Does monetary policy affect lending standards

1
Does monetary policy affect bank credit
standards?Angela Maddaloni, José Luis Peydró
Silvia ScopelEleventh Conference of the ECB-CFS
Research Network The market for Retail
Financial Services Development, Integration,
and Economic Effects 21 October 2008
2
Outline

• Motivation
• The euro area Bank Lending Survey (BLS)
• Results

3
Outline

• Motivation
• The euro area Bank Lending Survey (BLS)
• Some results

4
Main questions

• Does monetary policy affect bank lending
  standards?
• If so, how?
• bank/supply driven
• borrower quality driven
• banks risk taking
• Relevant questions for research and policy

5
Monetary policy and lending standards

• The impact of monetary policy
• loan demand (interest rate channel)
• loan supply (credit channel)

Balance sheet channel (firms)
Bank lending channel (banks)
Risk-taking channel
6
Monetary policy and lending standards

• There are two main identification problems
• demand versus supply
• channels of transmission of MP
• The BLS data, in particular the answers to the
  sub-questions can help to cope with these
  problems

7
Monetary policy and credit markets channels of
transmission

• If policy rates are lower,
• more loan demand (interest rate channel)
• more bank lending (bank lending channel)
• Bernanke and Blinder,1988 1992, Kashyap
  Stein, 2000
• borrowers balance sheets improve (balance sheet
  channel)
• Bernanke, Gertler, Gilchrist, 1996 1999,
  Matsuyama, 2007
• more risk-taking
• Smith, 2002, Diamond Rajan, 2006, Rajan,
  NBER 2006, Borio Zhou, 2007, Greenwald
  Stiglitz, 2003

8
The risk-taking channel

• Several papers argue that monetary policy has an
  impact on banks risk taking (liquidity risk or
  credit risk)
• Impact of short-term rates on risk-taking is a
  non-trivial yet unaddressed empirical question

9
The risk-taking channel

• Jimenez, Ongena, Peydró and Saurina (2008) find
  that lower short-term rates prior to loan
  origination imply loans with higher hazard rate,
  whereas higher rates during the life of the loan
  increase hazard rate
• Ioannidou, Ongena and Peydró (2007) find in a
  dollarized system (Bolivia) that banks take more
  credit risk and reduce the loan spreads when
  rates are low
• We use the data from the euro area Bank Lending
  Survey, so we are able to exploit cross-country
  heterogeneity

10
Outline

• Motivation
• The euro area Bank Lending Survey (BLS)
• Some results

11
The BLS questionnaire

• Lending standards are provided by the banks
• We can disentangle demand from supply (or at
  least we can control for it!) thanks to the
  richness of the survey
• Possible to control for the quality of borrowers
  (balance sheet channel) and for banks financing
  (bank lending channel)
• Exploit cross-country differences in monetary
  policy stance and business cycles

12
The BLS questionnaire

• The questionnaire ask about supply and demand of
  banks loans to enterprises and households (for
  house purchase and consumer credit)
• 18 regular questions, backward-looking and
  forward-looking, multiple-choice with 5 possible
  answers
• Data for 12 countries over the entire period
• Qualitative survey starting on 2002 q4, no
  information on the banks characteristics, except
  bank size (bank identifier, not identity!)

13
The BLS questionnaire

• Over the past three months, how have your banks
  credit standards as applied to the approval of
  loans
• or credit lines to enterprises changed?
• (overall, SMEs, large, ST, LT)
• to households changed?
• (house purchase, consumer credit)
• Factors affecting credit standards
• (cost of funds and balance sheets constraints,
  pressure from competition, perception of risk)
• Changes in conditions and terms
• (price, other)

14
The BLS questionnaire

• Over the past three months, how has the demand
  for loans changed at your bank, apart from normal
  seasonal fluctuations?
• to enterprises
• (overall, SMEs, large, ST, LT)
• to households
• (house purchase, consumer credit)
• Factors affecting demand for loans
• (financing needs, use of alternative finance)

15
Monetary policy and the BLS

• Estimate GLS panel regressions with country fixed
  effects

16
Our approach

• The BLS variables are the net percentage of banks
  reporting a tightening of standards or an
  increase in the demand for loans
• As a robustness check we also use individual bank
  answers
• Ordered probit
• Bank fixed effects
• Bank size
• But we dont have the last 4 quarters of data

17
Cross-country variation
Credit standards for loans to enterprises
18
Cross-country variation
Credit standards for loans to households for
house purchase
19
Cross-country variation
Real GDP growth
20
Outline

• Motivation
• The euro area Bank Lending Survey (BLS)
• Results

21
Results

• Lower policy rates soften bank credit standards
• Interestingly, the softening of credit standards
  is over and above an improvement of the quality
  of borrowers industry collateral (over and
  above the balance sheet channel)
• Banks reduce spreads on average loans, reduce
  collateral requirements and covenants
• Controlling for factors related to loan demand,
  monetary policy affects the volume and maturity
  of supplied loans

22
Results

• Rates too low for too long soften even further
  credit standards
• Securitization softens banks lending standards
  and makes the impact of policy rates stronger

23
Results credit standards

• Lower policy rates soften bank credit standards

24
Results credit standards

• Higher levels of overnight rates imply tighter
  credit standards
• The effect is stronger for loans to firms than
  for loans to households (is it related to demand
  or supply?)
• Need to control for borrowers net worth and
  quality of collateral

25
Results credit standards

• BLS ask about the reasons why banks change their
  credit standards
• Economic cycle
• Expectations on general economic activity
• Creditworthiness of the borrower
• Industry or firm-specific outlook
• Risk on collateral demanded
• Financial conditions of the bank
• Banks capital position
• Access to market financing
• Banks liquidity position

26
Results credit standards

• For borrowers with similar creditworthiness,
  lower policy rates soften bank credit standards.
  This suggests that banks take on higher credit
  risk when monetary policy is expansive (risk
  taking channel of monetary policy)

27
Results credit standards

• Even controlling for the balance sheet channel of
  monetary policy and improvement of financing
  conditions of banks (bank lending), EONIA remains
  significant in explaining changes in credit
  standards
• The impact tend to be larger for loans to
  enterprises and statistical significance is weak
  for loans to households for consumer credit

28
Results credit standards

• Banks tighten credit standards by changing
  conditions and terms of the loans, in particular
• margins
• collateral
• size
• maturity

29
Results credit standards

• Tightening takes place with changes in
  conditions...

30
Results credit standards

• The effect of policy rates is significant in
  explaining spreads applied to average loans and
  to riskier loans
• Banks adjust loan amounts, collateral
  requirements, loan covenants and maturity

31
Results credit standards

• Banks change their lending standards by changing
  terms and conditions of the loans
• In particular, when short term rates are low
• banks reduce margins on average loans, reduce
  collateral requirements and covenants
• controlling for factors related to loan demand,
  expansionary monetary policy increases the volume
  and maturity of supplied loans

32
Monetary policy and the BLS4th result

• The results are not only demand driven since
• All of the lending standards soften (or tighten),
  which is not consistent with demand driven
• We control for the risk of collateral to control
  for a different pool of borrowers
• Monetary policy directly affects BLS through bank
  net worth (i.e. through supply)

33
Results credit standards

• Pure supply factors, linked to financing
  conditions
• for banks, confirm the results....

34
Results the policy stance

• Consider the difference between the policy rates
  (common across countries) and a rate implied by a
  simple Taylor-rule
• We introduce a variable which measures the too
  low for too long stance
• Rates too low for too long soften even further
  credit standards

35
Results the policy stance

• Low rates relax credit standards at least for
  loans to households...

36
Results the policy stance

• ...but also for firms if interbank rates are used

37
Results business model

• Short-term rates are economically more important
  than long-term rates in explaining credit
  standards (business model of banks)
• Term spreads are less important than short-term
  rates

38
Results the yield curve
Short-term versus long-term rates...
39
Results securitisation

• Securitisation activity softens the lending
  standards
• Securitisation activity amplifies the impact of
  monetary policy
• The impact of lower short term rates on the
  softening of lending standards is larger the
  higher the securitization

40
Results securitisation
41
Results securitisation
42
Results loan amount

• Higher policy rates reduce the size and the
  maturity of loans
• Higher securitization increase the size and the
  maturity of the loans
• The impact remains significant even after
  controlling for loan demand

43
Results loan amount
44
Results loan maturity
45
Results size of loan
46
Summary

• Monetary policy affects lending standards (BLS)
• The transmission works through credit markets
  (i) the balance sheet channel (ii) bank lending
  channel (iii) and risk taking channel
• Banks change their lending standards by changing
  all the terms and conditions of the loans
• The results are not (only) demand driven

47
Summary

• Rates too low for too long soften even further
  credit standards
• Short-term rates are more important than other
  yield curve measures
• Securitisation activity amplifies the impact of
  monetary policy

48
(No Transcript)
49
Summary

• Lower overnight interest rates soften bank credit
  standards
• supply driven
• not only through borrowers balance sheet channel
• risk appetite/ability
• Rates too low for too long soften even further
  credit standards
• Securitization makes the impact of policy rates
  stronger

50
Whats next

• Monetary policy affects bank lending standards
• Impact on the real economy
• New paper, new results point to the BLS having
  predictive power for macro variables

51
Future research
52
Results credit standards
53
Results credit standards
54
Results credit standards
Controlling for factors related to economic
cycle, borrower s net worth and financing
conditions...
55
Results credit standards
56
Future research
57
Results demand for credit
Policy rates affect also the demand for credit...
58
Results demand for credit
Policy rates affect also the demand for credit...
59
Results demand for credit

• The impact of policy rates is bigger for credit
  demand for house purchase than for consumer
  credit and it is not significant for loans to
  enterprises
• When controlling for factors affecting credit
  demand, monetary policy retains a significant
  effect on the demand for house mortgages

60
Results demand for credit

• In particular, we are controlling for
• demand due to financing needs MA, fixed
  investment, inventories and working capital, debt
  restructuring
• demand due to the lack of alternative sources of
  financing securities issuance, loans from other
  banks and non-banks

61
Results demand for credit
62
Results the yield curve
Short-term versus long-term rates...