Int. J. Corporate Governance, Vol. 8, No. 1, 2017
Income diversification, bank stability and owners
identity: international evidence from emerging
economies
Naima Lassoued*
Higher School of Business,
University of Manouba,
Campus Universitaire 2010,
Manouba, Tunisia
Email: naima.lassoued@sesame.com.tn
*Corresponding author
Houda Sassi
Institute of High Business Studies,
Carthage University, 2016,
Tunisia
Email: houdasassi01@yahoo.fr
Abstract: This paper examines how ownership structure affects the
relationship between income diversification and bank stability in emerging
markets. Examining data on more than 171 banks from 13 Middle East and
North Africa countries examined over the 2006–2012 period, the impact of
foreign and state ownership on income diversification and bank stability is
jointly studied by a 3SLS model. The main finding is that diversification,
across interest and non-interest activities and non-traditional banking activities,
increases bank stability. State-owned banks seem to take more risk by
diversifying less their income. Inversely, foreign ownership decreases risk by
diversifying their income. The findings have significant strategic implications
for bank managers, regulators and investors who share a common interest in
boosting bank stability.
Keywords: bank stability; emerging economies; foreign ownership; income
diversification; state ownership.
Reference to this paper should be made as follows: Lassoued, N. and Sassi, H.
(2017) ‘Income diversification, bank stability and owners identity: international
evidence from emerging economies’, Int. J. Corporate Governance, Vol. 8,
No. 1, pp.61–80.
Biographical notes: Dr. Naima Lassoued is an Associate Professor of
Accounting and Finance at Higher School of Business Tunisia. She holds her
Habilitation in Finance from University of Manouba. Her main research
interests include corporate governance, financial reporting, Islamic finance, risk
management and earnings management and firm performance. She has
published in Research in international business and finance (Elsevier); Journal
of Accounting Auditing and Performance Evaluation (Inderscience), corporate
governance: The international journal of business in society, Research in
Accounting in Emerging Economies (Emerald Publishing).
Copyright © 2017 Inderscience Enterprises Ltd.
61
62
N. Lassoued and H. Sassi
Houda Sassi is a Teaching Assistant. She received her PhD in finance from the
Institute of High Business Studies (University of Carthage). Her research
interests are in corporate governance, banking institutions and earnings
management. She has published in Research in international business and
finance (Elsevier), Economics bulletins.
1
Introduction
Over the last few decades, financial institutions had been subject to increased
competition, intensified by information technology development, globalisation and
deregulation. These changes incited banks to take more risk in order to maximise their
expected profit and to adopt new lines of businesses in addition to their traditional
interest-based activities, like loan granting (Keeley, 1990; Meslier et al., 2014). More
precisely, banks diversify their activities by moving towards non-interest income
activities like trading securities, brokerage, investment and other activities.
By referring to modern portfolio theory, income diversification seems to be a
valuable technique to reduce banking risk when all income components either negatively
or only slightly correlate (Chiorazzo et al., 2008; Busch and Kick, 2009). However, most
empirical studies predominantly in the US and Europe1 found that non-interest activities
are often associated with higher risk because of their unstable nature. These findings may
not apply to emerging/developing banking sectors, because the latter often have less
mature financial systems and different banking market structures and institutional and
regulatory backgrounds (Meslier et al., 2014). Relatively, little recent evidence addresses
this issue in developing countries and often reach mixed findings (Sanya and Wolfe,
2011a; Meslier et al., 2014; Zhou, 2014). The mixed empirical results may be attributed
to the omission of reliable variables like ownership structure. Indeed, according to
Pennathur et al. (2012), studying income diversification and banking risk in emerging
economies must take into account ownership structure since these countries are known by
dominant controlling shareholders. These latter have a significant impact on corporate
decisions because of their sizable capital stakes and weak law enforcement (LaPorta
et al., 1998). In this regard, risk-taking is expected to be more practiced in firms with
concentrated ownership than in firms with dispersed ownership. Accordingly,
diversification may be an interesting channel to control risk (Jensen and Meckling, 1976).
However, when majority stakeholders do not own diversified portfolios, they will not be
incited to increase banking risk. Thereby, controlling shareholders do not have the same
motivations, objectives, means and control efficiency. It would be appropriate to study
the identity of controlling shareholders.
Our purpose in this study to extend the literature on the effect of ownership structure,
including state and foreign ownership, on the relationship between revenue
diversification and banking risk in the developing Middle East and North Africa
(henceforth MENA countries). Indeed, the banking sector in the MENA region seems to
be an interesting area to test our research problem, since it has undergone financial
openness and liberalisation aiming at limiting state intervention and establishing a legal
and an institutional framework that would attract foreign investments (Ben Naceur et al.,
2007). This policy has shifted the ownership of large shares of banks from government to
private control and from domestic to foreign control. Such changes took place as
Income diversification, bank stability and owners identity
63
governments privatised many of their state-owned banks. The most important
international bank groups have generally set up a wide presence in the MENA region
through subsidiaries and are competing with domestic banks. Foreign presence improves
human capital skills and may lead to more diverse products, better use of up-to-date
technologies and knowledge transfer forcing domestic banks to schedule major structural
reforms in order to compete on an equal basis with their counterparts. In addition to the
intense competition of foreign banks, banks in the MENA region also face increased
competition from Islamic banks, likely to encourage income diversification (Almarzoqi
et al., 2015). Consequently, the study of income diversification and risk in the banking
sector assumes particular relevance in developing countries, since it would shed light on
the eventual consequence of the liberalisation and deregulation process, contributing thus
to an increase in financial risk-taking. Therefore, policy makers need an accurate
understanding of the impact of liberalisation and privatisation on income diversification
and bank risk, in order to identify the proper financial measures.
We examine the effect of ownership structure on the relationship between income
diversification and bank risk using data on more than 171 banks from 13 MENA
countries over the period 2006–2012. We specify a 3SLS model in which the impact of
foreign and state ownership on income diversification and banking risk is jointly
examined to solve the endogeneity problem that arises from the relationship between
risk-taking and diversification. Our findings indicate that diversification, across interest
and non-interest activities and non-traditional banking activities, decreases bank risk.
State-owned banks seem to take more risk by diversifying less their income. Inversely,
foreign ownership decreases risk through diversifying their income. Then, we came to
confirm that diversification is a channel through which controlling shareholders influence
banking risk, and diversification level depends on the monitoring practiced by different
shareholders.
Our study contributes to the literature in several ways. First, we extend the recent line
of research to provide international evidence on the impact of income diversification on
banking risk by studying unexplored emerging markets. Second, we extend the results of
previous studies (Pennathur et al., 2012; Sanya and Wolfe, 2011b). Examining a sample
of Indian banks, Pennathur et al. (2012) studied the effect of income diversification on
banking risk for state-owned and foreign-owned banks. Unlike their study, we investigate
the effect of the type of ownership (foreigner and state) on both diversification and risk
while controlling for potential endogeneity between risk and diversification. Unlike
Sanya and Wolfe (2011b) who examined the effect of ownership concentration on
income diversification and banking risk in European banks, we explore the effect of
shareholder identities as most banks in the MENA region already have a concentrated
structure (Rocha et al., 2011; Lassoued et al., 2016). Thus, examining shareholder
identity would provide more and relevant results.
The rest of the paper is structured as follows. Section 2 presents banking systems in
the MENA region. In Section 3, we review the relevant literature. In Section 4, we
describe our variables and methodology. In Section 5, we present and discuss our
findings. Section 6 concludes the paper.
64
2
N. Lassoued and H. Sassi
Banking systems in MENA countries
In most MENA economies (like other emerging economies), financial repression policies
had been adopted for many decades. The banking sector, which is the main financing
source, was traditionally a highly protected industry living on spreads achieved on
regulated deposit and lending rates and pervasive restrictions on foreign entry (Almarzoqi
et al., 2015). During 80 years, several important reforms have been undertaken in the
region under the supervision of the international monetary fund. Most of MENA
governments have undertaken a comprehensive financial reform agenda by nationally
deregulating the banking industry and opening the financial markets to foreign
competition. As a result of opening economies, many state banks have been privatised;
banks have more freedom to expand their activities and entry barriers for foreign
investors have been abandoned.
As shown in Table 1, banks dominate MENA economies. Indeed, in most countries
banking assets reach near 100%. Lebanese banking assets account more than three times
the Gross Domestic Product (GDP). The smallest banking sectors are those of Algeria,
Oman and Saudi Arabia.
Table 1
Some features of banking sectors in MENA countries
Bank
assets
to
GDP
Bank
concentration
Stateowned
banks
Foreignowned
banks
Limitations
on foreign
bank entry
Noninterest
income to
total
income
Restrictions
on banking
activities
Algeria
0.622
0.753
0.930
0.047
4
0.3600
5.5
Bahrain
–
0.819
0.024
0.617
4
0.3515
7.33
Egypt
1.070
0.577
0.661
0.131
4
0.4401
7.33
Jordan
2.000
0.918
0.000
0.472
3.5
0.2844
8
Kuwait
0.980
0.818
–
–
3
0.3186
5.67
Lebanon
3.290
0.493
0.010
0.345
3.5
0.3155
8
Morocco
1.100
0.711
0.293
0.204
4
0.3296
8
Oman
0.680
0.741
0.000
0.115
2.5
0.2941
8.33
Qatar
1.100
0.850
–
0.149
–
0.3546
3.67
Saudi
Arabia
0.710
0.542
0.204
0.201
4
0.3323
8.5
Tunisia
1.090
0.424
0.420
0.157
4
0.3302
8
Turkey
0.940
0.477
0.314
0.089
4
0.3164
7.5
UAE
1.220
0.539
0.350
0.270
3
0.3566
7
Bank assets to GDP, Bank concentration and Non-interest income to total
income are extracted from Global Financial Development (2014). State-owned
banks, Foreign-owned banks, Limitations on foreign bank entry and
Restrictions on banking activities are collected from Barth et al. (2013).
State-owned banks column indicates the percentage of bank assets that are 50%
or more government-owned during 2006 to 2011 period.
Foreign-owned banks column displays the percent of the assets of banks where
foreigners own 50% between 2006 and 2011.
Income diversification, bank stability and owners identity
65
Limitations on foreign bank entry column indicates whether there are any
limitations placed on the ownership of domestic banks by foreign banks and
whether there are any limitations placed on the ability of foreign banks to enter
the domestic banking industry. Lower values point to greater restrictions on
foreign entries.
Restrictions on banking activities column measures the number of restrictions
imposed by national regulatory authorities on banking activities including
securities, insurance and real estate activities plus restrictions on banks owning
and controlling non-financial firms.
MENA banking is highly concentrated2. Indeed, the Jordanian banking sector is the most
concentrated. In fact, the three largest banks own, on average, more than 90% of the total
banking assets. The three largest banks in Qatar, Bahrain and Kuwait control more than
80% of banking assets. The Tunisian and Turkish banking sector are the less
concentrated. Indeed, the three largest banks own, on average, 42.2% of the total banking
assets in Tunisia and 47.7% in Turkey.
Column 3 displays percentage of the banking assets that are 50% or more
government-owned during 2006 to 2011 period. It seems that the Algerian state
monopolises an important part of banking assets in many countries (more than 90%)
followed by the Egyptian State, holding more than two-thirds of banking assets. By
contrast, Lebanon and Bahraini States own an insignificant part of banking assets while
the Jordanian and Omani states do not own any banking assets.
As for foreign ownership in column 4, on average, the percent of banking assets
indicates banks where foreigners owned 50% between 2006 and 2011. More than 50% of
Bahraini banking assets are owned by foreigners. Similarly, about the half of the banking
assets in Jordan are owned by foreigners. While, in Algeria foreigners own the lowest
banking assets (less than 5%) followed by Turkey where 16.45% are foreigner owned.
The remaining MENA nations fall in the middle, allowing only a partial ownership to
foreign owners. The growing presence of foreigners in MENA countries resulted in part
from abandoning barriers on foreign entries. Indeed, seven countries do not impose any
form of foreign entry (column 5) like Algeria, Bahrain, Egypt, Morocco, Saudi Arabia,
Tunisia and Turkey, according to the limitations on foreign banking index3. Oman is
more stringent on foreign entry and imposes more barriers on foreign banks. However,
foreign participation became considerably present in the banking sector in many
countries4.
The growing presence of foreign banks and the concentration of the banking sector
have significantly increased competitive pressures on banks. This has led to deep changes
in banking strategies inciting them to diversify their activities. Indeed, average noninterest income represents about the third of total income gained by banks. Egypt has the
highest portion of non-interest income (more than 40%) while Jordan has the lowest
portion.
Policy makers in the MENA region do not promote similar diversification activities.
In fact, many of them impose the same restrictions on the extent to which banks may
engage in many banking activities5. Saudi banks are the most restricted in terms of their
banking diversification. On average, there are more than 8 restrictions (out of 16).
Similarly, Lebanon, Morocco and Tunisia impose on average 8 restrictions on
diversification activities. However, Qatar restricts less banking activities (3.67).
As discussed above, banking sectors play an important role in financing MENA
economies. We notice many disparities in term of ownership structure. While many
66
N. Lassoued and H. Sassi
countries allow foreign entries, others still restrict them. Similarly, because of these
restrictions, income structure varies across these countries. Accordingly, there is a need
to investigate the effect of these disparities in term of ownership structure and
diversification activities on banking risk.
3
Literature review
3.1 Income diversification and banking risk
Modern portfolio theory stipulates that diversification reduces portfolio total risk when
assets are less correlated. Therefore, banks should diversify their revenue sources to
reduce their risk. Furthermore, Gurbuz et al. (2013) argue that banks can increase
shareholders’ value by shifting their focus from traditional income activities to noninterest income activities. Then, banks combine appropriately interest and non-interest
income to reduce risk in banking operations.
Revenue diversification brings many benefits for banks. First, the dramatic changes
in the financial industry, brought about by technological advances and deregulation, have
driven banks to build new skills so that they can capitalise on the first mover advantage in
choosing diversification activities (Elsas et al., 2010). Second, diversification generates
economies of scale for banks that allows improvements in cost/income ratios and
efficiency in input sharing (e. g. labour, technology and information) which represent a
major source of cost savings. At the same time, wider customer bases are reached
reducing thus cost per unit (Stein, 2002). Third, banks have an information advantage
about their customers compared to its competitors, thanks to the lending relationship.
Indeed, the lending relationship and knowledge about customer needs and the additional
requirements for non-interest financial products can be efficient incentives to match
customer needs. Fourth, diversification compensates for lower competitive interest
margins on traditional deposit/lending markets (Valverde and Fernandez, 2007).
Empirically, Sanya and Wolfe (2011a) examine a sample of 226 banks in 11
emerging economies during 2000 and 2007, and they find that diversification activities
decrease risk. In contrast to the traditional view, many studies found evidence indicating
that diversification is a risk source. Studying a sample of 472 large US commercial
banks, DeYoung and Roland (2001) found that non-interest income increases more bank
income volatility than traditional interest-based activities do. Similarly, Stiroh (2006)
shows that non-interest income is positively correlated with bank income volatility but
not with returns. Furthermore, Demirguç-Kunt and Huizinga (2009) examine the effect of
banking activity type and short-term funding strategies on banking risk and returns.
Examining a sample of 1334 banks in 101 countries, they found that universal banking
can be beneficial to diversification with very low risk levels and increasing returns.
However, banking strategies exclusively relying on generating non-interest income or
attracting non-deposit funding are very risky. Under this perspective, banks must focus
on single line of business because diversification costs might outweigh the benefits.
Chiorazzo et al. (2008) find that income diversification increases risk-adjusted returns for
the Italian banking market between 1993 and 2003.
One possible explanation to the harmful effect of income diversification advanced is
given by Acharya et al. (2006). The authors argue that there are diseconomies of scale
Income diversification, bank stability and owners identity
67
caused by poor monitoring incentives and a poorer loan portfolio quality when a risky
bank expands into additional industries and sectors. Moreover, extensive diversification
of income sources can reduce banking financial performance (Sahoo and Mishra, 2012).
Income diversification requires specialised managerial expertise to manage new risks
made possible by these additional activities. Finally, diversification may lead to higher
firm-specific risk because of increased complexity and lack of experience in managing
the newly diversified areas.
These mixed results may be explained by economic, cultural and institutional
dimensions of the studied markets or by omitting significant variables like ownership
structure, in particular in emerging market known by concentrated ownership.
3.2 Does ownership structure affect the income diversification and banking risk
relationship?
Controlling shareholders with a diversified portfolio seem to encourage managers to take
more risk for a higher expected profit (Shleifer and Vishny, 1986), and they can
compensate for losses by diversifying their portfolios. However, when controlling
shareholders own undiversified portfolios they will take less risk. Thereby, controlling
shareholders do not have the same motivations, objectives, means and control incentives.
In this regard, it will be more relevant to investigate the attitude of different types of
controlling shareholders. In the present study, we examine state and foreign ownership
that dominates the banking sector of the MENA region.
Regarding state-controlled banks, government ownership is expected to preserve
banks’ financial soundness and enhance good governance. Through its participation in
banks, the government achieves its social and political objectives. In the case of stateowned banks, the government finances projects create more jobs especially when its
projects could not obtain private financing (La Porta et al., 2002). State ownership is
considered as a source of inefficiency because of government bureaucracy and lack of
capital market monitoring (Lang and So, 2002). Therefore, managers may pursue their
own interests rather than acting in the best interest, which may result in negative effects
on performance and high risk level.
Many arguments have been advanced to explain excessive risk in state-owned banks.
For instance, Demirgüç-Kunt and Detragiache (2002) pointed out that state-owned banks
enjoy government protection because bankers may take more risk because losses and
excess costs are invariably covered by the government. Faccio et al. (2006) add that the
government can protect banks by either implicit or explicit financial and regulatory
support. Moreover, Dong et al. (2014) argue that the lending policy of state-owned banks
may target social than financial objectives. For example, they finance unprofitable and
risky projects, because it has social objectives like those undertaken by state-owned
firms. Finally, state-owned banks are essentially controlled by politicians, who may
transfer resources to their supporters (Shleifer and Vishny, 1986; Iannota et al., 2013).
The last two arguments suggest that state-owned banks might be seen as vehicles for
raising capital to finance projects with high social and political returns, but possibly with
high-risk and low financial returns.
Income diversification in public banks has not been sufficiently investigated, except
in the study of Pennathur et al. (2012). The authors conclude that public banks earn
significantly less fee-based income, and that government ownership negatively correlates
68
N. Lassoued and H. Sassi
with non-interest income sources. They also find that fee-based income significantly
reduces risk.
Two opposite hypothesis may explain the relationship between risk and income
diversification in public banks. According to the first hypothesis, lack of monitoring in
these banks give managers the opportunities to pursue their own interests rather than
acting in the best interest of owners. In this regard, income diversification is not an
important concern for state-controlled banks and managers deploy less effort than their
private counterparts to improve efficiency and attract customers to these banks.
The undiversified activities of state-owned banks may lead them to be more aggressive
risk-takers. However, under the second hypothesis, state-owned banks present high level
of risk generated by lending inefficient projects, thus managers may diversify income to
hedge risk.
As for foreign ownership in emerging markets, it brings about many benefits. Indeed,
entry of foreign banks improves human capital, skills and may lead to more diversified
products, better use of up-to-date technologies and knowledge transfer (Berger et al.,
2005). Foreign participation improves financial services and allows for an easier access
to international financial markets (Levine, 1996). Then, these comparative advantages
could favour foreign banks in managing operational and financial leverage, when shifting
towards non-interest activities. Conversely, foreign banks may also suffer from distance
problems, and large banks are disadvantaged in terms of the quality of ‘soft’ information
(Stein, 2002). Thus, foreign-owned banks may avoid diversifying their activities, because
they have fewer local connections and ineffective incentive schemes for managers to
maximise shareholder wealth (Berger et al., 2010). Meslier et al. (2014) find that foreignowned banks in developing countries generally suffer from lack of knowledge of the
local market and seem to specialise in non-interest income-activities rather than
traditional banking activities.
As for the impact of foreign ownership on risk-taking, it is perceived as a stimulator
of risk-taking for many reasons. First, foreign owners have the opportunity to diversify
better their risk internationally. Second, foreign banks are perceived more efficient and
take more risk compared to their domestic counterparts, because they have better access
to the capital market and are better able to serve the international clientele that is not
easily served by domestic banks (Berger et al., 2005). Third, in emerging countries, they
may have more expertise in collecting, evaluating and analysing quantitative information
on financial statements and be less exposed to political pressure.
Pennathur et al. (2012) study the impact of ownership structure on income
diversification-risk relationship in Indian banking sector. The authors add that foreign
banks report higher fee income that increases insolvency risk.
4
Empirical design
4.1 Data
Our original sample included 202 banks (listed and unlisted bank) operating in 13 MENA
countries (Algeria, Saudi Arabia, Bahrain, Egypt, Jordan, Kuwait, Lebanon, Morocco,
Oman, Qatar, Tunisia, Turkey and the United Arab Emirates) observed from 2002 to
2012. We apply some filtering rules to ensure data availability and sample homogeneity.
Income diversification, bank stability and owners identity
69
First, we exclude Central and cooperative banks. Second, we measure banking risk by
standard deviations of ROA (return on assets) and ROE (return on equity) over a moving
window of 4 years. By using data from 2002 to 2012, we are able to compute standard
deviations of returns for a seven-year period from 2006 to 2012. Thus, the study period is
reduced to 2006–2012. Third, we exclude banks with missing data of more than 4 years.
In order to accurately calculate the standard deviations of some variables, we defined risk
indicators. We ended up with an unbalanced panel of 171 banks distributed over the
2006–2012 period.
Bank data are collected from the Bankscope database. Regulation data are obtained
from Barth et al (2013) database and macroeconomic data are extracted from the World
Bank’s website.
4.2 Measurement of key variables
4.2.1 Diversification proxies
Following Mercieca et al. (2007), we construct two Herfindahl-Hirschman Indices. The
first (HHIREV) is based on a breakdown of net operating revenue into net interest income
(NET) and non-interest income (NON)
2 ⎛
2
⎛
⎞
⎞
HIIREV = ⎜⎝ NONNETOP ⎟⎠ + ⎜⎝ NET NETOP ⎟⎠
(1)
The second index (HHINON) measures diversification into non-interest income-generating
activities: Fee and commissions (COM), trading income (TRAD) and other operating
income (OTHOP) for each bank in each year. We note that NON = COM + TRAD +
OTHOP.
⎛
HII NON = ⎜ COM
⎝
⎞
NON ⎟⎠
2
⎛
+ ⎜ TRAD
⎝
⎞
NON ⎟⎠
2
⎛
+ ⎜ OTHO
⎝
⎞
2
NON ⎟⎠
(2)
When HHIREV and HHINON rise, bank income becomes more concentrated and less
diversified.
4.2.2 Ownership structure proxies
We create two ownership variables, which represent each bank type in our sample:
STAT_OWN: the capital stake held by the state (in percentage)
FOR_OWN: the capital stake held by foreigners (in percentage)
4.2.3 Banking risk proxies
We use two proxies of risk SD_ROA and SD_ROE that consist of the standard deviation
of ROA and the standard deviation of ROE computed over a moving window of 4 years.
Using data from 2002 to 2012, we were able to compute returns volatility for a sevenyear period, from 2006 to 2012.
We also compute a Z-score for bank i at year t indicating bank stability (Boyd
et al.,1993) as follows::
70
N. Lassoued and H. Sassi
ZSCORE it =
ROA i,( t − 4,t ) + CAR it
σ ( ROA )i,( t − 4,t )
(3)
With capital-to-asset ratio being the capital asset ratio.
4.3 Model specification
Portfolio theory suggests that diversification may influence risk. However, diversification
per se might share the same determinants of banking risk, since it is a decision taken by
the bank. Ownership structure may also be an endogenous outcome. Indeed, controlling
shareholders may affect diversification according to their portfolio diversification degree
and the risk aversion level. To address this endogeneity between banking risk, ownership
structure and the decision to diversify, we construct a 3SLS simultaneous equation
model6 described by two equations shown in Equations 4 and 5. Diversification, risk and
ownership are treated as endogenous variables.
Risk it = β 0 + β1Ownnershipit + β 2 Diversification it
+ β 3 Control variablesit + Year Dummies + µit
Diversification it = α 0 + α1Ownnershipit + α 2 Risk it
+ α 3 Control variables it + Year Dummies + ϑit
(4)
(5)
Equation (4) has either one of the banking risk measures (SD_ROA, SD_ROE and
ZSCORE) as the dependent variable. The model is estimated separately for the two
ownership structure variables STAT_OWN and FOREIGN_OWN. Two revenue
diversification measures are used (H_REV and H_NON). The model examines the
impact of ownership and diversification on banking risk.
We include a set of bank-and country-specific variables in our analysis to control the
sources of risk taking differences. Consistent with previous studies (Mercieca et al.,
2007; Deng et al., 2013; Lassoued et al., 2016), we include the control variables of bank
profitability, measured by ROA, capital-to-asset ratio (CAR) computed by the ratio of
equity book value to total assets and SIZE (natural logarithm of total assets).
We expect that ROA is negatively related to bank risk, since performance reflects
management quality. Then, poor management practices, in unprofitable banks, lead to an
increase of non-performing loans and then of bank risk. Moreover, low capital-to-assets
ratio indicates that banking risk is high, because capital protects banks when asset values
decline, on the other hand, higher capital ratio prevents bank failure (Deng et al., 2013).
The effect of bank size is expected to be positive under the “too big to fail” hypothesis
suggesting that larger banks tend to take more excessive risks.
As country-level control variables, we include GDP Growth (GDP_GR) and we
expect that the estimated coefficient is negative. Indeed, during periods of economic
expansion, borrowers need sufficient funds to service their debts, but during a recession,
the ability to service debt declines. Therefore banking risk will increase in recession
periods. Finally, we include deposit insurance (DEP_INS) as a dummy variable that takes
1 or 0 indicating whether the country has explicit DEP_INS or not (yes = 1; no = 0).
Deposit insurance can limit banking risk. Indeed, a DEP_INS system reduces bank risktaking incentives (Gropp and Vesala, 2001). However, bank managers may be
encouraged to take more risks in order to generate higher profits, and insurance will
Income diversification, bank stability and owners identity
71
cover a large part of the bank’s debts in case of non-payment (Angkinand and Wihlborg,
2010).
Equation (5) has either one of the income diversification measures (H_REV and
H_NON) as its dependent variable. We test the effect of ownership and banking risk on
income diversification. We include a set of control variable as factors that may affect
banking risk.
Following Campa and Kedia (2002) and Deng et al. (2013), we include profitability,
measured by ROA, bank-size (SIZE) and CAR as control variables in the diversification
model (Equation 5). We expect that profitability is inversely related to diversification,
because poorly performing banks seek to diversify to improve their income (Campa and
Kedia, 2002). Furthermore, CAR is expected to be negatively related because, wellcapitalised banks will take less risk and may therefore be less diversified (Stiroh and
Rumble, 2006). For SIZE indicators, we expect that the estimated coefficients will be
positive, because large banks have a greater capacity to diversify their income since they
have more opportunities to pursue a wider variety of loans and other activities (Sullivan
and Spong, 2007). Finally, we include GDP Growth (GDP_GR) to control for the effect
of economic growth on diversification. Indeed, Sanya and Wolfe (2011b) argue that it is
more profitable to diversify during periods of rapid economic growth.
We include three instruments in our model. The first instrument is the restrictions on
banking activities index, which indicates whether policy makers impose the same
restrictions on banking activities. The index, taken from Barth et al. (2013) influences
directly income diversification activities, since it describes the extent to which banks may
engage in many activities7 The second and third instruments are related to ownership
variables and refer to regulatory quality since a country’s institutional environment is
exogenous and closely related to ownership structure (La Porta et al., 1999) and might be
considered as an external governance influence that acts at the banking industry level
(Ciancanelli and Reyes, 2001; Lassoued et al., 2016). In this regard, we use the
regulatory part of The Worldwide Governance Indicator developed by Kaufmann et al.
(2010), which refers to the capacity of the government to effectively formulate and
implement sound policies. The first index is Government Effectiveness that captures
perceptions of public services quality, civil services quality and their degree of
independence from political pressure, policy formulation and implementation quality and
the credibility of the government’s commitment to such policies. The second is
Regulatory Quality, which captures perceptions of the ability of the government to
conceive and implement sound policies and regulations that encourage private sector
development.
5
Results and discussion
5.1 Descriptive statistics
Table 2 reports the descriptive statistics for the whole sample of 171 banks in the studied
MENA countries. Returns volatility measured by SD_ROAT is on average 0.006
(SD_ROET 0.075) with a great concentration below the mean (since the median does not
exceed the mean). In contrast to insolvency risk, the median of the ZSCORE is located
below the mean (59.02) indicating an important risk level.
72
N. Lassoued and H. Sassi
Table 2
Descriptive statistics
Variable
Mean
Median
Std. Dev.
Min
Max
SD_ROAT
0.006
0.003
0.011
0.0000
0.14
SD_ROET
0.075
0.033
0.191
0.0005
2.029
ZSCORE
59.02
38
67.28
0.02
591.25
H_REV
0.596
0.569
0.107
0
1
H_NON
0.483
0.5
0.207
0
1
STAT_OWN
0.144
0.012
0.289
0
1
FOR_OWN
0.333
0.041
0.383
0
1
ROA
0.015
0.013
0.015
–0.114
0.130
CAR
0.107
0.128
0.072
0
0.997
GDP_GR
5.257
4.802
3.902
–0.070
0.261
SIZE
9.665
7.834
0.770
5.760
11.04
REST_DIV
7
7.500
1.586
3
9
REG_REQ
0.681
0.707
0.1521
0.454
0.954
GOV_EFF
0.5
0.521
0.0701
0.5
0.75
The mean of the diversification index H_REV is 0.596 indicating a relative concentration
of bank income into interest-generating activities. Moreover, banks seem to diversify
within the range of non-interest income activities, since the mean and the median of
H_NON are, respectively, 0.483 and 0.500.
For ownership structure, states own on average 14.4% of bank capital and foreigners
about 33.3%. It seems that this type of owners is significantly present since ownership
shares exceed 0.012 for state ownership and 0.041 for foreign owners.
5.2 Basic model
Table 3 reports the results of our model using state ownership with two panels. In panel
A, diversification is measured by H_REV and in panel B by H_NON. The results of risk
estimation (SD_ROA, SD_ROE and ZSCORE) are displayed in columns (1, 2 and 3) and
those of diversification in columns (4, 5 and 6).
Across all model specifications, the null hypothesis (suggesting that Ordinary least
square (OLS) is an appropriate estimation technique) is rejected through Hausman test of
endogeneity. We validate that the relevance of estimating our instrumental variables.
Furthermore, the over-identifying Hansen-Sargan test does not reject the null hypothesis
and supports the validity of instruments.
73
Income diversification, bank stability and owners identity
Table 3
Three-stage least squares regression (3SLS) regression results of Bank risk,
diversification and state ownership
Panel A
H_REV
STAT_OWN
SD_ROAT
SD_ROET
ZSCORE
0.010
(2.50)**
0.031
(2.03)**
0.545
(3.49)***
0.998
(2.16)**
–9.175
(–2.19)**
–200.764
(–2.72)***
SIZE
CAR
GDP_GR
DEP_INS
H_REV
H_REV
0.246
(3.71)***
0.620
(22.99)***
–0.382
0.236
(3.10)***
0.637
(23.70)***
–0.481
–0.126
–5.478
618.126
0.245
(3.71)***
0.618
(22.99)***
–0.383
(–3.64)***
(–4.78)***
(2.36)**
(–1.47)
(–1.47)
(–1.82)*
–0.0001
–.0022
–1.556
0.0005
0.0005
0.0006
(–0.83)
(–0.25)
(–0.47)
(0.33)
(0.33)
(0.40)
–0.001
–0.631
135.759
0.050
0.050
0.043
(–0.12)
(–2.01)**
(1.88)*
(2.15)**
(2.21)**
(2.94)***
L.H_REV
ROA
H_REV
–0.0004
–0.007
3.109
–0.001
–0.001
–0.001
(–2.72)***
(–1.47)
(2.74)***
(1.06)
(–1.06)
(–0.94)
(546.84)***
(546.84)***
(581.6)***
0.003
0.094
–17.775
(2.29)**
(2.09)**
(–1.87)*
Chi2
(23.67)***
(33.71)***
(17.62)***
Hausman test
(16.21)**
(13.47)**
(14.3)**
(2.788)
(2.271)
(3.703)
(2.788)
(2.271)
(3.703)
SD_ROAT
SD_ROE
ZSCORE
H_NON
H_NON
H_NON
Hansen-Sargan
statistic
Panel B
H_NON
STAT_OWN
0.005
0.261
–34.255
(2.04)**
(3.24)***
(–2.39)**
0.006
0.347
–90.655
0.281
0.280
0.279
(1.80)*
(1.94)*
(–2.18)**
(2.87)***
(2.85)***
(2.96)***
L.H_NON
ROA
SIZE
CAR
GDP_GR
DEP_INS
0.563
0.562
(13.93)***
(13.47)***
–0.106
–3.513
490.683
–1.226
–1.228
–0.942
(–5.35)***
(–5.15)***
(2.26)**
(–1.83)*
(–1.83)*
(–2.11)**
–0.0001
–0.0027
3.580
–0.010
–0.010
–0.0107
(–0.85)
(–0.47)
(1.07)
(–2.48)**
(–2.45)**
(–2.49)**
–0.015
–0.421
–33.837
0.254
0.255
0.274
(–3.25)***
(–2.64)***
(–0.72)
(2.07)**
(2.07)**
(2.05)**
–0.0001
–0.002
1.744
–0.003
–0.003
–0.003
(–2.77)***
(–0.64)
(2.45)**
(–1.19)
(–1.19)
(–1.24)
(194.08)***
(195.32)***
(183.11)***
0.0001
0.060
1.609
(0.21)
(2.16)**
(0.21)
Chi2
(64.66)***
(56.03)***
(226.76)***
Hausman test
(17.01)***
(12.31)*
(13.74)**
(8.024)
(5.797)
(7.096)
Hansen-Sargan
statistic
0.561
(13.89)***
74
N. Lassoued and H. Sassi
In panel A, the coefficients for H_REV and H_NON are positive in SD_ROA and
SD_ROE regression (negative for the ZSCORE regression) and statistically significant
across all specifications. This implies that diversification into and within non-interest
income-generating activities correlates with lower income volatility and lower insolvency
risk8. More specifically, we found that risk declines as the bank diversifies across interest
and non-interest activities and diversifies within non-traditional banking activities (like
trading, brokerage…). Our finding is in line with the modern portfolio theory suggesting
that diversification decreases risk. This result for the MENA banks is consistent with
those of many studies on different samples .For instance, Sanya and Wolfe (2011a) and
Baele et al. (2007) show that income diversification reduces risk in a sample of European
banks. Deng et al. (2013) found the same for US banks. However, Stiroh (2006) find the
opposite relationship in a sample of US banks.
In the same regression (Equation 4), the coefficient of state ownership on risk is
positive (negative) and significant at the 1% level in the two first (third) columns in
panel A. Panel B presents the same result but less significant. These findings imply that
governments in the MENA countries encourage banks to take more risk.
To see whether income diversification and banking risk can be determined through
the impact of state ownership on the decision to diversify income, we report the results of
Equation (5) in columns (5–6) of Table 3. Panel A reports the results of H_REV as
diversification proxy. The coefficients of state ownership are positive and significant
(at the 5% and 1% levels) and imply that governments seem to discourage income
diversification, thus generating high risk level. The finding is very interesting and
explains previous studies showing that state ownership increases banking risk like Farazi
et al. (2013) and Lassoued et al. (2016) for MENA countries, Berger et al. (2005) for
Argentinean banks and Angkinand and Wihlborg (2010) for a sample of banks selected
from 32 countries. In fact, reluctance to diversify income (either across interest and noninterest activities or within non-traditional banking activities) for state-owned banks
generates high risk level.
Among the control variables in the risk equation, we found that banking performance
ROA is negatively associated with SD_ROA and SD_ROE risk and positively associated
to ZSCORE (in panels A and B) suggesting that performing banks are less risky and
more stable, consistent with the findings of Sanya and Wolfe (2011b). The coefficient of
CAR is negative and significant in column 2 and positive in column 3 for Panel A.
Similarly, the coefficient of CAR is negative and significant in column 1 and positive and
significant in column 3 for Panel B. In line with the findings of Mercieca et al. (2007)
and Meslier et al., 2014), our results indicate that when equity levels are low, banking
risk is high because capital protects banks when asset values decline. GDP_GR has a
negative and a significant effect on SD_ROA and a positive and a significant effect on
ZSCORE (Panels A and B) indicating that banks operating more growing economies are
less risky than other banks.
For DEP_INS, we found that it significantly increases risk in all specifications of
Panel A. The effect is significant only in column 2 of Panel B. Against the conventional
wisdom that DEP_INS reduces banking risk Gropp and Vesala, 2001), it seems to give
bank managers the incentive to take more risks, since insurance will cover a large part of
the bank’s debts in case of non-payment (Angkinand and Wihlborg, 2010).
As for the control variables in the diversification equation presented in columns
(4, 5 and 6), lagged diversification exhibits positive and significant coefficients in all
Income diversification, bank stability and owners identity
75
specifications for Panels A and B. This is consistent with the findings of Deng et al.
(2013). ROA has a negative significant coefficient in column 6 (panel A) and columns 4,
5 and 6 (Panel B). This finding is consistent with that of Campa and Kedia (2002) who
argue that poorly performing (less profitable) industrial firms tend to diversify. Finally,
CAR coefficients are positive and significant in all specifications, indicating that wellcapitalised banks are less diversified. Stiroh and Rumble (2006) found that banks with
high capitalisation will take less risk and may therefore be less diversified (Stiroh and
Rumble, 2006).
Table 4 presents results our model for foreign ownership with two panels. In panel A,
diversification is measured by H_REV and in panel B by H_NON. The results of risk
estimation (SD_ROA, SD_ROE and ZSCORE) are displayed in columns (1, 2 and 3) and
those of diversification in columns (4, 5 and 6).
Across all model specifications, we report similar statistics for the Hausman test of
endogeneity and the over-identifying Hansen-Sargan test. According to these tests, we
support the relevance of estimating the instrumental variables and the validity of the
instruments.
The coefficients for H_REV in panel A and for H_NON are positive in SD_ROA and
SD_ROE regression (negative for the ZSCORE regression). They are statistically
significant across all specifications. Once again, we found that diversification into and
within non-interest income-generating activities is associated with lower income
volatility and lower insolvency risk.
In the same regression (Equation 4), the coefficient of foreign ownership on risk is
negative (positive) and significant at the 1% level (at 5 % level) in the two first (third)
columns in panel A. Panel B reports the same results, indicating that foreign ownership
decreases banking risk in MENA countries.
Like above, we investigate whether income diversification and banking risk can be
determined through the impact of the foreign ownership on the decision to diversify
income. We present the result of Equation (5) in columns (5–6) of Table 3. Panel A
displays the results of using H_REV as a diversification proxy. The coefficients of
foreign ownership are negative and significant at the 1% level in all specifications. In
light of this finding, foreign participants seem to encourage income diversification, which
contributes to reducing banking risk.
Ours results are consistent with those of Crystal et al. (2002), Mian (2003) and
Lassoued et al. (2016) which found that foreign banks decrease banking risk in emerging
markets. This finding can be explained by the importance of income diversification
(either across interest and non-interest activities or within non-traditional banking
activities) for foreign-owned banks.
Finally, we found the same results for the control variables in the risk and
diversification equations. Indeed, ROA is negatively associated with SD_ROA and
SD_ROE risk and positively associated to ZSCORE (in panels A and B). CAR has a
negative effect on banking risk. GDP_GR seems to reduce risk while DEP_INS
significantly increases risk.
Regarding the control variables in the diversification equation presented in columns
(4, 5 and 6), lagged diversification displays positive and significant coefficients in all
specifications and CAR coefficients are positive and significant in all specifications
indicating that well-capitalised banks are less diversified.
76
N. Lassoued and H. Sassi
Table 4
Three-stage least squares regression (3SLS) regression results of bank risk,
diversification and foreign ownership
SD_ROA
SD_ROE
ZSCORE
H_REV
H_REV
H_REV
PANEL A
H_REV
FOR_OWN
0.002
0.297
–52.848
(1.97)**
(3.16)***
(–2.27)**
–0.005
–0.143
17.879
0.235
0.236
0.227
(–3.65)***
(–2.92)***
(2.45)**
(3.43)***
(3.44)***
(2.83)***
L.H_REV
ROA
SIZE
CAR
GDP_GR
DEP_INS
0.613
0.633
(23.65)***
(24.34)***
–0.091
–4.424
577.943
–0.257
–0.257
–0.401
(–5.47)***
(–7.76)***
(3.73)***
(–1.09)
(–1.09)
(–1.63)*
–0.003
–0.021
1.651
0.0007
0.0007
0.0025
(–0.95)
(–1.38)
(0.46)
(0.12)
(0.12)
(0.44)
–0.014
–0.156
33.465
0.010
0.010
0.009
(–5.57)***
(–1.89)**
(1.60)
(2.51)**
(2.51)**
(2.42)**
–0.0001
–0.001
1.492
–0.0004
–0.0004
–0.0002
(–3.39)***
(–0.51)
(3.76)***
(–0.57)
(–0.57)
(–0.52)
0.001
0.028
–5.462
(3.04)***
(2.26)**
(–1.74)*
Chi2
(94.40)*** (119.64)*** (45.64)***
Hausman test
(15.67)**
(12.98)**
(13.25)**
(7.949)
(10.095)
(13.771)
SD_ROA
SD_ROE
ZSCORE
Hansen-Sargan statistic
0.613
(23.66)***
(565.1)***
H_NON
(564.81)*** (600.68)***
H_NON
H_NON
PANEL B
H_NON
FOR_OWN
–0.006
–0.180
–46.243
(–2.60)***
(–3.00)***
(–2.26)**
–0.001
–0.069
6.761
–0.173
–0.173
–0.175
(–2.66)***
(–2.32)**
(2.39)**
(–3.14)***
(–3.14)***
(–3.16)***
L.H_NON
ROA
SIZE
CAR
GDP_GR
DEP_INS
0.532
0.527
(14.00)***
(13.72)***
–0.116
–3.092
471.653
–0.234
–0.235
–0.143
(–5.67)**
(–5.36)**
(2.13)**
(–0.37)
(–0.37)
(–0.19)
–0.0005
–0.0087
3.709
–0.0128
–0.0128
–0.0127
(–1.08)
(–0.71)
0.84
(–1.02)
(–1.02)
(–1.07)
–0.013
–0.283
3.009
0.183
0.188
0.191
(–4.57)**
(–3.45)**
(0.11)
(2.73)***
(2.71)***
(2.48)**
–0.0002
0.0002
2.293
–0.0002
–0.0004
–0.0003
(–4.11)***
(0.26)
(4.27)***
(–0.14)
(–0.14)
(–0.17)
0.001
0.031
–4.415
(1.21)
(2.43)**
(–1.01)
Chi2
(68.24)***
(84.79)***
(29.4)***
Hausman test
(14.93)**
(11.28)**
(12.51)**
(6.927)
(2.873)
(8.572)
Hansen-Sargan statistic
0.531
(13.97)***
(226.14)*** (226.82)*** (218.73)***
Income diversification, bank stability and owners identity
6
77
Conclusion
The aim of this paper is to examine how bank ownership structure influences the
relationship between income diversification and banking risk in the MENA countries. We
used data on171 banks from 13 MENA countries observed over the 2006–2012 period.
We specified a 3SLS model in which the impact of foreign and state ownership on
income diversification and banking risk is jointly examined to solve the endogeneity
problem that arises from the relationship between risk-taking and diversification.
We found that diversification, across interest and non-interest activities and within
non-traditional banking activities, decreases banking risk. Owner identities affect risk
level through diversification. Indeed, state-owned banks seem to take more risks by less
diversifying their income. Inversely, foreign ownership decreases risk by diversifying
their income. Therefore, we confirm that diversification is a channel through which
controlling shareholders influence banking risk and diversification level depends on the
monitoring exerted by the different shareholders.
Our findings have several implications for regulators and investors in the MENA
region. Policy makers should, first, encourage banks to diversify more their income by
removing restrictions on banking activities. Second, regulators should accelerate the
banking privatisation process. Third, it will be more appropriate to alleviate barriers to
foreign investment to decrease bank risk-taking. Investors may use ownership structure
and income diversification in their decisions.
In conclusion, our study highlighted that ownership structure affects bank risk-taking
through income diversification. Future research could provide additional insights by
jointly examining the effect of other types of diversification.
References
Acharya, V.V, Hasan, I. and Saunders, A. (2006) ‘Should banks be diversified? evidence from
individual bank loan portfolios’, Journal of Business, Vol. 79, No. 3, pp.1355–1411.
Almarzoqi, R., Ben Naceur, S. and Scopelliti, A. (2015) How Does Bank Competition Affect
Solvency, Liquidity and Credit Risk? Evidence from the MENA Countries, IMF Working Paper
Series Nos. 15/210
Angkinand, A. and Wihlborg, C. (2010) ‘Deposit insurance coverage, ownership, and banks’ risktaking in emerging markets’, Journal of International Money and Finance, Vol. 29, No 2,
pp.252–274.
Baele, L., De Jonghe, O. and Vennet, R.V. (2007) ‘Does the stock market value bank
diversification?’, Journal of Banking and Finance, Vol. 31, No. 7. pp.1999–2023.
Barth, J.R., Caprio, J. and Levine, R. (2013) Bank Regulation and Supervision in 180 Countries
from 1999 to 2011, National Bureau of Economic Research Working Paper 18733.
Ben Naceur, S., Ghazouani, S. and Omran, M. (2007) ‘The performance of newly privatized firms
in selected MENA countries: the role of ownership structure, governance and liberalization
policies’, International Review of Financial Analysis, Vol. 16, No. 4, pp.332–353.
Berger, A.N., Clarke, G.R., Cull, R., Klapper, L. and Udell, G. (2005) ‘Corporate governance and
bank performance: a joint analysis of the static, selection, and dynamic effects of domestic,
foreign, and state ownership’, Journal of Banking and Finance, Vol. 29, No. 8, pp.2179–2221.
Berger, A.N., Hasan, I. and Zhou, M.M. (2010) ‘The effects of focus versus diversification on bank
performance: evidence from Chinese banks’, Journal of Banking and Finance, Vol. 34,
pp.1417–1435.
78
N. Lassoued and H. Sassi
Boyd, J.H., Graham, S. and Hewitt, S. (1993) ‘Bank holding company mergers with nonbank
financial firms on the risk of failure’, Journal of Banking and Finance, Vol. 17, No. 1,
pp.43–63.
Busch, R. and Kick, T. (2009) Income Diversification in the German banking Industry. Deutsche
Bundes bank, Discussion Paper Series 2: Banking and Financial Studies No 09/2009.
Campa, J.M. and Kedia S. (2002) ‘Explaining the diversification discount’, The Journal of
Finance, Vol. 57, pp.1731–1762.
Chiorazzo, V., Milani, C. and Salvini, F. (2008) ‘Income diversification and bank performance:
evidence from Italian banks’, Journal of Financial Services Research, Vol. 33, No. 3,
pp.181–203.
Ciancanelli, P. and Reyes, J. (2001) Corporate Governance in Banking: A Conceptual Framework,
Working Paper SSRN, Available at: http://ssrn.com/abstract=253714.
Crystal, J.S., Dages, B.G. and Goldberg, L.S. (2002) ‘Has foreign bank entry led to sounder banks
in Latin America?’, Current Issues in Economics and Finance, Vol. 8, No. 1, pp.1–6.
Demirgüç-Kunt, A. and Detragiache, E. (2002) ‘Does deposit insurance increase banking system
stability? An empirical investigation’, Journal of Monetary Economics, Vol. 49, No. 7,
pp.1373–1406.
Demirguç-Kunt, A. and Huizinga, H. (2009) ‘Bank activity and funding strategies: the impact on
risk and returns’, Journal of Financial Economics, Vol. 98, No. 3, pp.626–650.
Deng, S., Elyasiani, E. and Jia, J. (2013) ‘Institutional ownership, diversification, and riskiness of
bank holding companies’, Financial Review, Vol. 48, No. 3, pp.385–415.
DeYoung, R. and Roland, K.P. (2001) ‘Product mix and earnings volatility at commercial banks:
evidence from a degree of leverage model’, Journal of Financial Intermediation, Vol. 10
No. 1, pp.54–84.
Dong, Y., Meng, C., Firth, M. and Hou, W. (2014) ‘Ownership structure and risk-taking:
comparative evidence from private and state-controlled banks in China’, International Review
of Financial Analysis, Vol. 36, pp.120–130.
Elsas, R., Hacketha, A. and Holzhäuser, M. (2010) ‘The anatomy of bank diversification’, Journal
of Banking and Finance, Vol. 34, No. 6, pp.1274–1287.
Faccio, M., Masulis, R.W. and McConnell, J.J. (2006) ‘Political connections and corporate
bailouts’, Journal of Finance, Vol. 61, No. 6, pp.2597–2635.
Farazi, S., Feyen, E. and Rocha, R. (2013) ‘Bank ownership and performance in the middle east
and North Africa Region’, Review of Middle East Economics and Finance, Vol. 9, No. 2,
pp.159–196.
Global Findex Database (2014) World Bank.
Gropp, R. and Vesala, J. (2001) Deposit Insurance and Moral Hazard: Does the Counterfactual
Matter?, European Central bank Working Paper 41.
Gurbuz, A.O., Yanik, S. and Ayturk, Y. (2013) ‘Income diversification and bank performance:
evidence from Turkish banking sector’, Journal of BRSA Banking and Financial Markets,
Vol. 7, No. 1, pp.9–29.
Iannota, G., Giacomo, N. and Sironi, A. (2013) ‘The impact of government ownership on bank
risk’, Journal of Financial Intermediation, Vol. 22, No. 2, pp.152–176.
Jensen, M. and Meckling, W. (1976) ‘Theory of the firm: managerial behavior, agency costs and
ownership structure’, Journal of Financial Economics, Vol. 3, No. 4, pp.305–360.
Kaufmann, D., Kraay, A. and Mastruzzi, M. (2010) The Worldwide Governance Indicator: A
Summary of Methodology, data, and Analytical Issues, WorldBank Policy Research Working
Paper No. 5430.
Keeley, M.C. (1990) ‘Deposit insurance, risk and market power in banking’, American Economic
Review, Vol. 80, No. 5, pp.1183–1200.
La Porta, R., Florencio L., Andrei, S. and Robert, W. (2002) ‘Government ownership of banks’,
The Journal of Finance, Vol. 57, No. 1, pp.256–301
Income diversification, bank stability and owners identity
79
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1999) ‘Corporate ownership around the world’,
Journal of Finance, Vol. 54, No. 2, pp.471–517.
Lang, L. and So, R. (2002) Bank Ownership Structure and Economic Performance, SSRN Working
Paper.
LaPorta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R.W. (1998) ‘Law and finance’,
The Journal of Political Economy, Vol. 106, No. 6, pp.1113–1155.
Lassoued, N., Sassi, H. and Ben Rejeb-Attia, M. (2016) ‘The impact of state and foreign ownership
on banking risk: Evidence from the MENA countries’, Research in International Business and
Finance, Vol. 36, No. 1, pp.167–178.
Lepetit, L., Nys, E., Rous, P. and Tarazi, A. (2008) ‘Bank income structure and risk: an empirical
analysis of European banks’, Journal of Banking and Finance, Vol. 32, No. 9, pp.1452–1467.
Levine, R. (1996) ‘Foreign banks, financial development and economic growth’, in Barfield, C.E.
(Ed.): International Financial Markets: Harmonization Versus Competition, Washington, DC,
A EI Press.
Mercieca, S., Schaeck K. and Wolfe, S. (2007) ‘Small European banks: benefits from
diversification?’ Journal of Banking and Finance, Vol. 31, No. 7, pp.1975–1998.
Meslier, C., Tacneng, R.C. and Tarazi, A. (2014) ‘Is bank income diversification beneficial?
evidence from an emerging economy’, Journal of International Financial Markets,
Institutions and Money, Vol. 31 No. 3, pp.97–126.
Mian, A. (2003) Foreign, Private Domestic, and Government Banks: New Evidence from Emerging
Markets, University of Chicago, Mimeo.
Pennathur, A.K., Subrahmanyam, V. and Vishwasrao, S. (2012) ‘Income diversification and risk:
Does ownership matter? An empirical examination of Indian banks’, Journal of Banking and
Finance, Vol. 36, No. 8, pp.2203–2215.
Rocha, R., Arvai, Z. and Farazi, S. (2011) Financial Access and Stability: A Road Map for the
Middle East and North Africa, World Bank, Washington, DC.
Sahoo, D. and Mishra, P. (2012) ‘Operational diversification and stability of financial performance
in indian banking sector: a panel data investigation’, Research Journal of Finance and
Accounting, Vol. 3, No. 3, pp.70–87.
Sanya, S. and Wolfe, S. (2011a) ‘Can banks in emerging economies benefit from revenue
diversification?’, Journal of Financial Services Research, Vol. 40, pp.79–101.
Sanya, S. and Wolfe, S. (2011b) Ownership Structure, Revenue Diversification and Insolvency Risk
in European Banks Amiable at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1787476.
Shleifer, A. and Vishny, R. (1986) ‘Large shareholders and corporate control’, The Journal of
Political Economy, Vol. 94, No. 3, pp.461–488.
Stein, J.C. (2002) ‘Information production and capital allocation: decentralized versus hierarchical
firms’, Journal of Finance, Vol. 57, No. 5, pp.1891–1921.
Stiroh, K.J. (2006) ‘A portfolio view of banking with interest and noninterest activities’, Journal of
Money Credit and Banking, Vol. 38, No. 5, pp.1351–1361.
Stiroh, K.J. and Rumble, A. (2006) ‘The dark side of diversification: the case of US financial
holding companies’, Journal of Banking and Finance, Vol. 30, No. 8, pp.2131–2161.
Sullivan, R.J. and Spong (2007) ‘Manager wealth concentration, ownership structure, and risk in
commercial banks, Journal of Financial Intermediation, Vol. 16, pp.229–248.
Valverde, S. and Fernandez, F. (2007) ‘The determinants of bank margins in European banking’,
Journal of Banking and Finance, Vol. 31, No. 7, pp.2043–2063.
Zhou, K. (2014) ‘The effect of income diversification on bank risk: evidence from China’,
Emerging Markets Finance and Trade, Vol. 50, p.201–213.
80
N. Lassoued and H. Sassi
Notes
1
2
3
4
5
6
7
8
DeYoung and Roland (2001); Stiroh et al. (2006); Lepetit et al. 2008), Stiroh and Rumble
(2006).
The assets of the 3 largest banks to the total assets of the banking sector.
The index indicates whether foreign banks are prohibited from entering through subsidiary
acquisition and joint ventures. Lower values point to greater restrictions on foreign entries.
According to Farazi et al. (2013), assets of foreign banks increased from 18% of total banking
assets in 2001 to 20% in 2008.
Like underwriting, brokering and securities trading and all aspects of the mutual fund industry
in terms of insurance underwriting and selling in real estate investment, development and
management.
Our empirical approach is widely inspired from Sanya and Wolfe (2011b).
like underwriting, brokering and dealing in securities… and all aspects of the mutual fund
industry in insurance underwriting and selling in real estate investment, development and
management.
Lower HHI indices show increased diversification.