International Journal of Economic Issues, 1: 2 (2009) : 1-8
© International Science Press
BANK REGULATION AND CREDIT QUALITY IN
INDIAN BANKING: A QUANTITATIVE EVALUATION
HOSAMANE M.D.
DOS in Economics & Co-Operation, University of Mysore, Manasagangotri, Mysore
DINABANDHU BAG
Research Scholar, DOS in Economics & Co-Operation, University of Mysore, Manasagangotri, Mysore
There exists meaningful inter-linkages among economic and financial variables
such that the variation in the credit quality at the macro level can be explained. Both
the economic factors and also the bank level factors play a critical role in determining
the credit quality of assets. This study undertakes an empirical analysis for finding
the impact of economic and financial factors on banks? non-performing loans. It
provides a framework for analysis of underlying default behavior of borrowers? at
the bank level in terms of banks? individual characteristics and also due to the
presence of other macroeconomic indicators. We conclude that a positive outlook on
economic growth on banks would favour loan repayment response of borrowers in
order to maintain credit worthiness and credit quality. Rising capital adequacy
ratio and higher credit deposit ratio can jointly help improve the portfolio credit
quality. The results of the study are in line with banking literature and provide an
important insight for banks? lending behavior.
Keywords: Credit Quality, NPA, Factors, Credit Deposit Ratio
1. INTRODUCTION
Credit risk is the risk of loss that may occur due to the failure of any counterparty to
abide by the contract with the Bank, principally the failure to make required payments
as per the contract. The purpose of this study is to provide meaningful analysis of
inter-linkages among economic and financial variables such that the variation in the
credit quality at the macro level can be explained. The credit quality scenario has
undergone significant changes in the last decade amidst reform of the financial sector
by rapid increase in globalization and advances in information technology. Overall,
these developments have led to structural change in the financial sector, which has
created conducive environment for market mechanism, in general, and economic
factors, in particular, for playing a critical role in influencing the portfolios of banks
and financial institutions. It is in this context that this study has considered it
imperative to undertake an empirical analysis for evaluating the impact of economic
and financial factors on banks? non-performing loans. The distinguishing feature of
/ INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
the study is that it provides a framework for analysis of underlying default behavior
of borrowers? at the bank level in terms of banks? individual characteristics and also
due to the presence of other macroeconomic indicators. Are the credit policies of
Indian banks adequate to maintain the credit quality of their portfolios? Do they
have comprehensive risk management strategies in place? To answer such questions
we highlight the risk policy of a leading Indian bank in the next section.
A leading bank called ICICI bank is being referred to in this context. The credit
policy in ICICI bank, vary across product groups and they include drivers such as
income of the borrower, loan to value ratio, demographic and stability of the
borrower. For small businesses, it involves industry catergory, geography, etc.
Therefore, the management of credit risk in the Indian banks is governed by a
credit policy approved by bank board. These credit policies and credit processes of
the Banks are prepared with the broad risk management objective. Hence, the
oversight of the credit policy remains with the senior management. Banks generate
plethora of information on the cause of riskiness1, the severity and the risk mitigation
actions, both as an internal management requirement and also a regulatory
requirement. For retail exposures the credit limits are determined subjectively. Bank
periodically monitors the utilization for each single exposure and also for a group of
exposures. Therefore, as exemplified above, Indian banks does have a broader credit
policy framework in place. However, the efficacy of these processes needs to be
examined in the light of their impact on rising or falling rates of Non Performing
Assets (NPAs)2. The banks? non-performing loans, also otherwise known as credit
loss, assumes critical importance since it reflects on the asset quality, credit risk and
efficiency in the allocation of resources to productive sectors. The next section outlines
the performance of Indian banks over the time period and draws conclusions.
2. REGULATION & CREDIT QUALITY IN INDIA
As shown in Chart 1, the credit quality of the Banking assets in India have improved
over a period of 7 years starting 2002-2008. For few Banks such as Dena Bank, from
a very high of 26.5% to as low as 4.1% in 2007 and subsequently 3.8% in 2008. The
reasons for such a fall in the NPA rate could be attributed to the prudential norms
and adhering to the RBI guidelines on both identifying and managing NPAs.
Unfortunately, the bank wise NPA (%) by asset class is not available and could
throw a completely different picture. For example, even if the bank level overall
NPA rates have come down, however, in recent times for many Banks, for retail
assets such as revolving assets, auto loans, mortgages, etc, the increasing NPA rates
have been of concern to the Banks. The reason for such deteriorations can be seen
from Chart 1 which provides the progress in retail banking in India. A rising trend
is seen since 2006 onwards across all the four categories of from mere 10 crores to
100 cores which is a big rise.
The success of the real estate sector in India is not unknown and so does the
exposure of banks to these sectors. Hence, it is interesting to note that indian banks
are not as risk averse to lending. The average CDR (Credit to Deposit) confirms the
aggressive lending, CDR levels from around 67% to 80% for the industry. However,
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / !
Chart 1: Gross NPA (%) in Indian Banks 2002 to 2007
30.0
Gross NPA(%)
Allahabad Bank
Andhra Bank
Bank of Baroda
X Bank of Maharashtra
Bank of Maharashtra
+ Citi Bank
Corporation Bank
20.0
Dena Bank
Hdfc Bank
Indian Bank
Oriental Bank of Commerce
10.0
X
+
X
0.0
+
+
+
2002
2003
2005
2004
X
2006
2007
Source: Statistical Tables
Chart 2: Credit Growth by Loan Type 1996 to 2008; Statistical Tables
100,000,000
X Other Retail Loans
90,000,000
X
Loans for Housing
80,000,000
Consumer Durable Loans
70,000,000
60,000,000
X
50,000,000
40,000,000
X
30,000,000
20,000,000
10,000,000
X
Personal Loans
X
X
X
0
2002
2003
2004
2005
2006
2007
2008
compared to Banks abroad the CDR of Indian Banks are much lower. As shown in
the Table 1 here, for the top 17 bank (mentioned in the sources of the table), which
includes a major share of the banking business in India. The profile of these banks
during the period 2002-2007 shows a mean capital adequacy ratio (CAR) of 12%,
" / INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
Table 1
Credit Profile of 17 Scheduled Banks
Bank?s Variable
RESERVES & SURPLUS
Growth in Reserves & Surplus (%)
Total Deposits
Total Advances
Growth in Advances (%)
Credit Deposit Ratio
Growth in Credit Deposit Ratio (%)
Total Assets
Growth in Total Assets (%)
Total Expenditure with Provisions
Expense To Assets Ratio
Growth in Expense To Assets ratio
Expense To Advances Ratio
Growth in Expenses To Advances Ratio
Capital Adequacy Ratio
Mean
Standard Deviation
436,235
34%
6,488,157
4,103,898
585,523
49%
7,959,941
5,293,340
61%
9%
8,265,728
16%
25%
10,423,140
713,401
9%
-1%
19%
-7%
12
906,139
1%
36%
5%
33%
2
Source: Statistical Tables Relating to Banks in India: Data On, Allahabad Bank, Andhra Bank, Bank of Baroda,
Bank of India, Bank of Maharashtra, Citi Bank, Corporation Bank, Dena Bank, Hdfc Bank, Icici Bank, Indian
Bank, Oriental Bank of Commerce, Punjab National Bank, State Bank of India, Union Bank of India, United
Bank of India ,Vijaya Bank
asset growth rate of 10%, capital and reserves growth rate of 34%, credit to deposit
rate growth of 9%, expenses to assets growth ratio of (-1)% , expenses to advances
growth rate of 19%. All of the above parameters depicts a health growth in the
business of banking in India in the form of both credit and deposit expansion,
combined with reduction in expenses to improve efficiency in intermediation.
All of the above parameters depicts a health growth in the business of banking
in India in the form of both credit and deposit expansion, combined with reduction
in expenses to improve efficiency in intermediation.
Therefore, the problem of falling NPAs even while increase in the recent time
periods could be related to several internal and external factors confronting the
borrowers. In the next section, we revisit the credit literature in both Indian and
international context to find explanations on credit quality changes.
3. CREDIT RISK LITERATURE IN INDIAN AND GLOBAL BANKING
In India little research has been devoted to the issue of credit risk management and
credit quality models. Nachane V. M. (2005) examined the impact of capital regulations
on Indian banking Sector and has concluded that Indian Banks do respond positively
to a captive requirement regime by making provisions for more capital. Nachane?s
study on the risk weighted assets and capital reserves for banks in India establishes
a negative and significant effect also concluding that bigger banks increased their
CAR less than other banks. The study concluded that regulatory effects does impact
the bank level CAR. In the banking literature, the problem of NPAs has been revisited
in several theoretical and empirical studies. Reddy (2004) highlighted terms of credit
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / #
issues on Indian banks and suggested that increasing NPAs in banks could be due to
rational economic decision. Mohan (2003) emphasized on banks? lending terms and
mentioned ?lazy banking?. Rajaraman and Vasishtha (2002) in an empirical study
provided an evidence of significant vicariate relationship between an operating
inefficiency indicator and the problem loans of banks. Subsequently, Ranjan and
Dhal (2001) proposed a model of panel regression using the bank specific factors
and economy information to identify and interpret the change in both NPA and
NNPA. They attribute the changes in NPA to few of the bank groups such as public
sector banks, private sector and other banks, assuming similar behavior within a
bank group. The study focuses on good measures created summarizing various ratio
of the bank for a limited time period (1999-2002). Hence Ranjan (2005)?s work would
be the most comprehensives empirical model to explain NPA formation in Indian
banking.
In this context it will be worthwhile to mention few of the seminal work relating
to other economies. Sergio?s (1996) study of Italy, McGoven?s (1993) study on US
loan losses, Bercoff, Giovanniz and Grimardx?s (2002) study on Argentina, Fuentes
and Maquieira?s (1998) study on Chille, etc, are the significant worldwide studies
on NPAs. All of the above work points to the significance of the concept of bank
level parameters and credit portfolio, and terms of credit, particularly, cost conditions,
economy, etc.
4. MODEL OF CREDIT QUALITY
Ranjan and Dhal (2003) proposed model of panel regression using the bank specific
factors and economy information to identify and interpret the change in both GNPA
and NNPA. Their factors include bank groups such as public sector banks, private
sector and other banks, assuming similar behavior within a bank group and also
measures (financial ratios) summarizing credit characteristics of the banks.
We propose to simplify the Ranjan?s (2005) model for credit quality into a model
of the following form:
NPAi,t = Function (CARI,t, GDPt, Maturity, Expenses to Total Assetsi, SENSEXt)
Where
CARI,t is the capital adequacy of the Bank ?I? in time?t?.
NPAi,t is defined as ith bank?s gross non-performing assets to gross advances or
net non-performing assets to net advances in period t.
GDPt captures the growth rate of aggregate economic activity in time ?t? (GDP at
factor Cost).
CDRI,t the credit to deposit ratio of a bank in time ?t?.
Expenses to Total Assetsi is the ratio of total expenses (inclusive of Provisions) to
the Total Assets of the Bank ?I? in time ?t?.
SENSEXt is the annualized return on Sensex in time ?t?.
We apply this approach for 17 of the top Banks in India for the time period 20022007 using linear regression approach. The bank specific fixed factors are assumed
unimportant because a single regulator RBI controls the banking policy of the economy
$ / INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
which are complied with by all banks in India. The bank level factors as well as GDP
is also considered in this approach. We consider the absolute measure, growth
measure and lagged measure for each of the variables mentioned in this model.
5. MODEL RESULTS
The regression methodology recognizes individual bank level ratios (characteristics)
in order to establish a meaningful relationship between different economic and
financial variables. Since the emphasis of the study is on analysis of borrower?s loan
repayment response to terms of credit, the appropriate approach entails an empirical
evaluation of the ratio of NPAs to advances rather than NPAs to assets ratio. We
analyse two regression models for NPA and NNPA.
The estimation of both the GNPA and NNPA model are summarized in Table 2
and Table 3 above. For both the model, the adjusted R2 is greater than 0.4, which
demonstrate a good fit given the ordinary linear regression model over a 8 year
recent data period for 17 banks. This is because we assume that the bank specific
fixed effects are absent, which means there is no intercept component for the RBI
categories of banks such as Foreign Banks, PSBs, private banks etc. The seemingly
unrelated regression between various banks such as the rise or fall in NPA of a
given bank may affect the rise or fall in NPA of a different bank may be in the same
time period or a different time period, could not be detected here. Therefore, the
bank specific factors are explained with the help of bank level parameter such as
CAR, Expense Ratio, etc. Both the NPA and NNPA model selects only two bank
factors and one economic factor called CAR, CDR and DP growth in Constant Prices.
As shown in Table 3, and Table 4, both CAR and CDR are negatively related which
is in line with theory. A 0.73 percent rise in a Bank?s CAR (could due to a regulatory
exercise), can impact a drop in GNPA over 1 percent. Similarly a 10.5 percent rise in
CDR can impact a drop in GNPA over 1 percent. Thus healthier banks with an
oversight of the effective implementation of credit policies by their senior
management are less likely to have higher NPA. Similarly, a better credit growth
culture and customer orientation of the bank which is measured in higher CDR
actually results in lower GNPAs. This phenomenon has also been established by
Ranjan and Dhal (2003). In Table 4, we find a 0.44 percent rise in CAR and a 4
percent rise in CDR can cause a percent drop in NNPA. The difference in the
parameter estimates between the two models are explained by the fact that GNPA
model explains borrower behavior where as NNPA model also explains the loan
loss provisioning by the lender on the top of the borrower behavior. The interesting
aspect of the results is the fact that GDP Growth plays a significant role in lowering
of both GNPAs and NNPAs. However, the impact of GDP growth rate to GNPA is
much smaller even if it is significant. A 92 percent increase in GDP can cause a
percent fall in GNPA and a 59 percent rise in GDP can cause a percent fall in NNPA.
6. CONCLUSIONS AND POLICY IMPLICATIONS
We proposed a simple regression model for understating the credit quality changes
in the Indian economy and the resulting impact of both bank level and economic
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / %
Table 2
Gross NPA Model Results
Model Parameters Estimates
Variable
Estimate
Intercept
Capital Adequacy Ratio
Credit Deposit Ratio
GDP Growth at Constant Prices
29.43
-0.73
-10.58
-92.08
R-Square
t-Value
11.03
-4.42
-4.64
-5.04
Root Mean Square Error
Adjusted R-Square
0.47
P Value
© International Science Press
BANK REGULATION AND CREDIT QUALITY IN
INDIAN BANKING: A QUANTITATIVE EVALUATION
HOSAMANE M.D.
DOS in Economics & Co-Operation, University of Mysore, Manasagangotri, Mysore
DINABANDHU BAG
Research Scholar, DOS in Economics & Co-Operation, University of Mysore, Manasagangotri, Mysore
There exists meaningful inter-linkages among economic and financial variables
such that the variation in the credit quality at the macro level can be explained. Both
the economic factors and also the bank level factors play a critical role in determining
the credit quality of assets. This study undertakes an empirical analysis for finding
the impact of economic and financial factors on banks? non-performing loans. It
provides a framework for analysis of underlying default behavior of borrowers? at
the bank level in terms of banks? individual characteristics and also due to the
presence of other macroeconomic indicators. We conclude that a positive outlook on
economic growth on banks would favour loan repayment response of borrowers in
order to maintain credit worthiness and credit quality. Rising capital adequacy
ratio and higher credit deposit ratio can jointly help improve the portfolio credit
quality. The results of the study are in line with banking literature and provide an
important insight for banks? lending behavior.
Keywords: Credit Quality, NPA, Factors, Credit Deposit Ratio
1. INTRODUCTION
Credit risk is the risk of loss that may occur due to the failure of any counterparty to
abide by the contract with the Bank, principally the failure to make required payments
as per the contract. The purpose of this study is to provide meaningful analysis of
inter-linkages among economic and financial variables such that the variation in the
credit quality at the macro level can be explained. The credit quality scenario has
undergone significant changes in the last decade amidst reform of the financial sector
by rapid increase in globalization and advances in information technology. Overall,
these developments have led to structural change in the financial sector, which has
created conducive environment for market mechanism, in general, and economic
factors, in particular, for playing a critical role in influencing the portfolios of banks
and financial institutions. It is in this context that this study has considered it
imperative to undertake an empirical analysis for evaluating the impact of economic
and financial factors on banks? non-performing loans. The distinguishing feature of
/ INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
the study is that it provides a framework for analysis of underlying default behavior
of borrowers? at the bank level in terms of banks? individual characteristics and also
due to the presence of other macroeconomic indicators. Are the credit policies of
Indian banks adequate to maintain the credit quality of their portfolios? Do they
have comprehensive risk management strategies in place? To answer such questions
we highlight the risk policy of a leading Indian bank in the next section.
A leading bank called ICICI bank is being referred to in this context. The credit
policy in ICICI bank, vary across product groups and they include drivers such as
income of the borrower, loan to value ratio, demographic and stability of the
borrower. For small businesses, it involves industry catergory, geography, etc.
Therefore, the management of credit risk in the Indian banks is governed by a
credit policy approved by bank board. These credit policies and credit processes of
the Banks are prepared with the broad risk management objective. Hence, the
oversight of the credit policy remains with the senior management. Banks generate
plethora of information on the cause of riskiness1, the severity and the risk mitigation
actions, both as an internal management requirement and also a regulatory
requirement. For retail exposures the credit limits are determined subjectively. Bank
periodically monitors the utilization for each single exposure and also for a group of
exposures. Therefore, as exemplified above, Indian banks does have a broader credit
policy framework in place. However, the efficacy of these processes needs to be
examined in the light of their impact on rising or falling rates of Non Performing
Assets (NPAs)2. The banks? non-performing loans, also otherwise known as credit
loss, assumes critical importance since it reflects on the asset quality, credit risk and
efficiency in the allocation of resources to productive sectors. The next section outlines
the performance of Indian banks over the time period and draws conclusions.
2. REGULATION & CREDIT QUALITY IN INDIA
As shown in Chart 1, the credit quality of the Banking assets in India have improved
over a period of 7 years starting 2002-2008. For few Banks such as Dena Bank, from
a very high of 26.5% to as low as 4.1% in 2007 and subsequently 3.8% in 2008. The
reasons for such a fall in the NPA rate could be attributed to the prudential norms
and adhering to the RBI guidelines on both identifying and managing NPAs.
Unfortunately, the bank wise NPA (%) by asset class is not available and could
throw a completely different picture. For example, even if the bank level overall
NPA rates have come down, however, in recent times for many Banks, for retail
assets such as revolving assets, auto loans, mortgages, etc, the increasing NPA rates
have been of concern to the Banks. The reason for such deteriorations can be seen
from Chart 1 which provides the progress in retail banking in India. A rising trend
is seen since 2006 onwards across all the four categories of from mere 10 crores to
100 cores which is a big rise.
The success of the real estate sector in India is not unknown and so does the
exposure of banks to these sectors. Hence, it is interesting to note that indian banks
are not as risk averse to lending. The average CDR (Credit to Deposit) confirms the
aggressive lending, CDR levels from around 67% to 80% for the industry. However,
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / !
Chart 1: Gross NPA (%) in Indian Banks 2002 to 2007
30.0
Gross NPA(%)
Allahabad Bank
Andhra Bank
Bank of Baroda
X Bank of Maharashtra
Bank of Maharashtra
+ Citi Bank
Corporation Bank
20.0
Dena Bank
Hdfc Bank
Indian Bank
Oriental Bank of Commerce
10.0
X
+
X
0.0
+
+
+
2002
2003
2005
2004
X
2006
2007
Source: Statistical Tables
Chart 2: Credit Growth by Loan Type 1996 to 2008; Statistical Tables
100,000,000
X Other Retail Loans
90,000,000
X
Loans for Housing
80,000,000
Consumer Durable Loans
70,000,000
60,000,000
X
50,000,000
40,000,000
X
30,000,000
20,000,000
10,000,000
X
Personal Loans
X
X
X
0
2002
2003
2004
2005
2006
2007
2008
compared to Banks abroad the CDR of Indian Banks are much lower. As shown in
the Table 1 here, for the top 17 bank (mentioned in the sources of the table), which
includes a major share of the banking business in India. The profile of these banks
during the period 2002-2007 shows a mean capital adequacy ratio (CAR) of 12%,
" / INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
Table 1
Credit Profile of 17 Scheduled Banks
Bank?s Variable
RESERVES & SURPLUS
Growth in Reserves & Surplus (%)
Total Deposits
Total Advances
Growth in Advances (%)
Credit Deposit Ratio
Growth in Credit Deposit Ratio (%)
Total Assets
Growth in Total Assets (%)
Total Expenditure with Provisions
Expense To Assets Ratio
Growth in Expense To Assets ratio
Expense To Advances Ratio
Growth in Expenses To Advances Ratio
Capital Adequacy Ratio
Mean
Standard Deviation
436,235
34%
6,488,157
4,103,898
585,523
49%
7,959,941
5,293,340
61%
9%
8,265,728
16%
25%
10,423,140
713,401
9%
-1%
19%
-7%
12
906,139
1%
36%
5%
33%
2
Source: Statistical Tables Relating to Banks in India: Data On, Allahabad Bank, Andhra Bank, Bank of Baroda,
Bank of India, Bank of Maharashtra, Citi Bank, Corporation Bank, Dena Bank, Hdfc Bank, Icici Bank, Indian
Bank, Oriental Bank of Commerce, Punjab National Bank, State Bank of India, Union Bank of India, United
Bank of India ,Vijaya Bank
asset growth rate of 10%, capital and reserves growth rate of 34%, credit to deposit
rate growth of 9%, expenses to assets growth ratio of (-1)% , expenses to advances
growth rate of 19%. All of the above parameters depicts a health growth in the
business of banking in India in the form of both credit and deposit expansion,
combined with reduction in expenses to improve efficiency in intermediation.
All of the above parameters depicts a health growth in the business of banking
in India in the form of both credit and deposit expansion, combined with reduction
in expenses to improve efficiency in intermediation.
Therefore, the problem of falling NPAs even while increase in the recent time
periods could be related to several internal and external factors confronting the
borrowers. In the next section, we revisit the credit literature in both Indian and
international context to find explanations on credit quality changes.
3. CREDIT RISK LITERATURE IN INDIAN AND GLOBAL BANKING
In India little research has been devoted to the issue of credit risk management and
credit quality models. Nachane V. M. (2005) examined the impact of capital regulations
on Indian banking Sector and has concluded that Indian Banks do respond positively
to a captive requirement regime by making provisions for more capital. Nachane?s
study on the risk weighted assets and capital reserves for banks in India establishes
a negative and significant effect also concluding that bigger banks increased their
CAR less than other banks. The study concluded that regulatory effects does impact
the bank level CAR. In the banking literature, the problem of NPAs has been revisited
in several theoretical and empirical studies. Reddy (2004) highlighted terms of credit
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / #
issues on Indian banks and suggested that increasing NPAs in banks could be due to
rational economic decision. Mohan (2003) emphasized on banks? lending terms and
mentioned ?lazy banking?. Rajaraman and Vasishtha (2002) in an empirical study
provided an evidence of significant vicariate relationship between an operating
inefficiency indicator and the problem loans of banks. Subsequently, Ranjan and
Dhal (2001) proposed a model of panel regression using the bank specific factors
and economy information to identify and interpret the change in both NPA and
NNPA. They attribute the changes in NPA to few of the bank groups such as public
sector banks, private sector and other banks, assuming similar behavior within a
bank group. The study focuses on good measures created summarizing various ratio
of the bank for a limited time period (1999-2002). Hence Ranjan (2005)?s work would
be the most comprehensives empirical model to explain NPA formation in Indian
banking.
In this context it will be worthwhile to mention few of the seminal work relating
to other economies. Sergio?s (1996) study of Italy, McGoven?s (1993) study on US
loan losses, Bercoff, Giovanniz and Grimardx?s (2002) study on Argentina, Fuentes
and Maquieira?s (1998) study on Chille, etc, are the significant worldwide studies
on NPAs. All of the above work points to the significance of the concept of bank
level parameters and credit portfolio, and terms of credit, particularly, cost conditions,
economy, etc.
4. MODEL OF CREDIT QUALITY
Ranjan and Dhal (2003) proposed model of panel regression using the bank specific
factors and economy information to identify and interpret the change in both GNPA
and NNPA. Their factors include bank groups such as public sector banks, private
sector and other banks, assuming similar behavior within a bank group and also
measures (financial ratios) summarizing credit characteristics of the banks.
We propose to simplify the Ranjan?s (2005) model for credit quality into a model
of the following form:
NPAi,t = Function (CARI,t, GDPt, Maturity, Expenses to Total Assetsi, SENSEXt)
Where
CARI,t is the capital adequacy of the Bank ?I? in time?t?.
NPAi,t is defined as ith bank?s gross non-performing assets to gross advances or
net non-performing assets to net advances in period t.
GDPt captures the growth rate of aggregate economic activity in time ?t? (GDP at
factor Cost).
CDRI,t the credit to deposit ratio of a bank in time ?t?.
Expenses to Total Assetsi is the ratio of total expenses (inclusive of Provisions) to
the Total Assets of the Bank ?I? in time ?t?.
SENSEXt is the annualized return on Sensex in time ?t?.
We apply this approach for 17 of the top Banks in India for the time period 20022007 using linear regression approach. The bank specific fixed factors are assumed
unimportant because a single regulator RBI controls the banking policy of the economy
$ / INTERNATIONAL JOURNAL
OF
ECONOMIC ISSUES
which are complied with by all banks in India. The bank level factors as well as GDP
is also considered in this approach. We consider the absolute measure, growth
measure and lagged measure for each of the variables mentioned in this model.
5. MODEL RESULTS
The regression methodology recognizes individual bank level ratios (characteristics)
in order to establish a meaningful relationship between different economic and
financial variables. Since the emphasis of the study is on analysis of borrower?s loan
repayment response to terms of credit, the appropriate approach entails an empirical
evaluation of the ratio of NPAs to advances rather than NPAs to assets ratio. We
analyse two regression models for NPA and NNPA.
The estimation of both the GNPA and NNPA model are summarized in Table 2
and Table 3 above. For both the model, the adjusted R2 is greater than 0.4, which
demonstrate a good fit given the ordinary linear regression model over a 8 year
recent data period for 17 banks. This is because we assume that the bank specific
fixed effects are absent, which means there is no intercept component for the RBI
categories of banks such as Foreign Banks, PSBs, private banks etc. The seemingly
unrelated regression between various banks such as the rise or fall in NPA of a
given bank may affect the rise or fall in NPA of a different bank may be in the same
time period or a different time period, could not be detected here. Therefore, the
bank specific factors are explained with the help of bank level parameter such as
CAR, Expense Ratio, etc. Both the NPA and NNPA model selects only two bank
factors and one economic factor called CAR, CDR and DP growth in Constant Prices.
As shown in Table 3, and Table 4, both CAR and CDR are negatively related which
is in line with theory. A 0.73 percent rise in a Bank?s CAR (could due to a regulatory
exercise), can impact a drop in GNPA over 1 percent. Similarly a 10.5 percent rise in
CDR can impact a drop in GNPA over 1 percent. Thus healthier banks with an
oversight of the effective implementation of credit policies by their senior
management are less likely to have higher NPA. Similarly, a better credit growth
culture and customer orientation of the bank which is measured in higher CDR
actually results in lower GNPAs. This phenomenon has also been established by
Ranjan and Dhal (2003). In Table 4, we find a 0.44 percent rise in CAR and a 4
percent rise in CDR can cause a percent drop in NNPA. The difference in the
parameter estimates between the two models are explained by the fact that GNPA
model explains borrower behavior where as NNPA model also explains the loan
loss provisioning by the lender on the top of the borrower behavior. The interesting
aspect of the results is the fact that GDP Growth plays a significant role in lowering
of both GNPAs and NNPAs. However, the impact of GDP growth rate to GNPA is
much smaller even if it is significant. A 92 percent increase in GDP can cause a
percent fall in GNPA and a 59 percent rise in GDP can cause a percent fall in NNPA.
6. CONCLUSIONS AND POLICY IMPLICATIONS
We proposed a simple regression model for understating the credit quality changes
in the Indian economy and the resulting impact of both bank level and economic
BANK REGULATION AND CREDIT QUALITY
IN
INDIAN BANKING: A QUALITATIVE EVALUATION / %
Table 2
Gross NPA Model Results
Model Parameters Estimates
Variable
Estimate
Intercept
Capital Adequacy Ratio
Credit Deposit Ratio
GDP Growth at Constant Prices
29.43
-0.73
-10.58
-92.08
R-Square
t-Value
11.03
-4.42
-4.64
-5.04
Root Mean Square Error
Adjusted R-Square
0.47
P Value