Drivers of Credit Losses in Australasian Banking

1
Drivers of Credit Losses in Australasian Banking

• Slides prepared by Kurt HessUniversity of
  Waikato Management School, Department of
  FinanceHamilton, New Zealand

2
Topics

• Motivation
• Literature review
• Credit loss data Australasia
• Methodological issues
• Results
• Conclusions

3
Motivation

• Stability and integrity of banking systems are of
  utmost importance to national economies
• Credit losses, or more generally, asset quality
  problems have repeatedly been identified as the
  ultimate trigger of bank failures e.g. in
  Graham Horner (1988), Caprio Klingebiel
  (1996)

4
Motivation

• Prudential supervisory agencies need to
  understand drivers of credit losses in banking
  system
• Validation of proprietary credit risk models
  parameters under Basel II
• This is the first specific research of long term
  drivers of credit losses for Australian banking
  system

5
Literature review
Two main streams of research that analyse drivers
of banks credit losses (or more specifically
loan losses)

• Literature with regulatory focus looks at macro
  micro factors
• Literature looks discretionary nature of loan
  loss provisions and behavioural factors which
  affect them

6
Literature review

• Behavioural hypotheses in the literature on the
  discretionary nature of loan loss provisions
• Income smoothingGreenawalt Sinkey (1988)
• Capital management Moyer (1990)
• Signalling Akerlof (1970), Spence (1973)
• Taxation Management

7
Literature review

• Studies with global samples (using commercial
  data providers)
• Cavallo Majnoni (2001),Bikker Metzemakers
  (2003)
• Country-specific samples
• Austria Arpa et al. (2001)
• Italy Quagliarello (2004)
• Australia Esho Liaw (2002)(in this APRA
  report the authors study level of impaired assets
  for loans in Basel I risk buckets for 16
  Australian banks 1991 to 2001)

8
Literature review

• Research based on original published financial
  accounts is rare (very large effort to collect
  data).
• Pain (2003) 7 UK commercial banks 4 mortgage
  banks 1978-2000
• Kearns (2004)14 Irish banks, early 1990s to
  2003
• Salas Saurina (2002) Spain

9
Credit Loss Data Australasia

• The database includes extensive financial and in
  particular credit loss data for
• 23 Australian 10 New Zealand banks
• Time period from 1980 to 2005
• Approximately raw 55 data elements per
  institution, of which 12 specifically related to
  the credit loss experience (CLE) of the bank

10
Credit Loss Data Australasia

• Sample selection criteria
• Registered banks
• Must have substantial retail and/or rural banking
  business
• Exclude pure wholesale and/or merchant banking
  institutions

11
Credit Losses and GDP Growth (New Zealand Banks)
Provisioning/write-off behaviour correlated to
macro factors
Note chart for NZ Bank sub-sample only
12
Credit Loss Data Australasia
13
Drivers of Credit Losses in Australasian Banking

• Methodology

14
Principal Model
CLEit Credit loss experience for bank i in period
t xkit Observations of the potential explanatory
variable k for bank i and period t uit Random
error term with distribution N(0,?), ? Variance-co
variance matrix of ?it error terms n Number of
banks in sample T Years in observation
period K Number of explanatory variables zk Maximu
m lag of the explanatory variable k of the
model q Maximum lag of the dependent variable of
the model
Kurt Hess, WMS kurthess_at_waikato.ac.nz
14
12-Nov-09
15
Principal Model

• Principal model on previous slide allows for many
  potential functional forms.
• There are choices with regard to
• Dependent CLE proxy
• Suitable drivers of credit losses and lags for
  these drivers
• Estimation techniques

Kurt Hess, WMS kurthess_at_waikato.ac.nz
15
12-Nov-09
16
Determinants of Credit Losses
Macro Factors (1)
17
Determinants of Credit Losses
Macro Factors (2)
18
Determinants of Credit Losses
Bank Specific Factors (1)
19
Determinants of Credit Losses
Bank Specific Factors (2)
20
Determinants of Credit Losses
Bank Specific Factors (3)
21
Determinants of Credit Losses
Bank Specific Factors (4)
22
Pooled regression model as per equation 1 in
paper

• Dependent
• Impaired asset expense as CLE proxy
• Determinants (as per table next slide)
• Alternative macro factors GDP growth,
  unemployment rate
• Alternative asset shock proxies share index,
  house prices
• Misc. bank-specific proxies
• Bank past growth

Kurt Hess, WMS kurthess_at_waikato.ac.nz
22
12-Nov-09
23
Dependent variables in model
Aggregate
Bankspecific
Kurt Hess, WMS kurthess_at_waikato.ac.nz
23
12-Nov-09
24
Drivers of Credit Losses in Australasian Banking

• Empirical results

25
Results macro state factors
see Table 8, 9,10 in paper

• GDP growth (GDPPGRW), change and level of the
  unemployment rate (UNEMP, DUNEMP) have expected
  effect (not all lags significant)
• Unemployment with best explanatory power for
  overall sample

26
Results macro state factors (2)
see Table 8, 9,10 in paper

• Country-specific differences between Australia
  and New Zealand
• Australias results show much greater
  sensitivities to GDP growth (see Table 9)
• New Zealand results are less significant and
  effects of GDP and UNEMP seem more delayed

Kurt Hess, WMS kurthess_at_waikato.ac.nz
26
12-Nov-09
27
Results asset price factors
see Table 8, 9,10 in paper

• Contemporaneous coefficient of share index return
  negative significant for overall and Australia.
  Less significant for NZ.
• Housing price index has less sigificanceIntuition
  early 90s crises not rooted in particular
  problems of the housing sector

28
Results CPI growth
see Table 8, 9,10 in paper

• Positive, but not significant coefficients for
  most regressions, i.e. inflationary pressure
  tends to lift credit losses
• Contemporaneous term negative and significant for
  Australian sub-sample, in line with evidence
  elsewhere that inflation may lead to temporary
  improvement of borrower quality (Tommasi, 1994)

Kurt Hess, WMS kurthess_at_waikato.ac.nz
28
12-Nov-09
29
Results size proxy
see Table 8, 9,10 in paper

• Higher level of provisioning for larger banks
  no significance of coefficients, however
• Intuition portfolios of smaller institutions
  often dominated by (comparably) lower risk
  housing loans

Kurt Hess, WMS kurthess_at_waikato.ac.nz
29
12-Nov-09
30
Results net interest margin
see Table 8, 9,10 in paper

• Generally negative, contemporaneous and 2yr
  lagged term significant, i.e.
• Lower past margins lead to higher subsequent
  losses (induce risk taking)
• Difficult to explain contemporaneous negative
  term
• Inconclusive results also in comparable studies,
  e.g. Salas Saurina (2002) for Spain

Kurt Hess, WMS kurthess_at_waikato.ac.nz
30
12-Nov-09
31
Results net interest margin (2)
see Table 8, 9,10 in paper

• Endogenous nature of net interest margins as
  postulated by Ho Saunders (1981) dealership
  model. Spread increases with
• Market power (inelastic demand)
• Bank risk aversion
• Larger size of transactions (loans/deposits)
• Interest rate volatility
• Net interest margins may thus control for other
  bank specific market characteristics

Kurt Hess, WMS kurthess_at_waikato.ac.nz
31
12-Nov-09
32
Results cost efficiency (CIR)
see Table 8, 9,10 in paper

• High and increasing cost income ratios are
  associated with higher credit losses
• Results reject alternative hypothesis that banks
  are inefficient because they spend to much
  resources on borrower monitoring
• Not surprising as gut feel would tell that
  excessive monitoring might not pay

Kurt Hess, WMS kurthess_at_waikato.ac.nz
32
12-Nov-09
33
Results earnings proxy
see Table 8, 9,10 in paper

• Very clear evidence of income smoothing
  activities, i.e. banks increase provisions in
  good years, withhold them in weak years.
• Confirms similar results found in many other
  studies

Kurt Hess, WMS kurthess_at_waikato.ac.nz
33
12-Nov-09
34
Results past bank growth
see Table 8, 9,10 in paper

• Clear evidence of the fast growing banks faced
  with higher credit losses in future (lags beyond
  2 years)
• Managers seem unable (or unwilling) to assess
  true risks of expansive lending
• Much clearer results than in other studies.
  Possibly due to test design with longer lags
  considered.

Kurt Hess, WMS kurthess_at_waikato.ac.nz
34
12-Nov-09
35
Conclusions

• Model presented here is very suitable for
  assessing general / global effects on impaired
  assets in the banking sector
• The dynamics of this transmission seems to differ
  among systems
• A study of particular effects might thus call for
  alternative models

36
Conclusions (2)

• Income smoothing is a reality, possibly also with
  new tighter IFRS provisioning rules as this
  ultimately remains a discretionary managerial
  decision

37
Conclusions (3)

• Use data base for comparative studies of
  alternative CLE dependent variables
• First results show that they (in part) correlate
  rather poorly which means there must be caution
  comparing results of studies unless CLE is
  defined in exactly the same way

Kurt Hess, WMS kurthess_at_waikato.ac.nz
37
12-Nov-09
38
Credit Loss Experience of Australasian Banks

• Back-up Slides

39
Basel II Pillars

• Pillar 1
• Minimum capital requirements
• Pillar 2
• A supervisory review process
• Pillar 3
• Market discipline (risk disclosure)

40
Basel II Pillars
Pages in New Basel Capital Accord (issued June
2004)
41
Pro Memoria Calculation Capital Requirements
under Basel II
Unchanged
Total Capital Credit Risk
Market Risk Operational Risk
? 8
(Could be set higher under pillar 2)
Significantly Refined
Relatively Unchanged
New
Source slide inspired by PWC presentation slide
retrieved 27/7/2005 from http//asp.amcham.org.sg/
downloads/Basel20II20Update20-20ACC.ppt ,
42
Basel II IRB Approach

• Two approaches developed for calculating capital
  minimums for credit risk
• Standardized Approach (essentially a slightly
  modified version of the current Accord)
• Internal Ratings-Based Approach (IRB)
• foundation IRB - supervisors provide some inputs
• advanced IRB (A-IRB) - institution provides
  inputs

43
Basel II IRB Approach

• Internal Ratings-Based Approach (IRB)
• Under both the foundation and advanced IRB banks
  are required to provide estimates for probability
  of default (PD)
• It is commonly known that macro factor are the
  main determinants of PD

44
Primer Loan Loss Accounting
45
Primer Loan Loss Accounting
46
Credit Loss Data Australasia
BNZ 1984 - 2002
47
Credit Loss Data Australasia

• Banks in sample

AUSTRALIA Adelaide Bank, Advance Bank, ANZ,
Bendigo Bank, Bank of Melbourne, Bank West, Bank
of Queensland, Commercial Banking Company of
Sydney, Challenge Bank, Colonial State Bank,
Commercial Bank of Australia, Commonwealth Bank,
Elders Rural Bank, NAB, Primary Industry Bank of
Australia, State Bank of NSW, State Bank of SA,
State Bank of VIC, St. George Bank,
Suncorp-Metway, Tasmania Bank, Trust Bank
Tasmania, Westpac
NEW ZEALAND ANZ National Bank, ASB, BNZ,
Countrywide Bank, NBNZ, Rural Bank, Trust Bank
NZ, TSB Bank, United Bank, Westpac (NZ)
48
Credit Loss Data Australasia

• Data issues
• Macro level statistics
• Differing formats between NZ and Australiae.g.
  indebtedness of households / firms
• House price series back to 1986 only for
  Australia
• Balance sheets of M3 institutions only back to
  1988 for New Zealand (use private sector credit
  statistics instead)

Kurt Hess, WMS kurthess_at_waikato.ac.nz
48
12-Nov-09
49
Credit Loss Data Australasia

• Data issues (2)
• Micro / bank specific data
• Lack of reporting limits choice of
  proxies(particularly through the very important
  crisis time early 1990)
• Comparability due to inconsistent reporting(e.g.
  segment credit exposures)

Kurt Hess, WMS kurthess_at_waikato.ac.nz
49
12-Nov-09
50
Measuring CLE

• Dedicated nature of database allows tests for
  many proxies for a banks credit loss experience
  (CLE)
• Level of bad debt provisions, impaired assets,
  past due assets
• Impaired asset expense (provisions charge to
  PL)
• Write-offs (either gross or net of recoveries)
• Components of above proxies, e.g. general or
  specific component of provisions (stock or
  expense)

Kurt Hess, WMS kurthess_at_waikato.ac.nz
50
12-Nov-09
51
Measuring CLE
Histogram of selected CLE proxies
Median
Pooled observations of Australian and NZ Banks
1980 - 2005
52
Credit Loss Experience of Australasian Banks

• Selected References

53
Selected References

• Bikker, J. A., Metzemakers, P. A. J. (2003).
  Bank Provisioning Behaviour and Procyclicality,
  De Nederlandsche Bank Staff Reports, No. 111.
• Caprio, G., Klingebiel, D. (1996). Bank
  insolvencies cross-country experience.
  Worldbank Working Paper WPS1620.
• Cavallo, M., Majnoni, G. (2001). Do Banks
  Provision for Bad Loans in Good Times? Empirical
  Evidence and Policy Implications, World Bank,
  Working Paper 2691.

54
Selected References

• Esho, N., Liaw, A. (2002). Should the Capital
  Requirement on Housing Lending be Reduced?
  Evidence From Australian Banks. APRA Working
  Paper(02, June).
• Graham, F., Horner, J. (1988). Bank Failure An
  Evaluation of the Factors Contributing to the
  Failure of National Banks, Federal Reserve Bank
  of Chicago.

55
Selected References

• Kearns, A. (2004). Loan Losses and the
  Macroeconomy A Framework for Stress Testing
  Credit Institutions Financial Well-Being,
  Financial Stability Report 2004. Dublin The
  Central Bank Financial Services Authority of
  Ireland.
• Pain, D. (2003). The provisioning experience of
  the major UK banks a small panel investigation.
  Bank of England Working Paper No 177, 1-45.

56
Selected References

• Salas, V., Saurina, J. (2002). Credit Risk in
  Two Institutional Regimes Spanish Commercial and
  Savings Banks. Journal of Financial Services
  Research, 22(3), 203 - 224.