REVIEW OF ECONOMIC PERSPECTIVES – NÁRODOHOSPODÁŘSKÝ OBZOR,
VOL. 12, ISSUE 4, 2012, pp. 235–264 , DOI: 10.2478/v10135-012-0011-9
Structure, Conduct and Performance of Indian
Banking Sector
Deepti Sahoo1
and Pulak Mishra2
Abstract: In the context of initiation of economic reforms in general and changes in
policies and regulations of the banking sector in particular, the present paper attempts to
examine the structure-conduct-performance relationships in Indian banking sector. It is
observed that there have been changes in the market structure of Indian banking sector,
conducts of the banks and their performance in the post-reform era, especially during
the last decade, though the changes have not been significant in every aspect. Using a
panel dataset of 59 banks operating in India during 1999-2000 to 2008-2009 and
applying the two-stage least squares (2SLS) method of estimation, the paper finds that
there exist strong inter-linkages amongst structure of the market, banks’ conduct and
their financial performance. While market share of a bank depends directly on its
market size, asset base, selling efforts, and past financial performance, its selling efforts
vary directly with market share, asset base, and past financial performance. On the other
hand, returns on assets of a bank vary directly with its market share, but inversely with
its asset base and selling efforts. The regression results essentially suggest for
multidirectional and dynamic SCP relationships in Indian banking sector. It is also
found that the nature of ownership has significant influence on market share, selling
efforts and financial performance of the banks. As compared to the nationalised banks,
market share of the private banks (both domestic and foreign) is found to be lower. But
private banks make greater selling efforts and have better financial performance vis-àvis
their public sector counterparts.
Key words: Banking sector, market structure, conduct, performance, policy, India
JEL Classification: C33, G18, G21, L190
Introduction
The Indian banking sector has taken a new dimension with initiation of economic
reforms in 1991 in general, and changes in sector specific policies and regulations since
the mid-1990s in particular. The basic objectives of these measures were to make Indian
banking sector internationally competitive, especially in respect of capital adequacy and
other prudential norms. Accordingly, emphasis has been given on improving
productivity, efficiency, technology, and profitability of the banks. Given that achieving
greater efficiency and competitiveness requires enhanced market competition, a number
of deregulatory policy measures have been introduced. Introduction of these
1
Faculty in Economics, Department of Humanities and Social Sciences, RGUKT, IIIT Nuzvid,
India, Email: sahood9@gmail.com
2
Associate Professor in Economics, Department of Humanities and Social Sciences, Indian
Institute of Technology Kharagpur, India, Email: pmishra@hss.iitkgp.ernet.in
REVIEW OF ECONOMIC PERSPECTIVES
236
deregulatory polices is largely based on the assumption that efficiency of banks
increases in a competitive environment.
However, instead of drastic privatization of the existing public sector banks, Indian
government has chosen a gradual approach towards their restructuring to enhance
competitiveness. What is more important, perhaps, is that with deregulation leading to
competitive business environment in the sector, the banks have also applied a variety of
strategies to cope with the emerging market conditions. The major strategies applied by
the banks under the new business conditions include differentiating products/services
and creating strategic entry barriers through advertising, widening customer base
through promotional activities, diversification of product/service portfolio to reduce
risks of operation, mergers and acquisitions to consolidate business and enhance
efficiency, and floating joint ventures with foreign banks to bring in better management
and enhance capital base.
Thus, reforms in Indian banking sector have resulted in two opposite forces. While the
policy and regulatory changes are expected to bring in greater competition in the market
and, thereby, to enhance efficiency of the banks, the strategic responses of the banks are
likely to limit market competition. This raises some important issues. How have the
policies of the government and strategic responses of the banks affected structure of the
market in Indian banking sector? What are the factors that determine strategies of the
banks? How have the domestic players performed vis-à-vis their foreign counterparts
under the new business conditions? What regulatory role should the Reserve Bank of
India (RBI) play to ensure greater market competition and better efficiency of the banks?
The rationale for raising these issues lies in the structure-conduct-performance (SCP)
relationships, propounded initially by Mason (1939) and modified subsequently by Bain
(1959). The traditional SCP framework suggests that the possibility of collusive
behaviour increases when the market is concentrated in the hands of a few firms, and
the higher the market concentration, the larger will be the profitability of the firms. In
other words, there is a positive correlation between market concentration and
performance of firms. Many studies have attempted to test validity of the basic
proposition of the traditional SCP paradigm that the market concentration lowers the
cost of collusion between firms, and results in higher than normal profits. There are a
numerous studies (e.g. Weiss, 1974) that have found a positive relationship between
market structure and performance of firms.
The SCP relationships in banking sector are well explored in the literature. The existing
studies in general have attempted to test the hypothesis that market concentration
significantly influences performance of banks. A large number of these studies find
evidence in support of the hypothesis that market concentration enhances banks’
performance (Edward, 1964; Phillips, 1967; Brucker, 1970; Vernon, 1971; Gilbert,
1984; Podenda, 1986; Evanoff and Fortier, 1988; Molyneux and Thornton, 1992;
Lloyad-Williams et al., 1994; Molyneux and Forbes, 1995; Katib, 2004). These studies
in general show that market concentration in the sector affects the price which
consumers pay for banking services and this in turn influences their financial
performance. However, contrary to this observation, there are also studies (e.g.,
Smirlock, 1985; Miller and VanHoose, 1993) that either do not support or reject the
hypothesis that market concentration has a positive impact on performance of banks.
Volume 12, Issue 4, 2012
237
Similarly, in Indian context, the studies by Bhattacharya and Das (2003), Sathye and
Sathye (2004), and Varma and Sainir (2010) do not find evidence in support of this
traditional SCP hypothesis.
The inconsistencies and contradictions in the SCP relationships in the literature are
generally contributed to the methodology and data, and the assumptions relating to
measures of market structure (Osborne and Wendel, 1983; Gilbert, 1984). Further,
interpreting higher profits in concentrated markets as evidence of market power is also
problematic as market concentration may not necessarily be a reflection of collusive
behaviour of the banks, rather a consequence of their superior efficiency (Demsetz,
1974; Smirlock, 1985; and Bresnahan, 1989). There are studies in Indian context as well
(e.g., Sathye and Sathye, 2004; Varma and Sainir, 2010) which have found efficiency as
an important determinant of banks’ performance. This means that, in addition to use of
appropriate methodology and consistent data, a proper understanding of the SCP
relationships requires simultaneous consideration of both market power and efficiency.
But, this aspect has remained largely unexplored in the existing studies.
Second, although the literature on competition is based on two predominant views, viz.,
the static and the dynamic view, with innovations resulting in new products and new
markets, disequilibrium in the existing markets is very likely in the long run,
particularly when entry and exit are free (Mueller, 1990). This makes the players
engaged in dynamic competitive process through various business strategies to gain
advantage over the rivals (Kirzner, 1973; Shackle, 1971). Hence, competition should be
viewed as a dynamic process of rivalry (Vickers, 1995). But, the existing studies have
largely focused on static analysis, and thereby have failed to incorporate variations in
bank and service specific competencies, growth of banks and the sector, and entry into
and exit from the sector. This is very important as the dynamics of competition in Indian
banking sector are expected to be influenced by structure of the market, conducts of the
banks and their performance (Bhanumurthy and Dev, 2008).
Hence, the SCP relationships in the context of banking sector should be explored in a
multidirectional and dynamic framework. The present paper is an attempt in this
direction. Such an effort is very important as there is no consensus in the literature about
the possible impact of deregulation on efficiency and financial performance of the banks.
While the proponents of the beneficial role of financial services in a free economy (Cho,
1986) tend to undermine the scope of unruly behaviour, there is also every possibility of
over speculation, financial crisis and misallocation of savings and investment to the
detriment of growth and stability (Singh, 1993; Grabel 1995). Experiences of India and
other developing countries largely support the second view (Ram Mohan, 2001). The
rest of the paper is organized in the following way: The second section discusses the
state of market structure, conduct of banks and their performance in the last decade. The
multidirectional and dynamic SCP relationships in the sector are specified in the third
section. The fourth section of the paper describes the estimation techniques applied and
sources of data used, whereas the fifth presents and discusses the empirical findings.
The sixth section sums up the major findings and concludes the paper.
REVIEW OF ECONOMIC PERSPECTIVES
238
II. SCP in Indian Banking Sector: An Overview
As mentioned above, the traditional SCP framework postulates a unidirectional
relationship between market structure, conduct and performance with structure of the
market (concentration, conditions on entry, etc.) influencing performance of the firms
(profits, growth, etc.) via their conducts (price and non-price behaviour). It is observed
that higher market concentration results in higher prices and hence in higher financial
returns. A number of studies (e.g., Berger and Hannan, 1989; Hannan, 1992; Neumark
and Sharpe, 1992; Okeahalam, 1998; Prager and Hannan, 1998) support this traditional
SCP relationship.
However, there are a number of studies that cast doubt on the unidirectional SCP
relationship. According to Demsetz (1973) and Berger (1995), when larger market
shares are results of better efficiency and lower costs, adverse impact on welfare does
not arise. On the other hand, when higher market concentration raises prices, quantity
demanded declines and hence profits may not necessarily increase (Mullineux and
Sinclair, 2000).3
Successive developments in the industrial organization literature have
addressed these aspects in a multidirectional structure-conduct-performance-policy
framework (e.g., Scherer and Ross, 1990). In the modified SCP framework, interdependencies
amongst market structure, conducts of banks and their performance make
the variables endogenous in nature (Scherer and Ross, 1990; Neuberger, 1997).
Inclusion of public policy and feedback effects make the framework interlinked further.
Further, given that the banks are special players in the market for information, market
conditions are very important for determination of their relative position in the market,
strategic behaviour and performance. Therefore, incorporation of different aspects of
incomplete information is necessary for better understanding of the SCP relationships in
the banking sector (Swank 1996, Thakor 1995, Neuberger 1997). Integration of such
market imperfections arising from market uncertainties, asymmetric information,
transaction costs and information costs into the basic market conditions, and functioning
of the market seem to have a significant impact on structure of the banking sector,
strategic behaviour of the banks and their performance. The modified SCP framework
in the context of banking sector, as suggested by Neuberger (1997), is shown in Figure
1. Here, the basic market conditions include risks, attitudes towards risks and principalagent-relationships.
These basic conditions affect market structure, conduct,
performance, and public policies against market failures. Given this backdrop, what
follows next is an attempt towards empirical understanding of the SCP relationships in
Indian banking sector.4
3
The findings of Shaffer (1989) also do not support the basic proposition of the traditional SCP
framework.
4
Due to non-availability of necessary data, the present paper fails to capture the impact
information asymmetry on basic conditions of the market. However, here, we capture the
importance of transaction costs in terms of cost and operational efficiency of the banks.
Volume 12, Issue 4, 2012
239
Figure 1: SCP Framework for Banking Sector
Source: Neuberger (1994)
Conduct
Advertising and marketing
Price competition
Network and quality competition
Price discrimination
Mergers
Information gathering expenses, preference
behavior and risk avoidance
Diversification Innovations
Market Structure
Market segmentation
Product differentiation
Diversification
Cost Structure
Barriers to entry
Performance
Productive and allocative efficiency
Return-cost relationship
Profitability
Public Policy
Anti-trust provision
Protective regulations
Prudential regulations
Competition policy
Basic Conditions
Uncertainty
Asymmetric Information
Transaction costs
Supply Demand
Product/Service durability Elasticity of demand
Value Availability of substitutes
Business attitude Consumer behaviour
REVIEW OF ECONOMIC PERSPECTIVES
240
Structure of the Market
Generally, the structure of a market is examined in terms of the degree of sellers’
concentration as it is an important feature of the extent of imperfect competition in the
market. The industrial organization literature suggests several measures of market
concentration to infer about the structure of a market, such as market shares of the firms,
n-firm concentration ratio, Herfindahl-Hirschman Index (HHI), price-cost margin,
profitability, etc.
5
In the present paper, four indices have been used namely, market
shares of the banks, n-bank concentration ratio, the HHI, and the entropy index to
measure the extent of market concentration in Indian banking sector. While market
share will signal position of an individual bank in the market, the 3-bank concentration
ratio, the HHI, and the entropy index will measure the degree of concentration in the
market as a whole.6
In addition, the present paper also examines size of a bank in terms
of its gross fixed assets in order to shed some light on the scale of operations, and the
demand for its products/services in terms of income to understand the basic market
conditions.
The trends in composition of the sector in terms of the number of banks in the Prowess
database of the Centre for Monitoring Indian Economy (CMIE) are presented in Table 1.
It is observed that private banks, especially the foreign ones, have a very strong
presence in Indian banking sector and in terms of their number, nationalized banks
account for only around one-fourth of the sector. While the number of nationalized bank
has declined only marginally after 2006, that of other types of banks have fluctuated
over the years. Table 2 shows the average market size of the banks and their average
asset base.7
Both the average market size and asset base of the banks have increased
during 2000-01 to 2009-2010. However, there have not been considerable variations in
these measures implying that the market size or the asset base has been consistent across
the banks.
5
In Indian context also, the existing studies have largely used these measures to examine
the state of market concentration. For example, while Kambhampati (1996) has assessed structure
of the market in Indian industry sector on the basis of n-firm concentration ratio, Ramaswamy
(2006), Mishra and Behera (2007), and Mishra (2008) have used the HHI for the same. On the
other hand, Basant and Morris (2000), and Mishra (2005) have used n-firm concentration ratio
and price-cost margin along with the HHI to substantiate the findings. Contrary to this,
Pushpangadan and Shanta (2008) have examined the dynamics of market competition in terms of
persistence of the profit rates.
6
In this context, it should be mentioned that the n-firm concentration ratio has a number of
limitations. The magnitude of the ratio is largely influenced by the choice of n and also by how a
market is defined, and the measure does not take into account the entire number and size
distribution of the firms. Despite these limitations, the present study uses this measure as it is the
most commonly used index of market concentration because of the ease of computation and
interpretation of the results. However, the HHI satisfies all the desirable properties by combining
both the number and size distribution of firms in the industry and is considered as a better
measure of market concentration. Further, by squaring market shares, the HHI weights more
heavily the values for large firms than for small ones. A simultaneous consideration of the two
indices helps in substantiating the findings.
7
In the present study, the natural logarithm of gross fixed assets of a bank is used as a measure of
its size (asset base), whereas the natural logarithm of income is considered as a proxy for its
market size.
Volume 12, Issue 4, 2012
241
The average market share of the banks has been very low during the period under
consideration (Table 3).8
However, the market share has differed considerably across
the banks though the extent of variations has declined over the years. The other indices,
viz., the 3-firm concentration ratio, the HHI, and the entropy index show that the extent
of market concentration in the sector has been reasonably low.9
These measures also
show decrease in market concentration during this period, though the decline has not
been significant.
Conduct of the Banks
Under imperfect competition, financial performance of a bank depends to a large extent
on its selling effort that includes advertising and marketing of products/services, and
development of service networks. Advertising influences financial performance of a
bank by differentiating its products/services from that of the rivals, creating entry
barriers, and enhancing image. Pursuing the consumers in favour of these differentiated
products/services makes the demand inelastic over the alternatives and hence results in
increased control of the bank over the market. Advertising also creates barriers to entry
of new banks as well as that to the upward mobility of the smaller banks. Advertising by
the entrants can help them to become recognized in the market, but intensive counter
advertising by the incumbents drowns out entrants’ images and, thereby, lessens the
volume of market they can capture. All these limit competition in the market and
thereby improve financial performance of the banks.10
On the other hand, expenditure by the banks towards building up marketing and
distribution related complementary assets improves financial performance in two ways.
First, such efforts raise competitiveness of the banks by developing marketing and
service networks, which, in turn, facilitates appropriateness of products/services and
enhances efficiency. Second, such assets increase bargaining power of the banks in
equity linked foreign collaborations as they have wider services networks as compared
to the foreign banks.
In the present paper, strategic behaviour of the banks is examined in terms of their
selling efforts. Here, selling intensity (i.e., the ratio of total selling expenses to income)
of a bank is used as a measure of its selling efforts. The selling efforts refer to expenses
by the bank towards advertising, promotion of services, and developing service
networks. Table 4 presents the selling efforts of the banks operating in India. It is
observed that the average selling intensity has been very low but has recorded
increasing trends over the years. This means that the banks operating in India have
increasingly used selling strategies to strengthen their position in the market. However,
selling efforts have varied considerably across the banks.
8
Here, market share of a bank is measured in terms of the ratio of its income to total income of all
the banks in the sector.
9
The HHI is measured as the sum of the squares of market shares of all the banks existing in the
sector. On the other hand, the entropy index measures market concentration in terms of weighted
average of market share with logarithm of reciprocal of market share itself being the weight.
10
However, advertising can also facilitate entry of new banks by helping them in making their
product/services known to the consumers quickly so that the concentration increasing effect can
be dissipated or even reversed.
REVIEW OF ECONOMIC PERSPECTIVES
242
Performance of the Banks
There are two broad strands of examining the financial performance of an enterprise,
viz., the stock market approach, and the profitability. The stock market approach applies
valuations in the stock market to determine the financial performance of an enterprise.
This approach is based on the assumption that the stock market is efficient and assesses
performance in terms of changes in share prices controlling for movements in the
market in general and the systematic risks in particular. However, the stock price
approach may suffer from the problem of undervaluation or overvaluation if the share
prices incorporate random valuation errors. When it is so, the changes in share prices
may largely be due to market correction. In this perspective, the profitability approach is
considered as much simpler measure of financial performance (Rhoades, 1985; Evanoff
and Fortier, 1988).11
The present paper uses the profitability approach to assess financial performance of the
banks during the post-reform period. Three alternative indices of profitability, viz., the
ratio of profit before interest and taxes (PBIT) to total income, the ratio of PBIT to gross
fixed assets, and the ratio of PBIT to capital employed have been computed to
substantiate the findings. While the ratio of PBIT to total income is termed as
profitability, the ratio of PBIT to gross fixed assets is referred as returns on assets. On
the other hand, the ratio of PBIT to capital employed is termed as the returns on capital
employed. In addition, the present paper also assesses operating efficiency of the banks
in terms of their cost efficiency and the current ratio.
It is observed that financial performances of the banks have fluctuated during 2000-01
to 2009-10 and the fluctuations are high in the case of ROCE (Table 5). Further, the
financial performance has varied across the banks and the difference in quite high in
case of ROCE. The same can be said in respect of the current ratio (i.e., the ratio of
current assets to current liabilities) as well (Table 6). As regards the costs efficiency, the
banks have shown signs of improvement over the years and the difference across the
banks has not been very high except that in 2006-07. However, total costs efficiency has
varied considerably across the banks (Table 6).12
The above discussions, therefore, show some interesting dynamics relating to market
structure of Indian banking sector, strategic behaviour of the banks and their financial
performance. While dominance of the private banks has continued in terms of their
number vis-à-vis the public sector banks, average size (asset base) of a bank and its
market size have shown increasing tendency during 2000-01 to 2009-10. However,
despite introduction of a number of liberal policy measures, extent of market
concentration in the sector shows only marginal decline. Under the new business
conditions the banks have increasingly relied on non-price competitive strategies like
selling efforts, but, financial performance or efficiency does not show much
11
However, the profitability approach itself has problems as companies can use creative
accounting techniques especially in respect of sales, assets, and profits and, therefore, the
published accounts may not be a true or fair reflection of their financial performance (Griffiths,
1986).
12
In the present study, cost intensity of a bank, i.e. the ratio of its total expenditure to income is
used as a measure of cost efficiency. It is assumed that higher cost intensity stands for lesser costs
efficiency of the bank.
Volume 12, Issue 4, 2012
243
improvement. This raises a few important issues. Why do the policy reforms fail to
enhance the extent of competition in Indian banking sector? How do market structure
and strategies affect business performance of the banks? Does control over the market
or business strategy or performance of a bank depend on its nature of ownership?
Answering these questions requires identifying the determinants of market structure of
the banking sector, business strategies of banks and their financial performance in a
multidirectional and dynamic SCP framework. The next section of the paper is an
attempt in this direction.
III. SCP Relationships in Indian Banking Sector: Model Specification
The existing literature on industrial organization (e.g., Bain, 1951; Gupta, 1983;
Schmalensee, 1989; Baker and Woodward, 1998) proposes various econometric
techniques to empirically analyze the SCP relationships. However, in the present paper,
following Khambampati (1996), simultaneous equation approach is applied. This
approach contains three different models assuming that each of structure (S), conduct (C)
and performance (P) is a function of the other two aspects13
, i.e.,
),(),,(),,( 321 CSfPPSfCPCfS ===
In the present paper, market share (SHRARE) is taken as a proxy for market structure,
selling intensity (SELL) for conduct, and returns on assets (ROA) for financial
performance (PER). While market share reflects both efficiency and market power
(Smirlock, 1985; Evanoff and Fortier, 1988; Frame and Kamerschen, 1997), selling
intensity can capture product differentiation with price competition (Hannan, 1991b;
Ciappori et al., 1995). On the other hand, return on assets is a much simpler and more
widely used measure of financial performance of a bank (Rhoades, 1985; Evanoff and
Fortier, 1988). However, market share of banks may be influenced by other structural
aspects of the market like their market size (MSZ) and asset base (BSZ), whereas,
returns on assets can be affected by their cost inefficiency (COST). Similarly, selling
efforts by the banks can be influenced by other business strategies as well. Hence, the
system of equations mentioned above can be rewritten as:
Here, S’ stands for set of variables relating to market structure other than market share
of the banks, C’ for conduct other than their selling efforts, and P’ for their performance
other than returns on assets. However, the functional relationships may not necessarily
be instantaneous in nature Kambhampati (1996). Instead, lagged relationships amongst
many of the constituent variables are very likely. For example, lagged conduct and
13
The ideal way of modeling structure-conduct-performance-policy relationships is to
estimate a system of at least four equations, one for each market structure, business
strategies, performance, and policies. But, due to lack of adequate information on policy
changes, in the present paper, the relationships are envisaged with three equations.
Further, the impact of any policy change is unlikely to be significantly different across
the banks. There is a number of studies (e.g., Strickland and Weiss, 1976; Martin, 1979a
and 1979b; Delorme et. al., 2002) that analyze the SCP relationships by using three
equations.
)',,(),,',(),,,'(1 PCSROAPCSSELLPCSSHARE φφφ ===
REVIEW OF ECONOMIC PERSPECTIVES
244
lagged performance can affect market concentration (Kambhampati, 1996). Similarly,
conduct of a bank is likely to be influenced by lagged performance. Accordingly,
appropriate lag structure is incorporated in the models. The lag structure can also
control the dynamics of the envisaged relationships in addition to reducing the problem
of endogeneity.
Determinants of Market Structure
We assume that market share of a bank (SHAREit) is a function of its current bank size
(BSZit), lagged market size (MSZi,t-1), lagged selling intensity (SELLi,t-1), lagged
returns on assets (ROAi,t-1), and nature of ownership, i.e.,
)1..(..........).........,,,,,( 211,1,1,1 DDROASELLMSZBSZfSHARE tititiitit −−−=
Here, MSZi,t-1 and BSZit are proxy for structural aspects of the market other than
market share, SELLi,t-1 for strategy, ROAi,t-1 for performance, and D1 and D2 for
nature of ownership of the banks. All banks are divided into three categories, viz.,
public sector banks, private domestic banks, and private foreign banks. The public
sector banks are considered to be the base category. As such, D1 = 1 for private
domestic banks and D1 = 0 otherwise, whereas D2 = 1 for private foreign banks and
D2 = 0 otherwise. Assuming the functional relationship to be linear, (1) can be rewritten
as,
)2(26
151,41,31,21
Lit
tititiitit
uD
DROASELLMSZBSZSHARE
++
++++++= −−−
β
βββββα
Possible Impact of the Independent Variables
Current Bank Size (BSZit): Size of a bank, measured in terms of its asset base, can be
seen as a proxy for the scale of operation. It is expected that larger banks operate at
higher scale. This helps the banks to reap the benefits of economies of scale, and
thereby to have greater share in the market. Hence, one may expect direct influence of
asset base on market share of a bank.
Lagged Market Size (MSZi,t-1): In the present paper, market size is used to proxy
demand for services of a bank, i.e., to indicate its absolute position in the sector. Given
the size of the sector, the banks with greater demand for their services are expected to
have larger share in the market. In other words, market share of a bank is likely to vary
directly with its market size. However, when market size of other banks also changes,
the relative position of a particular bank in the market may not necessarily alter.
Lagged Selling Intensity (SELLi,t-1): On the one hand, a bank disseminates information
on its services as well as creates image advantage and barriers to entry through
advertising. On the other hand, promotional efforts and widening of service networks
help the banks in creating necessary complementary assets. All these give the bank an
opportunity to raise its market share. Hence, market share of a bank is expected to vary
directly with its selling efforts.
Lagged Return on Assets (ROAi,t-1): In general, better financial performance raises the
ability and willingness of a bank to grow, and thereby can result in its higher market
Volume 12, Issue 4, 2012
245
share. On the other hand, in the absence of entry barriers, better financial performance
of the existing banks can encourage entry into the sector resulting in lower market
share14. Hence, the nature of impact of financial performance of a bank on its market
share depends on the relative strength of these diverse forces.
Nature of Ownership (D1 and D2): As private banks are relatively new players in the
sector and generally operate in the urban areas, they have lesser penetration in the
market. Hence, the private banks are likely to have lesser market share vis-à-vis the
public sector banks. In others words, one can expect the coefficient of D1 and D2 to be
negative.
Determinants of Conduct
It is assumed that selling efforts of a bank (SELLit) is a function of its lagged market
share (SHAREi,t-1), current asset base (BSZit), lagged financial performance (ROAi,t-
1), and the nature of their ownership (D1 and D2), i.e.,
)3.(.............................., 211,1,2 ),DD,ROA,BSZ(SHAREfSELL tiittiit −−=
Here, SHAREit, and BSZit control for structural aspects of the market, ROAi,t-1, for
banks’ performance, and D1 and D2 for their nature of ownership. Assuming the
functional relationship to be linear, (3) can be rewritten as,
)4.......(25141,321,1 ittiittiit vDDROABSZSHARESELL ++++++= −− βββββα
Possible Impact of the Independent Variables
Lagged Market Share (SHAREit): A higher market share of a bank may increase the
threats of new entry as well as strategic competition from existing rivals. When it is so,
the bank needs to protect its market share through selling related efforts. Hence, the
banks with higher market share are expected to have greater selling intensity.
Current Bank Size (BSZit): A larger bank has greater ability to spend on selling. A
larger bank may also become complacent and, hence, may have lesser willingness to
make selling related efforts. The nature of impact of bank size on selling intensity,
therefore, depends on how these diverse forces dominate each other.
Lagged Return on Assets (ROAi,t-1): On the one hand, better financial performance can
induce a bank to spend more on selling to create entry barriers or image advantage in
the market. On the other hand, better financial performance can make the bank
complacent and, hence, may restrict its selling related efforts. The nature of impact of
financial performance on selling intensity, therefore, depends on the relative strength of
these two opposite forces.
Nature of Ownership (D1 and D2): The private banks are considered to be more strategy
oriented to raise their dominance over the market and, thereby, to enhance financial
14
However, on many occasions, higher profitability of the incumbents can discourage entry of
new banks. This is so particularly when higher profitability is a result of greater market power or
larger scale of operation and the potential entrants do not have the capability to counter such
dominance.
REVIEW OF ECONOMIC PERSPECTIVES
246
performance. Hence, the private banks are likely to make greater selling related efforts
vis-à-vis the public sector banks. In other words, one may expect the coefficients of D1
and D2 to be positive.
Determinants of Performance
It is assumed that the returns on assets of a bank (ROAit) depend on its current market
share (SHAREit), current bank size (BSZit), lagged selling intensity (SELLit), current
cost intensity (COSTit), and the nature of their ownership (D1 and D2), i.e.,
( ) )5......(..........,,,,, 2113 DDCOSTSELLBSZSHAREfROA ititititit −=
Here, SHAREit, and BSZit control for structural aspects of the market, SELLit for
conduct of the banks, COSTit, for their performance other than returns on assets.
Assuming the functional relationship to be linear, (5) can be rewritten as
)6....(26
1541321
it
ititititit
D
DCOSTSELLBSZSHAREROA
εβ
βββββα
++
++++++= −
Possible Impact of the Independent Variables
Current Market Share (SHAREit): It is commonly perceived that higher market share
enables a bank to enhance financial performance through greater efficiency and/or
higher market power. However, when a bank strengthens its position in the market
through various incentives to the customers, financial performance may not necessarily
improve, especially in the short-run. The nature of impact of market share on financial
performance of a bank, therefore, depends on the relative strengths of these diverse
forces.
Current Bank Size (BSZit): A larger bank can operate at a higher scale and, therefore,
can reap the benefits of scale economies. This helps the bank enhance its financial
performance. However, if larger size of the bank results in X-inefficiency, its financial
performance may not necessarily improve. Hence, the nature of impact of asset base of
a bank on its financial performance depends on how these diverse forces dominate each
other.
Lagged Selling Intensity (SELLi,t-1): Product differentiation and image advantage
through advertising and creation of marketing and distribution related complementary
assets are expected to enhance profitability of a bank. In other words, higher selling
intensity of a bank is likely to enhance its financial performance.
Current Cost Intensity (COSTit): In general, cost intensity of a bank is likely to have a
negative impact on its financial performance. The banks with higher cost intensity are
expected to have lower returns on assets, particularly if it fails to cover the additional
costs through greater services charges.
Nature of Ownership (D1 and D2): It is generally perceived that private banks are more
efficient, and hence are likely to have better financial performance vis-à-vis the public
sector banks. However, since private banks are relatively new in the sector, they operate
at a smaller scale, which results in lower returns on assets as compared to the public
Volume 12, Issue 4, 2012
247
sector banks. The impact of nature of banks on their financial performance, therefore,
depends on how these diverse forces dominate each other.
IV. Estimation Techniques and Data
The equations specified above are estimated with a pooled dataset of 59 banks operating
in India over the period from 2000-01 to 2008-09. The sample includes 27 out of 28
public sector banks, 18 out of 22 foreign banks and 14 out of 16 private sector banks.
Hence, a total of 59 banks are included in the sample. Cooperative banks and a few
other banks are excluded due to non-availability of continuous data and on the basis of
their peculiar operational system, too.
When there is no simultaneity bias, the method of ordinary least squares (OLS) results
in consistent and efficient estimators. On the other hand, if there is simultaneity problem
in the envisaged relationships, the OLS estimators are biased and inconsistent. The three
equations specified above contain independent variables that are endogenous in nature.
Therefore, the system cannot be estimated consistently by applying the OLS method of
estimation. This is so because one of the crucial assumptions of the OLS method is that
the explanatory variables are either non-stochastic or, even if stochastic, they are
distributed independently of stochastic disturbance term. When neither of these
conditions is satisfied, the OLS estimators become biased and inconsistent. This means
that even if the sample size increases indefinitely, the estimates do not converge with
their true values (Gujarati and Sangeetha, 2007).
Since in simultaneous equation models the endogenous variable in one equation appears
as an explanatory variable in another equation of the system, this endogenous
explanatory variable becomes stochastic and is usually correlated with the random
disturbance term making the OLS estimators inconsistent (Gujarati and Sangeetha,
2007). Hence, in the present paper, the method of two-stage least square (2SLS) is
applied in order to deal with the simultaneity problem. In the presence of simultaneity
bias, the methods of two stage least squares (2SLS) and instrumental variables give
estimators that are consistent and efficient.
In the present paper, the endogenous variables include current market share, current
selling intensity, current returns on assets, current market size, current bank size, and
current costs intensity. On the other hand, the lagged market size, lagged market share,
lagged bank size, lagged selling intensity, lagged profitability, and nature of ownership
are treated as the exogenous variables.15
In the first stage, the endogenous explanatory
variables are regressed on all the exogenous variables included in the system. In the
second stage, the predicted values of the endogenous explanatory variables from the
first stage regression are used as their instruments for estimating the three structural
equations. The lagged structure of the equations is also expected to control endogeneity
problem of the independent variables. Further, all the variables included in the above
models are converted into their natural logarithms. This has two advantages. First, the
logarithmic transformation converts the individual slope coefficients into respective
elasticity. This helps in determining the relative importance of the independent variables,
15
In addition, the present paper also uses natural logarithm of the current ratio as additional
instrumental variable.
REVIEW OF ECONOMIC PERSPECTIVES
248
and thereby makes their effects comparable. Second, such an approach also reduces the
scale of measurement of the variables and hence the problem of heteroscedasticity.
Further, since the present paper uses a pooled dataset of 59 banks over a period of 7
years, the time-series component is less as compared to the cross sectional units. This
reduces the possibility of existence of unit roots. Despite that, the Levin-Lin-Chu (2002)
panel data unit root test is carried out to examine if the dataset used in regression
analysis suffers from the problem of non-stationarity. The panel data unit root test
emerges from the time-series unit root tests. However, unlike the time-series unit root
tests, the asymptotic behaviour of both the time-series dimension T and the crosssectional
dimension N are taken into consideration in the panel unit root tests. The
asymptotic behaviour of the estimators depends on the way in which N and T converge
to infinity. The Levin-Lin-Chu Test (2002) panel unit root test is used to test the null
hypothesis that all the panels contain unit root against the alternative hypothesis that
each time series is stationary. If the null hypothesis is rejected for a variable, it implies
that the concerned variable does not suffer from non-stationarity problem.
The present paper uses secondary data and the necessary data are collected from the
Prowess database of the Centre for Monitoring Indian Economy (CMIE), Mumbai. The
prowess database uses the initial sample from the database of the Reserve Bank of India
(RBI) and then augments the data. There is a number of advantages connected with
using this database since 1989-90. First, the Prowess database contains time-series
information relating to the finances of the banks and the ratios based on them at bank
level. The database is updated every day with regular intra-day and end-of-day updates.
Second, Prowess is a normalized database. During the course of normalization, the data
that come from heterogeneous sources are transformed into a standardized format to
enable meaningful comparisons over time and across the banks. Such normalization is a
necessary precondition in making the data comparable. Third, a merger is the most
ambiguous observation of a bank as it ceases to exist as a separate legal entity following
the integration. In such cases, the Prowess database retains all the information available
regarding the bank till it was merged.
Necessary adjustments are made to make the dataset consistent. A three-year moving
average is taken for each of the variables to control for the measurement errors and the
process of adjustment. Accordingly, the variables are measured as simple average of the
values of the previous three years with the year under reference being the starting year.
Such approach of moving average with a lag of two years is expected to reduce the
potential simultaneity bias further in the envisaged relationships.
V. Results and Discussions
Table 7 shows the summary statistics of the variables included in the regression models.
Further, Levin-Lin-Chu (2002) panel data unit root test is carried out, and the results of
the test are presented in Table 8. It is observed that the adjusted t statistic is statically
significant at 1 percent for all the variables. This means that none of the variables
included in the regression analysis suffers from the problem of unit-root, i.e., nonstationarity.
The regression results show that the F-statistic is statistically significant for
all of the three estimated equations (Table 9). This means that all the estimated
equations are statistically significant. Furthermore, the value of adjusted the R2 is
Volume 12, Issue 4, 2012
249
reasonably high for the equation on market structure and conduct. However, it is very
low for the equation on performance, but this does not necessarily indicate that the
estimated models are not acceptable.16
As regards the individual coefficients, the tstatistics
are computed by using robust standard errors to control for the problem of
heteroscedasticity further. In addition, presence of multicollinearity, if any, is examined
in terms of the variance inflation factors (VIFs). It is found that the VIFs of all the
independent variables are less than 10. Hence, following Marquardt (1970) and Mason
et al. (1989), it can be said the estimated models do not suffer from the problem of
severe multicollinearity.17
In the estimated equation on the determinants of market structure, it is observed that the
coefficients of BSZit, SELLi,t-1, ROAit, D1 and D2 are statistically significant. While
the coefficients of BSZit, SELLi,t-1 and ROAi,t-1 are positive, that of D1 and D2 are
negative. This means that the banks with larger asset base, greater selling efforts and
better financial performance in the past have larger share in the market. On the other
hand, the negative coefficients of D1 and D2 suggest that the private domestic banks as
well as the private foreign banks have lower market share vis-à-vis their public sector
counterparts.
The direct relationships are well expected. The banks that are larger in size can reap the
benefits of scale economies and hence can raise their market share. On the other hand,
greater selling efforts help the banks in raising their market shares by creating image
advantage and barriers to entry. A positive association between bank size and market
share is consistent with experiences from other sectors. For example, in the context of
Indian pharmaceutical industry, Mishra and Vikas (2010) found a positive impact of
asset base of the firms on their market shares, though the impact of selling efforts was
not statistically significant.
Similarly, the banks with better financial performance in the past have higher ability as
well as greater willingness to grow and raise market share. On the other hand, it is
generally perceived that better financial performance of the incumbents attract new
entrants into the market. But, in many cases, better financial performance of the
incumbents can also act as an entry barrier and may discourage the potential entrants
from entering into the market. This is so particularly when the potential entrants do not
have the capability to enhance their financial performance to the level of that of the
incumbents. When it is so, better financial performance of the incumbents are likely to
result in higher their market shares in future. The finding of lower market share of the
private banks as compared to their public sector counter parts may largely be caused by
their lesser penetration in the market banks as they are relatively new players in the
sector and generally operate in the urban areas.
However, the coefficient of MSZi,t-1 is not statistically significant. This means that
increase in demand for services of a bank does not necessarily result in its greater
16
As pointed out by Goldberger (1991), the R2 has a very modest role in regression analysis. The
classical linear regression model does not necessarily require that the value of the R2 should be
high. Neither is a high value of the R2 evidence in favour of a model nor is a low value of the R2
evidence against it (Gujarati and Sangeetha, 2007).
17
While Marquardt (1970) uses a VIF greater than 10 as a guideline for serious multicollinearity,
according to Mason et al. (1989), a VIF of greater than 10 is as reason for concern.
REVIEW OF ECONOMIC PERSPECTIVES
250
market share. Although it is generally expected that the banks with larger demand for
services will have higher market shares, the finding of no statistically significant
relationship between the two in the present paper is not surprising. A possible
explanation for such finding may be the way the two variables are defined. While
market size measures absolute presence of a bank in the sector, market share refers to its
relative position. Therefore, when market size of other banks in the sector also increases,
market share of an individual bank may not necessarily enhance, despite the rise in its
own market size. However, any definite conclusion in this regard requires further
exploration.
Hence, the structure of the market in Indian banking sector measured in terms of market
shares of the banks depends on other structural aspects of the market like their asset
base, conduct like selling efforts, and past financial performance. In other words, the
SCP relationships in Indian banking sector is not unidirectional, as it was suggested by
Bain (1956), where structure of the market influences performance via conduct. Instead,
market structure is influenced by conducts of the banks and their financial performance,
in addition to other structural aspects of the market. Since lagged selling efforts and
lagged financial performance have significant impact on market share, the relationships
are not instantaneous, and hence should be viewed in a dynamic context. Further,
significant impact of bank size on market share suggests that different aspects of market
structure are interdependent, and therefore, should not be considered in isolation of one
another.
As regards the determinants of conducts, it is observed that all the estimated coefficients
are statistically significant and positive. This implies that the banks, which have larger
asset base and larger market share, and better financial performance in the past, make
greater selling efforts. Further, the private banks, be it domestic or foreign, spend more
for selling purposes as compared to the public sector banks. The banks with higher
market share use the route of advertising and promotion of services to strengthen their
position further in the market. These banks seem to aim at creating image advantage,
strategic entry barriers, and marketing and distribution related complementary assets
through their selling efforts to limit market competition in the long run. Similarly, the
larger banks (i.e., the banks with larger asset base) spend more towards selling as they
have greater ability to spend for this purpose, whereas better financial performance in
the past enhances both ability and willingness of the banks to spend more towards
selling. Greater selling efforts by the private banks may largely be due to their market
driven business motives and approaches.
Thus, similarly to the market structure, the regression model on the determinants of
conduct also suggests for multidirectional SCP relationships in Indian banking sector,
which is contradictory to Bain’s (1956) unidirectional SCP relationship where conducts
are determined by structural aspects of the market. The regression results show that, in
addition to structural aspects of the market like market share of the banks and their asset
base, conducts of the banks are also determined by their financial performance. Since
lagged market share and lagged financial performance have significant impact on
banks’ selling efforts, the relationships are not instantaneous, and hence should be
viewed in a dynamic context.
Volume 12, Issue 4, 2012
251
The results of the regression model on the determinants of financial performance show
that the coefficients of all the variables except COSTit are statistically significant.
While the coefficient of SHAREit, D1 and D2 are positive that of BSZit and SELLi,t-1
are negative. This means that the returns on assets are higher for the banks with larger
market share. Furthermore, private banks, be it domestic or foreign, perform better
financially as compared to the public sector banks. On the other hand, the banks which
are larger in size or make greater selling efforts experience lower returns on assets.
However, the coefficient of COSTit is not statistically significant which implies that
costs efficiency does not have any significant impact on financial performance of the
banks.
From the regression results it is, therefore, evident that financial performance of the
banks is influenced by various structural aspects of the market structure, conducts of the
banks and their past performance level. It is observed that higher shares of the banks in
the market results in larger returns on assets either through greater efficiencies, or
higher market power, or both. This is contradictory to the findings of Bhattacharya and
Das (2003), Sathye and Sathye (2004), and Varma and Sainir (2010) that do not find
evidence in support of the traditional SCP hypothesis that market concentration directly
influences banks’ financial performance in Indian context. However, such a positive
relationship between market share and financial performance is consistent with the
findings of Edward (1964), Phillips (1967), Brucker (1970), Vernon (1971), Gilbert
(1984), Evanoff and Fortier (1988), Molyneux and Thornton (1992) and Katib (2004).
On the other hand, statistically significant but negative coefficients of bank size and
selling intensity are surprising. This is so possibly because the size larger than the
optimal level results in X-inefficiency, and hence lowers level of financial performance.
Similarly, while it is generally expected that greater selling efforts would enhance
financial performance of the banks, the finding of a negative coefficient of selling
intensity is possibly due to failure on the part of the banks in creating image advantage
and effective entry barriers through advertising. However, any definite conclusion in
this regard requires further research. The regression results also suggest that the nature
of ownership has a significant influence banks’ finance performance, and in comparison
to their public sector counterparts, private banks are better performers. Such better
financial performance of the private banks may largely be caused by their greater capital
base and better managerial efficiency.
Thus, like market structure and conduct, financial performance of the banks is also
determined by a variety of variables relating to structure of the market, conducts of the
banks, and other aspects of performance. This means that there are strong inter-linkages
between market structure, business strategies and financial performance in Indian
banking sector. Not only financial performance of the banks is influenced by their
conducts and structure of the market; there are also strong feedback effects from
performance to conduct and from conduct to structure. In Indian context, all these
essentially make the SCP relationships multidirectional and dynamic in nature. In
addition, financial performance of the banks is significantly influenced by the nature of
ownership as well.
REVIEW OF ECONOMIC PERSPECTIVES
252
VI. Summary Findings and Conclusions
In the context of initiation of economic reforms in general, and changes in policies and
regulations of the banking sector in particular, the present paper attempts to examine the
structure-conduct-performance relationships in Indian banking sector. It is observed that
there have been changes in the market structure of Indian banking sector, conducts of
the banks and their performance during the post-reform period, though the changes may
not necessarily be significant. Furthermore, strong inter-linkages exist amongst structure
of the market, conduct of banks and their financial performance. While market share of
a bank depends directly on its market size, asset base, selling efforts, and past financial
performance, as well as its selling efforts vary directly with market share, asset base,
and financial performance. On the other hand, returns on assets of a bank vary directly
with its market share, but inversely with its asset base and selling efforts.
This means that the SCP relationships in Indian banking sector are not necessarily
unidirectional. Not only the market share of a bank influences its selling efforts, the
reverse causality also holds discarding the unidirectional relationships between market
structure and business strategies of the banks. Similarly, while market share of a bank is
influenced by its financial performance, the reverse causality also holds. In other words,
there exist bidirectional relationships between market structure of the banks and their
financial performance. The same can be said in respect of conduct-performance
relationships as well.
The present paper also finds that nature of ownership of the banks plays a significant
role in determining structure of the market in Indian banking sector, business strategies
of the banks, and their financial performance. While market share of the private banks
(both domestic and foreign) is lower as compared to that of the nationalised banks, they
make greater selling efforts vis-à-vis their public sector counterparts. Further, the
private banks have better financial performance when compared with that of the public
sector banks.
The findings discussed above have some important policy implications. First, even
though higher market share improves financial performance, it should also be ensured
that the banks do not use their market power to charge monopoly prices for services.
Rather, efforts should be made in encouraging the banks to emphasize on enhancing
efficiency to improve their financial performance. However, the decision on whether the
banks should be allowed to raise their market shares through selling efforts requires
serious scrutiny as aggressive selling strategies may not only affect their financial
performance adversely, but may also result in unfair and restrictive business practices.
Second, since larger asset base results in higher market share of the banks, their growth
in terms of assets is a matter a serious concern. Moreover, given that it affects banks’
financial performance adversely, possibly due to larger nonperforming assets and Xinefficiencies
arising out of the larger size, efforts should be made in guiding the banks
towards more efficient use of the assets and in managing X-inefficiencies. This is very
important for enhancing competitiveness of the banks as well as their financial
performance.
Third, the findings that the nature of ownership influences structure of the market,
banks’ conduct and their financial performance raises a few important policy concerns.
Volume 12, Issue 4, 2012
253
What should be the policy approach to the private banks, especially to the domestic
ones in respect of their business strategies? Should the RBI play a proactive regulatory
role in this regard, particularly in respect of their unfair and restrictive business
practices? Are customers and investors benefitted with the entry of private banks, and
hence the changing dynamics of the sector? What are the macroeconomic implications
of these changes? Addressing these issues is very important and requires further
research to design appropriate policies and regulations.
References:
BAIN J. S. (1951). Relation of Profit Rate to Industry Concentration: American
Manufacturing 1936-1940. Quarterly Journal of Economics. 65(3), pp. 293-325.
BAIN, J. S. (1959), Industrial Organization. New York: Wiley
BAKER, J. B., WOODWARD, P. A. (1998). Market Power and the Cross-industry
Behaviour of Price around a Business Cycle Trough. Working Paper No. 221. Bureau of
Economics. Federal Trade Commission.
BASANT, R., MORRIS, S. (2000). Competition Policy in India: Issues for a
Globalising Economy. Project Report Submitted to the Ministry of Industry,
Government of India, New Delhi.
BERGER, A. N., HANNAN, T. H. (1992). The Price Concentration Relationship in
Banking: A Reply. Review of Economics and Statistics. 74(2), pp.376–389.
BERGER, A. N. (1995). The Profit-Structure Relationship in Banking: Tests of MarketPower
and Efficient-Structure Hypotheses. Journal of Money, Credit and Banking.
27(2), pp. 404-431.
BHATTACHARYA, K., DAS, A. (2003). Dynamics of Market Structure and
Competitiveness of the Banking Sector in India and its Impact on Output and Prices of
Banking Services. Reserve Bank of India Occasional Papers. 24(3), pp. 124-159.
BRESNAHAN, T. F. (1989). Empirical Studies of Industries with Market Power.
Handbook of Industrial Organization. Vol. 2, pp. 1011-57.
BRUCKER, E. (1970). A Microeconomic Approach to Banking Competition. Journal
of Finance. 25(5), pp. 1133-1141.
CHO, Y.J. (1986). Inefficiencies from Financial Liberalization in the Absence of WellFunctioning
Equity Markets. Journal of Money, Credit and Banking. 18(2), pp.191-199.
CHIAPPORI, P.A., PÉREZ-CASTRILLO, D., VERDIER, F. (1995). Spatial
competition in the banking system, localization, cross-subsidies and the regulation of
interest rates. European Economic Review. 39(5), pp. 889-919.
DELOMER, C. D., KLEIN, P. G., KAMERSCHEN, D. R., VOEKS, L. F. (2002).
Structure Conduct and Performance: A Simultaneous Equations Approach. Applied
Economics. 34(17), pp.13-20.
DEMSETZ, H. (1973). Industry Structure, Market Rivalry, and Public Policy. Journal
of Law and Economics. 16(1), pp. 1-9.
REVIEW OF ECONOMIC PERSPECTIVES
254
DEMSETZ, H. (1974). Two Systems of Belief about Monopoly. In H. Goldschmid, H.
Michael Mann and J. Fred Weston, eds. Industrial Concentration: The New Learning.
Boston: Little, Brown and Co., pp. 164-84.
EDWARD, F. R. (1964). Concentration in Banking and Its Effect on Business Loan
Rates. Review of Economic and Statistics. 46(3), pp. 294–300.
EVANOFF, D., FORTIER, L. (1988). Re-evaluation of the Structure-ConductPerformance
Paradigm in Banking. Journal of Financial Services Research. 1(3), pp.
277-294.
FRAME, W. S., D. R. KAMERSCHEN (1997). The Profit-Structure Relationship in
Legally Protected Banking Markets Using Efficiency Measures. Review of Industrial
Organization. 12, pp. 9-22.
GILBERT, A. (1984). Studies of Bank Market Structure and Competition: A Review
and Evaluation. Journal of Money, Credit and Banking. 16, pp. 617-644.
GOLDBERGER, A. S. (1991). A course in Econometrics. Cambridge: Harvard
University Press.
GRABEL, I. (1995). Assessing the Impact of Financial Liberalization on Stock Market
Volatility in Selected Developing Countries. Journal of Development Studies. 31, pp.
903 – 917.
GUJARATI, D. N., SANGEETHA. (2007). Basic Econometrics (4th
ed.). New Delhi:
McGraw-Hill, Inc.
GUPTA, V. K. (1983). A Simultaneous Determination of Structure, Conduct and
Performance in Canadian Manufacturing. Oxford Economic Papers. 35, pp. 281-301.
HANNAN, T. H., BERGER, A. N. (1989). Price rigidity and market structure: theory
and evidence from the banking industry. Finance and Economics Discussion Series 59.
Board of Governors of the Federal Reserve System (U.S.).
HANNAN, T. H. (1991).The Functional Relationship between Prices and Market
Concentration: The Case of the Banking Industry. Finance and Economics Discussion
Series 169. Board of Governors of the Federal Reserve System (U.S).
JOSHI, V., LITTLE, I.M.D. (1996). India's Economic Reforms 1991-2000. Oxford:
Clarendon Press.
KATIB, N. M. (2004). Market Structure and Performance in the Malaysian Banking
Industry: A Robust Estimation. 8th
Capital Market Conference. Indian Institute of
Capital Market Paper.
KHAMBAMPATI, U. S. (1996). Industrial concentration and performance: a study of
the structure, conduct, and performance of Indian industry. New Delhi: Oxford
University Press.
KIRZNER, I. M. (1973). Competition and Entrepreneurship. Chicago: University of
Chicago Press.
LEVIN, A., LIN, C. F., CHU, C. (2002). Unit Root Tests in Panel Data: Asymptotic and
Finite-Sample Properties. Journal of Econometrics. 108, pp. 1–24.
Volume 12, Issue 4, 2012
255
LLOYAD-WILLIAMS, D. M., MOLYNEUX, P., THORNTON, J. (1994). Market
structure and performance in Spanish banking. Journal of Banking and Finance. 18(3),
pp. 433–443.
MARQUARDT, D. W. (1970). Generalized Inverses, Ridge Regression, Biased Linear
Estimation and Nonlinear Estimation. Technometrics. 12(3), pp. 591–612.
MARTIN, S. (1979a). Advertising, Concentration and Profitability: The Simultaneity
Problem. Bell Journal of Economics. 10, pp. 639-47.
MARTIN, S. (1979b). Entry Barriers, Concentration and Profits. Southern Economic
Journal, 46, pp. 471-88.
MASON, E. (1939). Price and Production Policies of Large-Scale Enterprise. American
Economic Review. 29, pp. 61-74.
MASON, R. L., GUNST, R. F., HESS, J. L. (1989). Statistical Design and Analysis of
Experiments: Applications to Engineering and Science. New York: John Wiley and
Sons, Inc.
MILLER, R. L., VANHOOSE, D. D. (1993). Modern Money and Banking. (3rd ed.).
New York: McGraw-Hill.
MISHRA, P., BEHERA, B. (2007). Instabilities in Market Concentration: An Empirical
Investigation in Indian Manufacturing Sector. Journal of Indian School of Political
Economy. 19(3), pp.119-149.
MISHRA, P. (2005). Mergers and Acquisitions in the Indian Corporate Sector in the
Post- Liberalisation Era: An Analysis. Unpublished Doctoral Thesis. Vidyasagar
University, West Bengal.
MISHRA, P. (2008). Concentration-Markup Relationship in Indian Manufacturing
Sector. Economic and Political Weekly. 43(39), pp.76-81.
MISHRA, P., VIKAS. (2010). Structure, Conduct and Performance in Indian
Pharmaceutical Industry: A Simultaneous Equations Investigation. Review of
Development and Change, 15(1), pp. 69-99.
MOLYNEUX, P., THORNTON, J. (1992). Determinants of European Bank
Profitability: A Note. Journal of Banking and Finance 16(6), pp.1173–1178.
MOLYNEUX, P., FORBES W. (1995). Market Structure and Performance in European
Banking. Applied Economics. 27(2), pp.155–159.
MUELLER, D. C. (1990) (ed.). The Dynamics of Company Profits, An International
Comparison. New York: Cambridge University Press.
MULLINEUX, A., SINCLAIR, P. (2000). Oligopolistic Banks: Theory and Policy
Implications Unpublished .
NEUBERGER, D. (1997). Structure, Conduct and Performance in Banking Markets.
Thuenen-Series of Applied Economic Theory 12. University of Rostock, Institute of
Economics, Germany.
REVIEW OF ECONOMIC PERSPECTIVES
256
NEUMARK, D., SHARPE, S. A. (1992). Market Structure and the Nature of Price
Rigidity: Evidence from the Market for Consumer Deposits. The Quarterly Journal of
Economics. 107(2), pp. 657-80.
OKEAHALAM, C.C. (1998). Concentration in the Banking Sector of the Common
Monetary Area of Southern Africa. Paper was presented at the Development Issues in
the New Economy Conference. University of Cape Town, South Africa. March 24-26
2002.
OSBORNE, D. K., WENDEL, J. (1983). Research in Structure, Conduct and
Performance in Banking 1964–1979. Research paper 83-003. College of Business
Administration, Oklahoma State University.
PRAGER, R., HANNAN. T. H. (1998). Do Substantial Horizontal Mergers Generate
Significant Price Effects? Evidence from the Banking Industry. Journal of Industrial
Economics. 46(4), pp. 433-452.
PHILLIPS, A. (1967). Evidence of Concentration in Banking Markets and Interest
Rates. Federal Reserve Bulletin. 53(6), pp. 916–926.
PODENDA, R. J. (1986). Structure and performance: Some evidence from California
banking, Economic Review. Federal Reserve Bank of Sanfrancisco (winter), 5–17.
PUSHPANGADAN, K., SHANTA, N. (2008). Competition and Profitability in Indian
Manufacturing Sector. Indian Economic Review. 43(1), pp. 103-123.
RAM MOHAN, T.T. (2001). Long run performance of public and private sector bank
stocks Economic And Political Weekly. 36(52), pp. 4865-4871.
RAMASWAMY, K. V. (2006). State of Competition in Indian Manufacturing Industry.
In Mehta, P. S. (Ed). A Functional Competition Policy for India. New Delhi: Academic
Foundation.
RHOADES, S. A. (1985). Market Share as a Source of Market Power: Implications and
Some Evidence. Journal of Economics and Business. 37(4), pp. 343–365.
SCHERER, F.M., ROSS, D. (1990). Industrial Market Structure and Economic
Performance. Boston: Houghton Mifflin.
SCHMALENSEE, R. (1989). Inter-industry Studies of Structure and Performance. In R.
Schmalensee and R. Willig (ed.) Handbook of Industrial Organization. Vol. 2, New
York: North Holland. pp. 952-1009.
SHACKLE, G. L. S. (1971). Economics for Pleasure. Cambridge: Cambridge
University Press.
SHAFFER, S. (1989). Competition in the U.S. banking industry. Economics Letters.
29(4), pp. 321-323.
SINGH, A. (1993). Stock Markets, Financial Liberalization, and Economic
Development. Economic Journal. 107(442), pp. 771-782.
SMIRLOCK, M. (1985). Evidence on the (non) Relationship between Concentration
and Profitability in Banking. Journal of Money, Credit and Banking. 17(1), pp.69-83.
Volume 12, Issue 4, 2012
257
STRICKLAND, A. D., WEISS, L. W. (1976). Advertising, Concentration and PriceCost
Margins Journal of Political Economy. 84(5), pp. 1109-1121.
SWANK, J. (1996). Theories of the Banking Firm: A Review of the Literature. Bulletin
of Economic Research. 48(3), pp.173-207.
THAKOR, A.V. (1995). Financial Intermediation and the Market for Credit. edited by
Jarrow, Handbooks in Operations Research and Management Science. Elsevier Science.
Vol. 9, pp. 1073-1103.
VERNON, J. R. (1971). Separation of ownership and control and profits rates, the
evidence from banking: Comment. Journal of Financial and Quantitative Analysis. 6(1),
pp. 615–625.
VERMA, S., SAINIR, R. (2010). Structure Conduct and Performance of commercial
bank in India. Prajnan. 39(2), pp. 39-45.
VICKERS, J. (1995). Concepts of Competition. Oxford Economic Papers. 47(1), pp. 1-
23.
WEISS, L. (1974). The Concentration-Profits Relationship and Antitrust. In Audretsch,
D., Yamawaki, H., (eds), Industrial Concentration: The New Learning Boston: Little
Brown and Company.
REVIEW OF ECONOMIC PERSPECTIVES
258
Appendix – I
Table 1: Composition of Indian Banking Sector
Year
Number of Banks
Nationalized
Private
Others Total
Domestic Foreign
2000-01 29(22.8) 19(15.0) 34(26.8) 45(35.4) 127(100)
2001-02 29(23.6) 23(18.7) 35(28.4) 36(29.3) 123(100)
2002-03 29(22.5) 24(18.6) 31(24.0) 45(34.9) 129(100)
2003-04 29(21.0) 22(15.9) 31(22.5) 56(40.6) 138(100)
2004-05 29(20.9) 21(15.1) 31(22.3) 58(41.7) 139(100)
2005-06 29(22.7) 21(16.4) 26(20.3) 52(40.6) 128(100)
2006-07 28(21.7) 26(20.2) 28(21.7) 47(36.4) 129(100)
2007-08 28(23.9) 23(19.7) 26(22.2) 40(34.2) 117(100)
2008-09 28(25.2) 22(19.8) 24(21.6) 37(33.3) 111(100)
2009-10 28(27.5) 16(15.7) 22(21.6) 36(35.3) 102(100)
Note: Figures in parentheses indicate percentage share to total
Source: Prowess (CMIE)
Table 2: Basic Conditions of Indian Banking Sector
Year
Market Size Bank Size
Average C.V. Average C.V.
2000-01 5.20 0.42 7.33 0.30
2001-02 5.36 0.39 7.53 0.29
2002-03 5.24 0.44 7.36 0.33
2003-04 5.19 0.45 7.41 0.32
2004-05 5.01 0.51 7.37 0.34
2005-06 5.19 0.49 7.68 0.32
2006-07 5.18 0.52 7.65 0.35
2007-08 5.69 0.46 8.14 0.33
2008-09 6.15 0.41 8.56 0.30
2009-10 6.57 0.37 8.90 0.28
Note: Market Size – Natural logarithm of sales; Bank Size – Natural logarithm of assets;
Net Worth – Natural Logarithm of Net Worth; C.V. – Coefficient of Variation
Source: Prowess (CMIE)
Volume 12, Issue 4, 2012
259
Table 3: Structure of Indian Banking Sector
year
No. of
bank
Market Share 3-firm
Concentration
Ratio
HHI
Entropy
IndexAverage C.V.
2000-01 127 0.01 2.65 0.32 0.06 3.58
2001-02 123 0.01 2.63 0.31 0.06 3.56
2002-03 129 0.01 2.67 0.31 0.06 3.57
2003-04 138 0.01 2.75 0.32 0.06 3.58
2004-05 139 0.01 2.77 0.32 0.06 3.54
2005-06 128 0.01 2.66 0.32 0.06 3.52
2006-07 129 0.01 2.59 0.32 0.06 3.51
2007-08 117 0.01 2.26 0.31 0.05 3.55
2008-09 111 0.01 2.21 0.31 0.05 3.54
2009-10 102 0.01 2.06 0.29 0.05 3.54
Note: C.V. – Coefficient of Variation
Source: Prowess (CMIE)
Table 4: Strategies of Indian Banks, 2000-01 to 2009-10
year
Selling Intensity
Average C.V.
2000-01 0.23 6.11
2001-02 0.28 1.92
2002-03 0.30 4.52
2003-04 0.34 2.16
2004-05 0.48 3.18
2005-06 0.65 2.69
2006-07 0.79 2.38
2007-08 0.75 2.09
2008-09 0.78 3.06
2009-10 0.48 3.59
Note: C.V. – Coefficient of Variation
Source: Prowess (CMIE)
REVIEW OF ECONOMIC PERSPECTIVES
260
Table 5: Performance of Indian Banks, 2000-01 to 2009-10
Year
PBIT/Income PBIT/Asset ROCE
Average C.V. Average C.V. Average C.V.
2000-01 65.67 0.32 7.43 0.36 7.21 2.98
2001-02 67.95 0.30 7.35 0.44 2.47 11.1
2002-03 67.85 0.37 7.31 0.65 8.16 1.75
2003-04 66.43 0.36 7.09 0.64 7.75 1.79
2004-05 65.25 0.66 6.24 0.54 5.38 10.0
2005-06 55.37 0.91 4.70 0.46 2.80 1.25
2006-07 62.06 0.22 5.80 0.47 5.28 1.48
2007-08 65.10 0.17 5.59 0.21 5.92 1.21
2008-09 70.30 0.19 6.48 0.23 3.48 13.5
2009-10 72.30 0.20 6.96 0.91 7.32 1.09
Note: C.V. – Coefficient of Variation; PBIT – Profit before Interests and Taxes; ROCE
– Return on Capital Employed
Source: Prowess (CMIE)
Table 6: Cost Efficiency of Indian Banks, 2000-01 to 2009-10
year
Operating
Expenditure/Income
Total
Expenditure/Income
Current Ratio
Average C.V. Average C.V. Average C.V.
2000-01 71.56 0.25 74.66 0.99 0.71 0.23
2001-02 71.86 0.22 74.87 1.42 0.74 0.21
2002-03 67.53 0.26 70.30 1.51 0.71 0.27
2003-04 62.35 0.23 65.35 1.50 0.69 0.20
2004-05 58.55 0.31 61.84 1.92 0.66 0.29
2005-06 59.06 1.87 63.39 2.71 0.94 1.78
2006-07 60.83 7.07 64.86 2.77 1.26 7.07
2007-08 61.11 0.23 64.87 1.78 0.66 0.21
2008-09 61.21 0.26 64.86 1.40 0.66 0.24
2009-10 61.20 0.26 64.86 2.50 0.65 0.24
Note: C.V. – Coefficient of Variation
Source: Prowess (CMIE)
Volume 12, Issue 4, 2012
261
Table 7: Summary Statistics of the Variables
Variable Mean Std. dev. Min Max
SHAREit -5.3363 1.9142 -10.6748 -1.5055
BSZit 2.0102 0.3932 0.5108 2.6031
SELLit 0.8540 2.6406 -5.7038 7.3891
PROAit -2.8152 0.2706 -4.6332 -1.8769
NWit 1.8807 0.2695 0.7099 2.3708
TOCit -2.1246 1.0614 -7.0062 0.7651
SHAREi,t-1 -5.3816 1.8857 -10.6748 -1.5055
MSZi,t-1 1.7522 0.3934 0.1824 2.3784
SELLi,t-1 0.6062 2.5728 -5.7038 7.3891
ROAi,t-1 -2.7967 0.2927 -4.6332 -1.8769
PROFi,t-1 -0.4566 0.2050 -1.7931 -0.1046
D1 0.2712 0.4451 0.0000 1.0000
D2 0.3220 0.4678 0.0000 1.0000
Source: Prowess (CMIE)
Table 8: Results of Panel Unit-Root Test
Variables Adjusted t-Statistic
SHAREit -48.13*
BSZit -23.93*
SELLit -140.00*
PROAit -38.39*
TOCit -31.04*
NWit -11.78*
SHAREi,t-1 -94.63*
MSZi,t-1 -45.21*
SELLi,t-1 -55.05*
ROAi,t-1 -27.19*
PROFi,t-1 -47.32*
Note: *Statistically significant at 1 percent
REVIEW OF ECONOMIC PERSPECTIVES
262
Table 9: Regression Results
Determinants of Structure
Variable Coefficient t-Statistic VIF
Intercept -8.7054 -10.98*
BSZit 3.3337 9.73*
1.54
MSZit-1 0.3566 1.06 4.16
SELLit-1 0.2214 5.88*
3.35
ROAit-1 1.2686 5.84*
1.02
D1 -0.7708 -5.66*
1.35
D2 -1.0130 -6.41*
1.93
F-Stat 207.56*
Adj-R2
0.71
No. of Observation 406
Determinants of Conduct
Variable Coefficient t-Statistic VIF
Intercept -3.7508 -2.17**
SHAREi,t-1 0.6803 5.90*
2.30
BSZit 4.8390 8.22*
1.52
ROAi,t-1 0.7867 2.06**
1.04
D1 1.4352 6.54*
1.40
D2 1.0649 3.32* 2.29
F-Stat 144.53*
Adj-R2
0.62
No. of Observation 404
Determinants of Performance
Variable Coefficient t-Statistic VIF
Intercept -1.7057 -4.37*
SHAREit 0.1157 3.86*
8.54
BSZit -0.2973 -2.26**
1.44
SELLit-1 -0.0404 -2.93*
5.54
COSTit -0.0265 -0.79 1.16
D1 0.1376 4.04*
2.00
D2 0.0956 1.78***
2.98
F-Stat 4.32*
Adj-R2
0.12
No. of Observation 397
Note: * 1% significance level; ** 5% significance level; *** 10% significance level
Volume 12, Issue 4, 2012
263
Appendix II
Measurement of the variables
As mentioned earlier, in the present paper, the system of equations are estimated by
using pooled data collected from the Prowess data base of the Centre for Monitoring
Indian Economy (CMIE). In order to control for measurement errors, if any, three years
moving average is taken for each component of the variable used.
Market Share: Market share of bank i in year t (SHAREit) is measured as the ratio of its
income (INC) to total income of all the banks in the sector, i.e.,












∑
+
∑
+
∑
=
=
−
−
=
−
−
=
n
i
ti
ti
n
i
ti
ti
n
i
it
it
it
INC
INC
INC
INC
INC
INC
SHARE
1
2,
2,
1
1,
1,
1
3
1
Market Size: Market size of bank i in year t (MSZit) is measured as the natural
logarithm of its income (INC), i.e.,
[ ])ln()ln()ln(
3
1
2,1, −− ++= titiitit INCINCINCMSZ
Bank Size: Size of bank i in year t (BSZ) is measured as natural logarithm of its total
assets (ASSET), i.e.,
[ ])ln()ln()ln(
3
1
2,1, −− ++= titiitit ASSETASSETASSETBSZ
Selling Intensity: Selling intensity of bank i in year t (SELLit) is measured as the ratio of
its total expenditure for selling purpose (advertisement, marketing and distribution) (SE)
to its income (INC), i.e.,








++=
−
−
−
−
2,
2,
1,
1,
3
1
ti
ti
ti
ti
it
it
it
INC
SE
INC
SE
INC
SE
SELL
Cost Intensity: Cost intensity of bank i in year t (COSTit) is measured by the ratio of its
total expenditure (TE) to total income (INC), i.e.,








++=
−
−
−
−
2,
2,
1,
1,
3
1
ti
ti
ti
ti
it
it
it
INC
TE
INC
TE
INC
TE
COST
Returns on Assets: Returns on assets of bank i in year t (PROFit) is measured as the ratio
of profit before interests and taxes (PBIT) of the bank to its total assets (ASSET), i.e.,








++=
−
−
−
−
2,
2,
1,
1,
3
1
ti
ti
ti
ti
it
it
it
ASSET
PBIT
ASSET
PBIT
ASSET
PBIT
ROA
Net Worth: In the present paper, data on net worth (NW) is sourced from the Prowess
database of CMIE. The database measures net worth of a firm in terms of what it owes
the equity share holders. This consists of the money put into the company by the equity
REVIEW OF ECONOMIC PERSPECTIVES
264
share holders in the form of equity capital and the profits generated and retained as
reserves by the firm. Like other variables, three years moving average is taken for net
worth as well, i.e.,
[ ])ln()ln()ln(
3
1
2,1, −− ++= titiitit NWNWNWNW