PreprintPDF Available
Preprints and early-stage research may not have been peer reviewed yet.

Abstract

We investigate the effectiveness of the bank lending channel, that is, whether, and if so how, the accommodative monetary policies of the European Central Bank (ECB) mitigated the disruption in bank lending between 2008 and 2014. We show that both standard and non-standard measures of the ECB's accommodating monetary policy alleviated banks' funding constraints, helping support their lending activities in the syndicated loan market. We highlight a cross-sectional asymmetry in the banks' responses to both measures. By differentiating banks according to their size, funding constraints, and financial strength, we find that lowering the EONIA increases loan amounts provided by large and less liquid banks while unconventional policies support effectively smaller and highly capitalized banks, as well as banks with a high deposits and short-term debt ratio, and weaker banks. After the 2008 shock, the standard measures quickly reached their limits, highlighting the need to develop new monetary policy tools to support the lending activities of banks that needed it the most, i.e., that are small and financially constrained. As such, we show that the ECB was successful in doing so, with the implementation of non-standard monetary policy tools significantly supporting the loan offer of these banks after the crisis.
1
Lenders asymmetric reaction to the ECB’s monetary policy:
The case of the syndicated loan market
Aurore Burietza*
Matthieu Picaultb
Abstract:
We investigate the effectiveness of the bank lending channel, that is, whether, and if so how, the
accommodative monetary policies of the European Central Bank (ECB) mitigated the disruption
in bank lending between 2008 and 2014. We show that both standard and non-standard measures
of the ECB’s accommodating monetary policy alleviated banks’ funding constraints, helping
support their lending activities in the syndicated loan market. We highlight a cross-sectional
asymmetry in the banks’ responses to both measures. By differentiating banks according to their
size, funding constraints, and financial strength, we find that lowering the EONIA increases loan
amounts provided by large and less liquid banks while unconventional policies support effectively
smaller and highly capitalized banks, as well as banks with a high deposits and short-term debt
ratio, and weaker banks. After the 2008 shock, the standard measures quickly reached their limits,
highlighting the need to develop new monetary policy tools to support the lending activities of
banks that needed it the most, i.e., that are small and financially constrained. As such, we show
that the ECB was successful in doing so, with the implementation of non-standard monetary policy
tools significantly supporting the loan offer of these banks after the crisis.
Keywords: Syndicated loans, financial crisis, bank lending channel, European Central Bank, Non-
standard monetary policies
JEL classification: E52, F34, G21
__________________________
aIESEG School of Management, Univ. Lille, CNRS, UMR 9221 - LEM - Lille Economie Management, F-59000 Lille,
France. Phone : +33 320 545 892. a.burietz@ieseg.fr. *Corresponding author.
bUniversité d’Orléans, LEO, UMR 7322, Rue de Blois BP 26739, Orléans Cedex 2 45067, France. Phone .: +33
155 911 010. matthieu.picault@univ-orleans.fr.
2
1. Introduction
In 2008, Lehman Brothers filed for bankruptcy, triggering one of the most significant financial
crises in banking history and deeply affecting the syndicated loan market with a 60% drop in the
issuance volume between 2007 and 2009, and a run on the lines of credit granted before the crisis
for a total of 26.8 billion dollars (Ivashina and Scharfstein, 2010; Cerutti et al., 2015). The
syndicated loan market is a major source of external financing for firms, and it represents more
than one-third of all international corporate financing, including money market instruments, bonds,
and equities (Gadanecz, 2004).
1
A large body of literature explores the impact and the international
transmission of the financial crisis on the syndicated loan market (De Haas and Van Horen, 2012).
2
Banks suffered from a run that affected banks’ balance sheets, damaging their liquidity position,
and reducing new-loan origination given to large corporations. The confidence crisis, combined
with an increase in uncertainty, made banks reluctant to lend money providing them with incentives
to increase loan spreads (Santos, 2011) and to start hoarding liquidity for precautionary purposes
because of the rise in their funding risk (Acharya and Merrouche, 2013).
Amid the magnitude of the financial shock and the increasing pressures on the banking industry,
central banks
3
intervened to reduce strains on the financial markets and provide credit institutions
with financial support using both standard and non-standard tools.
4
In front of the magnitude of
these unprecedented measures, the literature has been extended to better understand the impact of
1
A syndicated loan is a hybrid of a bank loan and public debt, and it gathers together commercial banks and other
financial institutions, implying both are responsible for monitoring and underwriting activities (Dennis and
Mullineaux, 2000; Chaudhry and Kleimeier, 2015).
2
In an extensive study, Kleimeier et al. (2013) analyse the impact of roughly 200 financial crises on the geographical
repartition of cross-border loans from 1995 to 2008. By distinguishing between banking, currency, and twin crises,
the authors highlight significant differences among the types of crises, with stronger effects emerging from twin
financial turmoil.
3
See Meier et al (2021) for an extensive review of all policy measures (fiscal, monetary, regulatory, …) used after
both the financial crisis and the sovereign debt crisis.
4
Fawley and Neely (2013) provide a precise description of the quantitative easing programmes implemented by the
Federal Reserve (Fed), the Bank of England (BoE), the ECB, and the Bank of Japan (BoJ).
3
these innovative tools but remains barely developed regarding the measures’ effects on the
syndicated loan market. The goal of the current paper is twofold. First, we assess the impact of the
ECB’s accommodating monetary policy on the syndicated loan market, disentangling between
standard and non-standard instruments. We hypothesise that the measures implemented by the ECB
supported syndicated bank lending, reducing the impact of the 2008 financial crisis. More
precisely, we estimate the effects of the ECB’s interest rates and balance sheet policies on the
issuance volume of syndicated loans (bank lending channel). By providing credit institutions with
funds, the ECB alleviated the constraints on banks’ balance sheets, providing them with liquidity
and more flexibility to allocate resources. Second, as credit institutions do not face the same costs
to access alternate sources of funding, we study the banks’ asymmetric response to monetary policy
shocks by differentiating banks according to five different financial indicators: size, capital level,
customers’ deposits dependence, short-term debt reliance, and Tier 1 capital ratio. One objective
of the ECB when implementing an accommodative monetary policy is to support banks’ liquidity.
As such, smaller institutions' constraints, including their reliance on deposits or debt, can be
alleviated by this monetary policy. However, the use of such a policy can further worsen banks’
capital and Tier 1 capital ratio and so their financial strength. Therefore, lowly capitalised banks
as well as banks with a financially weaker position may not use the liquidity obtained from the
central bank to increase their lending supply on the syndicated loan market.
To empirically test our hypotheses, we estimate a cross-section regression for a sample of nineteen
European banking groups between 2008 and 2014. We analyse the potential effects of two
monetary policy instruments (i.e., the interest rate and non-standard ECB policies) on syndicated
bank lending by using the LPC Dealscan database. Besides, we introduce interaction terms between
monetary policy measures and five distinct banks' financial ratios to assess the effectiveness of the
bank lending channel. The analysis is run at the bank level to investigate the asymmetric
4
transmission of the monetary policy. We pay particular attention to the microeconomic foundations
of bank lending activities by combining seven databases allowing us to use loan-specific data rather
than overall lending aggregates (Popov and Van Horen, 2015). We also control for the banks’
heterogeneity by using bank and time fixed effects.
The major identification challenge is to disentangle between credit supply and credit demand
because both can be affected by a change in monetary and economic conditions. To address this
identification challenge, we control for credit demand using macro and microeconomic variables
in addition to borrower fixed effects. First, we consider the change in the Gross Domestic Product
(GDP) of the Eurozone to account for variations in business cycle conditions (Jiménez et al., 2012)
as well as the annual growth rate of the borrower’s country GDP. Second, we use the bank lending
survey that is provided by the ECB every quarter to build a proxy for the banks’ anticipation of
credit demand (Del Giovane et al., 2011). Moreover, we control for potential demand effects using
microeconomic variables such as the industry of the borrower and its credit rating. The first
measure integrates any productivity shock occurring in one specific sector while the second
measure allows us to evaluate the financial position of the borrower. Both are key determinants for
borrowers’ demand for loans. Finally, we add borrower fixed effects to our model in order to
control for unobservable time-invariant borrower characteristics.
Overall, we find that both the standard and non-standard measures strengthened bank lending
activities by increasing syndicated loan volume. The analysis shows that a decrease in the
benchmark rate and an increase in the ECB balance sheet are associated with larger loan amounts.
However, we highlight a cross-sectional asymmetry in the banks’ response. Standard measures
contribute to the increase in loan amounts provided by large and less liquid banks while non-
standard measures are more efficient in supporting the credit supply of small and highly capitalized
banks as well as banks with more funding constraints or that are weaker. Our findings confirm the
5
existence and effectiveness of the bank lending channel for the syndicated loan market within the
studied period. The innovation is the nature of the instruments that are found to be effective in the
transmission of monetary policy. After the Lehman collapse, the ECB successfully alleviated the
impact of the 2008 crisis by expanding its balance sheet, hence limiting the consequences for the
real economy, with the ultimate recipient being the borrowing companies. Providing several ample
liquidity programmes and substituting the interbank market, the ECB participated in the reduction
of funding the costs of small and constrained banks. This result remains valid when we consider
the banks’ specific loan-attribution process and resist several robustness checks.
With the current paper, we add to the debate on the effectiveness of the bank lending channel
(Bernanke and Blinder, 1988; Bernanke and Gertler, 1995) by investigating whether the ECB’s
accommodating monetary policy that was implemented after the Lehman collapse contributed to
mitigating the disruption in the issuance volume of syndicated loans.
5
All measures carried out by
the ECB may have potentially affected the economy through several transmission channels
(Mishkin, 1996). Because banks are credit constrained, the bank lending channel is effective when
the monetary policy affects credit institutions’ external finance premium, subsequently altering
credit availability in the economy (Stein, 1998; Gan, 2007; Disyatat, 2011 among others).
6
Peek
and Rosengren (2014) emphasise the importance of understanding the role of credit institutions in
monetary policy transmission; the authors show that the development of new non-standard
measures triggered a shift in the objective of the monetary policy, requiring a re-assessment of the
bank lending transmission channel. Adelino and Ferreira (2016) explain that the decrease in bank
5
J.C. Trichet speech (11/23/2009): These non-standard measures started in October 2008 and were designed to…
enable banks to continue their lending to households and firms’.
6
Considering the credit channel in general, Kishan and Opiela (2000) highlight the importance of distinguishing
between the bank lending channel and the borrower’s net worth channel. They argue that the former depends on the
banks asset volume and capital.
6
lending was because of reduced access to wholesale funding and an increase in the cost of funding,
reinforcing the importance of studying this channel.
We also contribute to the literature that seeks to identify the effects of non-standard measures on
financial institutions (Lenza et al., 2010; Gambacorta et al., 2011; Chodorow-Reich, 2014; Acharya
et al., 2019; Alper et al., 2020; Crosignani et al., 2020; Demirgüç-Kunt et al., 2020). We
complement this literature by studying the transmission mechanism of the whole ECB’s
accommodative monetary policy, that is, we consider both the decrease in the benchmark rate to
the zero lower bound (Eggertsson et al., 2019; Heider et al., 2019; Altavilla et al., 2021) and the
different programmes implemented to provide banks with liquidity. In contrast to Gambacorta et
al. (2011), who investigate the transmission of both standard and non-standard monetary policies
on aggregated nominal bank lending, we focus on the syndicated loan market for which bank
dependency may be higher. The financial crisis brought up the importance of bank lending to C&I
customers as the commercial paper market faltered. Hence, this crisis emphasized the importance
of bank-dependency of specific loans such as syndicated loans where bank’s expertise may provide
borrowers with loans at slightly better rates or terms, than issuing bonds on capital markets. To the
best of our knowledge, the current paper is one of the first works that explore the impact of the
overall ECB accommodative monetary policy on the syndicated loan market, which is one of the
major sources of international finance for corporations.
Finally, we add to the literature on cross-sectional asymmetry in banks’ responses to monetary
policy changes. Shocks to financial and monetary conditions do not have the same impact on all
banks, especially when taking into account their size, level of capital, and liquidity (Van den
Heuvel, 2002; Ehrmann and Worms, 2004; Gambacorta and Mistrulli, 2004; Kishan and Opiela,
2006; Altunbas et al., 2009; Gambacorta et al., 2011; Jiménez et al., 2014). Overall, these studies
7
find that the composition and strength of banks’ balance sheets
7
play a significant role in the
transmission channel of monetary policy (Altunbas et al., 2010). As such, assessing the
effectiveness of monetary policy transmission through the bank lending channel requires a deeper
analysis of these fluctuations across banks that have different financial positions because the degree
of informational asymmetry between banks and investors impacts the transmission of monetary
policies (Kashyap and Stein, 1995, 2000; Stein, 1998 among others). Our findings confirm previous
studies that highlight cross-sectional asymmetry in banksresponses to both standard and non-
standard monetary policy measures. We contribute to this literature by enlarging the sample of
banks and considering the whole ECB accommodating monetary policy.
The remainder of the present paper is organised as follows: Section 2 reviews the literature. Section
3 presents our methodology. Section 4 describes our data, and section 5 provides descriptive
statistics. Section 6 investigates whether the measures of the ECB’s monetary policy helped support
syndicated bank lending. Section 7 deepens the analysis by considering banks' size, capital
structure, and financial strength separately. Section 8 is dedicated to robustness checks, and section
9 concludes the paper.
2. The influence of monetary policy on banks’ lending
Typically, the ECB targets short-term interest rates to conduct monetary policy, that is, buying or
selling short-term debt securities using the Main Refinancing Operations (MROs) and the Longer-
Term Refinancing Operations (LTROs).
8
One month after the collapse of Lehman Brothers, the
ECB implemented the Fixed-Rate, Full Allotment (FRFA) tender procedures to address the
deterioration of financial conditions while decreasing its main interest rate by 325 basis points
7
Jiménez et al. (2020) show, in credit expansion periods, that banks’ exposure to real-estate assets influences their
risk-taking behavior.
8
The maturity of the LTROs was extended from three to six months for the first time on March 28, 2008.
8
between October 2008 and May 2009. However, in 2009, concerns over counterparty risk remained
significant, disturbing the operations of European interbank markets (Drudi et al., 2012). With
short-term interest rates approaching the zero lower bound, the ECB adopted non-standard
measures to reduce financial distress and stimulate the economy. The ECB extended the maturity
of its LTROs to twelve months, satisfying credit institutions' demand for longer maturities. Besides,
the ECB announced the Covered Bond Purchase Programme (CBPP), which aimed at purchasing
euro-denominated covered bonds for a predetermined amount equal to 60 billion euros over 14
months. This programme contributed to alleviating the maturity constraints that the credit
institutions faced when lending long and borrowing short. These measures targeting banks helped
increase the monetary base.
9
Nevertheless, during the financial crisis of 2008, tax revenues decreased, and economic growth
slowed down, exacerbating budget, and debt problems. In 2010, European credit institutions
holding substantial amounts of sovereign debt had to face new difficulties linked to the crisis in the
monetary union. The ECB announced its Securities Market Programme (SMP),
10
which had the
following two objectives: ensure liquidity and restore an appropriate transmission mechanism for
monetary policy. Unfortunately, the European sovereign debt crisis continued to plague European
interbank markets, and the ECB had to intervene with additional measures in 2011
11
to restore
confidence. As a result, the size of the ECB balance sheet significantly increased between 2008
and 2012.
9
Fawley and Neely (2013) highlight a significant difference between the programmes implemented by the ECB and
the BOJ and those implemented by the Fed and the BoE. The difference lies in the reality that the economies of
Europe and Japan are more bank oriented while those of the U.S. and UK are more bond oriented.
10
On September 6, 2012, the ECB replaced the SMP with the outright monetary transactions programme to address
the lack of an enforcement mechanism for receiving support.
11
In 2011, a second CBPP was set up for 40 billion euros. Besides, the ECB announced an extension of the LTROs'
maturity of up to thirty-six months.
9
All these measures implemented by the ECB during the 2008 crisis to support financial institutions
were relatively new, providing the academic world with a new research area focused on non-
standard measures. Chodorow-Reich (2014) runs high-frequency event studies to measure the
impact of unconventional monetary policy announcements by the FOMC on the financial sector.
Crosignani et al. (2020) focus on the three-year LTRO of the ECB and document a positive
relationship between the implementation of this programme and Portuguese banks purchasing
short-term domestic government bonds. Lenza et al. (2010) describe and compare the non-standard
measures implemented by the ECB, the Federal Reserve, and the Bank of England; they argue that
these measures had a significant impact on the money market spreads and analyse the consequences
for the real economy. Studying the negative interest rates implemented by 13 central banks,
Molyneux et al. (2020) relate this monetary policy to a decrease of banks’ lending using a
difference-in-difference approach. Gambacorta et al. (2011) use both standard and non-standard
measures proxied by the change in the overnight rate and the level of total assets of central banks,
respectively, to evaluate the impact of the monetary policy on bank lending. Heider et al. (2019),
focusing on negative interest rates on deposits, show that banks’ relying on deposit funding take
on more risk and decrease their lending relative to other banks. Grosse-Rueschkamp et al. (2019)
study specifically the purchase of corporate bonds by the ECB (CSPP program), and find that this
program supports bond debt for eligible firms and therefore reduces banks’ lending constraints,
hence increasing credit to other firms.
Moreover, the literature shows that shocks to financial and monetary conditions do not have the
same impact on all banks, especially when taking into account their size, level of capital, access to
sources of funding and liquidity. Van den Heuvel (2002), Gambacorta and Mistrulli (2004) and
Gambacorta et al. (2011) argue that banks’ reactions to monetary policy are not homogenous and
depend on the banks’ capital levels. They show that weakly capitalised banks with a higher
10
dependence on market funding reduce their credit availability more significantly than other banks
during the financial crisis. Moreover, the use of new and innovative tools, such as securitisation
may alter the role of credit intermediaries with an impact on the bank lending channel and the
banks’ ability to lend (Altunbas et al., 2009). Besides, Jiménez et al. (2014) highlight greater
vulnerability for Spanish banks with low capital or liquidity when monetary and macroeconomic
conditions worsen. Facing an increase in short-term interest rates or a decrease in GDP, these
weakly capitalised banks grant fewer loans than strongly capitalised banks, thereby worsening the
credit crunch. This increased vulnerability can be explained by the higher costs of borrowing
endured by financial institutions with a low level of capitalisation (Hubbard et al., 2002). Kishan
and Opiela (2006) investigate the asymmetry of banks’ reactions to both contractionary and
expansionary monetary policies. They find that when compared with small high-capital banks,
small low-capital banks decrease more total loans when facing a contractionary monetary policy
and are less able to increase total loans when an expansionary policy is implemented by the central
bank. The authors argue that the transmission of expansionary monetary policy during economic
recoveries can be supported by small banks when these banks are well capitalised. Ehrmann and
Worms (2004) set forth that another determinant, namely banks’ network, may affect bank lending
especially for small institutions providing them with an alternative solution to access the interbank
market, hence reducing the impact of a monetary contraction. Overall, these studies find that the
composition and strength of banks’ balance sheets play a significant role in the transmission
channel of monetary policy (Altunbas et al., 2010). In line with this analysis, several papers
(Angeloni et al., 2003; Gambacorta, 2005 among others) investigate how the relationship between
monetary policy and the level of deposits can disturb banks’ lending activities. Considering
deposits, Jacewitz, and Pogach, (2018) show that larger banks tend to pay lower premium for
uninsured deposits. Jayaratne & Morgan (2000) focus on banks’ level of insured deposits. They
11
find a positive correlation between the bank change of deposits and its loan growth with a stronger
correlation for lowly capitalized banks. Gambacorta (2005) studies a sample of Italian banks and
shows that a tightening monetary policy leads to a decrease in deposits and loans afterward, with
the effect being more significant in smaller banks that are unable to raise uninsured funds. Aside
from banks’ balance sheet composition, Grandi (2019) highlights the sovereign stress as another
determinant influencing the transmission of the monetary policy. Studying banks located in the
euro area between 2007 and 2016, the author argues that both conventional and unconventional
measures do not affect bank lending homogenously depending on banks’ exposure to sovereign
risk. He finds a weaker impact of the monetary policy on credit supply for banks located in
countries where the sovereign debt risk is higher. As such, assessing the effectiveness of monetary
policy transmission through the bank lending channel requires a deeper analysis of these
fluctuations across banks that have different characteristics because the degree of informational
asymmetry between banks and investors impacts the transmission of monetary policies (Kashyap
and Stein, 1995, 2000; Stein, 1998 among others).
3. Empirical methodology
Following a financial shock, credit institutions may experience higher funding constraints,
resulting in a contraction in bank lending in general and in syndicated loans in particular. Our
objective is two-fold. First, we estimate to what extent the ECB’s standard (proxied by the Euro
OverNight Index Average EONIA) and non-standard policies (proxied by the size of the balance
sheet) mitigated the impact of the 2008 financial crisis by supporting lending in the syndicated loan
market.
Insert Figures 1 and 2 here.
12
Figures 1 and 2 display the volume of syndicated loans (reported as the sum of the annual volume
of syndicated loan amounts in millions of euros) with the variation of the EONIA and of the size
of the ECB balance sheet respectively. The loan volume is rising on average over the period despite
two disruptions due to the financial and the sovereign debt crises. More precisely, we can observe
that a decrease (increase) of the EONIA (of the size of the ECB balance sheet) is followed by an
increase in the annual amount of syndicated loans, leading us to the following hypothesis:
Hypothesis 1: The overall expansionary monetary policy implemented by the ECB through both
standard and non-standard measures contributed to support bank lending activities, through an
increase in the amount of loans supplied by banks to borrowers.
Second, the literature highlights that the impact of monetary policy measures may differ across
banks. Banks with limited access to funding, such as deposits, debt or equity will rely on the
liquidity provided by the central bank. This limited access tends to reduce the quantity of funding
available and/or increases its costs, therefore reducing lending. Expansionary monetary policies
will tend to alleviate this funding constraints and so affect differently banks depending on the size
and structure of their balance sheet. As such, our objective is to investigate the three following
hypotheses:
Hypothesis 2a: The impact of monetary policy measures differs with respect to banks’ size.
Hypothesis 2b: The impact of monetary policy measures differs with respect to banks’ funding
constraints, measured through the capital, the customers deposits, and the short-term debt ratios.
Hypothesis 2c: The impact of monetary policy measures differs with respect to banks’ financial
strength, proxied by the Tier 1 capital ratio.
13
In our model, we analyse the bank lending channel and assess both the direct effect of the ECB’s
policies (Hypothesis 1) and whether this effect is different with respect to the specific bank’s size,
funding constraint, and financial strength (Hypotheses 2a, 2b and 2c). Our model is specified at the
tranche level of a syndicated loan; we manually matched the loan amount granted by each bank
(intensive margin) with the associated explanatory variables. This allows us to disentangle between
credit supply and credit demand by introducing firm-level variables that control for credit demand
and correctly identify the credit supply effect (Jiménez et al., 2012). We also control for borrower,
bank, and year fixed effects.
Besides, we contribute to the literature on syndicated loans by considering all credit institutions
that are part of the syndicate. In the syndicated loan market, a syndicate is divided into two distinct
groups of lenders, depending on their roles. The lead arrangers are responsible for structuring,
administering, and monitoring loans while the participants behave as investors and provide funds.
Although the literature focuses on loans provided by lead arrangers, we consider each bank's
individual decision to lend. Even if the bank is only a participant, it still has the choice to invest or
not at the beginning of the syndication process, and this decision may also be influenced by the
bank’s monetary conditions. However, we control for lead arrangers specific behaviour in our
model.
To test our first hypothesis, we model the amount of each syndicated loan (taken as a logarithm)
provided by lender to borrower at time as follows:
                
 
where  is a quarterly change in the monetary policy proxied by a quarterly change in the
EONIA () (Jiménez et al., 2014 among others) accounting for standard policy, and a
(1)
14
quarterly change in the size of the ECB balance sheet () (Gambacorta et al., 2011), accounting
for non-standard measures.
12
The variable representing the size of the ECB balance sheet contains
the MROs that are fulfilled at a fixed rate with full allotment after the Lehman collapse, the LTROs
that benefited from an extension in their maturity, and the securities held for monetary purposes
through the different programmes (e.g., CBPP, SMP, etc.). In the following estimations, we test
the joint effect of each monetary policy on the loan amount. In line with the theory, an
accommodating monetary policy, either through a decrease in the EONIA or an expansion of the
size of the ECB balance sheet, should contribute to an increase in bank lending, that is, would
be negative for standard measures but positive for non-standard policies.
Disentangling credit supply from credit demand is key in our analysis because both can be affected
by a change in monetary and economic conditions. To address this identification challenge, our
analysis contains macro and microeconomic variables in addition to borrower fixed effects () to
account for time-invariant borrower characteristics.
13
Jiménez et al. (2012) show that economic
conditions have a significant impact on bank loans. As such, we include the macroeconomic
context in our model with the quarterly change in the Eurozone’s GDP () , the annual
GDP growth rate of the country where the borrower is located, and the banks’ anticipation of credit
demand based on question 9 in the bank lending survey (),
14
which is provided quarterly
12
As an alternative proxy for ECB monetary policies, we tested two aggregated indicators, namely two ECB
shadow rates’ that are estimated by Wu and Xia (2016, 2017) and Krippner (2013), respectively. However, our
findings are less conclusive. The shadow rates provide a useful benchmark for a central bank monetary policy based
on forward rates. Nevertheless, with the negative values during our sample period, they cannot properly explain the
borrowing/lending decisions because they do not represent realistic borrowing costs in the international syndicated
loan market.
13
Considering the size of our sample, we cannot use borrower*year fixed effects due to a lack of degree of freedom.
14
The main objective of the BLS is to provide the ECB's Governing Council with information regarding the
financing conditions in the Eurozone, and this is done using questionnaires sent out to banks and enterprises to gauge
their opinions about the market’s appetite for loans. In our model, we use the information from question 9 (Please
indicate how you expect demand for loans or credit lines to enterprises to change at your bank over the next three
months [apart from normal seasonal fluctuations]’). We consider the quarterly variation of the overall category, that
is, all loans (short and long term) to all companies (small, medium, and large), and we include the balance of
opinions in our model (between -100 and +100).
15
by the ECB (Del Giovane et al., 2011). We also control for any productivity shock occurring in
one specific sector by building variables to account for the industry of the borrower and the risks
associated with this industry.
15
Moreover, we add the borrower’s credit rating when the loan is
issued
16
and whether the borrower is located in the same country as the lender, hence controlling
for possible home bias, as the key determinants of credit demand (Giannetti and Laeven, 2012
among others). These variables are included in the matrix , which also considers the
characteristics of the loan, that is, its maturity, whether the loan is secured, its type, and seasonal
effects. Our model contains a dummy variable reflecting whether the lender is the lead arranger.
We also consider the lender’s strategy in terms of industry portfolio diversification. A bank may
develop expertise in one specific industry because of often lending to companies in this industry
(Burietz and Ureche-Rangau, 2020). As such, the bank can save on information gathering and
monitoring costs. Giannetti and Saidi (2019) find that lenders to a specific industry in distress will
tend to provide more loans to this industry, especially if it is prone to fire sales. However, the risk
of this focus strategy (Acharya et al., 2006) is a lack of diversification, which may sometimes
push banks to lend more to companies in other industries. Finally, in our analysis, we integrate the
relationship between the lender and borrower,
17
as well as banks’ characteristics (i.e., total assets,
capital ratio, deposits ratio, short-term debt ratio, and Tier 1 capital ratio), and bank*time fixed
effects ( ), to control for time-variant bank heterogeneity.
15
The industry risk may affect a bank's portfolio of loans, especially during a crisis, when investors become risk-
averse. We compute a Value-at-Risk (VaR) per industry to control for this risk by using industry indices produced by
Datastream. Then, we manually match the industry of the borrower with these indices to associate one VaR per loan.
16
DealScan provides credit ratings produced by the three leading U.S. Credit-Rating Agencies (CRAs): Standard &
Poor's, Moody's, and Fitch. These ratings are automatically reported in the database when they appear. In our sample,
we consider for each loan the rating each time it is provided by one of the three CRAs. For rated loans with more
than one rating, we apply the ‘worst of 2 and median of 3 ratings’ rule (Bongaerts et al., 2012). We then categori se
borrowers as investment grade, junk grade, or unrated. In the regression, we use the group of unrated loans as the
reference.
17
The presence of the bank in the borrower’s board can also influence lending amounts (Ferreira and Matos, 2012;
Acharya et al., 2018).
16
To test our second hypotheses, we group our lenders depending on their financial structure. The
banks’ characteristics () are collected with a quarterly frequency and represent the size (total
assets), the capital level (common equity to total assets ratio), the dependence on customers’
deposits (customers’ deposits to total assets ratio), the reliance on short-term debt (short-term debt
to total assets ratio), and the financial strength (Tier 1 capital ratio) of each lender at time . To
match both the change in monetary policy and banks’ characteristics with the issuance date of the
loan (time ), we consider the quarter during which the loan is granted.
The nineteen banking groups in our sample are then sorted based on the five financial indicators
previously mentioned: (1) total assets, (2) capital ratio, (3) deposits ratio, (4) short-term debt ratio,
and (5) Tier 1 capital ratio. To rank these banking groups, we compute the average of each indicator
between 2008 and 2014 for each financial institution. We focus the analysis on the bottom or top
four banks (lower or higher quintile)
18
versus the other banks.
19
We run five distinct tests that
alternatively consider each indicator separately to assess how financial and monetary shocks affect
the two subgroups of lenders.
Building on Equation (1), we estimate the following:
                
          
The two dummy variables  and , included in the matrix , are equal to one for
banks that belong to the bottom and top quintile, respectively, in terms of size, capital ratio, deposits
ratio, short-term debt ratio, or Tier 1 capital ratio, the five specifications being estimated separately.
18
For the analysis of the deposits and the ST debt ratios, we still consider the lower and higher quintile but they
contain only the bottom and top three banks as our sample of banks is reduced due to missing data.
19
To assess our results, we run two robustness tests. First, we use the continuous variables representing banks’
characteristics instead of the subgroups of banks. Second, we consider banks belonging to the bottom or top tercile.
The results remain similar and are available upon request.
(2)
17
As such, the interaction term assesses whether the effect of a change in monetary policy differs
across financial institutions according to these five different financial indicators. Because the
lending strategy of a bank may evolve and to control for this time-variant bank heterogeneity,
(Lender*Year) fixed effects are implemented.
20
To take into consideration the asymmetric transmission of the monetary policy (Gambacorta,
2005), we run the model using a cross-section estimation method per loan and per credit institution,
rather than per country. We perform our regressions with clustered standard errors at the lender
level.
4. Data
Focusing on the monetary policy implemented by the ECB, our analysis considers all credit
institutions that can benefit from the ECB’s open market operations and non-standard programmes.
According to European Directive 2000/12/EC (European Parliament March 20, 2000), a credit
institution shall mean an undertaking whose business is to receive deposits or other repayable
funds from the public and to grant credits for its own account. The ECB establishes a list of
Monetary Financial Institutions (MFIs) that fall within the scope of this definition.
21
From this list,
we select only credit institutions that must satisfy the ECB’s reserve requirement, restricting our
list to 5,294 MFIs. To run our analysis on a quarterly basis, we restrict our sample to MFIs for
which we have access to quarterly financial information, and which are active in the syndicated
loan market. Our final sample contains 148 credit institutions located in eleven Eurozone countries
20
This fixed effect also control for the direct link between the bank characteristics and lending amounts as in Chu et
al. (2019).
21
MFIs are defined by the ECB as central banks, resident credit institutions as defined in community law, and other
resident financial institutions whose business is to receive deposits and/or close substitutes for deposits from entities
other than MFIs and, for their own account (at least in economic terms), to grant credit and/or make investments in
securities. Money market funds are also classified as MFIs (Regulation (EC) No. 25/2009 ECB/2008/32). On
February 29, 2016, this list contained 7,959 MFIs. The list is updated every month.
18
(Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands,
and Spain). We collect financial data from banks using Bloomberg, and we complete our series
with information from the banks’ balance sheets. The ECB provides the data on monetary policy
instruments.
Besides, we use the LPC DealScan database to collect data on syndicated loans provided by each
MFI. In the LPC DealScan, we obtain all of the loans’ characteristics and the industries involved,
as well as the credit rating and nationality of the borrower. The industry risk is computed using
data from Datastream. LPC DealScan also provides access to the MFI’s role in the syndicate, its
strategy in terms of industry specialisation, its relationship with the borrower, and the bank
allocation, i.e. how much each MFI has invested per loan. Finally, the quarterly change of the Euro
Area GDP, the annual GDP growth rate of borrowers’ countries, and the results of the bank lending
survey are extracted from Eurostat, the World Bank, and the ECB website, respectively.
To investigate the effect of the ECB’s accommodative measures on syndicated bank lending, we
run our baseline analysis from January 2008 to December 2014. For the sake of our study, we group
the 148 MFIs under the names of their parents and collect information about the latter on a quarterly
basis. Our final sample contains 21,314 unique loans provided by nineteen banking groups to 8,280
borrowing companies between 2008 and 2014. One specificity of the syndicated loan market is that
several banks participate in one syndicated loan, so that one loan may appear several times in our
database, resulting in a total of 51,446 observations in our sample. Our dataset allows us to make
good inferences on how accommodative monetary policy instruments impact credit supply
depending on the banks size, capital, deposits, short-term debt, and Tier 1 capital ratios.
5. Descriptive statistics
Table I provides the definition and descriptive statistics of each variable included in our analysis.
19
Insert Table I here.
Ranked according to the total quantity of loans provided, Table II presents the nineteen banking
groups included in our sample. In Table II, we report the countries in which these banking groups
have MFIs involved in syndicated loans, as well as some descriptive statistics of the average loan
characteristics and the quarterly average of total assets, capital ratio, deposits ratio, short-term debt
ratio, and Tier 1 capital ratio of each banking group over the 2008-2014 period.
Insert Table II here.
Table III displays the description of our sample in terms of the geographical repartition of the
borrowers, type, objective, and maturity of the loans. This table highlights that the nineteen banking
groups lend to companies that are mainly located either in Western Europe or North America, with
the two regions representing more than 80% of our sample. Our objective is to study the lending
behaviour of banks active in the international syndicated loan market. As such, we do not limit our
analysis to a sample of European borrowers; instead, we control for the geographical location of
the borrowers in our estimates. Regarding the most common loan characteristics, term loans and
lines of credit dominate the sample and are used mainly to finance general corporate functions,
LBOs, project finance, and takeovers, with more than 50% of the loans maturing in one to five
years.
Insert Table III here.
6. Estimation results
Table IV reports the estimation results for our baseline model testing the impact of the monetary
policy of the ECB on bank lending (Hypothesis 1). In model (1), we assess how the overall ECB
monetary policy influences syndicated loan amounts by considering the standard and non-standard
20
policies simultaneously, that is,  with . We include the control variables for the
characteristics of the loan, borrower, lender, and lender-borrower relationship to better consider
the credit institutions’ lending process.
Insert Table IV here.
We show that the ECB monetary policy through both standard and non-standard measures reduces
constraints on banks’ lending, increasing credit supply. We find a negative and significant
coefficient for the change in the EONIA. A decrease in the benchmark rate significantly supports
the supply of syndicated loans by all banks, the latter being able to provide loans of larger amounts.
Considering non-standard policies, the coefficient of the size of the ECB balance sheet is positive
and significant. Hence, we show that the programs newly implemented by the ECB complement
the standard measures, being successful in supporting bank lending further. These results confirm
the hypothesis that ECB non-standard measures contribute to mitigating lender funding constraints
and support lending activities. Beyond the statistical significance, the economic impact is also
significant. Our results show that a decrease (increase) of the EONIA (the size of the ECB balance
sheet) by 1 standard deviation contributes to increase the loan amount by 1 million euros.
22
In Table IV, we also report the results of five alternative specifications ((2) to (6)) of our main
model, allowing us to test the sensitivity of our results. In the loan contract, the different
characteristics may be co-determined with the loan amount, limiting the use of these variables as
explanatory variables. In model (2), we remove the control variables accounting for these loan
characteristics. Moreover, in our main estimation, we control for changes in the bank lending
behaviour using both lenders’ characteristics, and lender*year fixed effects to account for time-
22
To compute the economic impact of the monetary policy measures, we multiply the coefficient by the standard
deviation of the independent variable, and take the exponential, providing us with 0.976 million euros
(=exp(0.053*0.45)) for the standard measures, and 1.025 million euros (=exp(0.123*0.2)) for the non-standard
measures.
21
varying unobservable characteristics. In models (3) and (4), we try to capture these unobservable
factors using quarter fixed effects instead of year fixed effects, and lender country*quarter fixed
effects, respectively. Finally, to control for credit demand we combine macro and microeconomic
variables with borrower fixed effects. To assess the robustness of our results, we estimate our
model using an alternative control for credit demand based on Degryse et al.’s (2017) approach. In
line with other studies’ attempts to better distinguish between credit supply and credit demand
(Khwaja and Mian, 2008 among others), we build fixed effects per group of borrowers located in
the same country. We then multiply these borrower’s country fixed effects by year fixed effects to
account for the time-variant characteristics of these groups of borrowers (model (5)). In model (6)
we multiply the borrower’s country fixed effects by the borrower’s industry fixed effects to account
for any specificity related to the industry of the borrower at the national level. In all these five
alternative specifications, our conclusions remain strictly identical. In line with the literature, we
confirm the significant role played by non-standard measures in addition to standard ones in
supporting credit supply during the crisis period, highlighting the importance of adjusting monetary
policy tools during exceptional times (Gambacorta et al., 2011 among others). With alternative
programmes implemented for the first time since its creation, the ECB managed to limit the impact
of the financial crisis on the real economy.
7. The asymmetric effects of the ECB accommodative monetary policy
Our previous results emphasize the positive effects of the ECB’s accommodative monetary policy
on banks lending activities in the syndicated loan market. With eight financial institutions in the
Eurozone listed as global systemically important banks by the Financial Stability Board in 2014
(FSB, 2014),
23
the heterogeneity encompasses banks' structures, business models, and nationality.
23
BNP, Deutsche Bank, BBVA, Crédit Agricole, ING, Santander SA, Société Générale and Unicredit Bank. The list
was published on November 6, 2014.
22
To investigate further the asymmetric transmission of accommodating monetary policies, we
distinguish banks with respect to their size, funding constraints, and financial strength.
Table V provides the estimated coefficients of Equation (2) when the banks are ranked according
to their size (i.e., total assets (1)), their funding constraint (i.e., capital ratio (2), deposits ratio (3),
and short-term debt ratio (4)), and their financial strength (i.e., Tier 1 capital ratio (5)).
Insert Table V here.
Our findings for monetary policy measures are in line with the literature. We find that the effects
vary across banks with different characteristics, as these banks do not face the same restrictions.
More precisely, we find that, on average, a decrease of the main interest rate has a positive and
significant effect on loan amounts provided by large banks while the effect is negative for small
banks (specification (1)). The interaction term is positive and significant when considers the banks
of our sample with the lowest total assets. In other words, a decrease in the EONIA does not support
the lending amounts provided by small banks. While Jiménez et al. (2012) and Gambacorta and
Shin (2018) show that the negative effect of increased interest rates on lending is larger for banks
that are lowly capitalized, we argue that the effect of decreased interest rates is also reduced for
small banks. On the contrary, non-standard expansionary policies have a larger and positive effect
on the credit supply of these small banks when compared with the large banks. The implementation
of unconventional measures by the ECB reduces the funding constraint by directly providing
additional liquidity to smaller financial institutions, which may have more difficulties to raise funds
than larger institutions. As such, these institutions can maintain their credit supply by providing
loans of larger amounts: monetary policy programs provide large amounts of funding that can
therefore substitute equity, deposits or short-term borrowing as a mean to finance banks assets.
23
In specifications (2) to (4), we differentiate the sources of funding obtainable (banks’ capital,
deposits and short-term debt ratios) and their effects on the transmission of monetary policy. We
also observe a significant difference in the transmission of the monetary policy across banks.
Interestingly, we find that standard measures support bank lending provided by institutions that are
weakly capitalized, with a low level of deposits, and of short-term debt. Weakly capitalized banks
as well as banks with a low level of liquidity rely more on market funding to raise funds and finance
their activities (Van den Heuvel, 2002; Gambacorta and Mistrulli, 2004; Gambacorta et al., 2011).
As such, they are more sensitive to a change in borrowing costs with a direct impact on their lending
behaviour (Hubbard et al., 2002). On the opposite, non-standard measures contribute to increase
the loan amounts when granted by highly capitalized banks, with a high level of short-term debt.
Considering the level of deposits, the overall impact of an increase of the size of the ECB balance
sheet on bank lending is positive for all banks even for those with a lower level of deposit, albeit
less economically significant. In other words, we show that banks that are well capitalized and
liquid benefit from the unconventional measures of the monetary policy to raise additional funds
and increase the size of their loans. In line with Kishan and Opiela (2006), we confirm the level of
bank capital as being a significant determinant of the efficiency in the transmission of an
expansionary monetary policy during economic recoveries. Looking at the specificities of the
Eurozone banking sector, we know that small banks tend to have a higher capital ratio.
24
Hence,
our results are in line with the conclusions of previous literature (Peek and Rosengreen, 1995a,
1995b; Bliss and Kaufman, 2003; Kopecky and VanHoose, 2004; Peydró et al., 2021) when
considering highly capitalised banks, highlighting important implications regarding the
relationship between the transmission of monetary policy and capital regulations. We conclude that
24
We can see in Table II that two of the four smallest banks (Banca Popolare di Milano, Alpha bank AE) belong to
the sample of banks with the highest capital ratio while the two other banks (Bankinter, Sabadell SA) have a capital
ratio above the average, i.e. 5.24.
24
while restrictive policies tend to amplify the funding constraints, non-standard accommodative
policies successfully reduce these constraints.
Finally, specification (5) presents the results of our analysis of banks’ reactions to monetary policy
measures when we disentangle them using their financial strength (Tier 1 capital ratio). We find
that both monetary policy measures significantly contribute to support bank lending with sizeable
effect of non-standard tools on banks with a lower Tier 1 capital ratio. The literature highlights that
financial institutions that are financially weaker are expected to start hoarding cash instead of
lending more not to deteriorate their financial position further (Kishan and Opiela, 2006). Our
results show that, on the syndicated loan market, expansionary monetary policies alleviate this
constraint and lead to a larger increase of lending from banks with a weaker financial position (i.e.,
a lower Tier 1 capital ratio) relative to other banks.
Overall, these results confirm the existence and effectiveness of the bank lending channel in the
studied period for the syndicated loan market. The ECB had to intervene with additional major
measures in 2008 to limit the crisis from spreading to the real economy, which enabled banks to
maintain credit supplies, with the ultimate recipient being the borrowing parties. But more than
that, we shed light on the critical importance of the ECB non-standard measures to complement
and overcome the limits of standard measures during crises at the Zero Lower Bound. During the
period 2008-2014, the EONIA sharply decreased, from 4% in January 2008 to 0.33% in July 2009,
and remained low. As such, the standard measures quickly reached their limits, highlighting the
need to develop new monetary policy tools to support the lending activities of banks that needed it
the most, i.e., that are small and financially constrained. In our analysis, we show that the ECB was
successful in doing so, with the implementation of non-standard monetary policy tools significantly
supporting the loan offer of these banks after the crisis. As such, we argue that considering the
whole monetary policy is important to clearly understand its transmission mechanism.
25
8. Robustness Checks
Alternative measure of the ECB non-standard monetary policies
In our main estimations, we measure the non-standard tools of the ECB by using the quarterly
change in the size of the ECB balance sheet (). However, the variable  may be biased by
the presence of the MROs, which are considered standard measures before the crisis and FRFA
implementation. Accordingly, we built a more restrictive variable, called non-standard (), in
which we remove these MROs, focusing exclusively on non-standard policies implemented by the
ECB.
Tables VI and VII below provide the results for hypothesis 1 and hypothesis 2 (a, b, c), respectively.
Insert Tables VI and VII here.
In Table VI, our conclusions regarding the impact of the change in EONIA as well as the general
effect of non-standard measures on average remain the same, both contributing to support bank
lending after the crisis. Looking at the asymmetric effects of the ECB monetary policy in Table
VII, we confirm that the effect of the standard measures is fully captured by weakly capitalized
banks, with a low level of deposits and of short-term debt while the non-standard tools support
lending by small and strongly capitalized banks, banks with a high level of deposits and short-term
debt, and banks that are financially weaker. This is in line with our main estimations. Smaller banks
or banks with higher funding constraints or a low Tier 1 capital ratio benefit more from the
unconventional measures of the monetary policy implemented by the ECB after 2008, as they
significantly increase the amounts on granted loans.
26
Alternative banks ranking
In our analysis of the asymmetric effects of the ECB monetary policy, we rank banks using the
average of each indicator (i.e., total assets (1), capital ratio (2), deposits ratio (3), short-term debt
ratio (4), and Tier 1 capital ratio (5)) between 2008 and 2014 for each institution. However, we
want to ensure that our conclusions are not driven by the determinants of this ranking. As such, we
assess the robustness of our results using the ranking of each indicator based on the banks situation
before the financial shock, i.e., on the average for the year 2008 only. Table VIII below provides
the results.
Insert Table VIII here.
We confirm the positive effects of a decrease of the EONIA on syndicated loan amounts provided
by large banks while we do not observe significant differences across banks with respect to the
level of their funding constraint. Considering the non-standard measures, our conclusions remain
highly robust with a positive impact of an increase of the ECB balance sheet on bank lending when
banks are small, well capitalized, with a high level of short-term debt, and a low level of Tier 1
capital ratio. Overall, we provide evidence of the efficiency of the monetary policy implemented
by the ECB. More precisely, we emphasize the limits of standard measures that do not have any
significant effect anymore on syndicated loan amounts while unconventional measures contribute
to support lending by small and financially constrained banks.
ECB and Federal Reserve Monetary Policies
Following the Lehman collapse, several major central banks implemented exceptional measures to
limit the liquidity crisis in the financial industry. The Federal Reserve intervened massively with
its quantitative easing programmes aimed at lowering interest rates. The ECB and the Fed acted
simultaneously, complicating the task of disentangling between the effects of each monetary
27
policy. Moreover, dealing with the syndicated loan market implies that international banking and
the Fed’s monetary policy may affect the credit supply given by international banks. Banks in our
sample have subsidiaries located abroad, notably in the U.S. As such, these subsidiaries may have
benefited from the programmes implemented by the Fed, changing their lending behaviour
accordingly. To ensure that we correctly capture the effects of the ECB’s actions, we horserace the
ECB’s and Fed’s monetary policies in a model that focuses on hypothesis 1. We test whether the
overall expansionary monetary policy implemented by the ECB through both standard and non-
standard measures contributed to support bank lending activities while taking into account the
Fed’s monetary policy with two proxies: the evolution of the Fed’s funds rate and the size of the
Fed’s balance sheet. Table IX presents the effects of both the ECB’s and Fed’s monetary policies,
focusing on standard (interest rates) and non-standard (balance sheet) policies.
Insert Table IX here.
In line with previous estimations, we show that a decrease of the EONIA significantly supports the
supply of syndicated loans when we estimate the model with both the standard and non-standard
measures of the two central banks. On the contrary, a decrease of the Fed’s funds rate does not
significantly impact syndicated loan amounts. Starting our analysis in 2008 emphasizes even more
the limits of the standard measures of the Fed compared with the ECB, with the Fed’s funds rate
decreasing from 4% in January 2008 to 0.16% in December 2008, hence remaining close to 0 until
January 2016. However, looking at non-standard tools, the two coefficients of both central banks
measures are positive and significant. In other words, we confirm the support provided by the ECB
to bank lending. In addition, we highlight the additional contribution of the Fed’s unconventional
programs to the increase in syndicated loan amounts provided by the European banks of our
sample. Despite a lower statistical level of significance, the economic impact remains stable with
an increase in the loan amount by 1 million euros associated to a decrease by 1 standard deviation
28
of the change in the size of the ECB balance sheet. The economic impact of the Fed’s non-standard
measure is similar to the ECB. As such, while we complement our main analysis by showing a
positive and complementary impact of the Fed’s monetary policy on European bank lending
behaviour, our findings confirm the robustness of our results with respect to hypothesis 1 and show
that our variables correctly capture the ECB’s monetary policy in the main estimation.
9. Conclusion
The objective of the current paper is to assess the impact of the accommodative monetary policies
implemented by the ECB on syndicated bank lending disentangling the role played by both
standard and non-standard measures on banks credit supply. The use of these measures at
unprecedented levels requires a reassessment of the bank lending channel as a transmission
mechanism for monetary policy.
We run an empirical analysis on syndicated loan amounts from a sample of nineteen European
banking groups for the period between 2008 and 2014. The use of seven different databases allows
us to integrate the precise characteristics of all players involved with banking transactions. Through
a cross-sectional regression of 51,446 loans, we study the influence of monetary policies on the
amount of loans. We control for loans, borrowers, and lenders' characteristics in addition to the
relationship between the lender and borrower. Our analysis includes borrower, and lender*year
fixed effects.
Considering the transmission of ECB’s monetary policy, our empirical analysis of the syndicated
loan market provides evidence of the existence of the bank lending channel. On average, the
instruments used by the ECB seem to play a significant role in reducing the constraints on financial
markets, supporting the supply of syndicated loans. We show that accommodative standard and
non-standard policies significantly stimulate lending on the syndicated loan market, but the size of
29
this stimulus depends on bank characteristics. On the one hand, a decrease in the central bank’s
interest rate leads to a significant increase in the amount of loans supplied by European large and
less liquid banks. On the other hand, we find that the innovative, accommodating ECB monetary
policy facilitates banks’ access to alternative sources of funds, supporting the credit supply of
small, and highly capitalized banks. Moreover, banks with a relatively high level of deposits and
of short-term debt increased their lending more than other banks. Finally, banks with a low Tier 1
capital ratio use the accommodative monetary policy stance to increase their lending in the
syndicated loan market, therefore contributing to the transmission of monetary policy to the real
economy. Our results are robust to alternative rankings of these banks.
These results contribute to the debate on the effectiveness of unforeseen measures. We argue that
by supporting bank lending activities, the ECB’s measures limited the spillover effects of the 2008
financial crisis into the real economy. A further extension of the current paper would involve
deepening the analysis using more detailed data on the ECB’s open-market operations to better
understand the mechanisms of each instrument in the monetary policy. Another future research
channel of the present paper could focus on the marginal effect of non-standard policies on non-
financial institutions’ investment strategies and the implications this would have on the
macroeconomic environment. Finally, having access to more detailed information regarding the
different programmes implemented by the different central banks and the extent to which banks
that are active on the syndicated loan market benefited from these programmes would allow to
investigate further the role played by the other central banks in this international market and the
impact on bank lending behaviour.
30
Acknowledgment
We want to thank Manuel Buchholz, Hicham Daher, Kenneth de Beckker, Hans Degryse, Alain
Durré, Alexandre Girard, Christophe Godlewski, Pierre-Guillaume Méon, Suren Pakhchanyan,
Mikael Petitjean, Michael Scharnagl and Daniel Thornton, as well as the participants at the IFABS
Conference, FMA Annual Meeting, MFS Conference, EFMA Conference, AFFI Conference,
ICMAIF Conference, GdRE Conference, and 3L workshop, for their insightful comments and
suggestions. We acknowledge funding from the People Programme (Marie Curie Actions) of the
European Union’s Seventh Framework Programme FP7/2007-2013/ under REA grant agreement
n°608129. All remaining errors are ours.
References
Acharya, V. V., Eisert, T., Eufinger, C. and Hirsch, C. (2018) Real effects of the sovereign debt
crisis in Europe: Evidence from syndicated loans, The Review of Financial Studies 31, 2855-
2896.
Acharya, V.V., Eisert, T., Eufinger, C. and Hirsch, C. (2019) Whatever it takes: The real effects
of unconventional monetary policy, The Review of Financial Studies 32, 3366-3411.
Acharya, V.V., Hasan, I. and Saunders, A. (2006) Should Banks be Diversified? Evidence from
Individual Bank Loan Portfolios, Journal of Business 79, 1355-1412.
Acharya, V.V. and Merrouche, O. (2013) Precautionary Hoarding of Liquidity and Interbank
Markets: Evidence from the Subprime Crisis, Review of Finance 17, 107-160.
Adelino, M. and Ferreira, M.A. (2016) Bank Ratings and Lending Supply: Evidence from
Sovereign Downgrades, Review of Financial Studies 29, 1709-1746.
Alper, K., Altunok, F., Çapacıoğlu, T. and Ongena, S. (2020) The Effect of Unconventional
Monetary Policy on Cross-Border Bank Loans: Evidence from an Emerging Market, European
Economic Review 127, 103426.
Altavilla, C., Burlon, L., Giannetti, M. and Holton, S. (2021) Is there a zero lower bound? The
effects of negative policy rates on banks and firms, Journal of Financial Economics,
forthcoming.
Altunbas, Y., Gambacorta, L. and Marques-Ibanez, D. (2009) Securitisation and the Bank
Lending Channel, European Economic Review 53, 996-1009.
31
Altunbas, Y., Gambacorta, L. and Marques-Ibanez, D. (2010) Bank Risk and Monetary Policy,
Journal of Financial Stability 6, 121-129.
Angeloni, I., Kashyap, A. K. and Mojon, B. (Eds.). (2003). Monetary Policy Transmission in the
Euro Area: A Study by the Eurosystem Monetary Transmission Network. Cambridge University
Press.
Bernanke, B.S. and Blinder, A.S. (1988) Credit, Money, and Aggregate Demand, American
Economic Review 78, 435-439.
Bernanke, B.S. and Gertler, M. (1995) Inside the Black Box: The Credit Channel of Monetary
Policy, Journal of Economic Perspectives 9, 27-48.
Bliss, R.R. and Kaufman, G. (2003) Bank Procyclicality, Credit Crunches, and Asymmetric
Effects of Monetary Policy, Journal of Applied Finance 13, 23-31.
Bongaerts, D., Cremers, M. and Goetzmann, W.N. (2012) Tiebreaker: Certification and multiple
credit ratings, Journal of Finance 67, 113-152.
Burietz, A. and Ureche-Rangau, L. (2020) Better the devil you know: Home and sectoral biases
in bank lending, International Economics 164, 69-85.
Cerutti, E., Hale, G. and Minoiu, C. (2015) Financial Crises and the Composition of Cross-border
Lending, Journal of International Money and Finance 52, 60-81.
Chaudhry, S.M. and Kleimeier, S. (2015) Lead Arranger Reputation and the Structure of Loan
Syndicates, Journal of International Financial Markets, Institutions & Money 38, 116-126.
Chu, Y., Zhang, D., & Zhao, Y. E. (2019). Bank Capital and Lending: Evidence from Syndicated
Loans. Journal of Financial & Quantitative Analysis 54, 667-694.
Chodorow-Reich, G. (2014) Effects of Unconventional Monetary Policy on Financial
Institutions, Brookings Papers on Economic Activity Spring, 155204.
Crosignani, M., Faria-e-Castro, M. and Fonseca, L. (2020) The (Unintended?) Consequences of
the Largest Liquidity Injection Ever, Journal of Monetary Economics 112, 97-112.
De Haas, R., & Van Horen, N. (2012). International shock transmission after the Lehman
Brothers collapse: Evidence from syndicated lending. American Economic Review 102, 231-37.
Degryse, H., De Jonghe, O., Jakovljevic, S., Mulier, K. and Schepens, G. (2017) The Impact of
Bank Shocks on Firm-level Outcomes and Bank Risk-Taking, Working Paper.
Del Giovane, P., Eramo, G. and Nobili, A. (2011) Disentangling Demand and Supply in Credit
Developments: A Survey-based Analysis for Italy, Journal of Banking & Finance 35, 2719-2732.
32
Demirgüç-Kunt, A., Horváth, B.L. and Huizinga, H. (2020) Foreign Banks and International
Transmission of Monetary Policy: Evidence from the Syndicated Loan Market, European
Economic Review 129, 103542.
Dennis, S.A. and Mullineaux, D.J. (2000) Syndicated Loans, Journal of Financial Intermediation
9, 404-426.
Disyatat, P. (2011) The Bank Lending Channel Revisited, Journal of Money, Credit and Banking
43, 711-734.
Drudi, F., Durré, A. and Mongelli, F.P. (2012) The Interplay of Economic Reforms and Monetary
Policy: The Case of the Eurozone, Journal of Common Market Studies 50, 881-898.
Eggertsson, G. B., Juelsrud, R. E., Summers, L. H. and Wold, E. G. (2019) Negative nominal
interest rates and the bank lending channel, National Bureau of Economic Research, No.
w25416.
Ehrmann, M. and Worms, A. (2004) Bank Networks and Monetary Policy Transmission, Journal
of the European Economic Association 2, 1148-1171.
Fawley, B.W. and Neely, C.J. (2013) Four Stories of Quantitative Easing, Federal Reserve Bank
of St. Louis Review (Jan./Feb.), 51-88.
Ferreira, M. A., & Matos, P. (2012). Universal banks and corporate control: Evidence from the
global syndicated loan market. The Review of Financial Studies 25, 2703-2744.
Financial Stability Board (2014) 2014 Update of List of Global Systemically Important Banks
(G-SIBs), Financial Stability Board, November 2014.
Gadanecz, B. (2004) The Syndicated Loan Market: Structure, Development and Implications, BIS
Quarterly Review (December), 75-89.
Gambacorta, L. (2005) Inside the Bank Lending Channel, European Economic Review 49, 1737-
1759.
Gambacorta, L., Marques-Ibanez, D., Spaventa, L. and Haliassos, M. (2011) The Bank Lending
Channel: Lessons from the Crisis, Economic Policy 26, 135-182.
Gambacorta, L. and Mistrulli, P. E. (2004) Does Bank Capital Affect Lending Behavior? Journal
of Financial intermediation 13, 436-457.
Gambacorta, L., & Shin, H. S. (2018). Why bank capital matters for monetary policy. Journal of
Financial Intermediation 35, 17-29.
Gan, J. (2007) The Real Effects of Asset Market Bubbles: Loan- and Firm-Level Evidence of a
Lending Channel, Review of Financial Studies 20, 1941-1973.
33
Giannetti, M. and Laeven, L. (2012) The Flight Home Effect: Evidence from the Syndicated
Loan Market during Financial Crises, Journal of Financial Economics 104, 23-43.
Giannetti, M. and Saidi, F. (2019) Shock propagation and banking structure, The Review of
Financial Studies 32, 2499-2540.
Grandi, P. (2019) Sovereign Stress and Heterogeneous Monetary Transmission to Bank Lending
in the Euro Area, European Economic Review 119, 251-273.
Grosse-Rueschkamp, B., Steffen, S. and Streitz, D. (2019) A capital structure channel of
monetary policy, Journal of Financial Economics 133, 357-378.
Heider, F., Saidi, F. and Schepens, G. (2019) Life below zero: Bank lending under negative
policy rates, The Review of Financial Studies 32, 3728-3761.
Hubbard, R. G., Kuttner, K. N., & Palia, D. N. (2002). Are there bank effects in borrowers' costs
of funds? Evidence from a matched sample of borrowers and banks. The Journal of Business 75,
559-581.
Ivashina, V. and Scharfstein, D. (2010) Bank Lending During the Financial Crisis of 2008,
Journal of Financial Economics 97, 319-338.
Jacewitz, S., & Pogach, J. (2018). Deposit rate advantages at the largest banks. Journal of
Financial Services Research, 53, 1-35.
Jayaratne, J., & Morgan, D. P. (2000). Capital market frictions and deposit constraints at banks.
Journal of Money, Credit and Banking 32, 74-92.
Jiménez, G., Ongena, S., Peydró, J.-L. and Saurina, J. (2012) Credit Supply and Monetary Policy:
Identifying the Bank Balance-sheet Channel with Loan Applications, American Economic Review
102, 2301-2326.
Jiménez, G., Ongena, S., Peydró, J.-L. and Saurina, J. (2014) Hazardous Times for Monetary
Policy: What do Twenty-three Million Bank Loans say About the Effects of Monetary Policy on
Credit Risk Taking? Econometrica 82, 463-505.
Jiménez, G., Mian, A., Peydró, J. L. and Saurina, J. (2020) The real effects of the bank lending
channel, Journal of Monetary Economics 115, 162-179.
Kashyap, A.K. and Stein, J.C. (1995) The Impact of Monetary Policy on Bank Balance Sheets,
Carnegie-Rochester Conference Series on Public Policy, Elsevier 42, 151-195.
Kashyap, A.K. and Stein, J.C. (2000) What do a Million Observations on Banks say About the
Transmission of Monetary Policy? American Economic Review 90, 407-428.
Khwaja, A. I. and Mian, A. (2008) Tracing the Impact of Bank Liquidity Shocks: Evidence from
an Emerging Market, American Economic Review 98, 1413-1442.
34
Kishan, R.P. and Opiela, T.P. (2000) Bank Size, Bank Capital, and the Bank Lending Channel,
Journal of Money, Credit and Banking 32, 121141.
Kishan, R.P. and Opiela, T.P. (2006) Bank Capital and the Loan Asymmetry in the Transmission
of Monetary Policy, Journal of Banking & Finance 30, 259285.
Kleimeier, S., Sander, H. and Heuchemer, S. (2013) Financial Crises and Cross-border Banking:
New Evidence, Journal of International Money and Finance 32, 884-915.
Kopecky, K. and Van Hoose, D. (2004) A Model of the Monetary Sector with and without
Binding Capital Requirements, Journal of Banking & Finance 28, 633-646.
Krippner, L. (2013) Measuring the Stance of Monetary Policy in Zero Lower Bound
Environments, Economics Letters 118, 135-138.
Lenza, M., Pill, H. and Reichlin, L. (2010) Monetary Policy in Exceptional Times, Economic
Policy 25, 295-339.
Meier, S., Gonzalez, M. R. and Kunze, F. (2021) The global financial crisis, the EMU sovereign
debt crisis and international financial regulation: Lessons from a systematic literature review,
International Review of Law and Economics 65, 105945.
Mishkin, F.S. (1996) The Channels of Monetary Transmission: Lessons for Monetary Policy,
NBER Working paper No. 5464.
Molyneux, P., Reghezza, A., Thornton, J., & Xie, R. (2020). Did negative interest rates improve
bank lending? Journal of Financial Services Research 57, 51-68.
Peek, J. and Rosengren, E.S. (1995a) The Capital Crunch: Neither a Borrower nor Lender be,
Journal of Money, Credit, and Banking 27, 626-638.
Peek, J. and Rosengren, E.S. (1995b) Bank Lending and the Transmission of Monetary Policy,
Federal Reserve Bank of Boston Conference Series No. 39.
Peek, J. and Rosengren, E.S. (2014) The Role of Banks in the Transmission of Monetary Policy.
The Oxford Handbook of Banking, 453.
Peydró, J. L., Polo, A. and Sette, E. (2021) Monetary policy at work: Security and credit
application registers evidence, Journal of Financial Economics 140, 789-814.
Popov, A. and Van Horen, N. (2015) Exporting Sovereign Stress: Evidence from Syndicated
Bank Lending during the Euro Area Sovereign Debt Crisis, Review of Finance 19, 1825-1866.
Santos, J. A. (2011) Bank Corporate Loan Pricing Following the Subprime Crisis, The Review of
Financial Studies 24, 1916-1943.
35
Stein, J.C. (1998) An Adverse-Selection Model of Bank Asset and Liability Management with
Implications for the Transmission of Monetary Policy, The Rand Journal of Economics 29, 466-
486.
Van den Heuvel, S. (2002) Does Bank Capital Matter for Monetary Transmission? Economic
Policy Review 8, 1-7.
Wu, J. C. and Xia, F. D. (2016) Measuring the Macroeconomic Impact of Monetary Policy at the
Zero Lower Bound, Journal of Money, Credit, and Banking 48, 253-291.
Wu, J. C. and Xia, F. D (2017) Time Varying Lower Bound of Interest Rates in Europe, Chicago
Booth Research Paper No. 17-06.
36
Appendices
Appendix A: Pearson Correlation Matrix
We compute the correlation matrix of all variables that are not dummies or interaction terms.
Insert Table A here.
37
Figure 1. Monetary policy and syndicated bank lending
Source: ECB website, EONIA stands for Euro OverNight Index Average.
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
2008 2009 2010 2011 2012 2013 2014
ECB key interest rate and syndicated lending
EONIA Syndicated Lending (rhs)
Million (euro)
in %
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
0
250,000
500,000
750,000
1,000,000
1,250,000
1,500,000
1,750,000
2,000,000
2,250,000
2,500,000
2008 2009 2010 2011 2012 2013 2014
ECB balance sheet assets and syndicated lending
Credits Marginal calls Fine-Tuning Main Refinancing Operations
Marginal Lending Long-Term Refinancing Operations Government Debts
Securities (Monetary Policy purpose) Securities (Others) Syndicated Lending (rhs)
Million (euro) Million (euro)
38
Table I. Variables definition
This table defines the variables included in the empirical analysis. The dependent variable, i.e. the amount of the loan, in addition to the characteristics of the loan, the
characteristics of the borrower (except the Value-at-Risk for borrower's industry computed from Datastream), and their relationship are computed by the authors using
data from the LPC DealScan database. The characteristics of the lender are extracted from Bloomberg. The instruments of monetary policy and the results of the bank
lending survey are provided by the ECB while the Euro Area and borrowers’ countries annual GDP growth rates are extracted from Eurostat and the World Bank
respectively. Our final sample contains 21,947 unique loans provided by 19 banking groups to 8,565 borrowing companies between 2008 and 2014. In Appendix A, we
provide the Pearson correlation matrix between all variables that are not dummies.
Variable
Definition
Unit
Average
St. Dev.
Min.
Max.
Dependent Variable

Amount of loan provided by credit institution to borrower at time (taken
as a logarithm)
Million euro
45.58
88.72
0.01
4,200
: Monetary policy instruments

The variation of the quarterly EONIA
Bps
-0.07
0.45
-1.82
0.81

The quarterly variation of the ECB balance sheet equal to total assets minus
general government debt denominated in euro, marginal lending facility, credits
related to marginal calls, and other securities
%
0.03
0.20
-0.28
0.75

The quarterly variation of the value of ECB unconventional policies (i.e. the
sum of LTROs and securities purchased for monetary policy purposes from the
balance sheet assets)
%
0.05
0.27
-0.33
1.05
Lender characteristics
Total Assets (Quarterly frequency)
Billion euro
1,227.31
606.28
43.54
2,305.34

Capital Ratio (Common Equity / Total Assets) (Quarterly frequency)
%
4.09
1.66
0
13.67
Deposits Ratio (Customer Deposits / Total Assets) (Quarterly frequency)
%
23.58
20.27
0
70.12
ST Debt Ratio (Short-Term Debt / Total Assets) (Quarterly frequency)
%
10.53
10.48
0
46.14
Tier 1 capital ratio (Quarterly frequency)
%
11.21
2.28
0
17.30

Variable equal to one when loan is provided by the four lenders with the
smallest total assets
Dummy
0.03
0.16
0
1
Variable equal to one when loan is provided by the four lenders with the
lowest level of capital
Dummy
0.47
0.50
0
1
Variable equal to one when loan is provided by the four lenders with the
lowest level of customer deposits
Dummy
0.44
0.50
0
1
Variable equal to one when loan is provided by the four lenders with the
lowest level of short-term debt
Dummy
0.35
0.48
0
1
Variable equal to one when loan is provided by the four lenders with the
lowest Tier 1 capital ratio
Dummy
0.08
0.28
0
1

Variable equal to one when loan is provided by the four lenders with the
largest total assets
Dummy
0.51
0.50
0
1
39
Variable equal to one when loan is provided by the four lenders with the
highest level of capital
Dummy
0.05
0.22
0
1
Variable equal to one when loan is provided by the four lenders with the
highest level of customer deposits
Dummy
0.03
0.18
0
1
Variable equal to one when loan is provided by the four lenders with the
highest level of short-term debt
Dummy
0.04
0.19
0
1
Variable equal to one when loan is provided by the four lenders with the
highest Tier 1 capital ratio
Dummy
0.23
0.42
0
1
Macroeconomic environment

Quarterly change in the Eurozone GDP taken with one lag
%
0.34
2.05
-5.50
2.80

Quarterly change in banks’ anticipations of credit demand based on question 9
of the bank lending survey
Numerical
0.49
11.83
-28.61
31.03
Loan characteristics

The loan’s maturity (taken as a logarithm)
Month
58.66
43.27
1
432.00

Variable equal to one when the loan is secured
Dummy
0.40
0.49
0
1

Variable equal to one when the loan is a revolver loan (with a maturity lower
than 1 year)
Dummy
0.01
0.08
0
1
Variable equal to one when the loan is a revolver loan (with a maturity higher
than 1 year)
Dummy
0.38
0.49
0
1

Variable equal to one when the loan is a term loan
Dummy
0.33
0.47
0
1
Variable equal to one when the loan is issued during the fourth quarter of the
year (seasonal effect)
Dummy
0.26
0.44
0
1
Borrower characteristics

Annual GDP growth rate of the borrower’s country
%
1.34
2.77
-14.76
19.59

Variable equal to one when the borrower belongs to the manufacturing sector
Dummy
0.30
0.46
0
1
Variable equal to one when the borrower belongs to the financial sector
Dummy
0.11
0.32
0
1
Variable equal to one when the borrower belongs to the service sector
Dummy
0.12
0.32
0
1
Variable equal to one when the borrower belongs to the transportation sector
Dummy
0.10
0.30
0
1
Variable equal to one when the borrower belongs to the real estate sector
Dummy
0.02
0.15
0
1
Variable equal to one when the borrower belongs to the trade sector
Dummy
0.10
0.30
0
1
Variable equal to one when the borrower belongs to another sector
Dummy
0.30
0.46
0
1

Value-at-Risk of the industry
%
-0.02
0.01
0
-0.01

Variable equal to one if the borrower is investment grade
Dummy
0.17
0.37
0
1
Variable equal to one if the borrower is junk grade
Dummy
0.07
0.25
0
1

Variable equal to one when the borrower has the same nationality as the lender
Dummy
0.24
0.43
0
1
Lender additional characteristics

Variable equal to one when the lender is the lead arranger
Dummy
0.69
0.46
0
1

The total amount lent by the credit institution to the industry of the borrower
associated with loan the year before (taken as a logarithm)
Million euro
1,738
2,232
0.06
13,629
40
Lender-Borrower relationship

Variable equal to one when the lender has already lent to the borrower during
the previous year
Dummy
0.21
0.41
0
1
41
Table II. Sample of banking groups
This table provides descriptive statistics of the 19 banking groups included in our sample. BBVA stands for Banco Bilbao Vizcaya Argentaria. The second column contains
the countries where the MFIs are located, i.e. Austria (AU), Belgium (BE), Finland (FI), France (FR), Germany (GE), Greece (GR), Ireland (IR), Italy (IT), Luxembourg
(LU), Netherlands (NL), and Spain (SP). Number of loans represents the sum of all loans in which the banking group has participated. Average loan characteristics (i.e.
amount expressed in millions of euros, maturity expressed in months) and the quarterly average of total assets (expressed in billions of euros), capital ratio (i.e. common
equity to total assets), deposit ratio (i.e. customer deposits to total assets), ST debt ratio (i.e. short-term debt to total assets), and Tier 1 capital ratio and of each banking
group are computed for the 2008-2014 period.
Banking Group
Countries
Number
of loans
Amount
Maturity
Total
Asset
Capital
ratio
Deposit
ratio
ST Debt
ratio
Tier 1
ratio
BNP
FR/IR
8,442
53.19
55.41
1,979
3.47
.
.
11.10
Deutsche bank
GE/LU
7,166
66.39
53.87
1,896
2.81
27.78
9.03
14.05
ING
BE/FR/GE/IR/IT/LU/NL
5,678
39.27
58.37
1,189
3.85
41.47
15.38
11.29
Crédit Agricole
FI/FR/GE
5,258
44.37
63.93
1,646
2.64
3.62
2.81
10.62
Commerzbank
GE/IT/SP
3,895
37.59
50.25
671
3.11
33.79
25.47
11.31
Société Générale
FR/GE/LU
3,784
45.86
55.50
1,198
3.85
.
.
11.33
Natixis
FR/GE/LU
3,644
36.99
58.92
521
3.47
.
.
10.91
Unicredit bank
IT/LU
3,374
39.87
61.77
923
6.22
43.82
22.78
9.94
BBVA
FR/IT/SP
2,907
41.37
73.60
577
6.27
44.67
23.58
10.01
Santander SA
BE/SP
2,551
51.81
68.17
1,191
5.95
46.33
14.68
10.65
Intesa Sanpaolo
IT
1,823
49.34
52.37
642
7.86
41.33
15.21
9.87
KBC bank NV
BE/IR
1,216
26.46
52.96
284
3.65
39.84
16.64
11.48
Sabadell SA
SP
653
17.68
74.27
106
5.97
49.37
17.66
9.30
Banco Populare Espanol
SP
587
17.82
66.80
143
6.69
44.76
26.37
9.73
Bankinter
SP
474
11.01
67.22
56
5.26
38.60
27.48
9.49
Erste bank
AU/LU
473
22.34
51.72
209
5.44
56.67
4.55
9.80
Banca Monte dei Paschi di Siena
IT
341
18.59
61.08
219
5.90
48.65
26.98
7.82
Banca Popolare di Milano
IT
219
19.38
63.05
49
8.79
44.62
17.22
8.65
Alpha Bank AE
GR
70
33.54
90.13
66
8.50
60.24
18.05
10.96
42
Table III. Sample of loans
This table provides descriptive statistics of the sample of loans. The first column discloses the number of loans while
the second column contains the total amount expressed in millions of euros. The first panel provides the split of
borrowers according to their geographical region. The second, third, and fourth panels describe the sample of loans in
terms of loan type, loan objective, and loan maturity respectively.
Number of loans
Total loan amount
Borrowers' region
Africa
522
1%
19,797.76
1%
Asia Pacific
3,925
7%
134,297.79
6%
Eastern Europe/Russia
3,866
7%
119,153.05
5%
Latin America/Caribbean
1,497
3%
63,940.25
3%
Middle East
777
1%
32,611.24
1%
USA/Canada
11,981
23%
713,026.71
30%
Western Europe
29,987
57%
1312,849.5
55%
Loan type
Revolver (<1Y)
304
1%
25,786.44
1%
Revolver (>1Y)
20,203
38%
979,670.41
41%
Term loan
17,240
33%
642,939.24
27%
Others
14,808
28%
747,280.2
31%
Loan objective
General purposes
28,902
55%
1453,556.5
61%
Leveraged Buy-out (LBO)
2,955
6%
59,218.03
2%
Takeover
2,314
4%
238,894.56
10%
Project finance
4,469
9%
124,796.68
5%
Recapitalisation
524
1%
19,271.61
1%
Working capital
1,742
3%
65,453.3
3%
Acquisition
1,932
4%
102,648
4%
Commercial Paper backup
214
0%
22,899.25
1%
Others
9,503
18%
308,938.37
13%
Loan maturity
Short-Term (<1y)
1,200
2%
93,095.53
4%
Medium-Term (1y-5y)
22,930
44%
1093,505.7
46%
Long-Term (>5y)
28,425
54%
1209,075.1
50%
43
Table IV. Estimation results
We estimate the cross-section regression detailed in Equation (1) for 19 European banking groups. The dependent variable is the
loan amount granted by each MFI included in the sample and taken as a logarithm. Standard and non-standard policies are
assessed simultaneously using the quarterly change of the EONIA as standard policy with the quarterly change in the size of the
ECB balance sheet (BS) respectively. The table displays six different specifications. Model (1) includes all control variables in
addition to borrower and lender*year fixed effects. In model (2), we exclude the loan characteristics. Models (3) and (4) consider
lender*quarter and lender country*quarter fixed effects respectively. Models (5) and (6) consider borrower country*year and
borrower country*industry fixed effects respectively. All regressions are run with a constant term. Standard errors are clustered at
the lender level. Table 1 describes the variables. ***, **, * are significant at 1%, 5%, and 10%, respectively. Below the
coefficients, we provide the economic impact of the monetary policy instruments (i.e., we multiply the coefficient by the standard
deviation of the independent variable).
Variables
(1)
(2)
(3)
(4)
(5)
(6)
Monetary policy instruments

-0.053***
-0.050**
-0.024
-0.696*
-0.057**
-0.055**
(0.016)
(0.018)
(0.040)
(0.336)
(0.022)
(0.026)

0.123**
0.121*
0.262***
0.835***
0.172***
0.170***
(0.051)
(0.058)
(0.069)
(0.230)
(0.040)
(0.042)
Economic significance

-0.024
-0.023
-0.011
-0.313
-0.026
-0.025

0.025
0.024
0.052
0.167
0.034
0.034
Controls
Loan
Yes
Yes
Yes
Yes
Yes
Borrower
Yes
Yes
Yes
Yes
Yes
Yes
Macroeconomy
Yes
Yes
Yes
Yes
Yes
Yes
Lender
Yes
Yes
Yes
Yes
Yes
Yes
Relationship
Yes
Yes
Yes
Yes
Yes
Yes
Borrower FE
Yes
Yes
Yes
Yes
Borrower country*Y FE
Yes
Borrower country*Industry FE
Yes
Lender*Y FE
Yes
Yes
Yes
Yes
Lender*Q FE
Yes
Lender country*Q FE
Yes
Observations
49,637
50,594
49,637
49,579
51,414
51,332
0.710
0.694
0.711
0.710
0.313
0.378
44
Table V. Asymmetric effects of the ECB accommodative monetary policy
We estimate the cross-section regression detailed in Equation (2) for 19 European banking groups distinguishing lenders based on
(1) size (   , (2) capital structure (   , (3) deposits level (   , (4) ST debt level
(   , and (5) financial strength (   . The dependent variable is the loan amount granted by each
MFI included in the sample as a percentage of the MFI total assets. Standard and non-standard policies are assessed simultaneously
using the quarterly change of the EONIA as standard policy with the quarterly change in the size of the ECB balance sheet (BS)
respectively. The interaction terms between monetary policies and lenders' characteristics (size, capital structure, deposit and short-
term debt levels, and financial strength) are presented separately considering either the four(three) lenders with the lowest or the
highest level of total assets, capital, deposit, and ST debt ratios, and T1 ratio respectively. All regressions are run with a constant
term, borrower and lender*year fixed effects. Standard errors are clustered at lender level. Table 1 describes the variables. ***, **,
* are significant at 1%, 5%, and 10%, respectively.
Variables
Size effect
Funding constraint
Financial
strength
(1)
(2)
(3)
(4)
(5)
  
  
  
 
  
Monetary policy instruments

-0.037
-0.035
-0.007
-0.018
-0.058***
(0.025)
(0.023)
(0.027)
(0.028)
(0.017)

0.162**
0.164**
0.205*
0.159
0.124**
(0.062)
(0.061)
(0.097)
(0.091)
(0.050)
Interaction Terms
 
0.155**
-0.043**
-0.068***
-0.063***
0.060
(0.062)
(0.018)
(0.021)
(0.017)
(0.037)
 
-0.035*
0.064**
0.066
0.131**
0.006
(0.019)
(0.024)
(0.061)
(0.050)
(0.025)
 
0.350***
-0.094*
-0.182**
-0.113
0.247***
(0.121)
(0.052)
(0.074)
(0.065)
(0.078)
 
-0.084*
0.161**
0.132
0.314**
-0.069
(0.048)
(0.073)
(0.138)
(0.118)
(0.059)
Controls
Loan
Yes
Yes
Yes
Yes
Yes
Borrower
Yes
Yes
Yes
Yes
Yes
Macroeconomy
Yes
Yes
Yes
Yes
Yes
Lender
Yes
Yes
Yes
Yes
Yes
Relationship
Yes
Yes
Yes
Yes
Yes
Borrower FE
Yes
Yes
Yes
Yes
Yes
Lender*Y FE
Yes
Yes
Yes
Yes
Yes
Observations
49,637
49,637
34,113
34,113
49,637
0.710
0.710
0.709
0.709
0.710
45
Table VI. Estimation results (Alternative measure of non-standard policy)
We estimate the cross-section regression detailed in Equation (1) for 19 European banking groups. The dependent variable is the
loan amount granted by each MFI included in the sample and taken as a logarithm. Standard and non-standard policies are assessed
simultaneously using the quarterly change of the EONIA as standard policy with a more restrictive proxy of non-standard policy
based on the balance sheet (NS) respectively. The table displays six different specifications. Model (1) includes all control variables
in addition to borrower and lender*year fixed effects. In model (2), we exclude the loan characteristics. Models (3) and (4) consider
lender*quarter and lender country*quarter fixed effects respectively. Models (5) and (6) consider borrower country*year and
borrower country*industry fixed effects respectively. All regressions are run with a constant term. Standard errors are clustered at
the lender level. Table 1 describes the variables. ***, **, * are significant at 1%, 5%, and 10%, respectively.
Variables
(1)
(2)
(3)
(4)
(5)
(6)
Monetary policy instruments

-0.052**
-0.055**
-0.024
-1.270***
-0.043
-0.054*
(0.020)
(0.024)
(0.041)
(0.406)
(0.027)
(0.027)

0.075
0.063
0.179**
1.388***
0.139***
0.103***
(0.055)
(0.061)
(0.069)
(0.382)
(0.040)
(0.033)
Controls
Loan
Yes
Yes
Yes
Yes
Yes
Borrower
Yes
Yes
Yes
Yes
Yes
Yes
Macroeconomy
Yes
Yes
Yes
Yes
Yes
Yes
Lender
Yes
Yes
Yes
Yes
Yes
Yes
Relationship
Yes
Yes
Yes
Yes
Yes
Yes
Borrower FE
Yes
Yes
Yes
Yes
Borrower country*Y FE
Yes
Borrower country*Industry FE
Yes
Lender*Y FE
Yes
Yes
Yes
Yes
Lender*Q FE
Yes
Lender country*Q FE
Yes
Observations
49,637
50,594
49,637
49,579
51,414
51,332
0.710
0.694
0.711
0.710
0.313
0.378
46
Table VII. Asymmetric effects of the ECB monetary policy (Alternative measure of non-standard
policy)
We estimate the cross-section regression detailed in Equation (2) for 19 European banking groups distinguishing lenders based on
(1) size (   , (2) capital structure (   , (3) deposits level (   , (4) ST debt level
(   , and (5) financial strength (   . The dependent variable is the loan amount granted by each
MFI included in the sample as a percentage of the MFI total assets. Standard and non-standard policies are assessed simultaneously
using the quarterly change of the EONIA as standard policy with a more restrictive proxy of non-standard policy based on the
balance sheet (NS) respectively. The interaction terms between monetary policies and lenders' characteristics (size, capital structure,
deposit and short-term debt levels, and financial strength) are presented separately considering either the four(three) lenders with
the lowest or the highest level of total assets, capital, deposit, and ST debt ratios, and T1 ratio respectively. All regressions are run
with a constant term, borrower and lender*year fixed effects. Standard errors are clustered at lender level. Table 1 describes the
variables. ***, **, * are significant at 1%, 5%, and 10%, respectively.
Variables
Size effect
Funding constraint
Financial
strength
(1)
(2)
(3)
(4)
(5)
  
  
  
 
  
Monetary policy instruments

-0.033
-0.027
0.015
-0.009
-0.050**
(0.030)
(0.029)
(0.036)
(0.038)
(0.019)

0.108*
0.116*
0.174*
0.118
0.093*
(0.060)
(0.065)
(0.092)
(0.092)
(0.046)
Interaction Terms
 
0.194**
-0.056*
-0.111**
-0.083*
0.079**
(0.085)
(0.027)
(0.041)
(0.040)
(0.037)
 
-0.042
0.066**
0.130*
0.156**
-0.030
(0.025)
(0.031)
(0.063)
(0.059)
(0.040)
 
0.321**
-0.092
-0.213**
-0.119
0.194***
(0.118)
(0.067)
(0.096)
(0.100)
(0.055)
 
-0.073
0.120*
0.225**
0.266**
-0.124
(0.054)
(0.062)
(0.086)
(0.096)
(0.074)
Controls
Loan
Yes
Yes
Yes
Yes
Yes
Borrower
Yes
Yes
Yes
Yes
Yes
Macroeconomy
Yes
Yes
Yes
Yes
Yes
Lender
Yes
Yes
Yes
Yes
Yes
Relationship
Yes
Yes
Yes
Yes
Yes
Borrower FE
Yes
Yes
Yes
Yes
Yes
Lender*Y FE
Yes
Yes
Yes
Yes
Yes
Observations
49,637
49,637
34,113
34,113
49,637
0.710
0.710
0.709
0.709
0.710
47
Table VIII. Asymmetric effects of the ECB monetary policy (Alternative banks ranking)
We estimate the cross-section regression detailed in Equation (2) for 19 European banking groups distinguishing lenders based on
(1) size (    , (2) capital structure (    , (3) deposits level (   , (4) ST debt level
(   , and (5) financial strength (    as of 2008. The dependent variable is the loan amount granted
by each MFI included in the sample as a percentage of the MFI total assets. Standard and non-standard policies are assessed
simultaneously using the quarterly change of the EONIA as standard policy with the quarterly change in the size of the ECB balance
sheet (BS) respectively. The interaction terms between monetary policies and lenders' characteristics (size, capital structure, deposit
and short-term debt levels, and financial strength) are presented separately considering either the four(three) lenders with the lowest
or the highest level of total assets, capital, deposit, and ST debt ratios, and T1 ratio respectively. All regressions are run with a
constant term, borrower and lender*year fixed effects. Standard errors are clustered at lender level. Table 1 describes the variables.
***, **, * are significant at 1%, 5%, and 10%, respectively.
Variables
Size effect
Funding constraint
Financial
strength
(1)
(2)
(3)
(4)
(5)
  
  
  
 
  
Monetary policy instruments

-0.037
-0.048*
-0.030
-0.040
-0.064***
(0.025)
(0.026)
(0.023)
(0.023)
(0.019)

0.162**
0.155**
0.164
0.111
0.111**
(0.062)
(0.062)
(0.094)
(0.073)
(0.049)
Interaction Terms
 
0.155**
-0.017
-0.021
-0.008
0.060**
(0.062)
(0.021)
(0.034)
(0.035)
(0.026)
 
-0.035*
0.097***
0.060
0.090**
0.022
(0.019)
(0.034)
(0.070)
(0.039)
(0.024)
 
0.350***
-0.084
-0.125
-0.035
0.209**
(0.121)
(0.056)
(0.092)
(0.091)
(0.074)
 
-0.084*
0.255**
0.136
0.283***
-0.038
(0.048)
(0.094)
(0.186)
(0.090)
(0.047)
Controls
Loan
Yes
Yes
Yes
Yes
Yes
Borrower
Yes
Yes
Yes
Yes
Yes
Macroeconomy
Yes
Yes
Yes
Yes
Yes
Lender
Yes
Yes
Yes
Yes
Yes
Relationship
Yes
Yes
Yes
Yes
Yes
Borrower FE
Yes
Yes
Yes
Yes
Yes
Lender*Y FE
Yes
Yes
Yes
Yes
Yes
Observations
49,637
49,637
34,113
34,113
49,637
0.710
0.710
0.709
0.709
0.710
48
Table IX. ECB and Federal Reserve Monetary Policies
We estimate the cross-section regression detailed in Equation (1) for 19 European banking groups. The dependent variable is the
loan amount granted by each MFI included in the sample and taken as a logarithm. The table displays two different specifications.
Standard and non-standard policies are assessed simultaneously using the quarterly change of the EONIA as standard policy with
the quarterly change in the size of the ECB balance sheet (BS) in specification (1) while combining with the change of the Fed
Funds rate and the change of the size of the Federal Reserve balance sheet (FBS) in specification (2). All regressions are run with a
constant term, borrower and lender*year fixed effects. Standard errors are clustered at the lender level. Table 1 describes the
variables. ***, **, * are significant at 1%, 5%, and 10%, respectively. Below the coefficients, we provide the economic impact of
the monetary policy instruments (i.e., we multiply the coefficient by the standard deviation of the independent variable).
Variables
(1)
(2)
Monetary policy instruments

-0.053***
-0.042**
(0.016)
(0.015)

0.074
(0.082)

0.123**
0.105*
(0.051)
(0.054)

0.335**
(0.152)
Economic Significance

-0.0239
-0.0189

0.0226

0.0246
0.0210

0.0470
Controls
Loan
Yes
Yes
Borrower
Yes
Yes
Macroeconomy
Yes
Yes
Lender
Yes
Yes
Relationship
Yes
Yes
Borrower FE
Yes
Yes
Lender*Y FE
Yes
Yes
Observations
49,637
49,637
0.710
0.710
49
Table A. Pearson Correlation Matrix
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(1) 
1.00
(2) 
-0.01
1.00
(3) 
0.03
-0.58
1.00
(4) 
0.01
-0.73
0.91
1.00
(5) 
0.18
-0.00
0.02
0.01
1.00
(6) 
-0.09
0.07
-0.06
-0.09
-0.53
1.00
(7) 
-0.09
0.01
-0.03
-0.03
-0.49
0.55
1.00
(8) 
-0.13
-0.03
0.02
0.04
-0.52
0.36
0.79
1.00
(9) 
0.14
0.15
-0.11
-0.20
0.29
-0.08
-0.02
-0.20
1.00
(10) 
-0.01
0.05
-0.15
-0.19
-0.02
0.05
0.02
-0.02
0.09
1.00
(11) 
0.05
0.34
-0.04
-0.13
0.02
0.00
-0.00
-0.01
0.00
-0.32
1.00
(12) Maturity
-0.07
0.01
0.00
0.01
-0.04
0.05
0.04
0.04
-0.07
-0.03
0.05
1.00
(13) 
0.09
0.18
-0.00
-0.08
0.10
-0.06
-0.07
-0.10
0.18
-0.07
0.49
-0.05
1.00
(14) Industry risk
0.03
0.15
-0.10
-0.12
-0.01
0.11
0.07
-0.01
0.29
-0.07
0.49
0.10
0.29
1.00
(15) Strategy
0.18
0.02
-0.01
-0.03
0.41
-0.20
-0.20
-0.25
0.24
0.01
0.05
0.02
0.10
0.10
1.00
(16) 
0.02
0.49
-0.42
-0.52
0.04
0.11
-0.00
-0.08
0.34
-0.00
-0.05
-0.06
0.02
-0.06
0.05
1.00
(17) 
-0.00
-0.41
0.41
0.44
-0.04
-0.06
0.02
0.08
-0.19
0.12
0.03
0.04
0.03
0.13
-0.03
-0.89
1.00
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Launched in Summer 2012, the European Central Bank’s (ECB) Outright Monetary Transactions (OMT) program indirectly recapitalized European banks through its positive impact on periphery sovereign bonds. However, the stability reestablished in the banking sector did not fully translate into economic growth. We document zombie lending by banks that remained weakly capitalized even post-OMT. In turn, firms receiving loans used these funds not to undertake real economic activity, such as employment and investment, but to build cash reserves. Creditworthy firms in industries with a high zombie firm prevalence significantly suffered from this credit misallocation, which further slowed the economic recovery. Received March 21, 2018; editorial decision November 13, 2018 by Editor Philip Strahan. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.
Article
Exploiting confidential data from the euro area, we show that sound banks pass negative rates on to their corporate depositors and that pass-through is not impaired when policy rates move into negative territory. We do not observe a contraction in deposits, reflecting a general increase in corporate liquidity during the sample period. When their banks charge negative rates on deposits, firms with ex ante high liquidity invest more than comparable firms that are not charged negative rates and increase their liquid holdings less. These results challenge the common view that conventional monetary policy becomes ineffective at the zero lower bound.
Article
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.
Article
To ensure the safety and soundness of the global financial system as well as individ- ual financial institutions and to reduce systemic risk, numerous policy measures and regulatory reforms have been brought forward as a reaction to the Global Financial Crisis and the European Sovereign Debt Crisis. Simultaneously, numerous academic works have critically reviewed these developments. Therefore, based on a dataset of 455 papers, this article intends to structure the multitude of publications and provide a comprehensive overview of post-crisis regulatory research publications. Studies can be roughly divided into three overarching clusters: publications identifying causes of the crisis, articles focusing on policy and reform reactions, and literature investigating whether these reforms fit their purpose. A holistic and systematic review allows us to extract relevant recommendations and areas of action to be taken to prevent such a crisis in the future.
Article
We explore whether lenders’ decisions to provide liquidity in periods of distress are affected by the extent to which they internalize the negative spillovers of industry downturns. We conjecture that high-market-share lenders are more likely to internalize negative spillovers and show that they provide liquidity to industries in distress when fire sales are likely to ensue. High-market-share lenders also provide liquidity to customers and suppliers of distressed industries when the disruption of supply chains is expected to be costly. Our results suggest a novel channel to explain why credit concentration may favor financial stability. Received November 3, 2017; editorial decision November 5, 2018 by Editor Itay Goldstein.
Article
We show that negative policy rates affect the supply of bank credit in a novel way. Banks are reluctant to pass on negative rates to depositors, which increases the funding cost of high-deposit banks, and reduces their net worth, relative to low-deposit banks. As a consequence, the introduction of negative policy rates by the European Central Bank in mid-2014 leads to more risk-taking and less lending by euro-area banks with a greater reliance on deposit funding. Our results suggest that negative rates are less accommodative and could pose a risk to financial stability, if lending is done by high-deposit banks. Received April 17, 2018; editorial decision September 18, 2018 by Editor Philip Strahan. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.
Article
This paper empirically investigates banks’ lending and the extent to which they are influenced by specific preferences in terms of geographical location and industry. We study whether banks develop a field of expertise and focus on it, or whether they prefer to grant loans quite evenly among countries and industries. We manually built an original database of syndicated loans for banks in the four major banking systems in the eurozone, to estimate the determinants of loans’ amounts between 2005 and 2013. Our findings highlight a domestic bias and a sectoral bias with banks lending larger amounts to their domestic borrowers and to industries they are more familiar with.
Article
This paper examines how the sensitivity of cross-border syndicated loan supply varies with the internationalization of borrower country banking sectors, banks and loan syndicates. A higher foreign bank presence in borrower countries mitigates the transmission of monetary policy. Prior lending experience of international banks in borrower countries also attenuates monetary transmission. In contrast, to the extent they become more international, the credit supply of banks and loan syndicates becomes more sensitive to lender-country monetary policy.