What drives banks’ credit standards? An analysis based on a large bank-firm panel

An analysis based on a large

Metadata

Provider:

ECB

URL:

https://www.ecb.europa.eu//pub/pdf/scpwps/ecb.wp2902~278590619d.en.pdf

Retrieved on:

onsdag, februari 7, 2024

Information

Tags:

NPL, Seven Bank, Xi, Name, AnaCredit, Data, Bank, ECB, SAFE, JEL, Bias, Knowledge, XB, P50, Ciccarelli, Credit, Regression analysis, Shock, Central bank, Mand, E51, Movement, E44, SNL, Government, Access, Environment, Literature, P90, BLS, P25, CDS, Darst, Risk, European Central Bank, Bureau van Dijk, OBS, Probability, Complementarity, ROA, Policy, COVID-19, Research, Standard score, Uncertainty, R2, Heider, E52, Holmström, CSB, Health, P75

Geotags:

PETRA, ALTAVILLA, ALTMAN, ESTONIA, NETHERLANDS, ITALY, AUSTRIA, IRELAND, UKRAINE, PORTUGAL, LITHUANIA, SPAIN, GERMANY, FRANCE, LUXEMBOURG

Sammanfattning

Key Points:

An analysis based on a large
bank-firm panel
No 2902

Disclaimer: This paper should not be reported as representing the views of the European Central Bank
(ECB).
We find
that weaker capitalised banks adjust their credit standards more than healthier banks, especially for
firms with a higher default risk.
Here we find t hat w eaker b anks r espond m ore f orcefully by
tightening their credit standards more than better capitalised banks.
On the contrary, weaker banks
may be more prone to adopt looser credit standards, with the aim of increasing their revenues.
To answer these questions, we analyse the determinants of banks? credit standards, i.e., their internal
guidelines or loan approval criteria applied when deciding on granting credit.
2 Altavilla
ECB Working Paper Series No 2902

2

area banks tighten their credit standards more when linked to riskier firms, measured via firms? leverage
and default risk.
We assess how euro area banks adjusted their credit standards in response to
the negative COVID-19 pandemic shock, after accounting for government support measures.
When deciding on their credit standards, banks assess risks
based on both their own loss absorption capacity and the credit risk of their borrowers.
On the contrary,
weaker banks may be more prone to adopt looser credit standards, with the aim of increasing their
revenues.
We provide evidence that
euro area banks tighten their credit standards more when linked to riskier firms, measured via firms?
leverage and default risk based on the Altman Z-score.
In
addition, they focus on a different research question and use data from the IBLS only as a control.
ECB Working Paper Series No 2902
5

capital position implies less tightening of lending criteria, possibly reflecting the fact that banks can
afford to adjust their credit standards more moderately.
Based on our results, this implies a stronger deterioration of their lending conditions compared
with less vulnerable firms.
We assess how euro area banks adjusted their credit standards in response to
the negative COVID-19 pandemic shock, after accounting for government support measures.
This is in line with the role of government support
measures such as loan guarantees mitigating banks? exposure to firms? credit risks as they shield banks
from firms? increased credit risks.
2
Related literature

Our paper is closely related to studies analysing credit supply based on BLS indicators and the impact
of monetary policy shocks on bank lending conditions in the euro area.
Hempell and Kok (2010) disentangle
pure loan supply based on the BLS factors and investigate the role played by such factors for loan growth.
Several other studies link confidential individual BLS data with actual bank-level data, but not firm
data, allowing an analysis of bank characteristics relevant for bank lending conditions.
They find that a short-term interest rate shock decreases both loan supply
and demand, but more for less healthy banks.
Their findings are consistent with the results of our paper on the favourable impact of bank health on lending standards.
Both papers tend to find no evidence of higher risk taking of banks as a result
of accommodative monetary policy.
More recent studies are based on
confidential bank and firm-level data from national credit registers.
(2012) who focus on the bank-firm-relationship in Spain, based on credit register data.
Ferrero, Nobili, and Sene (2019) arrive at a corresponding conclusion on the risk-taking
channel based on a confidential loan-level dataset of Italian banks.
In another paper, Altavilla, Boucinha, and Bouscasse (2022)
disentangle credit demand and supply based on euro area credit register data (AnaCredit) for the period
of the pandemic.
Our results emphasise the
mitigating impact of government guarantees on a tightening of credit standards during the pandemic.
This mitigating impact played a major role in loan demand and not credit supply being decisive for lending volumes during the pandemic.
Based on their model, accommodative monetary policy is part of the optimal policy mix, combined with social insurance.
To keep the wealth of information
available in the BLS, we run our analysis at the quarterly frequency of the survey.
of employees
101.4

2456.9

2.0

4.0

12.0

37.0

116.0

14944589

Panel (a): Banks
Credit standards

Loan loss provisions
Panel (b): Firms

Notes: Descriptive statistics for the bank-firm sample included in the regression analysis.
Specifically, a one
standard deviation increase in the CET1 ratio leads to 0.2 standard deviations lower credit standards,
i.e., easier credit standards.
In their lending decisions, banks assess risks based on both their own
loss absorption capacity and the credit risk of their borrowers.
?Credit supply and monetary policy: Identifying the bank balance-sheet channel with loan applications.? American Economic
Review 102 (5):2301?2326.
?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 (2):463?505.
?The credit cycle and the business cycle: new findings using
the loan officer opinion survey.? Journal of Money, Credit and Banking 38 (6):1575?1597.
guarantees: proxy from BLS, bank level
0

.1

.2

.3

.4

Government guarantees exposure

-.5

-.25

0

.25

.5

Government guarantees exposure

Notes: Based on results from columns (3) and (6) of Table 4.