed-country-originated debt, which includes

mortgage-backed securities, covered bonds, and other debt instruments, ap-

proximately one quarter of the total moved from strongly to weakly capitalized

banks.

Taken together, our findings do not support classical LOLR theory. Classical

LOLR theory predicts that LOLR lending stops bank runs by allowing banks

to continue financing their existing assets. This removes the need for banks

to sell assets at fire-sale discounts and enables them to continue lending to

firms and households, averting a credit crunch. Hence, classical LOLR theory

says that banks borrow because they suffer runs, not necessarily because they

have low capital, which is what we find. Indeed, our results hold when we

exclude the likely sources of runs, when we limit attention to banks located

in the nondistressed countries (which did not suffer runs), when we control

for country-specific factors (which were the likely source of runs), and when we

measure capital before the start of the financial crisis (to avoid runs causing low

capital). Moreover, whereas classical LOLR theory says that banks use LOLR

borrowing to finance existing asset holdings, we find that weakly capitalized

banks used LOLR funding to actively buy risky assets.

Our findings thus point toward alternative LOLR theories that emphasize

banks’ incentives to increase risk-taking. Specifically,weakly capitalized banks

have an incentive to borrow from the LOLR because they are close to default

and hence their equity holders do not internalize losses conditional on default.

Consequently,they want to buy risky assets, particularly those whose downside

is realized when the bank defaults. This theory explains the relationship we

find between a bank’s capital, a proxy for its proximity to default, and its LOLR

borrowing and collateral risk. It further explains why weakly capitalized banks

invested in distressed-sovereign debt.

Banks may also borrow from the LOLR because of political economy pres-

sures. In particular, regulators and politicians in the distressed countries may

have encouraged banks to buy their home country’s sovereign debt to help

fund their governments. Weakly capitalized banks are the most susceptible to

such pressure because they are the most reliant on regulatory approval. This