A Non-Parametric Index of Corporate Governance in The Banking Industry - An Application To Indian Data

Socio-Economic Planning Sciences xxx (xxxx) xxxx
Contents lists available at ScienceDirect
Socio-Economic Planning Sciences

journal homepage: www.elsevier.com/locate/seps
A non-parametric index of corporate governance in the banking industry: An

application to Indian data
Rachita Gulatia, Ruth Kattumurib, Sunil Kumarc,∗
a
Department of Humanities and Social Sciences, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
b
India Observatory, London School of Economics and Political Science, London, United Kingdom
c
Faculty of Economics, South Asian University, New Delhi, India
ARTICLE INFO ABSTRACT
Keywords: This paper presents a methodological framework for constructing a non-parametric index of corporate gov-
Corporate governance index ernance for banks. The index is constructed by aggregating six distinct dimensional indices capturing different
Data envelopment analysis dimensions of corporate governance, namely board effectiveness, audit function, risk management, remunera-
Benefit-of-the-doubt model tion, shareholder rights and information, and disclosure and transparency. For aggregation, a tailored version of
Indian banks
data envelopment analysis (DEA) approach which is popularly known as constrained ‘Benefit-of-the-Doubt
Composite indicators
(BoD)’ model is employed. This approach is unique and distinctive in the sense that it requires no a priori
JEL classification: knowledge of weights, and assigns endogenous weights obtained from actual data to individual dimensions of
G21
bank governance in order to construct a composite index of corporate governance. This methodological fra-
G30
mework has illustrated by applying it for a data set of 40 Indian banks operating in the year 2017. The data set
G38
has been compiled using 58 governance regulations as defined by relevant jurisdictions.
1. Introduction employed to derive a composite index of corporate governance for

banks using a wide array of qualitative and quantitative regulatory
In recent years, especially after the global financial crisis of 2007- norms to be adhered by the banks. In particular, we show how the
09, the issue of corporate governance of banks has attracted serious dimensional indices capturing different dimensions of corporate gov-
attention of policy makers and regulators across nations. It is argued ernance of banks can be combined using the idiosyncratic and en-
that banks are ailing due to multiple agency distortions relating to both dogenous weights generated from the BoD model to construct a com-
equity governance and debt governance [1]. There is no doubt that any posite index of corporate governance. The dimensions that are
laxity of banks in adhering to governance regulatory norms affects their considered in this study to construct a composite index of corporate
ability to withstand the shocks in the system and jeopardizes their governance for banks pertain to i) board effectiveness, ii) audit func-
growth momentum. In this context, an index of corporate governance tion, iii) risk management, iv) remuneration, v) shareholder rights and
for banks provides an indispensable tool, which is beneficial for all information, and vi) disclosure and transparency. An illustration is also
stakeholders. Such an index indicates how well banks are adhering to provided to show the construction of corporate governance index for 40
governance practices that are framed by relevant jurisdictions. A Indian banks using data related to 58 governance norms that banks
composite index of corporate governance compresses the range of have to adhere to under relevant regulatory jurisdictions.
governance indicators, which are defined in distinct dimensions of Note here that there exists an implicit institutional framework for
corporate governance such as board effectiveness, audit function, and governance of banks in India, which comprises the guidelines by the
others, into a single numerical score that could be easier for inter- Ministry of Corporate Affairs under Companies Act, 1956 (recently
pretation, and is a useful tool for benchmarking banks and enabling amended in 2013) [56], the Reserve Bank of India (India's central
effective policy analysis. In addition, the changes in the corporate bank), and the Securities and Exchange Board of India (SEBI) under
governance index facilitate regulators to judge how governance reforms Clause 49 [Listing Obligations and Disclosure Requirements] Regula-
are working [2]. tions, 2000 (recently amended in 2015) [57]. However, the formal at-
Thereby, the focus of this paper is to illustrate how data envelop- tempt to reinforce corporate governance codes in the banking industry
ment analysis (DEA) based benefit-of-the-doubt (BoD) model can be in India was initiated by the Reserve Bank of India with the constitution
∗
Corresponding author.
E-mail addresses: [email protected] (R. Gulati), [email protected] (R. Kattumuri), [email protected] (S. Kumar).
https://doi.org/10.1016/j.seps.2019.03.008
Received 16 July 2018; Received in revised form 20 March 2019; Accepted 28 March 2019
0038-0121/ © 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Rachita Gulati, Ruth Kattumuri and Sunil Kumar, Socio-Economic Planning Sciences,
https://doi.org/10.1016/j.seps.2019.03.008
R. Gulati, et al. Socio-Economic Planning Sciences xxx (xxxx) xxxx
of Standing Committee on International Financial Standards and Codes: generated corporate governance index to be a handy tool for examining
Advisory Group on Corporate Governance in 2000 (Chairman: R.H. the differences in the levels of governance compliance across banks.
Patil). The Committee recommended the Organization for Economic The paper is structured as follows. Section 2 presents the relevant
Co-Operation & Development (OECD) principles of corporate govern- literature review. Section 3 elaborates the DEA based constrained BoD
ance as a yardstick for Indian banks. Later, the Consultative Group of model used in this study. Section 4 focuses on the different dimensions
Director for Financial Supervision (Chairman: A.S. Ganguly) in 2002, of corporate governance and their aggregation for index construction.
and more recently the Committee to Review Governance of Boards of The final section concludes the paper.
Banks in India in 2014 (Chairman: P.J. Nayak) reviewed governance
norms for banks in India. The expert committee under the chairmanship
of P.J. Nayak recommended banks to comply with Clause 49 stated by 2. Corporate governance in banking: a relevant literature review
the SEBI regulations, which is applicable for all the companies listed on
the Bombay Stock Exchange (BSE) and National Stock Exchange (NSE). Importance of good governance in non-banking firms has long been
To avoid the conflict of interest with their respective statutes by the advocated by the researchers (see, for example, [22,23]. However, the
government, the banking companies remain protected and exempted on role of governance in banking firms has received significant attention
a set of governance norms by the listing agency. In India, banks follow a only in recent years, especially the aftermath of the global financial
one-tier board structure known as ‘Anglo-American model’, where the crisis of 2007-09. Although the vast literature exists on discussing the
board consists of both executive and non-executive directors. The ex- qualitative aspects of corporate governance in banking [24–28], yet the
ecutive directors perform a managerial role together with their ac- empirical works on the quantification of corporate governance for
countability as members of the board of directors. In principle, there is banking firms are very limited (see Ref. [29] for a detailed review). It is
a fair separation of ownership and management in India. argued that better corporate governance practices by banks could be
The key contribution of this research endeavour is that this is the able to restrain the multiple agency conflicts that may arise among
first study to the best of authors’ knowledge, which proposes the use of stakeholders, that is either between shareholders and managers, and/or
DEA based BoD approach to construct a corporate governance index for between controlling and minority shareholders, and/or between
banking firms. Before this study, there has been no attempt to obtain a shareholders and creditors [30]. The literature also points out that the
robust measure of corporate governance index by using data-driven governance in banking firms differ from that in non-banking firms,
endogenous weighting system based on non-parametric linear pro- mainly due to their opaque nature, asymmetry in information and
gramming methodology. The empirical research on the subject of moral hazard concerns [31]. Therefore, the recent studies have shifted
construction of corporate governance index for banking industry is the focus on quantification through the construction of an aggregate
limited and is at an embryonic stage. Thus, this study contributes to index of corporate governance for banks distinctly. Table 1 reports
scant literature on this subject. It is noteworthy here that earlier efforts empirical studies that made an attempt to construct corporate govern-
on corporate governance index in banking used traditional approaches ance indices for banking firms.
such as simple linear unweighted average method (see, for example [3], We note that the previous literature reflects two key issues that are
or principal component analysis (PCA) (see, for instance, Refs. [4–7]. inherent in the construction of the composite index for corporate gov-
The major limitation of unweighted method is that it relaxes the as- ernance for banking firms. These issues pertain to: (i) the choice of
sumption of non-compensability or non-substitutability of indicators. indicators/dimensions for the construction of aggregate index, and (ii)
However, the use of PCA approach is inappropriate when sample size is index methodology for aggregation of indicators/dimensions. Most of
not large, and the variation of a variable is very small [8,9]. the researchers relied on a self-structured framework, which is based on
In the literature on construction of composite indicators, the BoD one or few sets of principle dimensions of corporate governance for the
approach has emerged as a most relevant and successful approach due construction of composite index (see, for instance, Refs. [3–6,35]. There
to its desirable properties and advantages. A few notable advantages of are also some studies where researchers used a structured framework
BoD approach are: i) it allows the actual data to decide on the weights developed by third-party (generally the rating agencies) to obtain
[9], ii) it assigns a single numerical score to a range of dimensions composite indices for measuring the strength and quality of governance
[10,11], iii) it is appropriate for small samples, iv) it is independent of a in banking firms. For example, Peni and Vӓhӓmaa [32] used the Gov-
priori statistical assumptions and appropriate to aggregate the unit in- Score corporate governance index developed by Brown and Caylor
variant data, and v) it allows endogenously calculated differential [33], which was based on 51 governance indicators; Ellul and Yerra-
weighting and aggregation of dimensions simultaneously [8,12]. milli [5] adopted 24 factors G-Index of Investor Responsibility Research
Owing to these aforementioned properties and advantages, the BoD Center (IRRC) database that was developed by Gompers et al. [34].
has been applied for construction of composite indices in diverse fields. A deeper scrutiny of the literature on corporate governance index
For example, Despotis [13] used it for building a Human Development for banking firms helps us to make following observations. First, the
Index and Cherchye et al. [14] constructed a Robust Human Develop- majority of the studies have constructed a composite index of corporate
ment Index; Murias et al. [15] crafted an Economic Wellbeing Index; governance considering only one or few sets of internal governance
Zhou et al. [12] computed a Sustainable Energy Index; Hermans et al. mechanisms, and have not considered the full comprehensive set of
[16] obtained a Road-safety Index; Antonio and Martin [17] built a governance mechanisms in the construction of such an index. It is im-
Child Health Index; Badasyan et al. [18] worked out a Broadband portant to note that the contemporary literature on corporate govern-
Achievement Index; Giambona and Vassallo [19] developed a Financial ance shields a wide-array of governance mechanisms for banks covering
Development Index; Gaaloul and Khalfallah [20] devised a Digital Ac- both equity and debt governance concerns [1]. Second, the studies have
cess Index; Martin et al. [21] constructed a Travel-tourism Competi- mainly employed the traditional unweighted method to construct cor-
tiveness Index. Since as noted above, we are not aware of any literature porate governance index, which implicitly implies that weights are
wherein the BoD approach has been applied for the construction of equal to one, ignoring the fact that all governance components may not
corporate governance index for individual banks, this study contributes be sharing equal policy priorities by banks. Our study is an attempt to
to this knowledge. In particular, the present study proposes the use of rationally deliberate on the aforementioned issues in the literature on
BoD as a robust approach for constructing an index of corporate gov- corporate governance in banking industry and proposes a holistic way
ernance for banking firms and shows how dimensional indices can be to construct the corporate governance index for banking firms using the
aggregated using the data-driven weights. We believe that researchers, endogenously generated weights using the non-parametric methodo-
policy makers and regulators would find our non-parametrically logical framework.
2
Table 1
Studies on the construction of corporate governance index for banks.
Author (Year) Sample (Period) Governance mechanisms Methodological framework
Peni and Vӓhӓmaa [32] 61 US large publicly traded banks Gov-Score index by Brown and Caylor [33] based on 51 governance attributes, Linear unweighted average
(2005) with each attribute as defined as binary variables index
Song and Li [6] 48 nations Self-structured index based on 15 variables related to board structure, Linear unweighted average
ownership structure, executive compensation and transparency index index
Ellul and Yerramilli [5] 74 US bank holding companies G-Index of IRRC database developed by Gompers et al. [34] and Risk Principal component analysis
(1999–2007) Management Index based on six risk governance attributes (PCA)
Love and Rachinsky [3] Russian and Ukrainian banks Self-structured index based on 26 indicators relating to five categories- Linear unweighted average
commitment to corporate governance, shareholders' rights, supervisory bodies, index
audit, and transparency and disclosure
Zagorchev and Gao [35] 820 US financial firms including 539 CG41 index based on 41 governance components for which data are available Linear unweighted average
commercial banks (2002–2009) in Aggarwal et al. [36] and RiskMetrics' Corporate Governance Quotient index
(CGQ). This index comprises of four sub-groups: (a) board, (b) audit, (c) anti-
takeover provisions, and (d) compensation and ownership.
Andrieș and Brown [58] 156 banks from Central and Eastern Self-structured index based on four supervisory board and four risk Linear unweighted average
Europe (2005–2012) management variables index
Tarchouna et al. [7] 184 US commercial banks Self-structured index based on five corporate governance variables- board size, Principal component analysis
(2000–2013) board independence, CEO duality, majority ownership and directors and (PCA)
executive officers' ownership
Andrieș et al. [4] 17 CEE nations (2005–2012) Self-structured index based on four supervisory board and four risk Principal component analysis
management variables (PCA)
Source: Authors' elaboration.
3. Data envelopment analysis based constrained ‘benefit-of-the- To formulate the basic BoD model, we assume that j = 1, ..., n refers
doubt’ model to banks, i = 1, ..., m refers to corporate governance dimensions I, and w
refers to weights such that 0 wij 1 and i = 1 wij = 1. The linear
m
This section presents the constrained BoD model used in this study programming formulation of the basic BoD model that looks like a DEA
for constructing a corporate governance index for individual banks. model in the multiplier form is given below:
Based on Farrell's [37] seminal work, Charnes, Cooper and Rhodes [38]
developed the first DEA model to assess relative efficiencies of peer
CGIo = max
m (1)
decision-making units (DMUs) in a non-parametric framework, which i = 1 wi, o Ii, o
wi, o
allows multiple inputs and multiple outputs. Over the years, several subject to
theoretical contributions and extensions have been made in DEA m (2) (A)
i = 1 wi, o Ii, j 1 j = 1, ..., n ;
modelling. Amongst these extensions, one of the most notable extension wi, o 0 i = 1, ..., m (3)
is the class of ‘benefit-of-the-doubt’ models that generate optimised
endogenous weights to aggregate the various dimensions of perfor-
mance. The BoD modelling approach was originally proposed by Melyn The optimal solution of Model (A) provides the observed value of
and Moesen [39] and later developed by Cherchye et al. [40,41]. To be the composite index of corporate governance for the bank o in terms of
more precise, a BoD model is akin to DEA model and aims to aggregate all the underlying dimensions. A few things are noteworthy here. First,
linearly quantitative performance indicators to construct a single CGIo lies between 0 (the worst performance among the banks in the
composite index when exact weights are not known a priori [41]. In the sample) and 1 (the best performance). Second, we solve Model (A) n
BoD model, the composite index is constructed by treating all the di- times once for each bank to obtain a set of composite indices
mensions/indicators as outputs, thereby considering no inputs in the CGI1, CGI2, ..., CGIn for n sampled banks. Third, wi, o are non-negative
model [42]. In fact, a BoD model compares the actual performance of bank-specific endogenous weights. In Model (A), the weights are se-
the unit with an internal benchmark rather than an external benchmark lected in such a way as to maximise the value of the composite indicator
that could not be realistically achievable in the specific local context of the evaluated bank. This, in turn, guarantees that any other
[19]. In fact, in the absence of true weights, BoD automatically assigns weighting scheme would worsen the ranking of this bank. Moreover,
the benefit-of-the-doubt weights determined by the data to each char- when these weights are used by any other bank in the sample would not
acteristic in order to build up the composite score, one for each unit result in a composite indicator greater than one [44]. Fourth, in the
[43]. construction of CGIo , bank o has always the highest possible scores in
The constrained BoD model that has been used in the present study relation to other banks in the sample. Thus, a good or bad position of a
is an extension of the basic BoD model. This model retrieves the in- bank does not depend on a good or bad weighting system because the
formation on appropriate weights from the observed data, and ag- weights are optimised to produce possible results for each bank [19].
gregates the distinct dimensions of corporate governance of banks. The Fifth, Model (A) avoids the subjectiveness in determining weights, and
constrained BoD model is a bank-specific model and needs to be solved therefore, provides a relatively objective performance score for each
for each bank in the sample separately so that we can get endogenously sampled bank [12].
computed weights that vary across banks and dimensions. For a typical Cherchye et al. [40] note that since the BoD approach uses data-
bank, the essence of the BoD model is to maximise the weights such that generated weights, therefore, one or few dimensions may get over-
the bank's corporate governance performance is as high as possible. emphasized or ignored. This situation arises when optimization pro-
Thus, BoD provides the weights that maximise (minimise) the impact of cedure assigns zero weights to one or more dimensions and that get
the dimension of corporate governance where the bank performs rela- ignored in the aggregation procedure. Charles and Díaz [45] argue that
tively good (poor) compared to the other banks. Hence, endogenously this situation arises because the units (here banks) are evaluated in the
generated weights from our BoD model are optimal and yield the best possible light in the basic ‘BoD’ model. Therefore, the computed
maximum value of composite index of corporate governance perfor- composite indices tend to overfocus on the dimension in which the unit
mance for a bank. performs the best and completely discard the information of the others.
3
In order to avoid such situations in the construction of a composite corporate governance is constructed by aggregating the six dimensional
index, additional weight restrictions are imposed on endogenous indices. A discussion of the dimensional indices of corporate govern-
weights, which set lower and upper bounds on the contribution of a ance for banks is warranted and is elaborated below.
particular dimension (see, [46–48]; for more details). Inclusion of
constraint (4) as follows, in the basic BoD model (A) transforms it into I. Index for Board Effectiveness: A bank board with ideal size and op-
the constrained BoD model: timal combination of inside (full-time executive) and outside (non-
executive affiliated and non-affiliated) directors is expected to be
wi, j Ii, j
Li , j m Ui, j i = 1, ..., m ; j = 1, ..., n more effective in monitoring management and resolving agency
w I
k = 1 i, j i, j (4) conflicts and contribute to superior bank efficiency [49]. Board
where Li, j and Ui, j represent lower and upper bound on endogenous effectiveness is assessed on 20 indicators pertaining to its compo-
weight assigned to ith dimension for jth unit. The similar model is used sition, structure (presence of qualified and independent board level
by Badasyan et al. [18] and Giambona and Vassallo [19]. In the present committees), independence (well-trained and certified non-execu-
study, we set lower bound as 10%, and the upper limit is assigned ac- tive independent directors on the board) and conduct (culture of
cordingly. For example, if lower bound is 0.10 (i.e., 10%) for five di- reinforcing ethical board conduct).
mensions, the proportional contribution for sixth dimension will be at II. Index for Audit Function: An audit committee is essential for an in-
1-(5 × 0.10) = 0.5 (i.e., 50%). Incorporating weight restrictions in the dependent audit process, which is assumed to provide a better
above manner not only overcomes the major flaw of overfocusing on oversight of the bank's financial reporting process, effectively
the best performing dimension in the basic BoD model but also tackle monitor the internal and statutory audits, and the auditor's in-
the problem of the presence of outliers on composite index scores to a dependence. A total of 9 governance indicators related to auditing
large extent since no dimension is ignored in the aggregation process. and auditor functioning are used in this study to examine com-
As BoD is sensitive to the dimensional score of zero and one, therefore, pliance on this dimension. Higher value of this dimensional index
before aggregating each dimensional index Ii, j is normalised (at mean represents tight audit controls within the bank.
100 and standard deviation 10) to account for zero and one values of Ii, j III. Index for Risk Management: Based on the recommendations by the
dimension [54]1. In the present study, the weight constrained BoD “Consultative Group of Directors of Banks/Financial Institutions”
model is estimated using the command ‘ci_bod_constr’ in Compind (Chairman: A.S. Ganguly), Indian banks constitute the board level
package using R software (see Appendix A for R codes used for running stand-alone risk management committee to independently monitor
the constrained BoD model). the risk policy and strategy for a bank. Banks also appoint the Chief
Risk Officer to efficiently monitor and mitigate internal risks. To
construct a dimensional index of risk management, we use 5 risk
4. Construction of corporate governance index for Indian banks:
governance indicators. Higher value of this dimensional index re-
an illustration
presents a tight risk management structure within the bank.
IV. Index for Remuneration: It has been argued that independent non-
4.1. Data, governance norms and dimensional indices
executive directors are needed to be engaged in deciding payment
of incentives to whole-time executives, and shareholders be kept
In order to provide an illustration of how the corporate governance
informed of the remuneration policies and structures. This is im-
index is computed by using the constrained BoD model, the first step
perative to counteract the managers' natural risk-aversion, and re-
involves collection of data on distinct governance norms adhered to by
muneration gives them incentives to take risk and maximise their
banks in India. Here, we consider 58 governance norms. The relevant
wealth. It is also important that excessive risk taking by the man-
qualitative and quantitative information on 58 governance indicators
ager must be controlled. This dimension is assessed on 5 governance
for 40 listed banks operating in the year 2017 had been obtained from
guidelines pertaining to remuneration to directors.
two different sources. The first and primary source is ‘Corporate
V. Index for Shareholder Rights and Information: Large shareholders can
Governance Report’ of a sampled bank. Note here that this report is an
influence the decisions of management/board and control them to
integral part of the annual report of a bank, which is generally publicly
protect their investments [50]. In order to avoid potential agency
available on the home page of the bank's website. The second data
conflicts and protect interests of minority shareholders and in-
source is SANSCO database and the ‘Corporate Governance’ section of
vestors, the jurisdictions advised banks to constitute a separate
the NSE website. We use this source for obtaining and supplementing
stakeholder's grievances committee with a non-executive director as
missing annual reports and data information. In total, we collected 40
chairman and company secretary as compliance officer of the
annual reports and thoroughly read these reports to gather data on 58
committee. All the grievances or complaints of investors should be
governance indicators. Thus, we have 2320 (i.e., 40 × 58) observations
the responsibility of the committee. This dimension is captured by
in total.
using 11 relevant governance norms.
The definition, coding and grouping of the selected governance
VI. Index for Disclosure and Transparency: An adequate disclosure and
norms for the construction of corporate governance index is provided in
transparency of inside information to outside stakeholders is an-
Table B1 of Appendix B. In the second step, each governance indicator
other important principle of effective governance [51]. As man-
is coded as a binary variable with a value of one implying that a bank
dated by law and regulations, a bank is required to disclose the
complies with the governance regulation, and zero otherwise. In the
quality and quantity information in the annual report to mitigate
third step, we construct the dimensional indices. For this, all govern-
agency problems arising from information asymmetry and enhances
ance norms are grouped under six mutually exclusive categories based
market discipline of banks [52]. The compliance on disclosure and
upon the key principles of corporate governance for banks, including
transparency is evaluated on a set of 8 indicators.
board effectiveness, audit function, risk management, remuneration,
shareholder rights and information, and disclosure and transparency.
4.2. Construction of corporate governance index for Indian banks
Fig. 1 illustrates the framework for bank governance used in the present
study. Dimensional indices corresponding to each dimension of bank
In the present study, we calculate a composite index of corporate
governance are obtained. In the final step, a composite index of
governance for banks by combining six distinct dimensional indices
using a non-parametric methodological framework. In the first step, we
1
We are grateful to Francesco Vidoli, one of the developers of Compind adopt a linear unweighted average method for constructing dimen-
package, for this suggestion. sional indices. The use of this method is popular among researchers to
4
Fig. 1. The framework for bank governance.
construct not only dimensional indices, but also an overall index of 9

y
r=1 r
corporate governance (see, for example [29,36,53], in case of non- II. Index for Audit Function = CGAudit = I2 =
9
banking firms and [3] for Russian and Ukrainian banks). In the second
step, we differ from the aforementioned studies. In particular, we ag- 5
y
r=1 r
gregate dimensional indices that we obtained in the first step by using a III. Index for Risk Management = CGRisk Management = I3 =
5
DEA based BoD model that assigns endogenous weights to each di-
mensional index in the process of aggregation. A detailed discussion on
5
y
r=1 r
IV. Index for Remuneration = CG Remuneration = I4 =
the methodological framework used in the present study is elaborated 5
in subsequent sections.
V. Index for Shareholder Rights and Information = CGShareholder Rights = I5
4.2.1. Construction of dimensional indices of corporate governance 11
y
r=1 r
As noted above, we employ the widely used linear unweighted =
11
average method for computing dimensional indices. It is important to
note here that we rely on this method to obtain dimensional indices VI. Index for Disclosure and Transparency = CGDisclosure and Transparency = I6
because the response on many governance norms is qualitative in 8
nature. This approach provides an index value for each dimension, y
r=1 r
=
which varies from zero to one. The dimensional indices are calculated 8
by using the following formula: Table 2 reports non-normalised values of dimensional indices of
s
y corporate governance, where index value is suggestive of areas of high
r =1 r ,j
CGDimensional index = Ii (i = 1, ..., m) = or low governance priority for the banks. Among all the dimensions, on
se, j
an average, higher level of regulatory compliance by banks is observed
where yr = {y1 , y2 , ..., ys } and yr , j = 1 if a bank is compliant to the r th on the shareholder rights and information (dimension V) followed by
indicator/norm of corporate governance, se, j = total maximum ex- disclosure and transparency (dimension VI). This is clear from the value
pected score of the corporate governance for jth bank in mth dimension. of indices for dimensions V and VI as 0.927 and 0.903. An average
This approach is adopted to obtain six dimensional indices as defined index value for dimensions I and III is relatively low. This reflects that
below: there is weak governance by banks on board effectiveness and audit
20
function. We believe that the recently reported big-ticket fraud cases in
y
r=1 r Indian banking industry are primarily due to poor governance of banks
I. Index for Board Effectiveness = CGBoard Effectiveness = I1 =
20 in these two important dimensions of bank governance. Further, banks
5
Table 2
Dimensional indices of corporate governance using linear unweighted average method for the year 2017.
Bank code Dimensions→ I II III IV V VI
Bank name↓ Index of Board Index of Audit Index of Risk Index of Index of Shareholder Index of Disclosure
Effectiveness Function Management Remuneration Rights and Information and Transparency
B1 Allahabad Bank 0.800 0.778 0.600 0.800 0.909 1.000

B2 Andhra Bank 0.850 0.778 0.600 0.800 1.000 1.000
B3 Axis Bank Ltd. 0.850 1.000 1.000 1.000 0.909 0.875
B4 Bandhan Bank Ltd. 0.650 1.000 1.000 0.800 0.636 0.625
B5 Bank of Baroda 0.650 0.778 0.800 0.800 1.000 0.875
B6 Bank of India 0.725 0.556 0.600 0.800 0.909 1.000
B7 Bank of Maharashtra 0.750 0.778 0.600 0.800 1.000 1.000
B8 Canara Bank 0.725 0.667 1.000 0.800 1.000 0.875
B9 Catholic Syrian Bank 0.700 0.889 1.000 1.000 0.636 0.875
Ltd.
B10 Central Bank of India 0.600 0.778 0.600 0.800 0.909 1.000
B11 City Union Bank Ltd. 0.900 1.000 0.800 1.000 0.909 0.750
B12 Corporation Bank 0.700 0.889 0.600 0.200 1.000 1.000
B13 DCB Bank Ltd. 0.925 1.000 1.000 1.000 0.909 0.875
B14 Dena Bank 0.650 0.667 0.600 0.800 0.909 1.000
B15 Dhanalakshmi Bank 0.950 0.778 0.800 1.000 0.909 1.000
Ltd
B16 Federal Bank Ltd 0.900 1.000 1.000 1.000 1.000 0.875
B17 HDFC Bank Ltd. 0.900 1.000 0.800 1.000 0.909 0.875
B18 ICICI Bank Ltd. 0.750 0.889 0.800 1.000 1.000 0.750
B19 IDBI Bank Ltd. 0.750 0.889 1.000 0.800 0.909 0.750
B20 IDFC Bank Ltd. 0.825 1.000 1.000 1.000 1.000 1.000
B21 Indian Bank 0.800 0.778 1.000 0.800 0.818 1.000
B22 Indian Overseas Bank 0.725 0.778 0.400 0.800 1.000 1.000
B23 IndusInd Bank Ltd. 0.925 0.889 0.800 1.000 0.909 0.875
B24 Jammu & Kashmir 0.600 0.667 0.600 0.800 0.909 0.875
Bank Ltd.
B25 Karnataka Bank Ltd. 0.775 1.000 0.600 1.000 0.909 0.750
B26 Karur Vysya Bank Ltd 0.950 0.889 0.600 1.000 1.000 0.875
B27 Kotak Mahindra Bank 0.875 1.000 0.000 1.000 0.909 0.875
Ltd.
B28 Lakshmi Vilas Bank 0.850 0.778 0.800 1.000 0.909 0.750
Ltd.
B29 Oriental Bank of 0.700 0.889 0.600 0.800 1.000 1.000
Commerce
B30 Punjab and Sind Bank 0.650 0.778 0.600 0.800 1.000 1.000
B31 Punjab National Bank 0.800 0.889 0.800 0.800 1.000 0.875
B32 RBL Bank Ltd. 0.850 0.889 0.600 1.000 0.818 0.875
B33 South Indian Bank Ltd. 0.900 0.889 1.000 1.000 1.000 1.000
B34 State Bank of India 0.650 0.778 0.800 0.800 0.909 0.750
B35 Syndicate Bank 0.675 0.556 0.800 0.800 0.909 1.000
B36 UCO Bank 0.750 0.667 0.600 0.800 0.909 1.000
B37 Union Bank of India 0.600 0.889 0.600 0.800 1.000 1.000
B38 United Bank of India 0.725 0.889 0.600 0.800 0.909 0.875
B39 Vijaya Bank 0.800 0.778 0.800 0.800 1.000 0.750
B40 Yes Bank Ltd. 0.850 1.000 1.000 1.000 0.909 1.000
Mean 0.775 0.845 0.745 0.870 0.927 0.903
Source: Authors' calculations.
are 25.5% less efficient on risk governance in the sampled year. Only 18 practices. Falling short on all dimensions of bank governance could be a
banks (45% of sampled banks) attained the dimensional index score of challenge for very large public sector banks such as State Bank of India,
one on governance guidelines pertaining to disclosure and transpar- which however can be addressed through investing human and fi-
ency, 15 banks (37.5% of sampled banks) on shareholder rights and nancial resources for prioritising reporting.
information, 11 banks (27.5% of sampled banks) on risk management,
10 banks (25% of sampled banks) on audit quality while none on board 4.2.2. Aggregation of dimensional indices to construct a corporate
effectiveness. Further, although banks have performed well on pro- governance index
tecting shareholder rights and maintaining disclosures, 45% of sampled After obtaining six distinct dimensional indices, the constrained
banks performed below average on audit and risk functions. BoD model is used to generate idiosyncratic and endogenous weights,
We also note substantial differences in bank's obedience to gov- which aggregates normalised values of dimensional indices to obtain a
ernance regulatory provisions. Our findings suggest that a bank could corporate governance index (CGI) for individual banks. As noted above,
be better governed with respect to one dimension but relatively worse this approach is unique in the sense that it is simpler and generates
on another. For instance, Yes Bank Ltd. is compliant (Ii, j =1) on all di- weights based on actual data. The estimated values of corporate gov-
mensions of governance except on board effectiveness, while nine other ernance index for sampled banks and their rankings across different
banks were found to be outperforming this bank in this regard. In ad- methods of index construction are reported in Table 3. However, in this
dition, three banks, namely Jammu & Kashmir Ltd., United Bank of sub-section, we shall focus on the results obtained using BoD model
India and State Bank of India attained the index score less than one with 10% lower bound restriction. The weight's matrix with this re-
(Ii, j < 1) on all six dimensions. This may be due to lacunae in reporting striction is given in Table C1 of Appendix C. An index of corporate
6
Table 3
Corporate governance index and ranking of banks using alternative aggregation methodological frameworks.
Bank code Bank name Constrained BoD (10%) Rank Constrained BoD (5%) Rank Factor Analysis Rank Equal Weighting Rank
(CGIBOD_0.10) (CGIBOD_0.05) (CGIFactor) (CGIEqual)
B1 Allahabad Bank 0.9377 21 0.9678 21 −0.1031 21 0.8145 27

B2 Andhra Bank 0.9523 13 0.9756 13 0.1121 14 0.8380 21
B3 Axis Bank Ltd. 0.9812 8 0.9905 9 0.5097 6 0.9390 6
B4 Bandhan Bank Ltd. 0.8716 39 0.9314 37 −0.8912 40 0.7852 32
B5 Bank of Baroda 0.9314 27 0.9645 27 −0.107 22 0.8171 25
B6 Bank of India 0.9078 35 0.9517 35 −0.4472 36 0.7649 38
B7 Bank of Maharashtra 0.9428 17 0.9706 17 0.0474 18 0.8213 23
B8 Canara Bank 0.9366 22 0.9672 22 −0.0374 20 0.8444 18
B9 Catholic Syrian Bank Ltd. 0.9103 34 0.9531 34 −0.2837 32 0.8500 15
B10 Central Bank of India 0.9173 31 0.9569 31 −0.2325 28 0.7811 34
B11 City Union Bank Ltd. 0.9670 11 0.9832 11 0.1147 13 0.8932 10
B12 Corporation Bank 0.8760 38 0.9339 36 −0.4546 37 0.7315 40
B13 DCB Bank Ltd. 0.9984 4 1.0000 1 0.5583 5 0.9515 5
B14 Dena Bank 0.9128 32 0.9544 32 −0.348 35 0.7710 37
B15 Dhanalakshmi Bank Ltd 0.9912 6 1.0000 1 0.3686 8 0.9061 8
B16 Federal Bank Ltd 1.0000 1 1.0000 1 0.725 4 0.9625 3
B17 HDFC Bank Ltd. 0.9803 9 0.9901 10 0.3732 7 0.9140 7
B18 ICICI Bank Ltd. 0.9494 14 0.9740 14 0.0527 16 0.8648 13
B19 IDBI Bank Ltd. 0.9352 24 0.9665 24 −0.1671 26 0.8497 16
B20 IDFC Bank Ltd. 1.0000 1 1.0000 1 0.935 1 0.9708 1
B21 Indian Bank 0.9436 16 0.9710 16 0.0517 17 0.8660 12
B22 Indian Overseas Bank 0.9286 28 0.9630 28 −0.1376 24 0.7838 33
B23 IndusInd Bank Ltd. 0.9786 10 0.9911 8 0.2416 10 0.8997 9
B24 Jammu & Kashmir Bank 0.8634 40 0.8822 40 −0.6389 39 0.7418 39
Ltd.
B25 Karnataka Bank Ltd 0.9398 20 0.9689 20 −0.1351 23 0.8390 19
B26 Karur Vysya Bank Ltd. 0.9882 7 1.0000 1 0.2718 9 0.8856 11
B27 Kotak Mahindra Bank Ltd. 0.9195 30 0.9581 30 −0.3184 34 0.7765 36
B28 Lakshmi Vilas Bank Ltd 0.9400 19 0.9691 19 −0.2133 27 0.8478 17
B29 Oriental Bank of 0.9470 15 0.9728 15 0.1628 11 0.8315 22
Commerce
B30 Punjab and Sind Bank 0.9325 25 0.9651 25 −0.0173 19 0.8046 28
B31 Punjab National Bank 0.9554 12 0.9772 12 0.1378 12 0.8606 14
B32 RBL Bank Ltd. 0.9410 18 0.9696 18 −0.1586 25 0.8387 20
B33 South Indian Bank Ltd 1.0000 1 1.0000 1 0.8357 2 0.9648 2
B34 State Bank of India 0.8794 37 0.8973 39 −0.5485 38 0.7811 35
B35 Syndicate Bank 0.9121 33 0.9540 33 −0.3107 33 0.7899 30
B36 UCO Bank 0.9227 29 0.9598 29 −0.2833 31 0.7876 31
B37 Union Bank of India 0.9353 23 0.9666 23 0.0982 15 0.8148 26
B38 United Bank of India 0.9004 36 0.9106 38 −0.2624 29 0.7997 29
B39 Vijaya Bank 0.9322 26 0.9649 26 −0.2686 30 0.8213 24
B40 Yes Bank Ltd. 0.9955 5 1.0000 1 0.7683 3 0.9598 4
governance for each bank is expected to lie between a minimum of 0 transparency (dimension VI) alongside weakness on board effectiveness
(the worst governed) and the maximum of 1 (the best governed).2 We (dimension I). The Jammu & Kashmir Bank Ltd. and Bandhan Bank Ltd.
note that corporate governance index value at 10% weight restriction attain lower ranking on the basis of their values of composite index of
(CGIBOD_0.10) varies from a minimum of 0.8634 to a maximum of 1, corporate governance, and therefore these banks require greater efforts
implying a considerably high level of banks' adherence to governance to improve their compliance with stated governance regulatory provi-
regulations set by the jurisdictions in the sampled year. A persistent sions. We also note that majority of banks are still far from perfection,
regulatory oversight and a coercion to avoid any sort of penalties or and they can achieve the status of “good governed bank” by improving
strictures or restrictions in business operations in recent years have their adherence in different dimensions of bank governance.
compelled banks to show higher obedience to governance norms in the Overall, our findings suggest that Indian banks work along the
sampled year. Only three banks, namely South Indian Bank Ltd., IDFC conventional lines, and therefore, put more efforts in protecting
Bank Ltd., and Federal Bank Ltd. attain the status of well governed shareholder rights and information, and maintaining better disclosure
banks (with CGIBOD_0.10 score of 1) and are found to be relative efficient and transparency levels. The results indicate that banks in India focus
in terms of adherence to corporate governance framework. Although less on contemporary debt governance principles. In the light of our
these banks emerge as ideal benchmarks of bank governance in India, results, we feel that Indian banks are required to pay a serious attention
they do lack compliance on one or more dimensions of governance. For towards adherence of regulatory norms concerning board quality, audit
instance, South Indian Bank Ltd. is not well governed on the dimensions and risk management practices in order to attain the status of well
of board effectiveness (dimension I) and audit function (dimension II); governed banks, and ultimately, to improve their financial health.
IDFC Bank Ltd. lacks perfect governance on board effectiveness (di-
mension I); Federal Bank Ltd. is not well governed on disclosure and 4.3. Sensitivity analysis
4.3.1. Choice of aggregation method and its impact on ranking of banks on

2
In the BoD approach, the index score may take a maximum value of 1 CGI
(benchmark bank with a high level of governance). However, achieving a Corporate governance index obtained using constrained BoD model
minimum value of 0 (with the overall governance level of zero) is not a case. at 10% lower bound restriction (CGIBoD_0.10) is compared with three
7
Table 4
Correlation between composite indices estimated across different methodological frameworks.
Aggregation method Constrained BoD (5%) Constrained BoD (10%) Factor Analysis Equal Weighting
Constrained BoD (5%) 1.000 – – –

Constrained BoD (10%) 0.984∗∗∗ 1.000 – –
Factor Analysis 0.888∗∗∗ 0.954∗∗∗ 1.000 –
Equal Weighting 0.936∗∗∗ 0.983∗∗∗ 0.991∗∗∗ 1.000
Note: ∗∗∗ indicate significance at 1% level.

alternative composite indices –constrained BoD with 5% lower bound improving board quality and over-emphasizes remuneration policy. On
restriction (CGIBOD_0.05), factor analysis based CGI index (CGIFactor) and the other hand, setting a lower bound equivalent to equal weight in BoD
equal weight CGI index (CGIEqual). In CGIBOD_0.05 and CGIBOD_0.10 where eliminates the difference in the policy priorities by a bank in one di-
the lower bound of weight constraint is restricted at 5% and 10%, re- mension relative to other. Thus, weight allocation is an important as-
spectively. CGIFactor is based on factor analysis, which is a non-frontier pect because ultimately good corporate governance should, sooner or
method and assigns data generated weights to each dimension in ac- later, involve all the six dimensions to ensure long-term financial sus-
cordance to the proportion of the variance explained by the dataset tainability.
[54]. In the present study, the CGIFactor index of corporate governance All in all, the study observes that the constrained BoD approach is a
is estimated using the command ‘ci_factor’ in the Compind package. robust approach for index construction, and successfully overcomes
Equal weight approach assigns a weight of 0.16 (1/m = 1/6) to each major methodological flaws that are present in traditional approaches.
dimension, and all dimensions are linearly aggregated in a simpler way, The constrained BoD model constructs a reliable composite index of
implying every bank gives equal policy priority to each dimension of corporate governance for individual banks by allowing a sensible
bank governance. The index values and ranking of banks corresponding weighting scheme for aggregation of normalised values of dimensional
to different alternative methods are also reported in Table 3. The sta- indices. These index values can be used for identifying the areas needed
tistical significance of differences in the ranking of banks across dif- for improvement in the governance framework at the level of an in-
ferent alternative methodologies is tested using Spearman's rank cor- dividual bank as well as the banking industry as a whole.
relation test.
The matrix of correlation coefficients of the ranks corresponding to 5. Conclusions
four alternative indices reveals that the rank correlation coefficients are
very high and statistically significant at 1% level (see Table 4). This This paper suggests a methodological framework to construct a non-
indicates a greater rank concordance of banks on the corporate gov- parametric index of corporate governance for banks using the benefit-
ernance index obtained by employing different aggregation methods. of-the-doubt (BoD) approach. This approach is unique, completely ro-
The ranking of banks across alternative approaches remain stable. bust and distinctive in the sense that it requires no a priori knowledge of
Therefore, we can safely infer that BoD is a robust approach and the weights, and assigns endogenous weights obtained from actual data to
idea of this approach to assign endogenous generated weights to each individual dimensions of governance to construct a composite index of
dimension is more judicious than assigning fixed or equal or no weights corporate governance. The framework suggested here shows how to use
for aggregation purpose. Overall, the construction of corporate gov- constrained BoD model to eliminate the problem with giving equal or
ernance index for banks using BoD is far superior to unweighted subjective weights to key dimensions of governance of banks while
method. Our inference is in line with Wittrup and Bogetoft [55] who constructing a composite index of corporate governance. In addition,
conclude that weighted assessment is superior to unweighted, while we found that BoD is suitable for application to small samples, such as
determining the court workload. the case of our data. An illustration has been presented using a data set
of 40 Indian banks in the year 2017. The data set is compiled using 58
4.3.2. Lower bound weight choice and its impact on CGI qualitative and quantitative governance indicators as defined by re-
An increase of the lower bound reflects the sensitivity of assigning levant jurisdictions. In terms of managerial implications, the metho-
weights to individual dimensions of CGI. In the present study, we dological framework presented in this study provides the policy for-
capture the variations in weight restrictions under two limits – un- mulators with the opportunity to not only rank the banks in accordance
restricted with lower bound restriction of 5% to a maximum of 16% with their corporate governance performance but also identify strong
(assuming if equal weight of 1/m is assigned to each dimension). and weak dimensions of governance in each bank. This, in turn, pro-
Accordingly, twelve alternative corporate governance indices are cal- vides comprehensive guidance for policy formulators to assist them in
culated (see Table C2 in Appendix C for details on CGI scores corre- identifying areas of corporate governance, which might require im-
sponding to lower bound from 5% to 16%). Similar procedure is provements by the banks. Further, the non-parametric corporate gov-
adopted for conducting a sensitivity analysis by Giambona and Vassallo ernance index serves as a classification system to monitor each bank's
[19] in the construction of financial development index. From Table 5, progress on adoption of governance principles. This system is crucial for
we note that unlike traditional equal weight or unweighted approaches, designing and implementing targeted policies to improve overall gov-
constrained BoD approach rewards for policy priority by banks and ernance quality of banks.
penalizes under-performance in one or more dimensions. For instance, The empirical results reveal that considerable efforts have been
banks' adherence to dimensions V and VI is rewarded by higher average made by banks in adhering to corporate governance regulations in India
aggregate weightage of 43% and 39% in the construction of CGI0.05 and in the last decade. This is evident from an estimated value of corporate
CGI0.10, respectively. However, penalty is imposed on dimensions I governance index for sampled banks, which range between a minimum
(board effectiveness) and II (audit function) in terms of attaining lower of 0.8634 and maximum of 1. Only three banks, namely South Indian
9% and 13% weightage, respectively, as can be seen from the weight's Bank Ltd., IDFC Bank Ltd., and Federal Bank Ltd. tops the list and are
matrix. As discussed by Cherchye et al. [40], we also computed the CGI found governance efficient with a corporate governance index score of
score with no weight restriction in BoD model. We find that unrest- one. Unlike traditional equal weight or unweighted approaches, our
ricted BoD allocates defective weight of zero percent to board effec- chosen BoD approach with weights constrained fully rewards the suc-
tiveness dimension and ignores the priority given by banks in cess of banks in obeying governance norms pertaining to shareholders'
8
Table 5
Average weights corresponding to different weight restrictions.
Dimensions→ Board Effectiveness Audit Risk Management Remuneration (w4) Shareholder Rights and Disclosure and Number of banks
(w1) Function (w2) (w3) Information (w5) Transparency (w6) with CGI = 1
Changes in lower
bound restriction↓
Unrestricted (0%) 0.00 0.03 0.12 0.43 0.27 0.15 37

5% 0.09 0.13 0.21 0.14 0.26 0.17 7
6% 0.11 0.15 0.17 0.11 0.16 0.29 7
7% 0.13 0.14 0.20 0.12 0.24 0.16 6
8% 0.16 0.12 0.17 0.14 0.13 0.27 5
9% 0.15 0.14 0.18 0.13 0.24 0.16 4
10%a 0.16 0.14 0.18 0.13 0.23 0.16 3
11% 0.16 0.17 0.14 0.13 0.16 0.23 3
12% 0.17 0.15 0.17 0.13 0.16 0.22 3
13% 0.18 0.15 0.16 0.14 0.16 0.21 3
14% 0.17 0.16 0.17 0.15 0.16 0.19 2
15% 0.17 0.16 0.16 0.16 0.17 0.18 2
Equal weight (16%) 0.17 0.16 0.16 0.16 0.17 0.17 1

Note.
a
Bank-wise weights assigned to each dimension are reported in Appendix Table C1.
right and information, and imposed a penalty on board effectiveness However, the key limitation of the proposed framework is that it cannot
and audit function. This can be seen from the weights’ matrix. Thus, accommodate the negative or binary values of indicators or dimen-
optimal weight allocation is an important aspect because equal sional indices. The future research can be directed to overcome this
weighting eliminates the difference in the policy priorities by a bank in drawback. In the future, research can also be directed to use of this
one dimension relative to other. Also, the ranking of banks on the non- framework to assess how the governance performance of banks has
parametric corporate governance index remains robust enough and not evolved over the period.
very sensitive to the choice of aggregation method.
On concluding note, the application of this suggested framework not
only provides us with the ranking of banks in accordance of their ad- Acknowledgements
herence to governance regulations, but also helps us to identify strong
and weak dimensions of governance for each bank. This identification The authors would like to thank the guest editors, anonymous re-
could facilitate with redesigning of existing policies both at bank-level ferees, and editor of the journal for their valuable comments and sug-
and industry-level. We believe that this framework would be greatly gestions, which helped to improve the quality of the paper sub-
beneficial for improving corporate governance in the banking sector. stantially. Any errors remain the authors' responsibility.
Appendix A
R codes used for constrained BoD
Library (Compind)
Options (stringsAsFactors = F)
***********************************************************************
Importing the comma delimited (.csv) file
***********************************************************************
DF = read.csv (“2017.csv”)
Summary (DF)
***********************************************************************
Normalization of indicator/dimension/pillar
***********************************************************************
Data_norm = normalise_ci (DF,c(2:7), polarity = (“POS", “POS”, “POS”, “POS”, “POS”, “POS”), method = 1,z.mean = 100, z. std = 10)
***********************************************************************
BoD as the constrained version (low_w vary from 0.05 to 0.16).
***********************************************************************
CI_constr = ci_bod_constr (data_norm$ci_norm,c (1:6), up_w = 1,low_w = 0.1)
CIci_bod_constr_est = CI_constr$ci_bod_constr_est
CIci_bod_constr_weights = CI_constr$ci_bod_constr_weights
***********************************************************************
Exporting output in comma delimited (.csv) file
***********************************************************************
summary (CIci_bod_constr_est)
plot (density (CIci_bod_constr_est))
write.csv (CIci_bod_constr_est,“resultsbod.csv”)
write.csv (CIci_bod_constr_weights,“resultmatrixbod.csv”)
9
Appendix B
Table B1
Definition of dimensions and indicators of bank governance.
Dimensions Indicators in each dimension ( yr : r 1, ..., s ) Regulatory/Legal Clause

(Im: m 1, ..., 6 )
I. Board Effectiveness (20) Whether a bank has no more than 15 directors on the board SEBI Clause 49 (II.A.1), 2015;
Companies Bill Clause 149 (1.b), 2013
At least 50% of the board comprise of non-executive directors SEBI Clause 49 (II.A.1), 2015
Board has not exceeding two nominee directors SEBI Clause 49 (II.B), 2015
Board appoint at least one woman director SEBI Clause 49 (II.A.1), 2015
Executive director holds directorship/chairmanship on not more than three listed companies SEBI Clause 49 (II.B.2), 2015
Non-executive independent director hold directorship on not more than six listed companies SEBI Clause 49 (II.B.2), 2015
Board meets at least four times a year SEBI Clause 49 (II.D.1), 2015
Chairman of the board held the position for a minimum of five years Companies Act 2013
Chairman of the board is non-executive director SEBI Clause 49 (II.A.2), 2015
Chairman and CEO are two separate persons on the board SEBI Clause 49 (II.A.2), 2015
Bank provide the details about the committee's establishment, mandate and composition including members SEBI Clause 49 (II.D), 2015
who are independent in the corporate governance report; 0.5 if details are partially provided.
Board include at least one-third of the independent directors, in case of non-executive chairman and at least SEBI Clause 49 (II.A.2), 2015
one-half, in case of executive chairman
Suitable training is imparted to independent directors and is disclosed by bank in the report or on website SEBI Clause 49 (II.B.7), 2015
Independent director(s) has given a separate declaration of their independence included as part of corporate SEBI Clause 49 (II.B), 2015
governance report
Separate meeting of independent directors is held in financial year and details are disclosed in the report; 0.5 SEBI Clause 49 (II.B.6), 2015
if meeting held and details are not provided
Board constitute nomination committee for appointment of directors SEBI Clause 49 (IV); BCBS, 2015;
Companies Act Clause 178, 2013
Bank board publishes its separate corporate governance report as a part of annual report SEBI Clause 49 (X), 2015
Bank defines code of corporate governance SEBI Clause 49 (II.E), 2015
Board lays down a ‘code of conduct’ certificate from all board members and senior management SEBI Clause 49 (II.E), 2015
Bank establishes a vigil mechanism for directors and employees through whistle blower policy SEBI Clause 49 (II.F), 2015
II. Audit Function (9) Board constitute audit committee SEBI Clause 49 (III.A), 2015;
Companies Bill Clause 177 (1), 2013
Committee has minimum three members SEBI Clause 49 (III.A.1), 2015;
Companies Bill Clause 177 (2), 2013
Two-third members of audit committee are independent directors SEBI Clause 49 (III.A.1), 2015
Chairman is independent non-executive director SEBI Clause 49 (III.A.3), 2015
Internal audit procedure is defined in the report SEBI Clause 49 (III.D & E), 2015; RBI,
2002
Statutory auditor is appointed by board SEBI Clause 49 (III.D & E), 2015; RBI,
2002
Committee met at least four times a year SEBI Clause 49 (III.B), 2015
Company secretory act as a secretory of audit committee SEBI Clause 49 (III.A.6), 2015;
Companies Bill Clause 134.f, 2013
Bank include a certificate from either the auditors' or practicing company secretaries regarding compliance of SEBI Clause 49 (III.D & E), 2015
conditions of corporate governance and disclose it in the annual report
III. Risk Management (5) Presence of risk management committee RBI, 2002; SEBI Clause 49 (VI), 2015;
BCBS, 2015
Appointment of a chief risk officer RBI, 2002; BCBS, 2015
Bank disclose the size of risk management committee RBI, 2002
Bank disclose the number of meetings held by RMC RBI, 2002
Bank have non- executive director as chairman of RMC RBI, 2002
IV. Remuneration (5) Board constitute a remuneration for overall oversight of management's implementation of remuneration SEBI Clause 49 (IV), 2015; Companies
system. Bill Clause 178, 2013
Independent director as a chairman of the committee. Companies Act 2013
At least three members, with majority as non-executive directors. Companies Act 2013
Information on payments of remuneration/sitting fees to directors, if any paid is disclosed in the report. SEBI Clause 49 (II.C), 2015
All directors are non-executive Companies Act 2013
V. Shareholder Rights and Board constitute stakeholder's grievance committee. SEBI Clause 49 (VIII.E.4), 2015;
Information (11) Companies Bill Clause 178, 2013
Non-executive director act as chairman of stakeholder grievance committee. SEBI Clause 49 (VIII.E.4), 2015
Compliance officer reporting to company secretary RBI, 2002
Committee look into the matters relating to investor complaints and board discloses the number of complaints SEBI Clause 49 (VIII.E.4), 2015;
received and resolved in a financial year. Companies Bill Clause 178 (6), 2013
Board disclose the information on the last three annual general meeting held in the annual report. SEBI Clause 49 (XII.6), 2015
Bank board disclose the information regarding its listing on various stock exchanges in the report. SEBI Clause 49 (XII.9), 2015
Disclose its dividend policy and dividend paid, if any, during the year in the report. SEBI Clause 49 (XII.9), 2015
Bank disclose the information on market price of its share. SEBI Clause 49 (XII.9), 2015
The procedure of share transfer system is explained comprehensively in the report. SEBI Clause 49 (VIII.E.5), 2015
Disclosure on the information on the shareholding pattern of shares held by directors is made. SEBI Clause 49 (XII.9), 2015
Information about the proportion of dematerialised shares held by bank given in its annual report. SEBI Clause 49 (XII.9), 2015
Bank disclose the policy on dealing with Related Party Transactions. SEBI Clause 49 (VIII.A), 2015
(continued on next page)
10
Table B1 (continued)
Dimensions Indicators in each dimension ( yr : r 1, ..., s ) Regulatory/Legal Clause

(Im: m 1, ..., 6 )
VI. Disclosure and Transp- Bank disclose the significant accounting policies adopted in Schedule- 17 of the annual report. SEBI Clause 49 (VIII.B), 2015
arency (8) Bank has a separate section on Management Discussion and Analysis as a part of annual report. SEBI Clause 49 (VIII.D), 2015
CEO/CFO certify to the board that board has complied accounting standards and code of conduct set by the SEBI Clause 49 (IX), 2015
bank.
Details of non-compliance by the company, penalties, and strictures imposed on the company by Stock SEBI Clause 49 (XII.7.ii), 2015
Exchange or SEBI or any statutory authority, on any matter related to capital markets, during the last three
years is disclosed in the annual report.
Bank disclose information regarding the ways and means by which shareholders are informed. SEBI Clause 49 (XII.8), 2015
Bank disclose the details about resignation or cessation of directors along with the detailed reasons of SEBI Clause 49 (VIII.F), 2015
resignation in report.
A brief resume of new director or re-appointed director is included for the information to the shareholders SEBI Clause 49 (VIII.G), 2015
Total dimensions of corpo- Total corporate governance norms/indicators = 58
rate governance = 6
Source: Authors' compilation.
Appendix C
Tables C1 provides the optimal idiosyncratic weights (specific to dimension and for sampled banks) obtained using BoD model restricted at 10%
lower bound. Table C2 reports the corporate governance index values for banks obtained corresponding to different weight restrictions on lower
bound in BoD model.
Table C1
Bank-specific weights generated for individual dimensions of CGI0.10.
Dimensions I II III IV V VI
Bank Bank name Board Effectiveness Audit Function Risk Management Remuneration Shareholder rights and informa- Disclosure and
code (w1) (w2) (w3) (w4) tion (w5) Transparency (w6)
B1 Allahabad Bank 0.1 0.1 0.1 0.1 0.1 0.5

B2 Andhra Bank 0.139147 0.1 0.1 0.1 0.460853 0.1
B3 Axis Bank Ltd. 0.183612 0.156672 0.359716 0.1 0.1 0.1
B4 Bandhan Bank Ltd. 0.1 0.103993 0.496007 0.1 0.1 0.1
B5 Bank of Baroda 0.1 0.1 0.1 0.1 0.5 0.1
B6 Bank of India 0.1 0.1 0.1 0.1 0.1 0.5
B7 Bank of Maharashtra 0.1 0.1 0.1 0.1 0.5 0.1
B8 Canara Bank 0.1 0.1 0.1 0.1 0.5 0.1
B9 Catholic Syrian Bank 0.1 0.1 0.5 0.1 0.1 0.1
Ltd.
B10 Central Bank of India 0.1 0.1 0.1 0.1 0.1 0.5
B11 City Union Bank Ltd. 0.218904 0.381096 0.1 0.1 0.1 0.1
B12 Corporation Bank 0.1 0.1 0.1 0.1 0.5 0.1
B13 DCB Bank Ltd. 0.443328 0.156672 0.1 0.1 0.1 0.1
B14 Dena Bank 0.1 0.1 0.1 0.1 0.1 0.5
B15 Dhanalakshmi Bank Ltd 0.5 0.1 0.1 0.1 0.1 0.1
B16 Federal Bank Ltd 0.191851 0.408149 0.1 0.1 0.1 0.1
B17 HDFC Bank Ltd. 0.191851 0.408149 0.1 0.1 0.1 0.1
B18 ICICI Bank Ltd. 0.1 0.1 0.1 0.1 0.5 0.1
B19 IDBI Bank Ltd. 0.1 0.1 0.5 0.1 0.1 0.1
B20 IDFC Bank Ltd. 0.1 0.1 0.5 0.1 0.1 0.1
B21 Indian Bank 0.1 0.1 0.5 0.1 0.1 0.1
B22 Indian Overseas Bank 0.1 0.1 0.1 0.1 0.5 0.1
B23 IndusInd Bank Ltd. 0.5 0.1 0.1 0.1 0.1 0.1
B24 Jammu & Kashmir Bank 0.1 0.1 0.1 0.1 0.5 0.1
Ltd.
B25 Karnataka Bank Ltd. 0.1 0.1 0.1 0.5 0.1 0.1
B26 Karur Vysya Bank Ltd 0.5 0.1 0.1 0.1 0.1 0.1
B27 Kotak Mahindra Bank 0.187731 0.156672 0.1 0.355596 0.1 0.1
Ltd.
B28 Lakshmi Vilas Bank Ltd. 0.139147 0.1 0.1 0.460853 0.1 0.1
B29 Oriental Bank of 0.1 0.1 0.1 0.1 0.5 0.1
Commerce
B30 Punjab and Sind Bank 0.1 0.1 0.1 0.1 0.5 0.1
B31 Punjab National Bank 0.1 0.1 0.1 0.1 0.5 0.1
B32 RBL Bank Ltd. 0.127231 0.1 0.1 0.472769 0.1 0.1
B33 South Indian Bank Ltd. 0.13294 0.1 0.46706 0.1 0.1 0.1
B34 State Bank of India 0.1 0.1 0.5 0.1 0.1 0.1
B35 Syndicate Bank 0.1 0.1 0.1 0.1 0.1 0.5
B36 UCO Bank 0.1 0.1 0.1 0.1 0.1 0.5
B37 Union Bank of India 0.1 0.1 0.1 0.1 0.5 0.1
B38 United Bank of India 0.1 0.5 0.1 0.1 0.1 0.1
(continued on next page)
11
Table C1 (continued)
Dimensions I II III IV V VI
Bank Bank name Board Effectiveness Audit Function Risk Management Remuneration Shareholder rights and informa- Disclosure and
code (w1) (w2) (w3) (w4) tion (w5) Transparency (w6)
B39 Vijaya Bank 0.1 0.1 0.1 0.1 0.5 0.1

B40 Yes Bank Ltd. 0.283951 0.24229 0.1 0.1 0.1 0.173759
Average 0.16 0.14 0.18 0.13 0.23 0.16
Table C2
Sensitivity analysis of corporate governance index – Alternative CGI for different weight restrictions on lower bound varying from 0.05 to 0.16.
Bank code CGI0.05 CGI0.06 CGI0.07 CGI0.08 CGI0.09 CGI0.10 CGI0.11 CGI0.12 CGI0.13 CGI0.14 CGI0.15 CGI0.16 Mean S.D
B1 0.9678 0.9617 0.9556 0.9495 0.9436 0.9377 0.9319 0.9261 0.9205 0.9149 0.9094 0.9039 0.9352 0.0210
B2 0.9756 0.9708 0.9661 0.9615 0.9569 0.9523 0.9478 0.9433 0.9389 0.9345 0.9301 0.9254 0.9502 0.0164
B3 0.9905 0.9887 0.9868 0.9849 0.9831 0.9812 0.9794 0.9776 0.9757 0.9735 0.9709 0.9682 0.9800 0.0071
B4 0.9314 0.9188 0.9065 0.8946 0.8830 0.8716 0.8606 0.8498 0.8393 0.8291 0.8191 0.8093 0.8678 0.0400
B5 0.9645 0.9577 0.9510 0.9443 0.9378 0.9314 0.9250 0.9188 0.9126 0.9065 0.9005 0.8945 0.9287 0.0229
B6 0.9517 0.9426 0.9336 0.9248 0.9162 0.9078 0.8995 0.8914 0.8834 0.8755 0.8678 0.8602 0.9045 0.0300
B7 0.9706 0.9649 0.9593 0.9538 0.9483 0.9428 0.9375 0.9322 0.9270 0.9218 0.9167 0.9116 0.9405 0.0193
B8 0.9672 0.9609 0.9547 0.9486 0.9425 0.9366 0.9307 0.9248 0.9191 0.9134 0.9078 0.9022 0.9340 0.0213
B9 0.9531 0.9442 0.9355 0.9270 0.9186 0.9103 0.9023 0.8943 0.8865 0.8788 0.8713 0.8639 0.9071 0.0292
B10 0.9569 0.9487 0.9407 0.9327 0.9250 0.9173 0.9098 0.9024 0.8951 0.8880 0.8809 0.8740 0.9143 0.0272
B11 0.9832 0.9799 0.9766 0.9734 0.9702 0.9670 0.9638 0.9606 0.9563 0.9506 0.9450 0.9395 0.9638 0.0139
B12 0.9339 0.9217 0.9098 0.8982 0.8870 0.8760 0.8652 0.8548 0.8445 0.8346 0.8248 0.8153 0.8721 0.0388
B13 1.0000 1.0000 1.0000 1.0000 1.0000 0.9984 0.9962 0.9940 0.9917 0.9882 0.9841 0.9800 0.9944 0.0070
B14 0.9544 0.9458 0.9373 0.9290 0.9208 0.9128 0.9049 0.8972 0.8896 0.8821 0.8747 0.8674 0.9097 0.0285
B15 1.0000 1.0000 1.0000 1.0000 0.9962 0.9912 0.9864 0.9815 0.9767 0.9720 0.9673 0.9602 0.9859 0.0143
B16 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9983 0.9957 0.9930 0.9989 0.0023
B17 0.9901 0.9881 0.9862 0.9842 0.9823 0.9803 0.9784 0.9765 0.9746 0.9711 0.9667 0.9624 0.9784 0.0086
B18 0.9740 0.9690 0.9640 0.9591 0.9542 0.9494 0.9446 0.9398 0.9352 0.9305 0.9259 0.9214 0.9473 0.0173
B19 0.9665 0.9601 0.9537 0.9475 0.9413 0.9352 0.9292 0.9232 0.9173 0.9115 0.9058 0.9002 0.9326 0.0217
B20 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.0000
B21 0.9710 0.9654 0.9598 0.9544 0.9490 0.9436 0.9383 0.9331 0.9279 0.9228 0.9177 0.9127 0.9413 0.0191
B22 0.9630 0.9559 0.9490 0.9421 0.9353 0.9286 0.9221 0.9156 0.9092 0.9029 0.8966 0.8905 0.9259 0.0238
B23 0.9911 0.9893 0.9876 0.9858 0.9827 0.9786 0.9745 0.9704 0.9664 0.9624 0.9584 0.9535 0.9751 0.0128
B24 0.8822 0.8784 0.8746 0.8708 0.8671 0.8634 0.8597 0.8560 0.8524 0.8488 0.8453 0.8418 0.8617 0.0132
B25 0.9689 0.9630 0.9571 0.9512 0.9455 0.9398 0.9341 0.9286 0.9231 0.9176 0.9123 0.9070 0.9373 0.0203
B26 1.0000 1.0000 1.0000 0.9981 0.9934 0.9882 0.9830 0.9779 0.9728 0.9678 0.9629 0.9567 0.9834 0.0156
B27 0.9581 0.9501 0.9423 0.9346 0.9270 0.9195 0.9122 0.9049 0.8978 0.8900 0.8818 0.8737 0.9160 0.0273
B28 0.9691 0.9631 0.9572 0.9514 0.9456 0.9400 0.9344 0.9288 0.9233 0.9179 0.9126 0.9069 0.9375 0.0203
B29 0.9728 0.9675 0.9623 0.9571 0.9520 0.9470 0.9420 0.9370 0.9321 0.9273 0.9225 0.9177 0.9448 0.0180
B30 0.9651 0.9584 0.9518 0.9453 0.9389 0.9325 0.9263 0.9201 0.9140 0.9080 0.9021 0.8962 0.9299 0.0226
B31 0.9772 0.9727 0.9683 0.9640 0.9596 0.9554 0.9511 0.9469 0.9427 0.9386 0.9345 0.9304 0.9535 0.0153
B32 0.9696 0.9637 0.9580 0.9522 0.9466 0.9410 0.9355 0.9300 0.9246 0.9193 0.9140 0.9088 0.9386 0.0199
B33 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9992 0.9999 0.0002
B34 0.8973 0.8937 0.8901 0.8865 0.8829 0.8794 0.8759 0.8725 0.8690 0.8656 0.8622 0.8589 0.8778 0.0126
B35 0.9540 0.9453 0.9368 0.9284 0.9202 0.9121 0.9041 0.8963 0.8886 0.8811 0.8737 0.8664 0.9089 0.0287
B36 0.9598 0.9521 0.9446 0.9372 0.9299 0.9227 0.9156 0.9087 0.9018 0.8950 0.8884 0.8818 0.9198 0.0256
B37 0.9666 0.9601 0.9538 0.9475 0.9414 0.9353 0.9293 0.9233 0.9175 0.9117 0.9060 0.9003 0.9327 0.0217
B38 0.9106 0.9086 0.9065 0.9044 0.9024 0.9004 0.8983 0.8963 0.8943 0.8923 0.8903 0.8884 0.8994 0.0073
B39 0.9649 0.9582 0.9516 0.9450 0.9386 0.9322 0.9259 0.9197 0.9136 0.9076 0.9016 0.8958 0.9296 0.0227
B40 1.0000 1.0000 0.9996 0.9982 0.9969 0.9955 0.9942 0.9928 0.9915 0.9901 0.9887 0.9871 0.9945 0.0045
References holding companies. J Financ 2013;68(5):1757–803.

[6] Song FM, Li L. Bank governance: concepts and measurements. Barth JR, Lin C,
Wihlborg C, editors. Research handbook on international banking and governance,
[1] Heremans D. Corporate governance issues for banks: a financial stability perspective vols. 17–41. Edward Elgar Publishing; 2012.
(February 2007). K.U. Leuven, Faculty of Economics, Department of Economics; [7] Tarchouna A, Jarraya B, Bouri A. How to explain non-performing loans by many
2007 CES- Discussion Paper Series (DPS) 07.07, Available at:. https://lirias2repo. corporate governance variables simultaneously? A corporate governance index is
kuleuven.be/bitstream/id/26493/;jsessionid= built to US commercial banks. Res Int Bus Finance 2017;42(C):645–57.
97BF8C5940A273D4EDB307044208213F. [8] Greco S, Ishizaka A, Tasiou M, Torrisi G. On the methodological framework of
[2] Sarkar J, Sarkar S, Sen K. A corporate governance index for large listed companies composite indices: a review of the issues of weighting, aggregation, and robustness.
in India. 2012 Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id= Soc Indicat Res 2018https://doi.org/10.1007/s11205-017-1832-9.
2055091. [9] OECD. Handbook on constructing composite indicators: methodology and users
[3] Love I, Rachinsky A. Corporate governance and bank performance in emerging guide. JRC European Commission; 2008 Available at: http://www.oecd.org/sdd/
markets: evidence from Russia and Ukraine. Emerg Mark Finance Trade 42495745.pdf.
2015;51(2):S101–21. [10] Puyenbroeck T. On the output orientation of the benefit-of-the-doubt-model. Soc
[4] Andries AM, Căpraru B, Nistor S. Corporate governance and efficiency in banking: Indicat Res 2018;139(2):415–31.
evidence from emerging economies. Appl Econ 2018;50(34–35):3812–32. [11] Rogge N, Puyenbroeck TV. An introduction to ‘benefit-of-the-doubt’ composite in-
[5] Ellul A, Yerramilli V. Stronger risk controls, lower risk: evidence from U.S. bank dicators. Soc Indicat Res 2007;82(1):111–45.
12
[12] Zhou P, Ang BW, Poh KL. A mathematical programming approach to constructing envelopment analysis: a handbook of empirical studies and application. New York:
composite indicators. Ecol Econ 2007;62(2):291–7. Springer Science+Business Media; 2016. p. 367–438 [Chapter 12].
[13] Despotis DK. A reassessment of the Human Development Index via data envelop- [45] Charles V, Díaz G. A non-radial DEA index for Peruvian regional competitiveness.
ment analysis. J Oper Res Soc 2005;56(8):969–80. Soc Indicat Res: Int Interdiscipl J Qual Life Measur 2017;134(2):747–70.
[14] Cherchye L, Ooghe E, Van Puyenbroeck T. Robust human development rankings. J [46] Allen R, Athanassopoulos AD, Dyson RG, Thanassoulis E. Weight restrictions and
Econ Inequal 2008;6(4):287–321. value judgements in DEA: evolution, development and future directions. Ann Oper
[15] Murias P, Martinez F, Miguel C. An economic wellbeing index for the Spanish Res 1997;73(0):13–34.
provinces: a data envelopment analysis approach. Soc Indicat Res [47] Athanassoglou S. Revisiting worst-case DEA for composite indicators. Soc Indicat
2006;77(3):395–417. Res 2016;128(3):1259–72.
[16] Hermans E, Van den Bossche F, Wets G. Combining road safety information in a [48] Wong YHB, Beasley JE. Restricting weight flexibility in data envelopment analysis.
performance index. Accid Anal Prev 2008;40(4):1337–44. J Oper Res Soc 1990;41(9):829–35.
[17] Antonio J, Martin R. An index of child health in the least developed countries [49] Financial Stability Board. Thematic Review on corporate governance. 2017
(LDCs) of Africa. Soc Indicat Res 2012;105(3):309–22. Available at: http://www.fsb.org/wp-content/uploads/Thematic-Review-on-
[18] Badasyan N, Shideler D, Silva S. Broadband achievement index: moving beyond Corporate-Governance.pdf.
availability. Telecommun Pol 2011;35(11):933–50. [50] Alchian AA, Demsetz H. Production, information costs, and economic organization.
[19] Giambona F, Vassallo E. Composite indicator of financial development in a benefit- Am Econ Rev 1972;62(5):777–95.
of-doubt approach. Econ Notes 2013;42(2):171–202. [51] Basel Committee on Banking Supervision (BCBS). Corporate governance principles
[20] Gaaloul H, Khalfallah S. Application of the ‘benefit-of-the-doubt’ approach for the for banks. 2015 Available at: http://www.bis.org/bcbs/publ/d328.html.
construction of a digital access indicator: a revaluation of the ‘Digital Access Index’. [52] Tadesse S. The economic value of regulated disclosure: evidence from the banking
Soc Indicat Res 2014;118(1):45–56. sector. William Davidson Institute; 2006 Working Paper Number 875, Available at:
[21] Martin JC, Mendoza C, Roman C. A DEA travel–tourism competitiveness index. Soc https://pdfs.semanticscholar.org/ce3c/5fed257bdaaf8dc1eb832b5c07d5c2c9a306.
Indicat Res 2017;130(3):937–57. pdf.
[22] Klapper LF, Love I. Corporate governance, investor protection, and performance in [53] Bhagat S, Bolton B, Romano R. The effect of corporate governance on performance
emerging markets. J Corp Financ 2004;10(5):703–28. The Robert W. Kolb Series in Finance In: Baker HK, Anderson R, editors. 2010.
[23] Shleifer A, Vishny RW. A survey of corporate governance. J Financ Corporate governance- a Synthesis of theory, research, and practiceNew Jersey: John
1997;52(2):737–83. Wiley & Sons; 2010. p. 97–122.
[24] Adams R, Mehran H. Is corporate governance different for bank holding companies? [54] Vidoli F, Fusco E. Compind: composite indicators functions based on frontiers in R
Econ Pol Rev 2003;9(1):123–42. (Compind package version 2.0). 2018 Available at: https://cran.r-project.org/web/
[25] Adams RB. Governance and the financial crisis. Int Rev Financ 2011;12(1):7–38. packages/Compind/vignettes/Compind_vignette.pdf.
[26] Caprio G, Levine R. Corporate governance of banks: concepts and international [55] Wittrup J, Bogetoft P. Benefit of doubt approach to case weighting: improving and
observations. 2002 Paper presented at the Global Corporate Governance Forum simplifying the assessment of workload in courts 2017 Technical report, Berlin,
Research Network Meeting, 5 April. Available at: http://citeseerx.ist.psu.edu/ Available at: http://research.cbs.dk/en/publications/benefit-of-doubt-approach-to-
viewdoc/download;jsessionid=EFCC60262B8C4CCAC723F6F414E33FC2?doi= case-weighting(f8cf1bb7-7473-4361-8047-e718ac08fa40).html.
10.1.1.19.254&rep=rep1&type=pdf. [56] Ministry of Corporate Affairs. The companies act 2013. New Delhi, India:
[27] De Haan J, Vlahu R. Corporate governance of banks: a survey. J Econ Surv Government of India; 2013 Available at: http://www.mca.gov.in/Ministry/pdf/
2016;30(6):228–77. CompaniesAct2013.pdf.
[28] Mehran H, Morrison A, Shapiro J. Corporate governance and banks: what have we [57] Securities and Exchange Board of India. Clause 49 [Listing Obligation and
learned from the financial crisis? Federal Reserve Bank of New York Staff Report Disclosure Requirements] Regulations 2015. 2015.https://www.sebi.gov.in/sebi_
No. 502 2011. p. 1–42. data/attachdocs/1441284401427.pdf. Circular No. CIR/CFD/POLICY CELL/2/
[29] Bozec R, Bozec Y. The use of governance indexes in the governance-performance 2014.
relationship literature: international evidence. Can J Admin Sci 2012;29(1):79–98. [58] Andries AM, Brown M. Credit booms and busts in emerging markets: the role of
[30] Mülbert PO. Corporate governance of banks after the financial crisis - theory, evi- bank governance and risk management. Econ. Transition 2017;25(3):377–437.
dence, reforms. ECGI working paper series in law. 2010. Working Paper No.151/
2010. Rachita Gulati is serving as Assistant Professor of Economics in the Department of
[31] Becht M, Bolton P, Röell A. Why bank governance is different. Oxf Rev Econ Pol Humanities and Social Sciences, Indian Institute of Technology (IIT), Roorkee,
2011;27(3):437–63. Uttarakhand, India since 2013. She has completed her post-doctoral research from
[32] Peni E, Vähämaa S. Did good corporate governance improve bank performance London School of Economics and Political Sciences, London, United Kingdom in 2018.
during the financial crisis? J Financ Serv Res 2012;41(1–2):19–35. She is teaching money, banking & financial markets, econometrics, and efficiency &
[33] Brown LD, Caylor ML. Corporate governance and firm valuation. J Account Public productivity analysis at masters' level. Her research interests include performance eva-
Policy 2006;25(4):409–34. luation and measurement using data envelopment analysis and its application in banking
[34] Gompers P, Ishii JL, Metrick A. Corporate governance and equity prices. Q J Econ and education sector; governance, regulation and stability of financial institutions; bank
2003;118(1):107–55. competition policy assessment; among others. She has published theoretical and em-
[35] Zagorchev A, Gao L. Corporate governance and performance of financial institu- pirical research papers related this area in the peer-reviewed journals.
tions. J Econ Bus 2015;82(C):17–41.
[36] Aggarwal R, Erel I, Ferreira M, Matos P. Does governance travel around the world?
Ruth Kattumuri is Co-Director of the India Observatory at London School of Economics
Evidence from institutional investors. J Financ Econ 2011;100(1):154–81.
[37] Farrell MJ. The measurement of productive efficiency. J R Stat Soc Ser A and Political Sciences, London, United Kingdom. Dr Kattumuri is responsible for devel-
1957;120(3):253–90. oping multi-disciplinary research and programmes of the India Observatory. With over
[38] Charnes A, Cooper WW, Rhodes E. Measuring the efficiency of decision making two decades of experience in higher education in UK and India, Ruth has pioneered
units. Eur J Oper Res 1978;2(6):429–44. several innovative knowledge development programmes. Her research interests are
technology and innovation; population and development issues including growth and
[39] Melyn W, Moesen W. Towards a synthetic indicator of macroeconomic perfor-
mance: unequal weighting when limited information is available. Public Economics inclusion; and climate change policy.
research paper- leuven. CES, KU Leuven; 1991. ZDB-ID 26175691, No. 17.
[40] Cherchye L, Moesen W, Van Puyenbroeck T. Legitimately diverse, yet comparable: Sunil Kumar is currently serving as Professor in the Faculty of Economics, South Asian
on synthesizing social inclusion performance in the EU. J Common Mark Stud University (SAU), New Delhi, India. He has more than 20 years of teaching experience in
2004;42(5):919–55. subjects related to Quantitative Methods, Macroeconomics, and Econometrics at both
[41] Cherchye L, Moesen W, Rogge N, Van Puyenbroeck T. An introduction to ‘benefit of undergraduate and graduate levels. Currently, he is teaching Efficiency and Productivity
the doubt’ composite indicators. Soc Indicat Res 2007;82(1):111–45. Analysis, and Topics in Applied Econometrics at graduate level. His research interests
[42] Lovell CAK, Pastor JT, Turner JA. Measuring macroeconomic performance in the include data envelopment analysis, stochastic frontier analysis, bank efficiency and in-
OECD: a comparison of European and non-European countries. Eur J Oper Res dustrial productivity. He has published more than three dozen research papers in journals
1995;87(3):507–18. of national and international repute. He authored a research book titled, “Deregulation
[43] Witte KD, Rogge N. Accounting for exogenous influences in performance evalua- and Efficiency of Indian Banks”. His professional memberships include The Econometric
tions of teachers. Econ Educ Rev 2011;30(4):41–653. Society (international society), Indian Society of Regional Science and Indian Economic
[44] Thanassoulis E, Witte KD, Johnes J, Johnes G, Karagiannis G, Portela CS. Association.
Applications of data envelopment analysis in education. In: Zhu J, editor. Data
13

A Non-Parametric Index of Corporate Governance in The Banking Industry - An Application To Indian Data

Uploaded by

Copyright:

Available Formats

A Non-Parametric Index of Corporate Governance in The Banking Industry - An Application To Indian Data

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Non-Parametric Index of Corporate Governance in The Banking Industry - An Application To Indian Data

Uploaded by

Copyright:

Available Formats

Socio-Economic Planning Sciences xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Socio-Economic Planning Sciences

A non-parametric index of corporate governance in the banking industry: An

ARTICLE INFO ABSTRACT

1. Introduction employed to derive a composite index of corporate governance for

Source: Authors' elaboration.

Fig. 1. The framework for bank governance.

construct not only dimensional indices, but also an overall index of 9

B1 Allahabad Bank 0.800 0.778 0.600 0.800 0.909 1.000

Source: Authors' calculations.

B1 Allahabad Bank 0.9377 21 0.9678 21 −0.1031 21 0.8145 27

Source: Authors' calculations.

4.3.1. Choice of aggregation method and its impact on ranking of banks on

Constrained BoD (5%) 1.000 – – –

Note: ∗∗∗ indicate significance at 1% level.

Unrestricted (0%) 0.00 0.03 0.12 0.43 0.27 0.15 37

Source: Authors' calculations.

R codes used for constrained BoD

Dimensions Indicators in each dimension ( yr : r 1, ..., s ) Regulatory/Legal Clause

Dimensions Indicators in each dimension ( yr : r 1, ..., s ) Regulatory/Legal Clause

Source: Authors' compilation.

B1 Allahabad Bank 0.1 0.1 0.1 0.1 0.1 0.5

B39 Vijaya Bank 0.1 0.1 0.1 0.1 0.5 0.1

Source: Authors' calculations.

Source: Authors' calculations.

References holding companies. J Financ 2013;68(5):1757–803.

You might also like