Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Singapore conference
1. A Qualitative Index of Risk
Management of Chinese Banks as an
Intermediate Input in Network DEA
Kent Matthews
Cardiff Business School
Global Science and Technology Forum, 23-24
May 2011, Singapore
2. Research Design
Obtain survey information on risk management
practice and organisation of individual banks.
Compare the responses from Chinese bank risk
managers with foreign bank risk managers
(benchmark).
Questionnaire – 3 areas of risk management.
First – Loan approval decision (large
enterprises, SMEs, consumers).
Second – quantitative and qualitative tools used in risk
management.
Third – organisation, training and staffing.
Chinese and foreign bank risk managers score various
questions on a 1 – 5 scale. 1 is worst and 5 is best.
3. Sample frame
24 bank executives interviewed over the period
2007-2008.
3 foreign banks, Big-4, 9 joint stock banks, 2 city
commercial banks – 15 Chinese banks.
Interviews in -
Beijing, Shanghai, Tianjin, Dalian, Guangzhou, Sh
enzhen.
While the numerical scores provided by the Risk
Managers was the primary objective much useful
soft information was also provided.
7. X17 – Risk Management Training .001*
X18 – Staffing levels .002*
X19 – Organisation of workloads .036*
X20 – Internal Recruitment .177
X21 – External Recruitment .001*
X22 – University background .286
X23 – Foreign University Training .530
X24 – Experience .100
X25 – Professional qualification .009*
X26 – Higher degree .038*
X27 – Retention Policy .834
8. Benchmarking
Foreign bank manager responses used
as benchmarks.
Negative scores – Chinese bank risk
manager score less than benchmark.
Positive score – Chinese bank risk
manager score greater than benchmark.
Negative scores are penalised but
positive scores are not penalised.
11. Principal Component
Analysis
Construct two measures using PCA.
First principal component.
(1) Risk Practice
(2) Risk Organisation.
12. Black Box DEA
Input 1 Output 1
Output 2
Input 2 Black Box
Output 3
Input 3
13. Network DEA
Recognises there are stages in the
production process
Each stage has an intermediate output
which is an intermediate input to a
further stage.
14. Three Stages
Stage 1 - Primary inputs: Operational costs
and fixed assets
Stage 2 - Intermediate inputs/outputs: Labour
costs, non-labour costs, interest costs,
branches, (Later we add Risk Management
Practice, Risk Management Organisation).
Deposits as a primary input.
Stage 3 - Final output: Interest earnings, non-
interest earnings and NPL as a bad output
Output is Revenue or Profit Efficiency
15. Three Exercises conducted
Exercise 1 - Black Box DEA where there
are only two stages – 3 inputs and 2
outputs
Exercise 2 – 3-stage DEA without
measures of Risk Practice and Risk
Organisation but NPL is a bad output.
Exercise 3 - Repeat exercise 2 but use
risk organisation index and risk practice
index in the process.
17. The network additions
RISKORG is an intermediate output at
stage 2 and an intermediate input at
stage 3
RISKPCA is an external input at stage 3
determined by the culture of the bank
and political and/or other pressures it
faces.
18. Contribution of Risk Measures
Accounting measure of bank
performance is return on assets (ROA) =
profits/assets.
Efficiency score from DEA is a measure
of efficiency relative to best practice.
Efficiency is revenue or profit efficiency.
Correlation between ROA and DEA
score should be high if it is a good
measure.
20. Contribution of Risk
Measures
Black box DEA Network DEA – without Network DEA – with Risk
risk measures Measures
Intercept 0.328 0.704*** 0.625***
(.197) (.000) (.000)
NPL_RATIO -.037*** -.032*** -.035***
(.002) (.005) (.001)
SCORE 0.911*** 0.877*** .725***
(.007) (.003) (.000)
.6378 .6755 .7803
R2
.2040 .1931 .1589
MSE
F(2,12) 13.33 15.57 25.86
21. Conclusion
The indices constructed from interviews
provide insights into the risk function in
Chinese banks relative to best practice
The combination of the risk practice and risk
organisation with the other inputs and outputs
of the banks in a NDEA framework can be
used to measure bank performance.
NDEA provides information to the manager
about the stages of production between inputs
and outputs.