i 
 
UNIVERSITY OF GHANA  
COLLEGE OF HUMANITIES 
 
 
FINANCIAL FREEDOM, COMPETITION AND BANK PERFORMANCE 
IN SUB-SAHARAN AFRICA 
 
 
                                   BY 
 
 
EMMANUEL A. SARPONG-KUMANKOMA 
(ID. 10016633) 
 
 
 
 
 
 
A THESIS SUBMITTED TO THE SCHOOL OF GRADUATE 
STUDIES IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR 
THE AWARD OF DEGREE OF DOCTOR OF PHILOSOPHY IN FINANCE 
 
 
 
 
 
 
 
DEPARTMENT OF FINANCE 
 
JUNE 2016 
University of Ghana http://ugspace.ug.edu.gh
ii 
 
DECLARATION 
I hereby declare that this thesis is my own work produced from research I carried out under 
supervision. This thesis has not been presented by anyone for any academic award, in this or any 
other institution. All references made to work done by other people have been duly acknowledged. 
I am solely responsible for any shortcomings in this work. 
 
 
……………………………………     …………………………………… 
Emmanuel A. Sarpong-Kumankoma      Date 
(Candidate) 
         
 
        
 
 
 
 
 
 
 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
iii 
 
CERTIFICATION 
We hereby certify that this thesis was supervised in accordance with procedures laid down by the 
University. 
  
        
……………………………………     …………………………………… 
Professor Joshua Yindenaba Abor       Date 
(Principal Supervisor) 
  
 
…………………………………….    ……………………………………  
Professor A. Q. Q. Aboagye        Date 
(Co-Supervisor) 
 
 
…………………………………….     …………………………………… 
Dr. Mohammed Amidu        Date 
(Co-Supervisor) 
 
 
 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
iv 
 
ACKNOWLEDGEMENTS 
I would like to express my appreciation for the financial support provided by the University of 
Ghana and the Business School to enable me undertake this study. 
 
Many people have also played important roles in the development of this thesis. Special mention 
should be made of my supervisors, Professor Joshua Yindenaba Abor, Professor A. Q. Q. Aboagye 
and Dr. Mohammed Amidu, who provided much guidance, encouragement and support throughout 
the preparation of the thesis. I would forever be grateful for your wonderful show of magnanimity. 
May the Almighty God, Jehovah reward the way you act. 
  
My colleagues in the Department of Finance have also shown good comradeship during the course 
of the study. My current Head of Department, Prof. Godfred Bokpin, in particular has shown much 
interest in getting this work completed, and so have Professor Kofi A. Osei, Dr. Simon K. Harvey, 
Dr. Charles Andoh and Dr. Vera Fiador. Considerable support have also been received from Dr. 
Kwaku Ohene-Asare, Dr. Patrick Asuming, Dr. Lord Mensah, Dr. Elikplimi Agbloyor, Mr. Frank 
Ametefe, Dr. Ibrahim Bedi, Dr. Agyapomaa Gyeke-Dako, Dr. Saint Kuttu, Miss. Esther Laryea, 
Miss. Sandra Ayiku and Mrs. Josephine Ofosu-Mensah Ababio, for which I am truly grateful.  
 
I am also grateful to Dr. Ebow Turkson and Mr. Solomon Aboagye from the Department of 
Economics, who provided much needed assistance with the data analysis. 
 
And finally, I very much appreciate the love and understanding shown by my wonderful wife, 
Beatrice, and my children, Nana Ama Sarpomaa, Maame Akosua and Kwabena Akosa. 
 
University of Ghana http://ugspace.ug.edu.gh
v 
 
ABSTRACT 
The inferences of competition for bank stability, profitability and efficiency have been the subject 
of much debate, yet they remain controversial and inconclusive. Identifying channels through 
which competition affects bank performance may improve our understanding of the inconsistent 
findings in the literature. Hence, this thesis analyses the implications of competition and financial 
freedom for bank performance. The study also examines the determinants of bank profit 
persistence. Using data from 139 banks in 11 Sub-Saharan African countries over the period, 2006-
2012, and the financial (or banking) freedom index from the Heritage Foundation, this study 
considers whether the degree of freedom in the banking sector affects the relationship between 
competition and bank performance. The results show a positive relationship between market power 
and bank stability, profitability and cost efficiency, suggesting that higher market power (or less 
competition) may improve bank profitability, efficiency, and stability. However, the relationship 
between market power and bank stability is found to be quadratic, implying that beyond a certain 
threshold, further increases in market power may damage bank stability. The evidence from this 
study shows that, the positive effect of market power is stronger on bank profitability, and weaker 
on bank efficiency with higher levels of financial freedom. We do not find evidence suggesting 
that financial freedom influences the relationship between competition and bank stability. On the 
determinants of bank profit persistence, the results of the study show differences across countries. 
The results of this study suggest that policies that allow banks to maintain some level of market 
power may be necessary to ensure banking system efficiency and overall stability. Also, while 
higher financial freedom improves bank profitability, it may be harmful for cost efficiency, 
especially for banks with higher market power. Thus, policies that ensure some restrictions on 
banking freedom may still be required to enhance bank efficiency.  
University of Ghana http://ugspace.ug.edu.gh
vi 
 
TABLE OF CONTENTS 
CONTENT                                                                                                                             PAGE 
DECLARATION ............................................................................................................................ ii 
CERTIFICATION ......................................................................................................................... iii 
ACKNOWLEDGEMENTS ........................................................................................................... iv 
ABSTRACT .................................................................................................................................... v 
TABLE OF CONTENTS ............................................................................................................... vi 
LIST OF TABLES .......................................................................................................................... x 
CHAPTER ONE ............................................................................................................................. 1 
INTRODUCTION .......................................................................................................................... 1 
1.1 Background of the Study .................................................................................................. 1 
1.2 Statement of the Research Problem ................................................................................. 7 
1.3 Research Questions ........................................................................................................ 12 
1.4 Research Hypotheses...................................................................................................... 13 
1.5 Objectives of the Study .................................................................................................. 16 
1.6 Significance of the Study ............................................................................................... 16 
1.7 Scope of the Study.......................................................................................................... 18 
1.8 Structure of the Thesis.................................................................................................... 19 
CHAPTER TWO .......................................................................................................................... 20 
LITERATURE REVIEW ............................................................................................................. 20 
2.1 Introduction ......................................................................................................................... 20 
2.2 Theoretical Framework and Determinants of Bank Competition ....................................... 21 
University of Ghana http://ugspace.ug.edu.gh
vii 
 
2.3 Bank Stability ...................................................................................................................... 27 
2.3.1 Theoretical Review of Bank Competition-Stability Relationship ................................... 27 
2.3.2 Empirical Evidence on Competition-Stability Relationship ............................................. 29 
2.3.3 Is There a Transmission Mechanism Between Competition and Stability? ................... 31 
2.4 Bank Profitability ................................................................................................................ 34 
2.4.1 Theoretical Review of Determinants of Bank Profitability .............................................. 34 
2.4.2 Empirical Evidence on Determinants of Bank Net Interest Margins .............................. 38 
2.4.3 Empirical Evidence on Determinants of Bank Profitability ............................................. 42 
2.5 Bank Competition and Efficiency ....................................................................................... 47 
2.5.1 Theoretical Review of Bank Competition-Efficiency Relationship ................................ 47 
2.5.2 Empirical Evidence on Competition-Efficiency Relationship ......................................... 50 
2.6 Bank Profit Persistence ....................................................................................................... 54 
2.6.1 Theoretical Background on Profit Persistence ................................................................... 54 
2.6.2 Empirical Evidence on Determinants of Bank Profit Persistence .................................... 55 
2.7 Economic Freedom and Bank Performance ........................................................................ 58 
2.8 Overview of the Banking Sector in Africa .......................................................................... 68 
2.8.1 Banking Systems in Africa .................................................................................................... 68 
2.8.2 African Banking Market Structure and Performance ........................................................ 71 
2.9 Chapter Summary and Gaps in the Literature ..................................................................... 76 
CHAPTER THREE ...................................................................................................................... 79 
METHODOLOGY ....................................................................................................................... 79 
3.1 Introduction ......................................................................................................................... 79 
3.2 Research Design .................................................................................................................. 79 
University of Ghana http://ugspace.ug.edu.gh
viii 
 
3.3 Population of Study ............................................................................................................. 81 
3.4 Sampling Technique and Sample Size ................................................................................ 81 
3.5 Measuring Bank Competition ............................................................................................. 82 
3.6 Measurement of Bank Performance .................................................................................... 86 
3.6.1 Measurement of Bank Stability ............................................................................................ 87 
3.6.2 Measurement of Bank Profitability ...................................................................................... 88 
3.6.3 Measurement of Bank Efficiency ......................................................................................... 88 
3.7 The Economic Freedom Index ............................................................................................ 91 
3.8 Estimating Effect of Competition and Financial Freedom on Bank Stability ..................... 93 
3.9 Modelling Competition, Financial Freedom and Bank Profitability ................................... 95 
3.10 Modelling Effect of Financial Freedom on Bank Competition ....................................... 100 
3.11 Estimating Effect of Competition and Financial Freedom on Bank Efficiency.............. 102 
3.12 Estimating the Determinants of Bank Profit Persistence ................................................ 105 
3.13 Method of Data Analysis ................................................................................................. 107 
3.14 Chapter Summary ............................................................................................................ 111 
CHAPTER FOUR ....................................................................................................................... 116 
RESULTS AND DISCUSSION ................................................................................................. 116 
4.1 Introduction ....................................................................................................................... 116 
4.2 Descriptive Statistics ......................................................................................................... 116 
4.3 Competition, Economic Freedom and Bank Risk ............................................................. 120 
4.3.1 Competition, Freedom and Bank Stability ........................................................................ 120 
University of Ghana http://ugspace.ug.edu.gh
ix 
 
4.3.2 Competition, Freedom and Bank Asset Quality ............................................................... 126 
4.4 Competition, Economic Freedom and Bank Profitability ................................................. 130 
4.4.1 Competition, Freedom and Bank Net Interest Margins .................................................. 133 
4.4.2 Competition, Freedom and Bank Return on Assets ......................................................... 136 
4.5 Competition, Economic Freedom and Bank Efficiency .................................................... 141 
4.5.1 Sources of Bank Market Power in Sub-Saharan Africa .................................................. 141 
4.5.2 Competition, Freedom and Bank Efficiency .................................................................... 144 
4.6 Bank Profit Persistence in Sub-Saharan Africa ................................................................. 153 
4.7 Chapter Summary .............................................................................................................. 155 
CHAPTER FIVE ........................................................................................................................ 166 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS............................................... 166 
5.1 Introduction ....................................................................................................................... 166 
5.2 Summary of Findings ........................................................................................................ 166 
5.3 Contributions of the Study ................................................................................................ 172 
5.4 Conclusions ....................................................................................................................... 174 
5.5 Policy Recommendations .................................................................................................. 175 
5.6 Limitations and Suggestions for Further Research ........................................................... 176 
REFERENCES ........................................................................................................................... 177 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
x 
 
LIST OF TABLES 
 
Table 3.1: Number of Banks for Each Country ............................................................................ 81 
Table 4.1: Summary Descriptive Statistics ................................................................................. 117 
Table 4.2: Effects of Competition and Freedom on Bank Stability (Conventional Lerner) ....... 121 
Table 4.3: Effects of Competition and Freedom on Bank Stability (Adjusted Lerner) .............. 122 
Table 4.4: Effects of Competition and Freedom on Asset Quality (Conventional Lerner) ........ 127 
Table 4.5: Effects of Competition and Freedom on Asset Quality (Adjusted Lerner) ............... 128 
Table 4.6: Effects of Competition and Freedom on Bank NIM (Conventional Lerner)............. 131 
Table 4.7: Effects of Competition and Freedom on Bank NIM (Adjusted Lerner) .................... 132 
Table 4.8: Effects of Competition and Freedom on Bank ROAA (Conventional Lerner) ......... 137 
Table 4.9: Effects of Competition and Freedom on Bank ROAA (Adjusted Lerner) ................ 138 
Table 4.10: Sources of Bank Market Power ............................................................................... 142 
Table 4.11: Bank Cost Efficiency Estimates .............................................................................. 145 
Table 4.12: Effects of Competition and Freedom on Cost Efficiency (Conventional Lerner)....146 
Table 4.13: Effects of Competition and Freedom on Cost Efficiency (Adjusted Lerner) .......... 147 
Table 4.14: Effects of Competition and Freedom on Cost Efficiency (IV with Conv. Lerner) . 148 
Table 4.15: Effects of Competition and Freedom on Cost Efficiency (IV with Adj. Lerner)….149  
Table 4.16: Determinants of Bank Profit Persistence…………………………………………..154 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
1 
 
 CHAPTER ONE 
INTRODUCTION 
 
1.1 Background of the Study 
 
The banking sector plays a critical role in every economy. Much of the work of transferring surplus 
funds to deficit units is carried out through the intermediation function of financial institutions, 
with banks as the dominant entity across many jurisdictions. A well-functioning banking system 
is widely recognized as essential to economic growth and development. Even so, reforms, 
globalization and technological advancement have changed the nature of the banking business over 
the last few decades. In addition, the 2007-2008 financial crisis that originated mainly in the U.S 
and spread to other developed countries and some emerging markets, and the recent banking crisis 
in Europe have re-ignited debate about regulation of the banking sector, and the level of freedom 
required to encourage innovation and efficiency and also ensure the stability of the financial 
system. 
 
Indeed, recent decades have witnessed a considerable increase in the number of financial reforms 
across the globe, in both developed and developing countries. This development has resulted in 
the liberalization of banking sectors in many countries with higher-income countries being more 
liberalized than lower-income countries. A major reason for these reforms was to make the 
financial markets in general, and the banking sector in particular, more competitive (Delis, 2012). 
A widely accepted view is that competition in the banking sector will engender efficiency in 
operations, and also enhance general performance of financial institutions. Yet, research has shown 
University of Ghana http://ugspace.ug.edu.gh
2 
 
that the relationship between competition and bank performance is very complex, and that the view 
that competition is unambiguously good may be more naive in banking than in other industries 
(Claessens & Laeven, 2004). Strong competition may not be best for performance in the financial 
sector. It is no wonder then, that several years after some of these reforms were initiated, 
fundamental questions are still being asked about the implications of competition for the 
performance of the banking sector. For instance, is competition good for the overall stability of 
the banking sector? Has the opening up of the banking sector to greater competition 
unambiguously led to greater efficiency in banking operations? Is the high profitability observed 
in many banking markets driven by improved efficiency, or by exploitation of market power by 
banks? What factors influence the effects of competition on bank performance? 
 
Finding answers to these questions remain relevant, because despite numerous research on the 
repercussions of competition for banking sector performance, there is no consensus on the effect 
of competition on bank stability, profitability or efficiency. A related issue that continues to puzzle 
researchers is the persistence of bank profits. While economic theory suggests that freedom of 
entry and exit in competitive markets should whittle away excess profits, this is not observed in 
the real world of banking. Bank profits continue to persist (Gschwandtner, 2005). So what drives 
the persistence of bank profits? Are there variations in these influencing factors across different 
countries?  
 
Nevertheless, profitability in the banking sector is important and could partly explain the stability 
observed in some banking markets. Highly profitable banks should be able to increase their capital 
reserve to absorb potential losses and remain resilient to shocks. And as suggested by the European 
University of Ghana http://ugspace.ug.edu.gh
3 
 
Central Bank (2010), the question of what is an acceptable level of bank profitability is likely to 
play a pivotal role in the post-crisis debate among banking executives, investors and regulators, 
considering the enormous losses in the financial crisis, and the huge government intervention 
required. Even so, recent events have shown that the most common measure for a bank’s 
performance, which is profitability, may be only part of the story, and that a more comprehensive 
assessment of bank performance may be warranted (European Central Bank, 2010). Another issue 
of concern, however, is the fact that higher bank profitability may be achieved at the expense of 
the rest of the economy, resulting in welfare losses and inefficient allocation of resources. 
 
Also, the need for a stable banking system cannot be overemphasized. Instability in the banking 
sector is likely to result in the malfunctioning of other sectors of the economy, which depend on 
the payment system and intermediation functions provided by the banking sector, besides other 
services. Indeed, frequent bank failures was one of the major reasons for the reforms carried out 
in the banking sectors in many countries. And since the implementation of the reforms, we have 
not seen as many bank failures as was witnessed in the 1980s and 1990s. Yet concerns remain 
about the possibility of bank failures, considering the continued integration of banking systems, 
and the uncertainty about the effects of policy reforms such as competition on bank stability. 
 
Efficiency in banking operations also has implications for the larger economy. For instance, the 
ability of banks to reduce costs through the use of modern technology should be beneficial to both 
their customers and shareholders, in terms of lower prices (and hence greater access to finance), 
and higher wealth creation.  
 
University of Ghana http://ugspace.ug.edu.gh
4 
 
The theoretical and empirical literature on the relationship between competition and bank 
performance (efficiency, profitability and stability) is quite extensive, yet it is obvious that more 
work needs to be done in view of the contrasting and even inconclusive results obtained in several 
studies. Generally, both the theoretical and empirical literature present ambiguous positions on the 
competition-stability, competition-profitability and competition-efficiency relationships. 
 
For instance, the theoretical literature on the relationship between competition and bank stability 
presents no consensus.  Two opposing views are theorized in the literature. One view, called the 
competition–fragility or competition-instability view, and pioneered by the influential work of 
Keeley (1990), asserts that competition in banking reduces market power (or the ability of banks 
to price their products/services above marginal cost), decreases profit margins, and results in 
reduced franchise value that encourages banks to take on greater risks. This position is supported 
by Hellmann, Murdock and Stiglitz (2000), Besanko and Thakor (1993), Allen and Gale (2000), 
and Allen and Gale (2004). In contrast, the competition–stability view came out strongly from the 
work of Boyd and de Nicoló (2005), who argued that competition leads to greater bank stability. 
They assert that as banking markets become more concentrated (and less competitive), banks tend 
to show market power by charging higher interest rates to borrowers. This worsens moral hazard 
incentives and makes it more difficult for borrowers to repay, as they also take on greater risks in 
search of more profits, and the result is an increase in the possibility of loan default and therefore 
the risk of bank portfolios. On the other hand, a reduction in interest rates on loans as a result of 
greater competition would encourage borrowers to seek safer investments, thus minimizing the 
risk of default and bank portfolio risk. This is also supported by Caminal and Matutes (2002) and 
Beck, Demirguc-Kunt and Maksimovic (2004), among others. 
University of Ghana http://ugspace.ug.edu.gh
5 
 
Also, two competing theories emerged from the industrial organization literature to explain the 
relationship between market structure and bank profitability. These are the Market Power (MP) 
and Efficiency Structure (ES) paradigms. These approaches provide different suggestions for the 
direction of causality between market structure and profitability.  
 
The literature on competition and bank efficiency is related to the hypotheses that explain the 
relationship between market structure and performance. A special case of the MP hypothesis, the 
Quiet Life hypothesis, is the focus of this study. The ‘Quiet Life’ hypothesis (QLH), originally 
proposed by Hicks (1935) suggests that the higher the market power enjoyed by a firm, the lower 
the effort put forth by managers to improve efficiency, resulting in a negative relationship between 
market power and efficiency (Berger & Hannan, 1998). As an alternative, the efficient structure 
hypothesis developed by Demsetz (1973), submits that more efficient banks are better armed to 
survive competitive pressures and gain market share at the expense of the less efficient ones. 
Schaeck and Cihak (2008) categorize these hypotheses as ‘competition-efficiency’ and 
competition-inefficiency hypotheses. Similar to the competition-stability, and competition-
profitability relationships, the empirical evidence on competition-efficiency relationship has been 
inconclusive. And only some recent studies have tried to test the quiet life hypothesis in banking 
(Coccorese & Pellecchia, 2010).  
 
Noticeably absent in the banking literature is an examination of the links between economic 
freedom and bank performance. The limited research in this area is somewhat surprising given the 
importance of bank lending in promoting economic development and the impact that economic 
freedom is likely to have on the banking sector (Sufian & Habibullah, 2010). Indeed, as noted by 
University of Ghana http://ugspace.ug.edu.gh
6 
 
Hafer (2013), a number of studies have found that financial development and higher levels of 
economic freedom are associated with (or cause) economic growth. The unanswered question, 
however, is whether the financial development-economic growth nexus reflects influences of 
economic freedom operating through the financial system (Hafer, 2013). Sufian and Habibullah 
(2010) provide new empirical evidence on the positive impact of economic freedom on banks’ 
performance in Malaysia. Chortareas, Girardone, and Ventouri (2013), possibly the first to directly 
investigate the dynamics between the financial freedom counterparts of the economic freedom 
index drawn from the Heritage Foundation database and bank efficiency levels, suggest that the 
higher the degree of an economy’s financial freedom, the higher the benefits for banks in terms of 
cost efficiency. Financial freedom is a measure of the degree of restrictions and controls in the 
financial sector. When financial institutions operate in a less restricted environment they are more 
likely to engage in competitive policies, resulting in higher levels of efficiencies. Related studies 
include Smimou and Karabegovic (2010) who show that changes in economic freedom have a 
positive impact on equity market returns, and Hafer (2013) who finds that countries with higher 
levels of initial economic freedom, on average, exhibit greater levels of financial intermediary 
development in subsequent years.  
 
Chortareas et al. (2013) has noted that studies that consider the effects of economic freedom on 
bank performance typically treat the freedom index as one of the control variables, and also focus 
on the aggregate (economic) freedom index and not on the specific financial freedom counterparts, 
which gives rise to the possibility of misspecification bias. Indeed, since it became available some 
twenty years ago, several studies have used the financial freedom index (sometimes called banking 
University of Ghana http://ugspace.ug.edu.gh
7 
 
freedom) either as control variable or instrumental variable, but hardly is there any focus on its 
effect on bank performance.  
 
Given the difficulty in understanding the channels between competition and bank stability, our 
study seeks to assess the effect of financial freedom on the competition-stability relationship. We 
also seek to provide additional insight into the ambiguous relationship between competition and 
bank profitability by considering the conditioning effect of financial freedom on this relationship. 
Further, this study seeks to test the quiet life hypothesis in the developing country context, which 
is largely ignored in the literature, and also provide new empirical evidence on the impact of 
financial freedom on the competition-efficiency relationship. In addition, we consider the 
determinants of bank profit persistence, and differences in determinants of profit persistence 
among countries. And though yet to be tested in the literature, there is some concern that excessive 
financial freedom may contribute to financial institutions’ propensity to take on greater risks, 
which in turn may have contributed to the recent global and European crises (Chortareas et al., 
2013).  
 
1.2 Statement of the Research Problem 
Recent studies that provide evidence for the competition–fragility hypothesis include Diallo 
(2015), Fungacova and Weill (2013), Berger, Klapper and Turk-Ariss (2009) and Beck, Demirguc-
Kunt and Levine (2006). They obtain similar findings, suggesting that increased competition may 
undermine bank stability. On the other hand, gains in market power will increase the stability and 
reduce the risk for the banking system. In contrast, a number of studies have found support for the 
University of Ghana http://ugspace.ug.edu.gh
8 
 
hypothesis that competition enhances bank stability. For example, Boyd and de Nicoló (2005), 
Uhde and Heimeshoff (2009), Schaeck and Cihak (2014) and Schaeck, Cihak and Wolfe (2009).  
 
However, recently, Martinez-Miera and Repullo (2010) questioned the theoretical predictions of 
Boyd and de Nicoló (2005) by arguing that lower loan rates resulting from greater bank 
competition also reduces the interest payments from performing loans (which provide a buffer 
against loan losses) because of imperfect correlation of loan defaults. They conclude that a U-
shaped relationship between competition and the risk of bank failure generally obtains. Studies 
supporting this position are Forssbæck and Shehzad (2014), Beck, de Jonghe, and Schepens 
(2013), Liu, Molyneux and Wilson (2013), and Jimenez, Lopez, and Saurina (2013). 
 
Empirical research on the impact of competition or market structure on bank interest margins have 
also produced conflicting results. Studies that have found a positive relationship between market 
power and net interest margins or a negative relationship between competition and net interest 
margins include Almarzoqi and Naceur (2015), Hawtrey and Liang (2008), Van Leuvensteijn, 
Sorensen, Bikker and Rixtel (2013), and Poghosyan (2013). Similarly, for Africa, while Aboagye, 
Akoena, Antwi-Asare, and Gockel (2008a) found that increases in bank market power (or 
concentration) significantly increase net interest margin in Ghana, Chirwa and Mlachila (2004) 
suggest that the observed high spreads in Malawi can be attributed to high monopoly power. 
 
In contrast, Maudos and de Guavara (2004) show that the fall of margins in the European banking 
system is compatible with a relaxation of the competitive conditions (increase in market power 
and concentration), as this effect has been counteracted by a reduction of interest rate risk, credit 
University of Ghana http://ugspace.ug.edu.gh
9 
 
risk, and operating costs. On the other hand, while Kasman, Tunc, Vardar and Okan (2010) found 
market power to be a vital determinant of bank interest margin in both new and old EU member 
countries, the effect was opposite for the two groups. Beck and Hesse (2009) also found little 
evidence for market structure explaining variation in spreads or margins over time for Uganda. 
 
Generally, the empirical evidence on the market structure and bank profitability relationship have 
been mixed. While some have found support for the Market Power (MP) hypothesis, other studies 
find no evidence of market structure or market power having an effect on bank profitability. 
Instead, some have found that the level of bank profitability is explained by efficiency, thus 
supporting the Efficiency Structure (ES) paradigm. Support for the MP hypothesis has been found 
by Tregenna (2009), Jeon and Miller (2002), Mirzaei, Moore and Liu (2013) and Fu and Heffernan 
(2009). In contrast, some studies report that the MP argument is not held in the banking industry. 
Examples are Seelanatha (2010) and Chortareas, Garza-Garcia and Girardone (2011). Instead, 
these studies suggest that efficiency seems to be the main driving force of increased bank 
profitability.  
 
Similar to the competition-stability, and competition-profitability relationships, the empirical 
evidence on competition-efficiency relationship has been inconsistent. Surprisingly, the quiet life 
hypothesis (QLH) in particular, has received relatively little attention in the empirical literature, 
and only some recent studies have tried to test the QLH in banking (Coccorese & Pellecchia, 2010). 
Moreover, the few studies in this area have largely centered on developed economies. However, 
given the differences in levels of efficiency and institutional arrangements between developed 
markets and developing markets (Eldomiaty, 2007), we cannot assume that the results obtained 
University of Ghana http://ugspace.ug.edu.gh
10 
 
will necessarily apply to developing economies.  Hence, studies in developing countries should 
enhance our understanding of this important issue in banking. Those reporting evidence of the 
quiet life hypothesis include Berger and Hannan (1998), Delis and Tsionas (2009), and Coccorese 
and Pellecchia (2010). In contrast, Weill (2004), Koetter, Kolari and Spierdijk (2012), Casu and 
Girardone (2006), Maudos and de Guavara (2007), Casu and Girardone (2009), and Williams 
(2012) provide support for a negative relationship between competition and efficiency in banking 
(a rejection of the QLH).  
 
Even so, Williams (2012) has noted that the contrast in the empirical evidence may be attributable 
to the simultaneous relationship between competition (market power) and bank efficiency, which 
is usually ignored in empirical investigations. The interdependence between competition and bank 
efficiency implies the possibility of reverse causality. The efficient structure hypothesis, for 
instance, suggests that efficiency may be driving market power (Turk-Ariss, 2010). The problem 
is demonstrated further by the empirical evidence that shows that the quiet life hypothesis is 
usually accepted in studies that do not control for simultaneity (Berger & Hannan, 1998), whilst it 
is rejected in studies that account for the problem (Maudos & de Guevara, 2007; Koetter et al., 
2012). As noted by Koetter et al. (2012), with the exception of Maudos and de Guevara (2007), 
Koetter and Poghosyan (2009), and Delis and Tsionas (2009), virtually all studies on market power 
and efficiency ignore the simultaneous relation between them. This points to a potential gap in the 
literature, which our study seeks to fill, given the seeming dependence of the outcome on whether 
or not simultaneity is taken into consideration.  
 
University of Ghana http://ugspace.ug.edu.gh
11 
 
Contestable markets theory and the new industrial organization literature also highlight the 
influence of potential and actual competition on profitability. The Persistence of Profit (POP) 
theory proposed by Mueller (1977), asserts that entry into and exit from an industry are sufficiently 
free to abolish any abnormal profit quickly, and that the profit rates of all the firms in an industry 
tend to converge towards the same long-run average value. However, this theory does not seem to 
find enough empirical support (Gschwandtner, 2005). Besides, while there is an extensive 
empirical Persistence of Profit (POP) literature based on manufacturing data, only a handful of 
studies investigate POP in banking (Goddard, Liu, Molyneux & Wilson, 2013), and in only a few 
countries (Goddard, Liu, Molyneux, & Wilson, 2011). Indeed, even fewer still consider the 
determinants of profit persistence where they exist. Also, previous studies have generally ignored 
differences in determinants of profit persistence across countries.  
 
Chortareas et al. (2013) has also noted that studies that consider the effects of economic freedom 
on bank performance typically treat the freedom index as one of the control variables, and also 
focus on the aggregate (economic) freedom index and not on the specific financial freedom 
counterparts, which gives rise to the possibility of misspecification bias.  
 
Moreover, most of the above-mentioned studies on competition and bank performance have 
concentrated on the developed economies, particularly the U.S and European banking markets. Of 
course, some studies have covered our broad areas of interest in contexts similar to our own. For 
the competition-stability relationship, studies similar to ours include Tabak, Fazio, and Cajueiro 
(2012) for Latin American countries, Turk-Ariss (2010) for developing economies, Amidu (2013) 
and Amidu and Wolfe (2013a) for emerging and developing countries. On the competition-
University of Ghana http://ugspace.ug.edu.gh
12 
 
profitability nexus, studies in similar context include Beck and Hesse (2009) for Uganda, Perera, 
Skully, and Chaudrey (2013) for South Asian banks, Al-Muharrami and Matthews (2009) for the 
Arab Gulf Cooperation Council (GCC) region, Hossain (2012) for Bangladesh, Chortareas, Garza-
Garcia, and Girardone (2012) for Latin American banking, Maudos and Solís (2009) for Mexico, 
Amidu and Wolfe (2013b) for emerging and developing countries, Ahokpossi (2013) for Sub-
Saharan Africa, Aboagye et al. (2008a) for Ghana, and Chirwa and Mlachila (2004) for Malawi. 
In regard to the competition-efficiency relationship, we also have studies such as Turk-Ariss 
(2010) for the developing economies, Mlambo and Ncube (2011) for South Africa, and Pruteanu-
Podpiera, Weill, and Schobert (2008) for Czech banks, among many others. Kouki and Al-Nasser 
(2014) also recently examined the impact of market power on bank efficiency and stability in 
Africa. Even so, these studies ignored the potential effect of financial freedom on the competition 
and bank performance relationships. 
 
1.3 Research Questions 
The study provides empirical evidence on competition, financial freedom and bank performance 
that seeks to answer the following key questions: 
i. What is the relationship between competition, financial freedom and bank stability in 
Sub-Saharan Africa?  
ii. What is the relationship between competition, financial freedom and bank profitability 
in Sub-Saharan Africa?  
iii. What is the relationship between competition, financial freedom and bank efficiency 
in Sub-Saharan Africa?  
University of Ghana http://ugspace.ug.edu.gh
13 
 
iv. What are the determinants of bank profit persistence in different countries in Sub-
Saharan Africa?   
 
1.4 Research Hypotheses 
Based on the theoretical predictions and the available empirical evidence, as well as the African 
context, ten main hypotheses related to the research questions above, are developed in this thesis. 
 
Generally, the few studies on the effect of competition on bank stability in emerging and 
developing countries (Turk-Ariss, 2010; Amidu, 2013; Kouki & Al-Nasser, 2014) suggest that an 
increase in the degree of market power (or less competition) leads to greater bank stability, and 
that increased competition may undermine bank stability. Given the similarities in economic and 
institutional context, we expect a similar result for this study on Sub-Saharan Africa. Hence, this 
leads to our first hypothesis: 
 
H1: Competition has a negative effect on bank stability. 
 
The effect of financial freedom on bank stability is yet to be tested in the literature, but there are 
suggestions that excessive financial freedom may contribute to financial institutions’ propensity 
to take on greater risks, which in turn may have contributed to the recent global and European 
crises (Chortareas et al., 2013). One could expect that higher financial freedom (or less restrictions 
in the banking sector) is likely to open up banking markets and engender greater competition, 
which could result in instability. Thus we formulate hypotheses 2 and 3 below: 
 
University of Ghana http://ugspace.ug.edu.gh
14 
 
H2: Financial freedom has a negative effect on bank stability. 
 
H3: The effect of competition on bank stability increases with financial freedom. 
 
Even though the empirical evidence on the market structure and bank profitability relationship 
presents no consensus, we expect that in the highly profitable developing country context, 
increasing competition is likely to result in reduced profits, as more banks compete for essentially 
the same customers. Our fourth hypothesis is thus formulated as: 
 
H4: Competition has a negative effect on bank profitability. 
 
On the other hand, higher financial freedom may allow banks to explore new markets and 
activities, resulting in advantages from economies of scale or scope, and better management of 
variability in earnings across product lines, thus increasing their profit potential (Goddard et al., 
2011). This leads us to the following hypothesis: 
 
H5: Financial freedom has a positive effect on bank profitability. 
 
Also, the negative effect of competition on bank profitability may be enhanced by the existence of 
higher financial freedom, but this is yet to be tested in the banking literature. In harmony with our 
expectation we suggest the following hypothesis: 
 
H6: The effect of competition on bank profitability increases with financial freedom. 
University of Ghana http://ugspace.ug.edu.gh
15 
 
The empirical evidence on competition-efficiency relationship has been inconsistent, but since we 
measure competition and efficiency using the same framework, we expect to find a negative 
relationship between competition and bank efficiency (a rejection of the QLH), just like previous 
studies that followed a similar approach (see Koetter et al., 2012). Our hypothesis is: 
 
H7: Competition has a negative relationship with bank efficiency. 
 
The effect of financial freedom on bank efficiency has only recently been tested in the banking 
literature and found to be positive (Chortareas et al., 2013; Lin, Doan & Doong, 2016). We 
anticipate a similar finding, and thus hypothesize as: 
 
H8: Financial freedom has a positive effect on bank efficiency. 
 
 
Further, we test whether financial freedom moderates the effects of competition on bank 
efficiency, which has so far not received attention in the literature. Our hypothesis is: 
 
H9: The effect of competition on bank efficiency increases with financial freedom. 
 
 
 
Finally, given the peculiar characteristics of individual economies, coupled with the regulatory 
and institutional environments, we anticipate that different factors account for bank profit 
persistence in various countries, leading us to the following hypothesis: 
 
H10: Determinants of bank profit persistence vary across countries. 
University of Ghana http://ugspace.ug.edu.gh
16 
 
1.5 Objectives of the Study 
The main objective of this study is to examine the effects of competition and financial freedom on 
bank performance. Specifically, the study seeks to: 
i. evaluate the effects of competition and financial freedom on bank stability. 
ii. ascertain the effects of competition and financial freedom on bank profitability. 
iii. assess the effects of competition and financial freedom on bank efficiency. 
iv. examine the determinants of bank profit persistence. 
 
1.6 Significance of the Study 
This study enhances our understanding of the ambiguous relationship between competition and 
bank performance, by considering the potential effect of financial freedom on this relationship. It 
differs from most of the other studies in the literature in the following ways: First, we assess the 
impact of financial freedom on the competition-efficiency, competition-profitability, and 
competition-stability relationships (which is new in the literature). Second, we carry out a joint-
estimation of competition and efficiency using the same model. Although very necessary, given 
the close relationship between competition and efficiency, only a handful of studies have jointly 
estimated competition and bank efficiency using the same framework (see Koetter et al., 2012 for 
the few exceptions). Most studies estimate competition and efficiency separately, and then assess 
their relationship. This approach confounds our understanding of the relationship between 
competition and bank efficiency. Thus using a common framework in this study should result in a 
more appropriate policy response.  Third, we test the quiet life hypothesis which has received very 
little attention in banking as noted by Coccorese and Pellecchia (2010), especially in developing 
countries. Policy makers often take the view that promoting competition or reducing bank market 
University of Ghana http://ugspace.ug.edu.gh
17 
 
power will enhance efficiency, but that may not necessarily be the case in the SSA region, given 
the differences in institutional arrangements between developed and developing economies. 
Fourth, we test for a non-linear relationship between competition and bank stability (a new 
development in the literature following the theoretical work of Martinez-Miera and Repullo, 
2010). Fifth, we assess the determinants of bank profit persistence and variations across countries, 
which is largely ignored in the literature. Sixth, we provide a more comprehensive assessment of 
bank performance by considering bank efficiency, profitability and stability together. Most studies 
consider only one or two of these at a time which does not provide a complete picture.  
 
The Sub-Saharan African (SSA) banking markets serve as a fertile ground for a study of the 
relationship between competition and bank performance. Generally, banking markets in this region 
are less competitive compared to other regions of the world. African banks are also well-
capitalized, quite liquid, more profitable and fairly stable (Beck & Cull, 2013; Honohan & Beck, 
2007; Moyo, Nandwa, Oduor, & Simpasa, 2014; Beck, Maimbo, Faye, & Triki, 2011). 
Competitive conditions continue to improve with the gradual relaxation of remaining restrictions 
on banking activities as pertains in the developed markets in the Western world. For instance, the 
index of financial freedom developed by the Heritage Foundation (2015) which gives an indication 
of the freedom with which banks conduct their activities clear of government intervention and 
control, has increased in many of these countries over the last decade (with an average of 51 by 
2012).  
 
But as competition gets more intense in SSA banking markets along with greater financial 
freedom, how will bank efficiency be affected? Will banks in these markets continue to be 
University of Ghana http://ugspace.ug.edu.gh
18 
 
profitable, well-capitalized and stable? Or are we likely to witness bank crises again, as 
experiences from the developed markets suggest, especially given the poor institutional 
environment in these areas? Will the comparatively high net interest margins decline or increase?  
Our study contributes significantly to the literature by addressing some of the issues raised above 
and other related matters. 
 
1.7 Scope of the Study 
This thesis is limited to a discussion of the effects of competition and financial freedom on bank 
cost efficiency, profitability and stability over the 7-year period from 2006 to 2012. The study 
period was largely influenced by the availability of banking data in the Bankscope database, which 
was the main source of data for the study. Our goal was to use as much data as possible, but we 
focused on the dominant financial institution, commercial banks, and specialized financial 
institutions whose nature and operations are akin to that of commercial banks. We excluded other 
financial institutions such as investment banks since their sphere of activities are dissimilar. 
 
Our initial sample comprised of banks in all SSA countries, but due to data limitations, especially 
inadequate data points at the country level required for some of the regression estimates, we settled 
on data from 139 banks operating in 11 countries in SSA. Also, banks with less than three 
consecutive years of observations were excluded. Besides, some banks did not have values needed 
for some of the key variables used in the study and had to be excluded. The final sample is an 
unbalanced panel with 700 bank-year observations. Even so, the results should generally reflect 
the situation of the banking industry in this region, and may be applicable to other developing 
University of Ghana http://ugspace.ug.edu.gh
19 
 
countries as well, given that our sample comprise most of the largest banking markets in SSA 
(www.eiu.com). 
 
1.8 Structure of the Thesis 
This thesis has been organized into five chapters. Chapter one provides a general introduction to 
the thesis, including background of the study, the problem statement, research questions, research 
hypotheses, research objectives, significance and scope of the study. In chapter two, we review the 
relevant theoretical and empirical literature related to competition and bank efficiency, 
profitability and stability. Chapter three covers the various methodologies employed to assess the 
relationships of interest, while chapter four deals with the analysis and discussion of the results 
obtained. Chapter five concludes the thesis with a summary of the findings, contributions of the 
study, conclusions and policy recommendations, as well as limitations and suggestions for further 
research. 
 
 
 
 
 
 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
20 
 
CHAPTER TWO 
LITERATURE REVIEW 
 
2.1 Introduction  
 
This chapter reviews relevant literature on the broad areas of competition, economic freedom and 
bank performance. Section 2.2 begins with a review of the theoretical framework, methods of 
measuring competitive conditions, and general determinants of competition in banking markets.   
Section 2.3 gives an overview of the bank stability literature, drawing on both the theoretical and 
empirical literature, as well as the more recent works aimed at addressing the ambiguity of the 
competition-stability relationship. Section 2.4 deals with the bank profitability literature, while 
section 2.5 provides literature on bank efficiency. Both sections begin with an overview of the 
theoretical literature, before considering the empirical evidence. We review bank profit persistence 
literature in section 2.6, and economic freedom and bank performance literature in section 2.7. In 
section 2.8, we provide a brief overview of the African banking sector. The chapter concludes with 
a summary and gaps in the literature in section 2.9. We focus particularly on recent empirical 
studies on areas closely related to our research objectives: effect of financial freedom and 
competition on bank stability; effect of financial freedom and competition on bank profitability; 
effect of financial freedom and competition on bank efficiency; as well as the determinants of bank 
profit persistence.   
 
 
 
University of Ghana http://ugspace.ug.edu.gh
21 
 
2.2 Theoretical Framework and Determinants of Bank Competition 
 
The seminal work of Klein (1971) provides the conventional framework for examining the 
competitive behavior of banks. Klein (1971) considers a bank as a monopolistic firm engaged 
principally in two activities: taking deposits and making loans. Thus the cost function of the 
intermediation services of a bank could be represented as: 
 
C = f (D, L),        (2.1) 
 
where D refers to the volume of deposits and L represents the volume of loans produced by the 
bank. The monopoly model of Klein (1971) assumes that the banking firm operates in an 
environment characterized by imperfect competition in both the deposit and credit markets. This 
implies that the bank has monopoly power to set prices (interest rates) in at least one of the markets 
where it operates, usually the credit market. According to Klein (1971), the monopoly power of 
the bank explains its scale of operations, its asset and liability structure, as well as its decisions on 
interest rates on assets (loans) and liabilities (deposits). Thus, in this model, the spread of a banking 
firm is basically a reflection of its ability to charge a price that is higher than its marginal cost of 
production in both the loan and deposit markets.  
 
As shown by Oreiro and de Paula (2010), in a context where r is the inter-bank market rate; rL is 
the interest rate charged on loans produced by the bank; rD is the interest rate paid on deposits with 
the bank; α is the reserve requirements as a proportion of the bank’s deposits; ԑL is the interest 
elasticity of loan demand; ԑD is the interest elasticity of deposit supply; CL is the marginal cost of 
loan services; and CD is the marginal cost of deposits services, and supposing that the banking firm 
University of Ghana http://ugspace.ug.edu.gh
22 
 
is risk neutral and that its behaviour is aimed at maximizing profits, then the optimal interest 
margin on loans and deposits is given by: 
 
*
*
*
)(1
L
LL
L r
Crr 
        (2.2) 
 
*
*
*
))1(1
D
DD
D r
rCr  
       (2.3) 
 
Equations (2.2) and (2.3) suggest that a banking firm that operates in a monopolistically 
competitive environment prices its loans and deposit services in such a way that the Lerner indices 
are equal to the inverse of the interest elasticity of the loan demand and deposit supply functions. 
This implies that the greater the sensitivity of loan demand and deposit supply functions to interest 
rate changes, the smaller the bank’s margin in both loan and deposit taking services, and, hence 
the smaller the bank margin. However, if the banking firm faces an oligopolistic market structure 
in granting loans and taking deposits, then the optimal interest margin on loans and deposits is 
given by: 
 
*
*
*
)(
L
LL
L r
Crrs 
        (2.4) 
 
*
*
*
))1(
D
DD
D r
rCrs  
,       (2.5) 
University of Ghana http://ugspace.ug.edu.gh
23 
 
where s is the market share of the nth bank. Equations (2.4) and (2.5) show the bank interest margin 
on loan supply and deposit taking as a growing function of its market share. This is understood to 
mean that a reduction in the number of banking firms due to mergers or acquisitions results in 
increased banking market concentration and thus interest margins. Thus a banking firm’s interest 
margin is a growing function of the banking sector market structure (da Silva, Oreiro, de Paula & 
Sobreira, 2007). 
 
The Klein (1971) monopoly model has been criticized for assuming (for the sake of simplicity) 
that interest rate changes reflect the key risk faced by banks, and for its inability to capture the 
hedging behavior of banks adequately (Ho & Saunders, 1981). Consequently, Ho and Saunders 
(1981) proposed a ‘dealership model’ to explain bank profitability.  
 
The Ho and Saunders (1981) model considers a bank as a risk-averse dealer between the users and 
suppliers of funds. Given that the arrival of deposits and loans are not synchronized, the bank has 
to set interest rates on loans and deposits in such a way as to compensate it for providing immediate 
liquidity service at the risk of losing money, as a result of exposure to changes in money market 
interest rates (i.e. interest rate risk).  
 
Suppose a new deposit (loan) is contracted at a long-term interest rate rD (rL). If this deposit (loan) 
arrives earlier than a new loan (deposit), the bank will have to invest (borrow) the funds at the 
short-term money market rate r. In providing these services, the bank faces a reinvestment 
(refinancing) risk at the end of the decision period should the short-term rate, r, fall (rise). Hence, 
University of Ghana http://ugspace.ug.edu.gh
24 
 
the bank will transfer these financial costs which arise from the uncertainty in the provision of 
deposit and loan operations to economic agents. 
 
As a result, each bank participating in the market will set a loan and deposit interest rate, rD (and 
rL) that reflects these financial costs as follows: 
          
arrD  ; brrL  ,         (2.6) 
 
where a and b represent the financial costs related to the provision of deposits and loans 
respectively, and r is the expected money market interest rate.  Thus, the net interest spread is: 
 
)( barr DL           (2.7) 
 
Maudos and de Guevara (2004), extended the Ho and Saunders (1981) model by explicitly taking 
into account, banks’ operating costs, and used a direct measure of the degree of competition, the 
Lerner index in addition to the degree of concentration of the market (the Herfindahl index). They 
postulate that, given the initial wealth of the bank as the difference between its assets and liabilities, 
and incorporating the production costs associated with intermediation (which are assumed to be a 
function of the deposits and loans), the final wealth of the bank will be: 
 
E[U(W)] =  U(W*) + U´(W*)E[W-W*] + ½U´´(W*) E[W-W*]²    (2.8) 
 
University of Ghana http://ugspace.ug.edu.gh
25 
 
This assumes that banks are maximizers of expected utility, and that a bank’s utility function, U, 
is approximated using the Taylor expansion around the expected level of wealth. It is also assumed 
that the bank’s utility function is continuous and doubly differentiable such that U´ > 0 and U´´ < 
0, so that the bank is risk averse. 
 
Hence, the optimal interest margin is determined principally by the following: 
• The competitive structure of the markets  
• Average operating costs  
• Degree of risk aversion  
• Volatility of money market interest rates  
• Credit risk 
• The covariance or interaction between interest rate risk and credit risk 
• The average size of the credit and deposit operations 
 
The subsequent sections discuss further extensions on the theoretical literature on competitive 
behavior of banks, and its implications for performance.  
 
The main drivers of competition identified in the literature include size, default risk, 
diversification, efficiency, equity, concentration, market share, ownership, monetary policy, 
inflation, and financial depth (Aboagye, Akoena, Antwi-Asare & Gockel, 2008b; De Guevara & 
Maudos, 2007; De Guevara, Maudos, & Pérez, 2005). For instance, De Guevara et al. (2005) have 
shown that bank size has a positive and significant effect on market power. Larger banks enjoy 
greater market power due to either cost advantages or to their capacity to impose higher prices.  
University of Ghana http://ugspace.ug.edu.gh
26 
 
Mirzaei and Moore (2014) report that banks located in countries with good-quality institutional 
development face greater competition for emerging and developing economies, whereas inter-
industry competition from insurance industries, together with financial freedom, seem to be the 
main drivers in increasing competition amongst developed economies. Chortareas et al. (2013) 
also suggest that an open and transparent banking environment facilitates access to financing and 
encourages competition to provide efficient financial intermediation between households and 
firms, as well as between investors and entrepreneurs.  
 
Similarly, Claessens and Laeven (2004) find that banking systems with greater foreign bank entry 
and fewer entry and activity restrictions tend to be more competitive. But they find no evidence 
that the competitiveness measure negatively relates to banking system concentration. On the 
contrary, they find some evidence that more concentrated banking systems are more competitive. 
Their findings confirm that contestability determines effective competition especially by allowing 
(foreign) bank entry and reducing activity restrictions on banks. The results on the lack of 
importance of market structure suggest that competition policy in the financial sector is more 
complicated than perhaps previously thought. Hence, they posit that developing proper 
competitiveness tests and methodologies will remain an important area of research and policy 
focus. Again, Casu and Girardone (2006) suggest that the degree of concentration is not necessarily 
related to the degree of competition in EU banking markets. They also find little evidence that 
more efficient banking systems are also more competitive. 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
27 
 
2.3 Bank Stability 
 
This sections covers a review of the theoretical and empirical bank stability literature with a focus 
on the competition-stability relationship.  
  
2.3.1 Theoretical Review of Bank Competition-Stability Relationship 
 
The theoretical literature on the relationship between competition and bank stability presents no 
consensus.  Two opposing views are theorized in the literature. One view, called the competition–
fragility or competition-instability view, and pioneered by the influential work of Keeley (1990), 
asserts that competition in banking reduces market power of banks, decreases profit margins, and 
results in reduced franchise value that encourages banks to take on greater risks. Using U.S data 
to illustrate his model, Keeley (1990) finds that the increased competition and deregulation that 
followed the relaxation of state branching restrictions in the U.S. in the 1980s, eroded monopoly 
rents and resulted in an increase of bank failures. On the other hand, higher market (or monopoly) 
power in the banking market leads to greater stability because the greater lending opportunities, 
higher profits, higher capital ratios and charter values of incumbent banks, put them in a better 
position to survive demand-side and supply-side shocks, which in turn provides a dis-incentive for 
excessive risk-taking (Keeley, 1990; Allen & Gale, 2000, 2004; Carletti, 2008). 
 
Hellmann et al. (2000) makes a similar argument, stating that financial liberalization increases the 
potential scope for gambling by banks, since it often leads to intense competition among banks, at 
the same time that banks gain greater freedom to allocate assets and set interest rates. They point 
out that in a dynamic economy, capital requirements may not be as powerful as previously thought, 
because in addition to a one-period capital-at-risk effect that minimizes the incentive to gamble, 
University of Ghana http://ugspace.ug.edu.gh
28 
 
there is a future-franchise-value effect that increases the incentives to gamble. Further, Besanko 
and Thakor (1993) used a theoretical model to show that lending rates decrease and borrowing 
rates increase as more banks enter the market and each tries to differentiate itself from the 
competition. This results in a reduction in bank margins and charter values giving banks greater 
incentives to gamble or take on greater risk. 
 
In addition, Allen and Gale (2000) and Allen and Gale (2004) suggest that greater competition in 
banking reduces incentives to properly screen borrowers and increases the risk of fragility, since 
banks earn less informational rent from their relationships with borrowers. Allen and Gale (2000) 
also show that a small number of large institutions in a concentrated banking system makes for 
more stability and resilience to shocks, since banks are easier to monitor and supervisors are less 
burdened. 
  
In contrast, the competition–stability view came out strongly from the work of Boyd and de Nicoló 
(2005), who argued that competition leads to greater bank stability. They assert that as banking 
markets become more concentrated (and less competitive), banks tend to show market power by 
charging higher interest rates to borrowers. This worsens moral hazard incentives and makes it 
more difficult for borrowers to repay, as they also take on greater risks in search of more profits, 
and the result is an increase in the possibility of loan default and therefore the risk of bank 
portfolios. On the other hand, a reduction in interest rates on loans as a result of greater competition 
would encourage borrowers to seek safer investments, thus minimizing the risk of default and bank 
portfolio risk. Similarly, while Caminal and Matutes (2002) show that banks with more market 
power (in less competitive markets), have to contend with monitoring costs and are more likely to 
University of Ghana http://ugspace.ug.edu.gh
29 
 
create risky loan portfolios, Beck, Demirguc-Kunt, and Maksimovic (2004) state that both greater 
concentration in banking markets and higher competition are more conducive to bank stability. 
 
Recently, Martinez-Miera and Repullo (2010) questioned the theoretical predictions of Boyd and 
de Nicoló (2005) by arguing that lower loan rates resulting from greater bank competition also 
reduces the interest payments from performing loans (which provide a buffer against loan losses) 
because of imperfect correlation of loan defaults.  According to Martinez-Miera and Repullo 
(2010), in addition to the risk-shifting effect in the Boyd and de Nicoló (2005) model, there is also 
a margin effect that goes in the opposite direction, so that the final effect on the risk of bank failure 
is in principle ambiguous. They show that the risk-shifting effect tends to dominate in monopolistic 
markets, whereas the margin effect dominates in competitive markets, so a U-shaped relationship 
between competition and the risk of bank failure generally obtains. 
 
2.3.2 Empirical Evidence on Competition-Stability Relationship 
 
Similar to the conflicting theoretical predictions of the competition and bank stability relationship, 
empirical evidence in support of the competition–fragility or competition–stability views is also 
inconclusive. 
 
Recent studies that provide evidence for the competition–fragility hypothesis include Beck et al. 
(2006) who show that crises are less likely in economies with more concentrated banking systems 
even after controlling for differences in commercial bank regulatory policies, national institutions 
affecting competition, macroeconomic conditions, and shocks to the economy. Berger et al. (2009) 
University of Ghana http://ugspace.ug.edu.gh
30 
 
also find evidence consistent with the traditional “competition-fragility” view that banks with a 
higher degree of market power in developed nations also have less overall risk exposure.  
 
Besides, Turk-Ariss (2010) finds that an increase in the degree of market power (less competition) 
leads to greater bank stability, and suggests that increased competition may undermine bank 
stability, and may have significant repercussions for stressed banking systems in developing 
economies. Amidu (2013) also suggests that the relatively low insolvency risk among banks in 
emerging and developing countries during 2000–2007 could be attributed to the high degree of 
market power and the use of internally generated funds. Similarly, Fungacova and Weill (2013) 
find that increase in bank competition could undermine financial stability in the case of Russian 
banks, for the period 2001–2007. For Africa, Kouki and Al-Nasser (2014) argue that gains in 
market power will increase the stability and reduce the risk for the banking system. 
 
More recently, Diallo (2015) reports that bank competition is detrimental to bank stability, and 
also shortens the survival time of banking systems, using data from 145 countries over the period 
1997–2010, and three measures of bank competition, namely the Boone indicator, the conventional 
Lerner and the adjusted Lerner indices.  
 
In contrast, a number of studies have found support for the hypothesis that competition enhances 
bank stability. For example, Uhde and Heimeshoff (2009) find that Eastern European banking 
markets which exhibit a lower level of competitive pressure are more prone to financial fragility, 
and that national banking market concentration has a negative impact on European banks’ financial 
soundness.  
University of Ghana http://ugspace.ug.edu.gh
31 
 
Also, using the Panzar and Rosse (1987) H-statistic as a measure of competition for 45 countries, 
Schaeck et al. (2009) find that more competitive banking systems are less prone to experience a 
systemic crisis and exhibit increased time to crisis. Again, with a sample of European Banks, 
Schaeck and Cihak (2014) report that competition is stability-enhancing, and that the stability-
enhancing effect of competition is greater for healthy banks than for fragile ones.  
 
2.3.3 Is There a Transmission Mechanism Between Competition and Stability? 
 
The results of some other recent studies suggest that the relationship between competition and 
bank stability is not straightforward and may actually depend on other factors previously ignored 
in the literature. For instance, Agoraki, Delis, and Pasiouras (2011) investigate whether regulations 
have an independent effect on bank risk-taking or whether their effect is channeled through the 
market power possessed by banks. Using data from the Central and Eastern European banking 
sectors over the period 1998–2005, the empirical results suggest that higher activity restrictions in 
combination with more market power reduce both credit risk and the risk of default, while official 
supervisory power has only a direct impact on bank risk. Overall, it appears that ignoring the 
interactions between regulations and market power leads to erroneous inferences about the impact 
of regulations on both credit risk and overall solvency risk. They contend that this provides 
stimulus for a more disaggregate analysis of the impact of bank regulations on risk-taking 
activities. In particular, it would be interesting to analyze which of the elements used to compose 
the regulatory indices are the most relevant in explaining bank risk. Also, another possible area for 
future research could be to provide a more detailed analysis of the different country-specific 
institutional characteristics that may affect bank risk and how these characteristics are affected by 
University of Ghana http://ugspace.ug.edu.gh
32 
 
the competitive conditions in the banking sector or of other industries that were involved in the 
recent financial turmoil, such as the insurance industry. 
 
Tabak et al. (2012) also corroborate the conclusions of Martinez-Miera and Repullo (2010) in 
finding that competition affects risk-taking behaviour in a non-linear way in Latin American 
countries as both high and low competition levels enhance financial stability, with the opposite 
effect for average competition. In addition, bank size and capitalization are found to be essential 
factors in explaining this relationship. On the one hand, the larger a bank is, the more it benefits 
from competition. On the other hand, a greater capital ratio is advantageous for banks that operate 
in collusive markets, while capitalization only enhances the stability of larger banks under high 
and average competition. 
 
Likewise, Liu et al. (2013) find evidence pointing to the fact that an inverted U-shaped relationship 
exists between regional bank competition and stability, using regional data to examine bank 
stability in 10 European countries over the period 2000-2008, thus providing support for the 
argument for a U-shaped relationship between competition and the risk of bank failure proposed 
by Martinez-Miera and Repullo (2010). Jimenez et al. (2013) also tested this hypothesis using data 
from the Spanish banking system and find support for this non-linear relationship using standard 
measures of market concentration in both the loan and deposit markets. However, when direct 
measures of market power, such as Lerner indices, are used, the empirical results are more 
supportive of the original franchise value hypothesis, but only in the loan market. They argue that 
the overall results highlight the empirical relevance of the Martinez-Miera and Repullo model, 
even though further analysis across other banking markets is needed. 
University of Ghana http://ugspace.ug.edu.gh
33 
 
Using a sample of 978 banks in 55 emerging and developing countries over 2000–2007, Amidu 
and Wolfe (2013a) report that competition increases bank stability as diversification across and 
within both interest and non-interest income generating activities of banks increases. Their analysis 
identifies revenue diversification as a channel through which competition affects bank insolvency 
risk in emerging countries.  
 
Further, Beck et al. (2013) document large cross-country variation in the relationship between 
bank competition and bank stability, and show that an increase in competition will have a larger 
impact on banks’ fragility in countries with stricter activity restrictions, lower systemic fragility, 
better developed stock exchanges, more generous deposit insurance and more effective systems of 
credit information sharing.  
 
Fu, Lin, and Molyneux (2014) also suggest that greater concentration in Asian banking markets 
nurtures financial fragility and that lower pricing power also induces bank risk exposure. In other 
words, the findings provide support for a neutral view of the competition-stability nexus, indicating 
that the competition-stability and competition-fragility theories can simultaneously apply to Asia 
Pacific banking markets.  
 
Forssbæck  and Shehzad (2014) also investigate the relationship between market power and risk 
for a large panel of banks worldwide and find that both loan and deposit market power  have a 
stable, monotonically negative effect on risk, irrespective of risk measure. The effect is larger for 
asset risk, and is independent of charter value and capital ratios. However, the effect on default 
University of Ghana http://ugspace.ug.edu.gh
34 
 
risk tends to decrease in the quality of banking regulation, whereas the conditioning effects of 
deposit insurance protection are mixed.  
 
2.4 Bank Profitability 
 
This section reviews the literature on bank profitability, including some basic theoretical concepts, 
empirical evidence on the competition-profitability relationship, and general determinants of 
profitability. However, we have separate discussions for bank profitability in general, and net 
interest margins in particular. 
 
2.4.1 Theoretical Review of Determinants of Bank Profitability 
  
A number of other extensions have been made to the basic Ho and Saunders (1981) model. As 
explained in Almarzoqi and Naceur (2015), some simplifying assumptions of the Ho and Saunders 
(1981) model have since been relaxed. For example, Allen (1988) provides for heterogeneity in 
the model (since banks offer different types of deposit and loan services), and show that pure 
interest margins may be reduced as a result of diversification of bank services and products. This 
approach deals with the cross–elasticities of demand between bank products and services. 
McShane and Sharpe (1985) replaced the volatility of the deposit or loan rates, as in Ho and 
Saunders (1981), with a more relevant variable, the volatility of the money market interest rate. 
Angbazo (1997) also extended the Ho and Saunders (1981) model by incorporating credit risk and 
its interaction with interest rate risk. Wong (1997) derived an alternative model to explain the 
bank’s optimal net interest spread in which banks are assumed to set the deposit and loan rate 
simultaneously. More recently, Carbo and Rodriguez (2007) further developed the model of Ho 
University of Ghana http://ugspace.ug.edu.gh
35 
 
and Saunders (1981) by including both traditional and non-traditional activities in order to study 
the effect of specialization on bank spreads, using a multi-output model for European banking.  
 
Besides the monopoly theory of Klein (1971) and the dealership model of Ho and Saunders (1981), 
two other competing theories explaining the relationship between market structure and bank 
profitability emerged from the industrial organization literature, specifically the Concentration-
Profit Theory proposed by Bain (1951).  The theory as applied to banking, asserts that higher 
concentration in banking markets results in collusion among banks who take advantage of their 
market power to set higher prices and gain substantial profits. However, Demsetz (1973) proposed 
an alternative explanation for the market structure-profitability relationship (the Efficient-
Structure-Hypothesis).  
 
Thus, as explained by Tregenna (2009), the traditional literature falls into two broad approaches: 
the Market Power (MP) and Efficiency Structure (ES) paradigms. These approaches provide 
different suggestions for the direction of causality between concentration and profitability. For the 
MP hypothesis, the direction of causality runs from market structure to behaviour, and then 
performance. In other words, a concentrated structure promotes the use of market power in ways 
that may enhance banks’ profitability. In contrast, the ES hypothesis perceives causality as running 
from individual firms’ efficiency to their market share and profitability. 
 
Further, within the MP paradigm, two different approaches emerge in the literature: the Structure–
Conduct Performance (SCP) hypothesis, and the Relative Market Power (RMP) hypothesis. The 
SCP hypothesis asserts that concentration in a banking market gives rise to potential market power 
University of Ghana http://ugspace.ug.edu.gh
36 
 
by banks which allows them to set prices that are less favorable to consumers (lower deposit rates, 
higher loan rates) which may then increase their profitability. A related theory is the relative-
market-power hypothesis (RMP), which posits that only firms with large market shares and well-
differentiated products are able to exercise market power in pricing their products and earn 
supernormal profits (Shepherd, 1982). Thus, whereas the SCP hypothesis would predict generic 
benefits to banks arising from higher concentration, the RMP hypothesis sees any benefits as 
accruing to individual banks based on their own market share. According to the latter approach, 
only large banks can influence prices and increase profits (Tregenna, 2009). 
 
Again, within the ES paradigm there are also two theories that explain the positive relationship 
between profits and either concentration or market share: the X-efficiency (ESX) and Scale-
Efficiency (ESS) hypotheses. The X-efficiency version of the Efficient-Structure hypothesis 
(ESX) asserts that firms with superior management or production technologies have lower costs 
and therefore higher profits. Such firms tend to gain larger market shares, which may manifest in 
higher levels of market concentration, but without any causal relationship from concentration to 
profitability. The scale efficiency approach within the ES paradigm emphasizes economies of scale 
rather than differences in management or production technology. Larger firms can obtain lower 
unit costs and higher profits through economies of scale. Again, as these firms have higher market 
shares, which may manifest in higher concentration, there may be an apparent—yet spurious—
relationship between concentration and profitability. Thus, according to the ES approaches, a 
positive correlation between concentration and profitability need not indicate a causal relationship, 
especially not through market power (Tregenna, 2009).  
 
University of Ghana http://ugspace.ug.edu.gh
37 
 
The two market-power (MP) hypotheses have radically contrasting implications from the two 
efficient-structure (ES) hypotheses for merger and antitrust policy. To the extent that the MP 
hypotheses are correct, mergers may be motivated by desires to set prices that are less favorable 
to consumers, which would decrease total consumer plus producer surplus. But to the extent that 
the ES hypotheses are correct, these mergers may be motivated by efficiency considerations that 
would increase total surplus. Thus, advocates of the MP hypotheses tend to see antitrust 
enforcement as socially beneficial, while ES advocates tend to see policies that inhibit mergers as 
socially costly (Berger, 1995).  
 
In a distinguished contribution to this literature, Berger (1995) entered the debate on the profit-
structure relationship in banking very strongly, by adding direct measures of both X-efficiency 
and scale efficiency to the empirical analysis, in order to test all four hypotheses, in a single 
specification. Using 30 separate cross-sectional datasets, he regressed profits against measures of 
concentration, market share, X-efficiency, and scale efficiency. He also regressed concentration 
and market share against the efficiency variables to test the necessary condition of the ES 
hypotheses that efficiency affects market structure. The empirical results provide some support for 
the RMP and partial support for the X-efficiency version of the ES hypothesis (ESX). However, 
support for the other necessary condition of ESX, that X-efficiency is positively related to 
concentration or market share so that it can explain the positive profit-structure relationship, is 
much weaker. 
 
While some prior studies did control for scale efficiency, they generally did not control for X-
efficiency, and so were unable to distinguish whether the profit-structure relationship observed 
University of Ghana http://ugspace.ug.edu.gh
38 
 
reflected superior management or greater market power of firms with large market shares. Prior 
studies also did not test whether efficiency has the predicted efficient-structure effects on market 
structure (Berger, 1995). 
 
2.4.2 Empirical Evidence on Determinants of Bank Net Interest Margins 
 
Empirical research on the impact of competition or market structure on bank interest margins have 
produced conflicting results. As noted by Saunders and Schumacher (2000), the effect of market 
structure on bank spreads appears to vary across countries. The more segmented or restricted the 
banking system, in terms of geographic restrictions on branching and universality of banking 
services, the larger appears to be the monopoly power of existing banks and the higher their 
spreads. Other studies that have found a positive relationship between market power and net 
interest margins or a negative relationship between competition and net interest margins include 
Almarzoqi and Naceur (2015) for Caucasus and Central Asia, Hawtrey and Liang (2008) for 
OECD countries, Hossain (2012) for Bangladesh, Chortareas et al. (2012) for Latin American 
banking, Van Leuvensteijn et al. (2013) for euro area banks, and Maudos and Solís (2009) for 
Mexico. 
 
In addition, Amidu and Wolfe (2013b) found that the high net-interest margins of banks in 
emerging and developing countries can be explained by the degree of market power. Poghosyan 
(2013) also analyze factors driving persistently higher financial intermediation costs in 48 low-
income countries-LICs (including some countries in Sub-Saharan Africa), relative to 67 emerging 
market-EM country comparators (including some countries in Northern and Southern Africa) for 
the period 1996–2010. Overall, he finds that concentrated market structures and lack of 
University of Ghana http://ugspace.ug.edu.gh
39 
 
competition in LICs banking systems and institutional weaknesses constitute the key impediments 
preventing financial intermediation costs from declining. The results provide strong evidence that 
policies aimed at fostering banking competition and strengthening institutional frameworks can 
reduce intermediation costs in LICs. 
 
For Africa, Ahokpossi (2013) examined the determinants of bank interest margins using a sample 
of 456 banks in 41 Sub-Saharan African (SSA) countries. The results show that market 
concentration is positively associated with interest margins, but the impact depends on the level of 
efficiency of each bank. In particular, compared to inefficient banks, efficient ones increase their 
margins more in concentrated markets. This indicates that policies that promote competition and 
reduce market concentration would help lower interest margins in SSA. Similarly, while Aboagye 
et al. (2008a) found that increases in bank market power (or concentration) significantly increase 
net interest margin in Ghana, Chirwa and Mlachila (2004) suggest that the observed high spreads 
in Malawi can be attributed to high monopoly power. 
 
In contrast, Maudos and de Guavara (2004) analyze the interest margin in the principal European 
banking sectors (Germany, France, the United Kingdom, Italy and Spain) in the period 1993–2000 
using a panel of 15,888 observations, and identifying the fundamental elements affecting this 
margin. They followed the methodology developed in the original study by Ho and Saunders 
(1981) and later extensions, but widened it to take banks’ operating costs explicitly into account. 
The results show that the fall of margins in the European banking system is compatible with a 
relaxation of the competitive conditions (increase in market power and concentration), as this 
effect has been counteracted by a reduction of interest rate risk, credit risk, and operating costs. 
University of Ghana http://ugspace.ug.edu.gh
40 
 
On the other hand, while Kasman et al. (2010) found market power to be a vital determinant of 
bank interest margin in both new and old EU member countries, the effect was opposite for the 
two groups. They suggest that the apparent difference might be the fact that increases in the degree 
of concentration in the old member countries influenced by the process of mergers and acquisitions 
triggered an upward pressure on the competition, thus narrowing the interest margins. Beck and 
Hesse (2009) also found little evidence for market structure explaining variation in spreads or 
margins over time for Uganda. 
 
The empirical literature on the relationship between bank size and interest margins presents 
contrasting results. Among those reporting a positive relationship are Almarzoqi and Naceur 
(2015), Aboagye et al. (2008a), and Maudos and Solis (2009). On the contrary, Kasman et al. 
(2010) Fungacova and Poghosyan (2011), and Maudos and de Guevara (2004) suggest a negative 
relationship between bank size and interest margins. 
 
Most studies have reported a positive relationship between credit risk and interest margins 
indicating that banks charge additional risk premiums to compensate for credit risk (Naceur & 
Omran, 2011; Hossain, 2012; Chortareas et al., 2012a; Amidu & Wolfe, 2013b; Ahokpossi, 2013). 
However, similar to Fungacova and Poghosyan (2011), Almarzoqi and Naceur (2015) found a 
negative correlation between credit risk and interest margins, and considers that as being reflective 
of inadequate interest spreads (mispricing of risks) to compensate for provisions for non-
performing loans.  
 
University of Ghana http://ugspace.ug.edu.gh
41 
 
Likewise, with regard to risk aversion (or capitalization), most empirical findings report a positive 
effect on net interest margins. Among others, Saunders and Schumacher (2000), Aboagye et al. 
(2008a), Hawtrey and Liang (2008), and Fungacova and Poghosyan (2011) found that the level of 
capitalization has a positive and significant impact on bank interest margins, suggesting an 
important policy trade-off between assuring bank solvency—high capital-to-asset ratios—and 
lowering the cost of financial services to consumers—low net interest margins. It appears that as 
banks deploy higher equity, they charge their customers more so as to earn enough to service the 
higher expected returns of shareholders (Aboagye et al., 2008a). 
 
Kasman et al. (2010) report that managerial efficiency (measured by the ratio of operating cost to 
income) is negatively and significantly related to net interest margins, implying that higher 
managerial efficiency stimulates banks to offer higher deposit rates and lower loan rates to their 
clients. Angbazo (1997), Maudos and de Guevara (2004), Hawtrey and Liang (2008), and Claeys 
and Vennet (2008) report similar findings, but Gischer and Juttner (2003) suggest that improved 
quality of management should narrow down the interest margin due to efficiency. Almarzoqi and 
Naceur (2015) note that more recent studies find that banks with well-developed non-interest 
income sources have lower net interest margins, suggesting that banks may tend to offer loans with 
small or even negative margins to attract clients and compensate with higher fees. In contrast, 
Chiorazzo, Milani, and Salvini (2008) and Elsas, Hackethal, and Holzhauser (2010) assert that 
revenue diversification enhances bank profitability via higher margins from non-interest 
businesses. Nguyen (2012) also found that net interest margin is not always inversely related to 
diversification. 
 
University of Ghana http://ugspace.ug.edu.gh
42 
 
Naceur and Omran (2011) suggest that macroeconomic and financial development indicators have 
no significant impact on net interest margins, except for inflation. On the other hand, regulatory 
and institutional variables seem to have an impact on bank performance. Likewise, Ahokpossi 
(2013) found inflation to be positively related to interest margins, but there is no conclusive 
evidence that economic growth has any impact on margins. However, as pointed out by Ahokpossi 
(2013), there is a dearth of research on interest margins in Sub-Saharan African (SSA).  
 
2.4.3 Empirical Evidence on Determinants of Bank Profitability 
 
A sound and profitable banking sector is better able to withstand negative shocks and contribute 
to the stability of the financial system. Therefore, the determinants of bank profitability have 
attracted the interest of academic research as well as of bank management, financial markets and 
bank supervisors (Athanasoglou, Brissimis, & Delis, 2008). Most recent studies on the relationship 
between bank competition and profitability test the SCP and/or the ES hypotheses. Generally, the 
results have been mixed. While some have found support for the SCP, other studies find no 
evidence of market structure or market power having an effect on bank profitability. Instead, some 
have found that the level of bank profitability is explained by efficiency. Of course, even when 
market structure or market power is found to affect profitability, the outcome is not always the 
same. In some cases profitability is positively affected by market structure, while the effect is 
negative in other situations.  
 
For instance, Tregenna (2009) analyses the effects of structure on bank profitability in the U.S 
from 1994 to 2005. He found evidence that market concentration increases bank profitability. This 
holds even when the largest banks are excluded from the sample, suggesting that the relationship 
University of Ghana http://ugspace.ug.edu.gh
43 
 
between concentration and profitability acts in a generalized structural way and that the higher 
profits arising from concentration are at the expense of the rest of the economy. The analysis points 
to various policy implications relevant to the current crisis, in particular in terms of the legitimacy 
of expectations of the restoration of pre-crisis profit rates and the need for much stronger regulation 
of the banking sector, especially in terms of the structure of the sector, pricing behaviour and use 
of profits. Closely related to this is Fu and Heffernan (2009), who investigate the relationship 
between market structure and performance in China’s banking system from 1985 to 2002, a period 
when this sector was subject to gradual but notable reform. Their results lend some support to the 
relative market-power hypothesis in the early period. 
  
Similarly, Perera et al. (2013) find that even though increasing competition (arguably driven by 
on-going deregulation and liberalization of the financial services industries) exerts negative 
pressure on bank profitability, high levels of industry concentration still allows South Asian banks 
to earn higher profits. And for the Arab Gulf Cooperation Council (GCC) region, Al-Muharrami 
and Matthews (2009) find that the performance in banking industry is best explained by the 
mainstream SCP hypothesis. Also, Jeon and Miller (2002) examined the evidence, if any, of the 
relationship between several measures of bank concentration at the state level (in the U.S) and the 
average performance of banks within that state. They find a robust positive correlation between 
bank concentration in a state and the average return on equity within that state. Moreover, the 
linkage appears to run from increasing bank concentration to increasing bank profitability, and not 
the reverse. Those observations, they conclude, imply that the market power, rather than the 
efficient-structure, hypotheses holds for the U.S. banking industry during the last quarter of the 
University of Ghana http://ugspace.ug.edu.gh
44 
 
20th century. They suggest that bank regulators need to monitor the consolidation process within 
the banking industry to head off the accumulation of monopoly power.   
 
Mirzaei et al. (2013) empirically investigate the effects of market structure on profitability of 1,929 
banks in 40 emerging economies (Eastern Europe and Middle East) and advanced economies 
(Western Europe) over 1999–2008 by incorporating the traditional Structure-Conduct-
Performance (SCP) and Relative-Market-Power (RMP) hypotheses. They observe that a greater 
market share leads to higher bank profitability being biased toward the RMP hypothesis in 
advanced economies, yet neither of the hypotheses is supported for profitability in emerging 
economies. The evidence also highlights that profitability increase with increased interest margin 
revenues in a less competitive environment for emerging markets. In addition, Goddard, 
Molyneux, and Wilson (2004) find some evidence of a positive association between concentration 
and profitability, but little evidence of a link between bank-level x-inefficiency and profitability. 
 
In contrast, Seelanatha (2010) suggest that the traditional SCP argument is not held in the banking 
industry in Sri Lanka, and that bank profitability does not depend on either market concentration 
or market power of individual firms but on the level of efficiency of the banking units. Likewise, 
Chortareas et al. (2011) suggest that despite the significant rise in takeovers from foreign banks 
and the increase in market concentration, banks’ profits do not seem to be explained by greater 
market power in Latin America. Instead, efficiency (particularly scale efficiency) seems to be the 
main driving force of increased profitability for most countries. The key implication is that policies 
aimed at removing the remaining barriers to competition should be expected to benefit the banking 
system without being detrimental to consumers.  
University of Ghana http://ugspace.ug.edu.gh
45 
 
 
And while Flamini, Schumacher and McDonald (2009) assert that bank profits are high in Sub-
Saharan Africa (SSA) compared to other regions, they do not obtain conclusive results as to 
whether market power influences bank returns. 
 
In regard to other determinants of bank profitability, Athanasoglou et al. (2008) show that bank 
capital is important in explaining bank profitability, while increased exposure to credit risk and 
high operating expenses lowers profits, showing that cost decisions of bank management are 
instrumental in influencing bank performance. Similarly, Liu and Wilson (2010) find evidence that 
well capitalized, efficient banks, with lower credit risks tend to outperform less capitalized, less 
efficient counterparts with higher credit risks in Japan. Perera et al. (2013) also reveal that well-
capitalized banks and those with relatively more efficient production processes in South Asia are 
the more profitable. On the contrary, Goddard et al. (2004) and Goddard et al. (2013) show that 
banks that maintain high capital-assets ratios tend to record relatively low profitability among 
some European Union countries, suggesting that the opportunity cost of holding high levels of 
capital tends to depress shareholder returns. 
 
Perera et al. (2013) report that South Asian banks seem to experience economies of scale as bank 
size is positively associated with profitability. This is similar to Flamini et al. (2009) who found 
that higher returns on assets are associated with larger bank size in Sub-Saharan Africa (SSA), but 
in contrast with what Athanasoglou et al. (2008) found on size in Greece.  
 
University of Ghana http://ugspace.ug.edu.gh
46 
 
On the effect of diversification, Goddard et al. (2013) provide consistent evidence that more highly 
diversified banks outperform their more highly focused counterparts, similar to what Flamini et al. 
(2009) found for SSA. Also, Chiorazzo et al. (2008) and Elsas et al. (2010) assert that revenue 
diversification enhances bank profitability via higher margins from non-interest businesses. 
However, other previous studies (e.g. Stiroh & Rumble, 2006) show that greater diversification of 
the banking business does not necessarily translate into an improvement of the bank’s profitability. 
In fact, such diversification may be detrimental to profitability.  
 
Other studies highlight the important role of environmental factors. For instance, with 795 
individual banks in 39 countries covering the period 1999–2006, Lee, Hsieh, and Dai (2012) report 
that a lower level of economic development of the host country enhances the positive effects of 
foreign bank ownership on the income, profit and cost of domestic banks. And Liu and Wilson 
(2010) find that Gross Domestic Product (GDP) growth and the extent of stock market 
development play an important role in determining the profitability of Japanese banks. Perera et 
al. (2013) indicate that slack legal systems in South Asian countries (leading to inferior contract 
enforcement) positively affect profits as banks probably require higher risk premiums on their loan 
contracts. Naceur and Omran (2011) examine the influence of bank regulation and financial and 
institutional development on commercial bank profitability across a broad selection of Middle East 
and North Africa (MENA) countries. The empirical results suggest that regulatory and institutional 
variables seem to have an impact on bank performance. 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
47 
 
2.5 Bank Competition and Efficiency 
 
In this section, we review the theoretical literature and empirical evidence on the relationship 
between competition or market power and bank efficiency. The literature on this issue is related 
to the hypotheses that explain the relationship between market structure and performance. Our 
focus is on the Quiet Life hypothesis (QLH), which is considered as a special case of one of the 
market structure and performance hypotheses, the market power hypothesis (Maudos & de 
Guevara, 2007). 
 
2.5.1 Theoretical Review of Bank Competition-Efficiency Relationship  
 
As discussed under section 2.4.1, there are two main theories explaining the relationship between 
market structure and performance. These are the Market Power (MP) and Efficiency Structure 
(ES) hypotheses. These theories provide different suggestions for the direction of causality 
between concentration and profitability. For the MP hypothesis, the direction of causality runs 
from market structure to behaviour, and then performance. In other words, a concentrated structure 
promotes the use of market power in ways that may enhance banks’ profitability. In contrast, the 
ES hypothesis perceives causality as running from individual firms’ efficiency to their market 
share and profitability, asserting that the most efficient banks obtain both greater profitability and 
market shares and, as a consequence, the market becomes more concentrated. Hence, the positive 
relationship observed between concentration and profitability is actually spurious and simply 
reflects the relationship between superior efficiency, market share and concentration (Maudos & 
de Guevara, 2007). Subsequently, Berger (1995) divided this hypothesis into the X-efficiency and 
scale efficiency hypotheses. 
 
University of Ghana http://ugspace.ug.edu.gh
48 
 
In addition, within the MP paradigm, two different approaches emerge in the literature: the 
Structure–Conduct Performance (SCP) hypothesis, and the Relative Market Power (RMP) 
hypothesis. The SCP hypothesis asserts that concentration in a banking market gives rise to 
potential market power by banks which allows them to set prices that are less favorable to 
consumers (lower deposit rates, higher loan rates) which may then increase their profitability. A 
related theory is the relative-market-power hypothesis (RMP), which posits that only firms with 
large market shares and well-differentiated products are able to exercise market power in pricing 
their products and earn supernormal profits (Shepherd, 1982). Thus, whereas the SCP hypothesis 
would predict generic benefits to banks arising from higher concentration, the RMP hypothesis 
sees any benefits as accruing to individual banks based on their own market share. According to 
the latter approach, only large banks can influence prices and increase profits (Tregenna, 2009). 
 
A special case of the market power hypothesis, is the ‘Quiet Life’ hypothesis (QLH), originally 
proposed by Hicks (1935). This hypothesis suggests that the higher the market power enjoyed by 
a firm, the lower the effort put forth by managers to improve efficiency, resulting in a negative 
relationship between market power and efficiency. Berger and Hannan (1998) summarize the 
reasons that may explain the influence of market structure, as a proxy for market power, on 
efficiency. First, according to the quiet life hypothesis, if banks that compete in a market with 
higher concentration can set prices above marginal costs, managers do not have incentives to work 
as hard to keep costs under control. In other words, monopoly power allows managers to relax 
their efforts. Second, market power may allow managers to pursue objectives other than profit 
maximization. Third, in a non-competitive market, managers may apply resources to obtain and 
maintain market power, raising costs and reducing efficiency. And fourth, if banks enjoy market 
University of Ghana http://ugspace.ug.edu.gh
49 
 
power, incompetent managers can survive without willfully shirking work efforts (Maudos & de 
Guevara, 2007). 
 
As an alternative, the efficient structure hypothesis submits that more efficient banks are better 
armed to survive competitive pressures and gain market share at the expense of the less efficient 
ones (Demsetz, 1973). Thus, while the quiet life hypothesis posits a negative relationship between 
market power and bank efficiency, the relationship is positive should the efficient structure 
hypothesis prevail.  
 
Schaeck and Cihak (2008) categorize these hypotheses as ‘competition-efficiency’ and 
competition-inefficiency hypotheses. Under the ‘competition-efficiency’ hypothesis, an 
adaptation from the efficient structure hypothesis postulated by Demsetz (1973), increases in 
competition lead to increases in bank efficiency. An exogenous shock such as deregulation, forces 
banks to minimize costs, offer services at lower prices, while compelling them to increase profits 
at the same time through changes in output strategy. In such a scenario, efficient banks (those with 
superior management and production technologies that translate into higher profits) will increase 
in size and market share at the expense of less efficient banks, leading to higher market 
concentration. On the other hand, in contrast, uncompetitive markets allow bank managers to enjoy 
a ‘quiet life’ whereby costs are not kept under control, leading to lower levels of efficiency. Under 
this hypothesis, we can expect competition to bring about higher bank efficiency. (Schaeck & 
Cihak, 2008). 
 
University of Ghana http://ugspace.ug.edu.gh
50 
 
The alternative, the Competition-Inefficiency’ hypothesis suggests that competition leads to a 
decline in bank efficiency. According to Schaeck and Cihak (2008), higher competition is likely 
to be associated with less stable, shorter relationships between customers and banks as customers 
tend to switch to other banks in more competitive environments. This situation worsens 
information asymmetry problems that require additional resources for screening and monitoring 
borrowers. Also, banks are likely to reduce efforts at building relationships with borrowers in 
competitive markets, since such relationships may have shorter durations, resulting in less 
reusability and value of information. Consequently, banks incur higher costs in retaining old and 
attracting new customers through investments in technology and marketing efforts. Thus, this 
alternative hypothesis implies that competition decreases bank efficiency. 
 
2.5.2 Empirical Evidence on Competition-Efficiency Relationship 
 
Similar to the competition-stability, and competition-profitability relationships, the empirical 
evidence on the competition-efficiency relationship has been inconsistent. Indeed, only some 
recent studies have tried to test the quiet life hypothesis in banking (Coccorese & Pellecchia, 2010).  
 
Berger and Hannan (1998) report evidence of the quiet life hypothesis for the U.S regional banking 
markets. Likewise, Delis and Tsionas (2009) provide an empirical framework for the joint 
estimation of efficiency and market power of individual banks. The model is applied to the EMU 
and U.S banking industries and the findings suggest that most banks are characterized by 
moderately competitive behavior. In addition, a clear negative relationship is identified between 
the level of market power and efficiency of individual banks, a result in line with the theory 
underlying the quite life hypothesis of Hicks (1935). Besides, Coccorese and Pellecchia (2010) 
University of Ghana http://ugspace.ug.edu.gh
51 
 
test the quiet life hypothesis (QLH) using data on the Italian banking industry. They estimate bank-
level cost efficiency scores and Lerner indices, and then use the estimated market power measures, 
as well as a vector of control variables, to explain cost efficiency. The empirical evidence supports 
the QLH, although the impact of market power on efficiency is not particularly remarkable in 
magnitude. For the developing economies context, Turk-Ariss (2010) investigates how different 
degrees of market power affect bank efficiency, and show that market power leads to significant 
cost efficiency losses.  
 
In contrast, Weill (2004) provides support for a negative relationship between competition and 
efficiency in banking (a rejection of the QLH), for a sample of EU countries, using the Rosse-
Panzar H-Statistic to measure competition, and stochastic frontier approach to obtain the efficiency 
scores. Similarly, Casu and Girardone (2006) found that the most efficient EU banking systems 
are also the least competitive, suggesting that the pro-competitive deregulation of the EU banking 
markets might have led to increased market power for the most efficient banks, therefore lowering 
the pressures initially resulting from increased competition. 
 
Further, Maudos and de Guavara (2007) show the existence of a positive relationship between 
market power and cost X-efficiency in the EU banking industry, allowing a rejection of the so-
called quiet life hypothesis. They also show that the welfare gains associated with a reduction of 
market power are greater than the loss of bank cost efficiency, showing the importance of 
economic policy measures aimed at removing the barriers to outside competition. They conclude 
that the discrepancy of their results from previous ones obtained for the U.S banking industry show 
the need for additional empirical evidence referring to other banking sectors, because there are 
University of Ghana http://ugspace.ug.edu.gh
52 
 
hardly any studies that estimate the costs of market power, whether in terms of social welfare or 
in terms of bank cost efficiency.  
 
With the use of quarterly data for Czech banks, Pruteanu-Podpiera et al. (2008) estimate the effects 
of banking competition in the Czech Republic. They found negative causality only running from 
competition to efficiency. They therefore reject the intuitive ‘quiet life’ hypothesis and indicate a 
negative relationship between competition and efficiency in banking. Somewhat similar results on 
the relationship between competition and efficiency were found by Casu and Girardone (2009) in 
the banking sectors of five EU countries by using Granger-type causality test estimations. They 
found positive causation between market power and efficiency, but the causality running from 
efficiency to competition was weak. This does not support Hick's (1935) quiet life hypothesis as 
they indicate that an increase in banks' monopoly power (market power) does not translate into a 
decrease in cost efficiency. However, they could not conclude that a rejection of the quiet life 
hypothesis in turn supports the efficient structure paradigm, as the results of the reverse causality 
running from efficiency to competition do not suggest that increases in efficiency foster market 
power. 
 
Poshakwale and Qian (2011) examine the relationship between industry competition, productive 
efficiency and economic growth in Egypt by employing an endogenous growth model. Their 
findings suggest that industry competition measures are negatively affecting the cost efficiency 
measure in the long run. They posit that this may be because cost efficiency for established banks 
in their sample may have declined over time due to reduced operational scale caused by the arrival 
of more competitors as a result of financial reforms. However, Andries and Capraru (2012) 
University of Ghana http://ugspace.ug.edu.gh
53 
 
investigating the relationship between competition and efficiency in EU 27 banking systems as a 
whole using Granger-type causality tests rejected the efficient structure hypothesis, but their 
findings provide little or no evidence to support or reject the “quite life” hypothesis. 
 
However, as noted by Williams (2012), the contrast in the empirical evidence may be attributable 
to the simultaneous relationship between competition (or market power) and bank efficiency. In 
econometric models, simultaneity creates a concurrent correlation between the regressors and the 
error term, which if not controlled for leads to inconsistent estimates. The interdependence 
between competition and bank efficiency implies the possibility of reverse causality. The efficient 
structure hypothesis, for instance, suggests that efficiency may be driving market power (Turk- 
Ariss, 2010). The problem is demonstrated further by the empirical evidence that shows that the 
quiet life hypothesis is usually accepted in studies that do not control for simultaneity (Berger & 
Hannan, 1998), whilst it is rejected in studies that account for the problem (Maudos & de Guevara, 
2007; Koetter et al., 2012).  
 
Williams (2012) also examines the relationship between bank efficiency and market power to test 
the quiet life hypothesis for a sample of 419 Latin American commercial banks between 1985 and 
2010. He draws the efficiencies from the stochastic frontier function model that treats unobserved 
heterogeneity, whilst conventional Lerner indices and the efficiency-adjusted Lerner index are 
used as the proxy for market power. From the results, the quiet life hypothesis is firmly rejected 
after various robustness checks. The evidence suggests that bank restructuring has promoted 
competition at the expense of market power and yielded efficiency gains at banks under conditions 
of monopolistic competition. 
University of Ghana http://ugspace.ug.edu.gh
54 
 
2.6 Bank Profit Persistence 
 
This section provides a brief review of the theoretical background of profit persistence, and the 
empirical evidence on the determinants of bank profit persistence. 
 
2.6.1 Theoretical Background on Profit Persistence 
 
Contestable markets theory and the new industrial organization literature highlight the influence 
of potential and actual competition on profitability. The Persistence of Profit (POP) theory 
proposed by Mueller (1977), asserts that entry into and exit from an industry are sufficiently free 
to abolish any abnormal profit quickly, and that the profit rates of all the firms in an industry tend 
to converge towards the same long-run average value. However, some incumbent firms may have 
the ability to prevent imitation, or hinder entry. In that case, surplus profits would tend to persist 
from year to year, creating differences in firm-level long-run average profit rates. The degree of 
first-order autocorrelation in firm- or industry-level time-series profit data provides an indication 
of the speed at which competition eliminates any above- or below-average profits that are earned 
in the short run (Goddard et al., 2013). 
 
Although the competitive process should eliminate differences in profits between different 
firms/industries in the long run, this does not seem to be what is observed in the real world. The 
theory states that if a firm has excess profits, competitors will enter the market to offer similar 
products at lower prices until the profitability in the market equals the competitive rate. On the 
other hand, if firms have profits below average, some investors will move to other markets with 
higher profits until at least normal profits are obtained. However, this theory does not seem to find 
enough empirical support, because profits seem to persist from year to year (Gschwandtner, 2005). 
University of Ghana http://ugspace.ug.edu.gh
55 
 
According to Berger, Bonime, Covitz, and Hancock (2000), upward shifts in industry profit 
persistence imply one or more of the following possibilities: (1) product markets have become less 
competitive; (2) the banking industry has become more opaque, and/or (3) the banking industry 
has become more sensitive to regional/macroeconomic shocks. 
 
Also, from an antitrust, regulatory and supervisory point of view, a dynamic model of bank 
profitability gives an indication of the effectiveness of competition in forcing the adjustment or 
convergence of profits (above or below the norm) towards their long-run equilibrium levels. This 
may be helpful to regulators, in particular, in distinguishing between situations in which a 
competitive equilibrium is likely to be achieved rapidly without regulatory intervention, and 
situations in which regulatory intervention may be required in order to achieve a competitive ideal 
(Goddard et al., 2011). 
 
2.6.2 Empirical Evidence on Determinants of Bank Profit Persistence 
 
There is an extensive empirical Persistence of Profit (POP) literature based on manufacturing data, 
but only a handful of studies investigate POP in banking (Goddard et al., 2013). Indeed, even fewer 
still consider the determinants of profit persistence where they exist. As explained by Jiang and 
Kattuman (2010), the fundamental notion is that intense competition will quickly evaporate any 
short run excess profit enjoyed by any company, and force each to revert to its own ‘normal’ level 
of profitability, as determined by its command over various strategic resources.  
 
Berger et al. (2000) investigate how banking market competition, informational opacity, and 
sensitivity to shocks have changed over the last three decades by examining the persistence of 
University of Ghana http://ugspace.ug.edu.gh
56 
 
firm-level rents. The analysis suggests that different processes underlie persistence at the high and 
low ends of the performance distribution. They find that impediments to competition and 
informational opacity continue to be strong determinants of persistence; that the reduction in 
geographic regulatory restrictions had little effect on competitiveness; and that persistence remains 
sensitive to regional/macroeconomic shocks. The findings also suggest reasons for the recent 
record profitability of the industry. 
 
Using dynamic panel regressions based on vector autoregressive (VAR) model, Goddard et al. 
(2004) estimate growth and profit equations for a sample of commercial, savings, and co-operative 
banks from five major European Union countries during the mid-1990s. The persistence of profit 
appears higher for savings and co-operative banks than for commercial banks. Stephan and Tsapin 
(2008) also studied the persistence of profit and its determinants in emerging markets. They 
applied Markov chain analysis, dynamic panel GMM estimation, and quantile regression 
techniques to a panel of approximately 3,000 Ukrainian companies. The empirical results show a 
moderate level of profit persistence, as well as a relatively low speed of adjustment to the steady-
state profit level, thus providing no support for the hypothesis that there is a lower persistence of 
profits in emerging markets due to more intense competition. 
 
Further, to account for profit persistence, Athanasoglou et al. (2008) apply a GMM technique to a 
panel of Greek banks over the period 1985–2001. The estimation results show that profitability 
persists to a moderate extent, indicating that departures from perfectly competitive market 
structures may not be that large. Flamini et al. (2009) also find moderate persistence in profitability 
in Sub-Saharan Africa (SSA) banks, but they do not test for determinants of profit persistence.  
University of Ghana http://ugspace.ug.edu.gh
57 
 
Goddard et al. (2011) examine the intensity of competition in 65 national banking industries. The 
persistence of bank profit appears to be weaker for banks in developing countries than for those in 
developed countries. The empirical evidence suggests that the persistence of bank profit is 
negatively related to the rate of growth in GDP per capita, suggesting that the business 
opportunities afforded by higher economic growth tend to enhance competition and weaken the 
ability of incumbent banks to sustain abnormal profits. On the other hand, the persistence of bank 
profit is positively related to the size of legal entry barriers, in accordance with the view that actual 
or potential entry is a key determinant of the intensity of competition. There is an association 
between several institutional and external governance covariates and the persistence of bank profit: 
persistence tends to be weaker, and competition stronger, in countries where businesses and 
individuals are afforded more freedom from government interference, where the level of 
institutional development is advanced, and where the protection of property rights is relatively 
strong. Their study, however, included only a few African countries. 
  
Gschwandtner (2012) analyzed and compared profit persistence during the periods 1950–66, 
1967–83 and 1984–99 in the U.S. The results point towards a constant increase of competition 
after the opening of the U.S economy to international competition in the 1960–80s. Key 
determinants of profit persistence seem to be the firm’s and industry size, industry growth, and 
more recently risk, advertising and exports. Perera et al. (2013) also found profit persistence in 
selected South Asian banks over 1992-2007, but they did not ascertain the factors driving profit 
persistence of these banks. 
 
University of Ghana http://ugspace.ug.edu.gh
58 
 
Similarly, Chronopoulos, Liu, McMillan, and Wilson (2013) examine the determinants of 
profitability for a large sample of U.S banks over the period 1984-2010. They specifically assess 
the extent to which short-run profits persist, and whether such persistence is affected by changes 
in regulation and the recent financial crisis. Using system GMM estimation, the findings suggest 
that the competitive process reduces positions of abnormal profitability, albeit this is not 
immediate. There is also evidence that changes in regulation enacted during the 1990s affected 
both the level and persistence of bank profitability. The financial crisis of 2007-2010 appears to 
have resulted in an increase in the persistence of bank profitability. 
 
Goddard et al. (2013) examine the determinants and convergence of bank profitability in eight 
European Union member countries, between 1992 and 2007, using a dynamic panel model. They 
also found evidence of persistence of excess profit from one year to the next. The persistence of 
profit was lower in 1999–2007 than it was in 1992–98 in all eight countries. This finding is 
consistent with the hypothesis of an increase in the intensity of bank competition as a result of an 
increase in the integration of EU financial markets following the introduction of the euro in 1999, 
and the implementation of the Financial Services Action Plan.  
 
2.7 Economic Freedom and Bank Performance 
There are two major measurements of economic freedom, the Economic Freedom of the World 
Index produced by the Fraser Institute (Gwartney, Lawson & Hall, 2014) and the Economic 
Freedom Index constructed by the Heritage Foundation (Heritage Foundation, 2015) in 
collaboration with the Wall Street Journal. The Fraser Institute’s annual report on Economic 
Freedom of the World (EFW) provides information about the degree of economic freedom in up 
University of Ghana http://ugspace.ug.edu.gh
59 
 
to 152 countries of the world. The EFW index is an annual composite measure at country level, 
encompassing the size of government, the legal structure and security of property rights, access to 
sound money, freedom to engage in international trade, and the regulation of credit, labor and 
business (Gwartney et al., 2014). The composite index falls within a range of 1 to 10, with greater 
numbers indicating greater level of economic freedom in a country.  
 
The Heritage Foundation’s Index of Economic Freedom (EF) covers over 180 countries around 
the world, ranking them with an economic freedom score based on 10 measures of economic 
openness, regulatory efficiency, the rule of law, and competitiveness. The basic principles of 
economic freedom emphasized in the Index are individual empowerment, equitable treatment, and 
the promotion of competition (Heritage Foundation, 2015). The index measures the ability of 
individuals to exercise their fundamental right to control their own labor and property. In an 
economically free society, individuals have the freedom to work, produce, consume, and invest in 
any way they please. Also, in such societies, governments allow labor, capital, and goods to move 
freely, and desist from coercion or restriction of liberty beyond the extent needed to protect and 
maintain liberty itself (Heritage Foundation, 2015). 
 
This Heritage Foundation’s Index of Economic Freedom is a composite measure of a country’s 
overall economic freedom based on 10 different factors. It comprises of Financial Freedom, 
Business Freedom, Property Rights, Monetary Freedom, Investment Freedom, Freedom from 
Corruption, Labor Freedom, Trade Freedom, Fiscal Freedom and Government Spending (Size). 
The index takes values in a scale from 0 to 100, with scores approaching 100 representing higher 
levels of freedom. Similarly, the higher the score on a factor, the lower the level of government 
University of Ghana http://ugspace.ug.edu.gh
60 
 
interference in the economy. The overall score for each country is derived as the average of these 
ten economic freedoms, with equal weighting to each category. 
 
Financial freedom, one of Heritage Foundation’s ten measures of economic freedom, is a measure 
of banking efficiency as well as a measure of independence from government control and 
interference in the financial sector. State ownership of banks and other financial institutions such 
as insurers and capital markets reduces competition and generally lowers the level of available 
services (Heritage Foundation, 2015). 
 
In an ideal banking and financing environment where a minimum level of government interference 
exists, independent central bank supervision and regulation of financial institutions are limited to 
enforcing contractual obligations and preventing fraud. Credit is allocated on market terms, and 
the government does not own financial institutions. Financial institutions provide various types of 
financial services to individuals and companies, and banks are free to extend credit, accept 
deposits, and conduct operations in foreign currencies. Foreign financial institutions operate freely 
and are treated the same as domestic institutions (Heritage Foundation, 2015). 
 
Carow and Kane (2002) review and extend event-study evidence about the distribution of the 
benefits and costs of relaxing long-standing geographic and product-line restrictions on U.S. 
financial institutions. The evidence indicates that the new financial freedoms may have 
redistributed rather than created value. And using data on over 1,400 banks across 72 countries 
while controlling for bank-specific characteristics, Demirgüç-Kunt, Laeven, and Levine (2004) 
University of Ghana http://ugspace.ug.edu.gh
61 
 
indicates that tighter regulations on bank entry, restrictions on bank activities, and regulations that 
inhibit the freedom of bankers to conduct their business boost bank net interest margins. 
 
Boubakri, Cosset, Fischer, and Guedhami (2005) examine the post-privatization performance of 
81 banks from 22 developing countries. The results suggest that on average, banks chosen for 
privatization have a lower economic efficiency, and a lower solvency than banks kept under 
government ownership. They observe that in the post-privatization period, profitability increases 
but, depending on the type of owner, efficiency, risk exposure and capitalization may worsen or 
improve. However, over time, privatization yields significant improvements in economic 
efficiency and credit risk exposure.  
 
Brissimis, Delis, and Papanikolaou (2008) examine the relationship between banking sector reform 
and bank performance – measured in terms of efficiency, total factor productivity growth and net 
interest margin, with bank panel data from ten newly acceded EU countries. The results indicate 
that both banking sector reform and competition exert a positive impact on bank efficiency, while 
the effect of reform on total factor productivity growth is significant only toward the end of the 
reform process. They suggest that a possible area for future research could be to provide a more 
detailed analysis of the different country-specific institutional characteristics that may affect bank 
performance and, more broadly, financial stability in emerging markets. 
 
Pasiouras, Tanna, and Zopounidis (2009) use stochastic frontier analysis to provide international 
evidence on the impact of the regulatory and supervision framework on bank efficiency. The 
dataset consists of 2,853 observations from 615 publicly quoted commercial banks operating in 74 
University of Ghana http://ugspace.ug.edu.gh
62 
 
countries during the period 2000–2004. They investigate the impact of regulations related to the 
three pillars of Basel II (i.e. capital adequacy requirements, official supervisory power, and market 
discipline mechanisms), as well as restrictions on bank activities, on cost and profit efficiency of 
banks, while controlling for other country-specific characteristics. The results suggest that banking 
regulations that enhance market discipline and empower the supervisory power of the authorities 
increase both cost and profit efficiency of banks.  
 
Hakenes and Schnabel (2011) show that capital regulation may destabilize the banking sector 
through its effect on banking competition. The ambiguous effect of competition on banks’ risk-
taking translates into an ambiguous effect of capital regulation. A stabilizing effect of capital 
regulation tends to obtain in those situations where competition has a destabilizing effect (i.e., the 
‘‘charter value effect’’ dominates), and vice versa. Thus, they suggest that capital regulation may 
not be suited in all circumstances to prevent excessive risk-taking in banking. Stricter capital 
requirements attenuate competition for loans, implying higher loan rates, and hence higher risk-
taking by firms. 
 
Delis, Molyneux, and Pasiouras (2011) examine the relationship between the regulatory and 
supervision framework, and the productivity of banks in 22 European countries over the period 
1999–2009. The results indicate that regulations and incentives that promote private monitoring 
(PMON) have a positive impact on productivity. Restrictions on banks’ activities relating to their 
involvement in securities, insurance, real estate, and ownership of nonfinancial firms also have a 
positive impact. They contend that future research regarding the impact of regulations on bank 
performance and productivity growth may benefit from focusing on the structure and risks facing 
University of Ghana http://ugspace.ug.edu.gh
63 
 
the banking industry or individual banks. In addition, an interesting question concerns whether 
government policies that restrict diversification substitute or complement the effects on PMON or 
other elements of bank regulation. They suggest that this can be achieved by examining the 
interplay between the forms of regulation examined separately in their study.  
 
Hassan, Sanchez, Ngene, and Ashraf (2012) examine the impact of foreign bank entry, among 
others, on the domestic banking sector. They find that the degree of openness of the market to 
foreign bank entry positively impact domestic banking system operating efficiency, capitalization, 
risk management, long-term soundness, financial performance as well as economic and financial 
development. However, foreign bank entry is associated with reduced profit margins and increased 
operating costs of domestic banks in countries with less developed capital markets.  
 
More recently, Chortareas, Girardone and Ventouri (2012) investigated the dynamics between key 
regulatory and supervisory policies and various aspects of commercial bank efficiency and 
performance for a sample of 22 EU countries over 2000–2008. The results show that strengthening 
capital restrictions and official supervisory powers can improve the efficient operations of banks. 
Evidence also indicates that interventionist supervisory and regulatory policies such as private 
sector monitoring and restricting bank activities can result in higher bank inefficiency levels. The 
evidence produced suggests that the beneficial effects of capital restrictions and official 
supervisory powers on bank efficiency are more pronounced in countries with higher quality 
institutions. The emerging challenge, though, is to consider which specific aspects of regulatory 
and supervisory policies affect bank performance and how their implementation and effectiveness 
is related to the broader institutional framework. 
University of Ghana http://ugspace.ug.edu.gh
64 
 
Ding, Wu, and Chang (2013) examine the dynamic changes in bank performance before and after 
government intervention during the global financial crisis that started from 2007 onwards. Using 
data collected from Bloomberg for banks of five major Asian economies over the eleven-quarter 
period from 4th quarter of 2007 to 2nd quarter of 2010, they find that, on average, the bank 
performance in terms of solvency, credit risk, and profitability improves after government 
intervention. Moreover, the influence of government intervention on bank performance depends 
on the evaluative financial indicator, the economy, and whether banks are internationalized.  
 
Barth, Caprio, and Levine (2013) examine whether bank regulation, supervision and monitoring 
enhance or impede bank operating efficiency. Based on an un-balanced panel analysis of 4,050 
banks observations in 72 countries over the period 1999–2007, they find that tighter restrictions 
on bank activities are negatively associated with bank efficiency, while greater capital regulation 
stringency is marginally and positively associated with bank efficiency. They also find that a 
strengthening of official supervisory power is positively associated with bank efficiency only in 
countries with independent supervisory authorities. Moreover, independence coupled with a more 
experienced supervisory authority tends to enhance bank efficiency. Finally, market-based 
monitoring of banks in terms of more financial transparency is positively associated with bank 
efficiency. 
 
In contrast, Gaganis and Pasiouras (2013) find that central bank independence has a negative 
impact on bank profit efficiency. Using a large sample of nearly 4,000 commercial banks operating 
in almost 80 countries over the period 2000–2006, and relying on stochastic frontier techniques, 
the results also show that efficiency decreases as the number of the financial sectors that are 
University of Ghana http://ugspace.ug.edu.gh
65 
 
supervised by the central bank increases. Additionally, banks operating in countries with greater 
unification of supervisory authorities are less profit efficient. These results could have important 
policy implications, as there is an on-going debate covering all aspects of the institutional design 
of supervision. Nevertheless, in all the cases there is a long list of pros and cons, and as a result 
the theoretical analysis of the potential effects of alternative supervisory regimes remains 
inconclusive.  
 
Noticeably absent in the banking literature is an examination of the links between economic 
freedom and bank performance. The limited research in this area is somewhat surprising given the 
importance of bank lending in promoting economic development and the impact that economic 
freedom is likely to have on the banking sector (Sufian & Habibullah, 2010). Indeed, as noted by 
Hafer (2013), a number of studies have found that financial development and higher levels of 
economic freedom are associated with (or cause) economic growth. The unanswered question, 
however, is whether the financial development-economic growth nexus reflects influences of 
economic freedom operating through the financial system (Hafer, 2013).  
 
Sufian and Habibullah (2010) provide new empirical evidence on the impact of economic freedom 
on banks’ performance. The empirical analysis is confined to the Malaysian banking sector during 
the period of 1999–2007. They find that overall economic freedom and business freedom exerts 
positive impacts, implying that higher (lower) freedom on the activities that banks can undertake 
and entrepreneurs to start businesses increases (reduces) banks’ profitability. The empirical 
findings seem to suggest that corruption has a corrosive impact on Malaysian banks’ profitability. 
Interestingly, though, the impact of monetary freedom is negative implying that higher (lower) 
University of Ghana http://ugspace.ug.edu.gh
66 
 
monetary policy independence reduces (increases) banks profitability, providing support for the 
importance of government intervention in determining the profitability of banks operating in the 
Malaysian banking sector. They suggest that future research could include more variables such as 
taxation and regulation indicators, exchange rates as well as indicators of the quality of the offered 
services. 
 
Smimou and Karabegovic (2010) examine the relationship between economic freedom index and 
equity market returns for the Middle East and North Africa (MENA) markets. The evidence shows 
that changes in economic freedom have a positive impact on equity market returns, which are not 
explained by business-cycle control variables related to expected returns, and that legal structure 
and security of property rights have the most significant impact. Hafer (2013) also finds that 
countries with higher levels of initial economic freedom, on average, exhibit greater levels of 
financial intermediary development in subsequent years. And if greater financial intermediary 
development engenders faster economic growth, the results of this study explain, at least partially, 
the observed link between economic freedom and economic growth. 
 
Further, cost inefficiency scores for banks in ten new EU member countries of Central and Eastern 
Europe are estimated by Mamatzakis, Kalyvas, and Piesse (2013), using stochastic frontier 
analysis for the period 2000-2010. These are then employed in both static and dynamic panels to 
estimate the impact of regulation on bank specific inefficiency in the transition economies. Using 
the Fraser Index of Economic Freedom, they find that, among all the indices of economic freedom, 
the composite regulation index that includes regulation in credit, labour and business is the one 
that has more importance for the banking sector as it exerts a negative and statistically significant 
University of Ghana http://ugspace.ug.edu.gh
67 
 
impact on bank inefficiency. By decomposing the regulation index, into its three components 
(credit, business and labour regulation) they find that strict labour regulation is associated with 
higher bank inefficiency while certain aspects of credit regulation such as foreign ownership and 
competition as well as private ownership are significantly associated with decreased bank 
inefficiency. The dynamic panel-VAR results using impulse response functions and variance 
decomposition support the validity of these results further.  
 
Interestingly, Chortareas et al. (2013), possibly the first to directly investigate the dynamics 
between the financial freedom counterparts of the economic freedom index drawn from the 
Heritage Foundation database and bank efficiency levels, suggest that the higher the degree of an 
economy’s financial freedom, the higher the benefits for banks in terms of cost advantages and 
overall efficiency. Relying on a large sample of commercial banks operating in the 27 European 
Union member states over 2001-2009, and using Data Envelopment Analysis (DEA) and a 
truncated regression model combined with bootstrapped confidence intervals, they also show that 
the effects of financial freedom on bank efficiency tend to be more pronounced in countries with 
freer political systems in which governments formulate and implement sound policies and higher 
quality governance. A further challenge that emerges is to consider whether ‘‘excessive’’ financial 
freedom may contribute to financial institutions’ propensity to take on greater risks, which was 
beyond the scope of their work. 
 
Lin et al. (2016) also examine how financial freedom moderates the effect of changes in bank 
ownership on cost efficiency in twelve Asian developing countries during the period 2003–2012. 
Using the stochastic frontier method for estimating bank efficiency scores, they report that foreign 
University of Ghana http://ugspace.ug.edu.gh
68 
 
presence improves bank efficiency, mainly in countries with high financial freedom. Additionally, 
they find evidence that increased government (domestic) ownership of banks seems to enhance 
(hamper) bank efficiency in economies with more financial freedom following the financial crisis. 
 
2.8 Overview of the Banking Sector in Africa 
 
This section provides a brief overview of the banking sector in Africa. We consider developments 
in the sector over the past 30 years or so, focusing on banking sector reforms, market structure and 
performance. 
 
2.8.1 Banking Systems in Africa 
 
Banks are the most important institutions of the financial system in sub-Saharan Africa (SSA). 
However, in many countries, other financial structures (such as bond and stock markets) are either 
underdeveloped or almost nonexistent (Kablan, 2010). Credit to the private sector as a percentage 
of gross domestic product (GDP) stands at 16% and 19% respectively in the West and East African 
sub-regions, whilst the Southern and North African sub-regions record approximately 55% and 
45% respectively (Fosu, 2013). Only one-third of countries in the region have stock markets, which 
are mostly small and illiquid. Corporate bond markets are almost non-existing in most countries, 
with many countries even having very shallow government bond markets if at all (Mu, Phelps, & 
Stotsky, 2013). 
 
Many African countries suffered banking crises in the 1980s and 1990s. Indeed, at the peak of 
regional financial distress (about 1995), as many as 27 African countries suffered from banking 
crises, with 20 of them being systemic (Laeven & Valencia, 2008). While the usual ingredients of 
University of Ghana http://ugspace.ug.edu.gh
69 
 
banking crises –macroeconomic boom and bust cycles and bad private banking – were also present 
in banking crises in Sub-Saharan Africa, it appears that government failures were the leading 
cause. For example, in many countries, such as Mozambique, Tanzania, Uganda, and Zambia, 
large government-owned banks had to be rescued and sold (Beck, Fuchs & Uy, 2009). 
 
Banking in Africa has undergone dramatic changes over the past 20 years. While dominated by 
government-owned banks in the 1980s and subject to restrictive regulation – including interest rate 
ceilings and credit quotas – financial liberalization, institutional and regulatory upgrades and 
globalization have changed the face of financial systems across the region. Today, most countries 
have deeper and more stable financial systems, though challenges of concentration and limited 
competition, high costs, short maturities, and limited inclusion persist (Beck & Cull, 2013). Recent 
reforms in the banking sector have led to the liberalization of interest rates and credit markets. For 
instance, interest rate controls, particularly in Kenya, Ghana and Tanzania, and directed credit in 
Uganda, have been replaced with open market operations.  
 
Although there is still strong government presence in African banking sectors (e.g. Algeria and 
Tunisia), a significant degree of success has been achieved in privatizing banks in a number of 
countries including Morocco, Kenya, Tanzania, Uganda, Rwanda and Zambia (Allen, Otchere & 
Senbet, 2011). These reforms have led to significant growth in the number of banks in many 
African countries, with a significant increase in the degree of cross-border banking as well. But 
with domestic credit to the private sector averaging about 32% of GDP, financial intermediation 
remains relatively low in a number of African countries. This feature of the banking sectors is 
coupled with strong government ownership and traditional banking activities. The poor 
University of Ghana http://ugspace.ug.edu.gh
70 
 
performance of the banking sector, especially record high levels of problem loans in the 1980s, 
led to significant financial sector reforms in many countries (Fosu, 2013). 
 
There is an increasing realization among policy makers and other stakeholders among African 
countries that deepening and broadening the financial sector is critical for increasing growth rates 
and reducing poverty and thus making progress towards the Millennium Development Goals, yet 
there is still significant debate about how to proceed. The controversies on financial sector reform 
focuses on two issues: the role of government in the financial sector and how African financial 
systems (and the economies in general) should adapt to increasing globalization. Debates on both 
issues have changed dramatically over the past decades. The approach of governments replacing 
markets was seen as necessary in the 1960s and 70s, changed into an almost laissez-faire approach 
focusing on liberalization and privatization in the 1980s and 90s, before the pendulum went back 
towards a more active, albeit different, role for government during the last decade (Beck et al., 
2009). 
 
Even so, many of the non-orthodox policies successfully implemented in East Asia have not 
worked in Sub-Saharan Africa. Financial repression and industrial policies have had little or no 
positive impact on financial and economic development in Sub-Saharan Africa. Instead, financial 
repression has inhibited savings and resulted in mis-allocation of credit. The interference in credit 
allocation and government ownership of banks have led to large losses and distorted credit markets 
and resource allocation, with harmful consequences for economic growth and equity (Beck et al., 
2009). 
 
University of Ghana http://ugspace.ug.edu.gh
71 
 
Of course, most African countries have opened their financial systems up to foreign bank entry, 
yet capital account restrictions are still in place in many countries, although often more de-jure 
than de-facto (Beck et al., 2009). Generally, though, the North African financial systems are 
dominated by government-owned financial institutions to a much larger extent than are systems in 
Sub-Saharan Africa, where foreign-owned banks now dominate the banking system, while 
governments dominate in other segments of the financial system, such as the pension sector and 
the bond market (Beck et al., 2011). 
 
2.8.2 African Banking Market Structure and Performance 
 
The ownership structure of Africa’s banking systems has undergone substantial changes over the 
past decades, with a larger number of countries now dominated by foreign banks and only few 
banking systems with mainly government-owned banks, a result of the privatization wave in Africa 
in the 1980s and 1990s. While foreign bank penetration has increased from already high levels 
over the past decade, the composition of the foreign bank population has changed substantially. 
Previously dominated by European banks, banks from emerging markets and – critically – from 
inside Africa have gained importance over the past years. For example, after the end of Apartheid, 
several South African banks, most notably Standard Bank and ABSA, started expanding through 
the continent. Also, two West African banks – Ecobank and Bank of Africa – have expanded 
throughout Sub-Saharan Africa. Similarly, Moroccan banks have started to expand to the South. 
Finally, and as a result of the recent consolidation wave in Nigeria, Nigerian banks started 
expanding throughout West Africa, but increasingly also throughout the rest of the continent. 
(Beck & Cull, 2013).  
 
University of Ghana http://ugspace.ug.edu.gh
72 
 
However, in spite of the record levels of new entry and foreign banks penetration, very high levels 
of concentration still characterize African banking sectors. For instance, Fosu (2013) found that 
during 2002-2009, the average Herfindahl–Hirschman Index (HHI) was as high as 2059, whilst 
the five-bank concentration ratio stood at 77.29% for the whole African region. Nonetheless, it 
was observed that banking market concentration assumed a downward trend across all the sub-
regions in Africa over the past few years. The HHI showed a dramatic and consistent downward 
trend in all sub-regional banking sectors except West Africa, where the trend was found to be 
moderate. The five-bank concentration ratios, also showed a similar trend. Similarly, Beck et al. 
(2011) reports that Africa’s banking systems are mostly concentrated, with a few banks sharing 
the small universe of clients. Besides, standard indicators of competition show significant market 
power among banks across the region.  
 
Of course, high interest rate spreads have accompanied greater concentration and lower 
competitiveness in African banking markets. While the share of the five largest banks was 81 
percent in the median African country in 2011, it was 64 percent outside Africa (Beck & Cull, 
2013). Greater concentration can also partly explain the lower degree of competition within 
African banking system. The median Lerner index, which is the mark-up between marginal 
revenue and costs, is 30 percent in the median African country, while it is 25 percent outside 
Africa. However, Beck and Cull (2013) notes that the correlation between concentration and the 
Lerner index of market power is relatively low within Africa, at only 11 percent, suggesting that 
market structure is only one, and maybe not the most important determinant of the lack of 
competition within African banking markets.  
 
University of Ghana http://ugspace.ug.edu.gh
73 
 
In a recent study, Fosu (2013) using both the static and dynamic versions of the Panzar–Rosse 
model, reports that banks in African sub-regional markets can be described as monopolistically 
competitive. In particular, the findings suggest that, with the exception of North Africa, African 
banks exhibit higher competition at interest-generating activities compared to total banking 
activities. Further, it was observed that the degree of competition in African banking markets is 
comparable to that existing in other emerging markets.  
 
Even so, Beck et al. (2011) notes that competition remains an important area for government 
action. Allowing competition within the banking system and from outside the banking system will 
foster the necessary financial innovation to push the financial system toward the frontier and 
exploit the possibilities that new methodologies, products, and technologies can offer. An 
increased focus on competition, however, has critical repercussions on regulation and government 
policy in general. It entails a sophisticated approach that has to balance (1) the need for innovation, 
(2) the need to avoid market dominance by new players that rapidly gain market shares in new 
products, and (3) the need to reduce the risks of fragility.  
 
Beck et al. (2009) suggests that the write-down of bad loans has led to a shrinking of the financial 
systems in some countries, but overall financial intermediation has improved in many countries. 
But in spite of these progresses, Africa’s financial systems are still characterized by their 
shallowness, by their high costs, exemplified in high interest rate spreads, and by limited access to 
finance. While an increase in interest spreads was to be expected after liberalization, their 
continued high level in most countries of the region has turned into a serious concern for policy 
makers.  
University of Ghana http://ugspace.ug.edu.gh
74 
 
Beck et al. (2011) report that African banking is very expensive as reflected by high interest 
spreads and margins. The high interest rate spreads and margins are driven by diseconomies of 
scale, risk, and lack of competition, as documented by cross-country and individual country 
studies. Decompositions of interest rate spreads and margins point to high overhead costs as the 
main driver. These high overhead costs can be explained to a large extent by the scale 
diseconomies suffered by African banks. However, they also reflect the generally high costs of 
doing business that affect banks and that include deficient energy and road infrastructure, as well 
as a lack of reliable credit information and infrastructure. 
 
Beck and Cull (2013) argue that, while there are many reasons why spreads and margins are higher 
in Africa, one important reason is higher operating costs. For instance, in a recent decade, Beck 
and Cull (2013) report that overhead costs in the median African financial system stood at 5.5 
percent of total assets, while they were 3.4 percent outside Africa. On the other hand, African 
banks are also more profitable than banks outside Africa. The return on assets (ROA) stood at 2.1 
percent in the median African country, while it was 1.5 percent outside Africa. Net interest margins 
in the median African country stood at 5.9 percent in 2011, while they stood at 4.7 percent outside 
Africa. Similarly, the interest rate spread between lending and deposit rate was 10.3 percent in 
Africa and 8.2 percent outside. Thus, generally, while being more profitable, African banks are 
less efficient and operate in less competitive environments (Beck & Cull, 2013).  
  
According to Honohan and Beck (2007), the high spread between deposit and lending interest rates 
provides dis-incentives for both savings and lending and is driven mainly by the absence of scale 
economies and very high risks due to weak and underdeveloped contractual frameworks and 
University of Ghana http://ugspace.ug.edu.gh
75 
 
economic and political volatility. Indeed, as noted by Senbet and Otchere (2006), savings 
mobilization and credit allocation have generally not improved by as much as expected, after years 
of reforms. Again, Honohan and Beck (2007) observe that in spite of high costs and high risks, 
banks in Africa are very profitable. In fact, subsidiaries of foreign banks in Sub-Saharan Africa 
have higher returns on assets and equity than subsidiaries of the same banks in other regions of the 
world, perhaps due to lack of competition. 
 
In terms of stability, African financial systems have made some progress. Beck et al. (2011) 
indicates, that while in 1995, a third of all countries on the continent were suffering from a systemic 
banking crisis, fragility has subsided across the continent. Today, most African banking systems 
are stable and well capitalized and have a good level of liquidity. Nevertheless, there is still hidden 
or silent fragility in several Central and West African countries. Systemic distress is concentrated 
in state-owned banks, and a number of small locally owned banks face liquidity problems because 
of their dependence on the public sector and wholesale funding (Beck et al., 2011). 
 
Further, Beck and Cull (2013) opine that while shallow, Africa’s banking systems have also proven 
stable and resilient over the past years. The shallowness of Africa’s banking systems appears to 
have helped them weather the Global Financial Crisis of 2008 better than some other regions of 
the world, with the impact of the crisis on Africa mostly working through real sector channels, 
such as lower demand for export goods, or through lower foreign direct investment. According to 
Beck and Cull (2013), the limited integration with global financial markets and exposure to “toxic” 
assets explains why financial institutions across the Africa region largely evaded the direct impact 
of the global financial crisis.  
University of Ghana http://ugspace.ug.edu.gh
76 
 
Greater stability is also illustrated in the aggregate balance sheet indicators of African banks. In 
2011, the capital to risk-weighted asset ratio was 19 percent in the median African country, 
compared to 17 percent outside Africa. On the systemic level, Africa has suffered few banking 
crises since the bout of systemic fragility in the 1980s and 90s (Laeven & Valencia, 2012). 
However, as highlighted by Moyo et al. (2014), recent global evidence has shown that deregulation 
of the banking industry can sometimes have the unintended effect of destabilizing the financial 
system, contributing to macroeconomic instability and, in some cases leading to a reversal of 
economic growth. A conspicuous feature of structural financial sector reforms is enhanced 
competition in the banking industry, with the attendant stability-fragility trade-off. According to 
Moyo et al. (2014), compared to other economies, Sub-Saharan Africa (SSA) financial system is 
broadly bank-based and weakly contestable, therefore, any systemic bank failures would have 
serious contagious ramifications in these economies. They argue that bank-specific, 
macroeconomic and institutional factors are important in predicting episodes of bank distress in 
SSA.  
 
2.9 Chapter Summary and Gaps in the Literature 
The theoretical and empirical literature present ambiguous positions on the competition-stability, 
competition-profitability and competition-efficiency relationships. For instance, the theoretical 
literature on the relationship between competition and bank stability presents no consensus.  Two 
opposing views are theorized in the literature. One view, called the competition–fragility or 
competition-instability view, and pioneered by the influential work of Keeley (1990), asserts that 
competition in banking results in greater risks. Recent studies that provide evidence for this include 
Beck et al. (2006), Amidu (2013), Turk-Ariss (2010) and Diallo (2015). In contrast, the 
University of Ghana http://ugspace.ug.edu.gh
77 
 
competition–stability view came out strongly from the work of Boyd and de Nicoló (2005), who 
argued that competition leads to greater bank stability. Among others, Uhde and Heimeshoff 
(2009), and Schaeck and Cihak (2014) provide support for this hypothesis. The results of some 
other recent studies suggest that the relationship between competition and bank stability is not 
straightforward and may actually depend on other factors previously ignored in the literature.  
 
Also, two competing theories explaining the relationship between market structure and bank 
profitability are the Market Power (MP) and Efficiency Structure (ES) paradigms. While some 
have found support for the MP hypothesis (Fu & Heffernan, 2009; Tregenna, 2009; Perera et al., 
2013; Mirzaei et al., 2013), other studies find that the level of bank profitability is explained by 
efficiency, and not by market power (Seelanatha, 2010; Chortareas et al, 2011).  
 
A special case of the MP hypothesis, is the ‘Quiet Life’ hypothesis (QLH), which suggests that the 
higher the market power enjoyed by a firm, the lower the effort put forth by managers to improve 
efficiency (Hicks, 1935). Again, the empirical evidence on the competition-efficiency relationship 
has been inconsistent. Those reporting evidence of the QLH include Delis and Tsionas (2009), 
Coccorese and Pellecchia (2010), and Turk-Ariss (2010). In contrast, Weill (2004), Casu and 
Girardone (2006), Maudos and de Guavara (2007), and Williams (2012) provide evidence 
rejecting the QLH.  
 
Contestable markets theory and the new industrial organization literature also highlight the 
influence of potential and actual competition on profitability. The Persistence of Profit (POP) 
theory (Mueller, 1977), asserts that entry into and exit from an industry are sufficiently free to 
University of Ghana http://ugspace.ug.edu.gh
78 
 
abolish any abnormal profit quickly, and that the profit rates of all the firms in an industry tend to 
converge towards the same long-run average value. However, this theory does not seem to find 
enough empirical support (Gschwandtner, 2005). Besides, while there is an extensive empirical 
POP literature based on manufacturing data, only a handful of studies investigate POP in banking 
(Goddard et al., 2013). Indeed, even fewer still consider the determinants of profit persistence 
where they exist.  
 
Given the difficulty in understanding the channels between competition and bank stability, our 
study seeks to assess the effect of financial freedom on the competition-stability relationship. We 
also aim to provide additional insight into the ambiguous relationship between competition and 
bank profitability by considering the conditioning effect of financial freedom on this relationship.  
Further, this study seeks to test the QLH in the developing country context, which is largely 
ignored in the literature, and also provide new empirical evidence on the impact of financial 
freedom on the competition-efficiency relationship. And finally, we assess differences in the 
determinants of bank profit persistence among countries, a relatively unexplored area in banking. 
 
 
 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
79 
 
CHAPTER THREE 
METHODOLOGY 
 
3.1 Introduction  
In this chapter, we present the various methodologies employed in carrying out the analyses 
required to achieve our study objectives. We begin with the research design and sampling, as well 
as the models and proxies used to measure competition and bank performance (stability, 
profitability and efficiency), and also describe the financial freedom and economic freedom 
indices. Next, we outline the estimation methods for assessing the effects of financial/economic 
freedom on the competition-stability, competition-profitability and competition-efficiency 
relationships. We also consider the control variables which typically affect these relationships. 
And finally, we consider the model for bank profit persistence and its determinants. 
 
3.2 Research Design  
The key research questions for this study encompass determining whether or not financial freedom 
moderates the effects of competition on bank stability, profitability and efficiency; and what 
different factors account for bank profit persistence in sub-Saharan Africa (SSA). Generally, 
addressing these issues called for quantitative data and analytical techniques. The research strategy 
was to obtain secondary data from financial statements of banks in this region, and to carry out 
various analyses on the relationships of interest. The analytical techniques used were similar to 
those common in the literature and could be relied upon to obtain valid conclusions on the research 
problem and questions. For instance, we follow Koetter, Kolari, and Spierdijk (2008) in modelling 
the underlying cost structure of the banking sector. This model allows us to estimate competition 
University of Ghana http://ugspace.ug.edu.gh
80 
 
(bank market power) and cost efficiency using the same framework. Koetter et al. (2008) used a 
similar model in assessing U. S. bank cost efficiency. Our approach in estimating the competition-
stability, competition-profitability and competition-efficiency relationships are similar to Liu et al. 
(2013), Dietrich and Wanzenried (2014), and Coccorese and Pellecchia (2010), respectively. We 
also, adapted Goddard et al. (2013) in assessing the determinants of bank profit persistence. 
 
The source of most of the banking data is the Bankscope database, which reports published 
financial statements from banks across the globe. The data has been standardized into a common 
format to facilitate comparison across countries and therefore suitable for a cross-country study. 
Unconsolidated financial accounts (income statement and balance sheet) available for the 7-year 
period from 2006 to 2012 were used. The financial freedom and economic freedom variables were 
obtained from the Heritage Foundation’s indices produced in collaboration with the Wall Street 
Journal annually since 1995. This data is available on the Heritage Foundation Website 
(www.heritage.org). The macroeconomic data, domestic credit to the private sector as a percentage 
of GDP, and GDP per capita are from the World Development Indicators produced by the World 
Bank, and available on the Website of the World Bank.  
 
Data extracted from the income statements and balance sheets of the banks include interest 
expenses, deposits and money market funds, personnel costs, total assets, other operating expenses 
and total income. Table 3.1 below provides details of the number of banks from each country used 
for the study. 
 
 
University of Ghana http://ugspace.ug.edu.gh
81 
 
Table 3.1: Number of Banks for Each Country 
COUNTRY 2006 2007 2008 2009 2010 2011 2012 
ETHIOPIA 0 0 2 5 8 7 7 
GHANA 2 12 16 19 21 21 16 
KENYA 2 6 22 24 27 27 25 
MALAWI 0 0 4 4 5 5 4 
MAURITIUS 6 7 9 9 10 9 7 
MOZAMBIQUE 0 0 6 7 8 8 7 
NAMIBIA 3 4 4 4 4 3 3 
SOUTH AFRICA 11 14 15 15 16 12 12 
TANZANIA 10 11 13 12 13 13 11 
UGANDA 0 5 9 12 13 13 12 
ZAMBIA 0 2 7 7 11 11 11 
TOTAL 34 61 107 118 136 129 115 
Source: Compilation from Bankscope database 
 
3.3 Population of Study  
All banks operating in sub-Saharan African countries for the period of 2006-2012 constituted the 
population for the study.  Our goal was to use as much data as possible in order to allow for 
generalization of our results. We focused on the dominant financial institution, commercial banks, 
and specialized financial institutions whose nature and operations are akin to that of commercial 
banks. We excluded other financial institutions such as investment banks since their sphere of 
activities are dissimilar. 
 
3.4 Sampling Technique and Sample Size 
Our initial sample comprised of banks in all Sub-Saharan African countries, but due to data 
limitations, especially inadequate data points at the country level required for some of the 
regression estimates, we settled on data from 139 banks operating in 11 countries in Sub-Saharan 
Africa. Also, banks with less than three consecutive years of observations were excluded. Besides, 
University of Ghana http://ugspace.ug.edu.gh
82 
 
some banks did not have values needed for some of the key variables used in the study and had to 
be excluded.  
 
The final sample is an unbalanced panel with 700 bank-year observations. Countries included in 
the study are Ethiopia, Ghana, Kenya, Malawi, Mauritius, Mozambique, Namibia, South Africa, 
Tanzania, Uganda and Zambia. According to the Economic Intelligence Unit (www.eiu.com), our 
sample comprise most of the largest banking markets in Sub-Saharan Africa, with the exception 
of Nigeria and Angola which could not be included for the reasons stated above. 
 
3.5 Measuring Bank Competition  
As explained in Weill (2013), the number of different tools used to measure bank competition can 
be categorized into traditional Industrial Organization (IO) and new empirical IO approaches. The 
traditional approach uses tests of market structure to assess bank competition based on the 
Structure-Conduct-Performance (SCP) model that states that greater concentration leads to less 
competitive bank behavior, which in turn leads to greater profitability of the bank. With this model, 
concentration indices (market share of the largest banks or the Herfindahl-Hirschman index) are 
used to measure competition. 
 
On the other hand, the new empirical IO approach uses non-structural tests to overcome the 
problems with competition measures in the traditional IO approach, such as the use of indirect 
proxies (market structure or market shares) to conjecture the degree of competition. The non-
structural measures such as the Lerner index and the Panzar–Rosse (1987) model measure bank 
conduct directly, rather than inferring the competitive conduct of banks through the analysis of 
University of Ghana http://ugspace.ug.edu.gh
83 
 
market structure. The Lerner index provides an individual measure of market power at the bank 
level, while the Panzar–Rosse model produces an aggregate measure of competition. Other 
approaches used to assess the competitiveness of the banking sector in the literature are the 
Persistence of Profitability (POP) model, the Conjectural Variation (CV) approach, and more 
recently the Boone (2008) indicator. 
 
In this study, we use a direct measure of bank competition (market power), the Lerner index. This 
allows us to measure competition and bank efficiency using the same framework, and thus obtain 
better insight into the closely related issues of bank competition and efficiency. Most previous 
studies assessing these relationships use different models to measure competition and efficiency 
and thereby fail to take into account the simultaneous relationship between them (Koetter et al., 
2012). Also, given that other key variables used to test the relationship between competition and 
bank performance are obtained at the bank level, it is best to use Lerner index which is estimated 
at the bank level, unlike other competing proxies for bank competition. For instance, the Panzar–
Rosse (1987) model which has also been a popular measure of competition in banking markets 
produces an aggregate measure of competition at the industry level. The Lerner index is the mark-
up of price over marginal costs, with higher values denoting higher pricing power and less 
competitive market conditions. 
 
Similar to Berger et al. (2009) and Turk-Ariss (2010), we measure bank competition using the 
Lerner indicator of market power defined as: 
 
itTAitTAitTAit PMCPLerner ,,, /)( 
,       (3.1)  
University of Ghana http://ugspace.ug.edu.gh
84 
 
   
where 
itTAP ,
 refers to the price of total assets (calculated as the ratio of total revenues to total assets) 
and 
itTAMC ,
 is the marginal cost of producing an additional unit of output. As is common in the 
literature, and similar to Koetter et al. (2008), we use the following translogarithmic function to 
model the underlying cost structure of the banking sector: 
  
 
 

3
1
3
1
,,
22
10 lnlnln)(ln2lnln k k itkitkitkkititit WTAWTATATOC 
 
 
 
  

2
1
3
1
3
1
3
1
2
,,, )(ln)2/(lnln
k
k
k
k j j
itjjitjitkij trendWWW 
+ 
itititj
j
j trendTAtrendW  

lnln ,
3
1
 ,      (3.2) 
 
where lnTOC  represents the natural logarithm of bank’s total costs (financial and operating costs) 
and lnTA is a proxy for bank output measured as total assets. 1W , 2W  and 3W  are the prices of 
funds, labour and physical capital, respectively. They are respectively calculated as the ratio of 
interest expenses to total deposits and money market funds, personnel expenses to total assets, and 
other operating expenses (excluding personnel expenses) to total assets. It should be noted that, 
scaling over total employees, instead of total assets is a better proxy for the price of labour, but the 
latter was chosen because the number of employees is not available for many observations. Trend 
is a time trend which captures movements in the cost function over time or technical change. We 
also scale the input prices and TOC by 
3W  in order to ensure homogeneity of degree one in input 
prices of the cost function. As an alternative to the translogarithmic functional form used in this 
University of Ghana http://ugspace.ug.edu.gh
85 
 
study, the Fourier flexible form has been found to produce results similar to the translogarithmic 
form (Berger & Mester, 1997). 
 
The marginal cost, 
itTAMC ,
is then derived from the translogarithmic cost function by taking the 
first derivative with respect to the output (total assets) for each bank in the sample as follows: 
     
 



  

3
1
,21 lnln
k
ititkkit
it
it
it trendWTATA
TOCMC 
    (3.3) 
 
The Lerner index computed from equation (3.1) is basically the conventional Lerner index. 
However, as noted by Turk-Ariss (2010), this approach is likely to result in biased estimates 
because of a bank’s ability to exercise some form of monopoly power in the deposit market which 
enables it to obtain funds cheaply. He argues that usually when pricing their loans, managers of 
banks endeavor to cover their cost of funds, and charge a premium to reflect their exercise of 
market power, in addition to charging a risk premium for the uncertainty of repayment. Hence, to 
obtain a ‘‘raw” or ‘‘clean” proxy of pricing power that is not distorted by market power which had 
previously originated in the deposit market while raising funds, cost of funds should be excluded 
to produce a funding-adjusted Lerner index. To check the robustness of our results, we estimate 
another version of the Lerner index by including only two inputs, the price of labor and the price 
of physical capital to obtain the funding-adjusted Lerner index using the same model.  
 
 
 
University of Ghana http://ugspace.ug.edu.gh
86 
 
3.6 Measurement of Bank Performance  
European Central Bank (2010) has noted that there are a multitude of measures used to assess bank 
performance, with each group of stakeholders having its own focus of interest. They assert that, 
among the large set of performance measures for banks used by academics and practitioners alike, 
a distinction can be made between traditional, economic and market-based measures of 
performance. The traditional performance measures they identified are similar to those applied in 
other industries, with Return on Assets (ROA), Return on Equity (ROE) or Cost-to-Income (CTI) 
ratio as the most widely used. In addition, they observe that, given the importance of the 
intermediation function for banks, Net Interest Margin (NIM) is typically monitored. In this study, 
we focus on some of these traditional measures of bank performance – NIM and ROA.  
 
The economic measures of performance mainly focus on efficiency as a central element of 
performance, but generally have high levels of information requirements. The most popular one is 
the economic value added (EVA) developed by Stern, Stewart and Chew (1995). EVA measures 
whether a company generates an economic rate of return higher than the cost of invested capital 
in order to increase the market value of the company. Another economic measure is RAROC (risk-
adjusted return on capital), but it is difficult to calculate without having access to internal data 
(European Central Bank, 2010). 
 
The most commonly used market-based measures of performance include: the total share return-
the ratio of dividends and increase in the stock value over the market stock price; the price-earnings 
ratio-the ratio of the financial results of the company over its share price; and the price-to-book 
value-which relates the market value of stockholders’ equity to its book value (European Central 
Bank, 2010). 
University of Ghana http://ugspace.ug.edu.gh
87 
 
Other popular measures of bank performance in the literature include bank efficiency, stability, 
and asset quality. In addition to the profitability measures mentioned above, our study also 
measures bank performance as bank stability (insolvency risk and asset quality) and bank cost 
efficiency.  
 
3.6.1 Measurement of Bank Stability 
As in Beck et al. (2013) and others (Turk-Ariss, 2010; Berger et al., 2009), we measure bank 
stability or risk by the Z-score. The Z-score is the most popular measure of bank stability or 
insolvency risk in the literature. The Z-score combines profitability, leverage, and volatility of 
bank returns in a single measure of overall risk or stability. It is measured by the ratio below: 
 
itititit ROATAEROAscoreZ ,,, /)/(        (3.4) 
 
Here, ROA is return on assets, E/TA refers to the bank equity to total assets ratio and ROA
represents the standard deviation of return on assets. We use a three-year rolling time window to 
obtain values of standard deviation of ROA, just like Beck et al. (2013). This approach allows for 
time variation in the denominator of the Z-score, and also avoids suggesting that the Z-scores are 
solely driven by variation in the levels of capital and profitability. And given the unbalanced nature 
of our panel dataset, it would be appropriate to avoid computing the denominator over different 
window lengths for different banks. The Z-score indicates the number of standard deviations by 
which profitability would have to fall from the mean before wiping out bank capitalization, with 
larger values reflecting greater bank stability and less bank risk potential. However, in order to 
reduce scale bias, we use the natural logarithm of the Z-score in the empirical estimation. 
University of Ghana http://ugspace.ug.edu.gh
88 
 
The Z-score increases with higher profitability and capitalization levels, and decreases with 
volatile earnings as indicated by a higher standard deviation of return on assets. However, since it 
is difficult to assess and capture bank stability using a single measure, as noted by Turk-Ariss 
(2010), the sensitivity of the results is also checked using bank asset quality. We measure bank 
asset quality using the ratio of loan loss reserves to total assets as a proxy. 
 
3.6.2 Measurement of Bank Profitability  
We measure bank profitability using two alternative measures that are commonly used in the bank 
performance literature - the Net Interest Margin (NIM) and the Return on Average Assets (ROAA). 
The NIM is used to assess how a bank generates income from its main function of intermediation. 
It is the ratio of net interest income (interest income minus interest expense) to total assets. The 
ROAA is measured as the bank’s net income for the year divided by average total assets, and 
indicates the bank’s ability to earn income on its investments in assets.  
 
3.6.3 Measurement of Bank Efficiency 
In recent years, methods employed to measure efficiency of banks are basically parametric 
(econometric) and non-parametric (linear programming-based) in nature. Stochastic Frontier 
Analysis (SFA) and the Distribution Free Approach (DFA) are commonly used types of the 
parametric method. On the other hand, most studies employing the non-parametric approach use 
the DEA (Data Envelopment Analysis) method. The DEA approach involves the application of 
mathematical programming to observed data in order to locate the frontier which can be used to 
evaluate the efficiency of each of the organizations responsible for the observed output and input 
quantities. The organizations responsible for the observed output and input quantities are referred 
University of Ghana http://ugspace.ug.edu.gh
89 
 
to as the decision making units (DMUs). DEA assigns an efficiency score to each DMU by 
comparing the efficiency score of each unit with that of its peers. DEA produces a frontier of best 
performers which is used as a yardstick to measure the efficiency of individual firms. Firms lying 
on the frontier are considered efficient and those not on the frontier are viewed as inefficient. To 
date the question of which method is the best is still not clear. However, we use the SFA to measure 
the cost efficiency of banks in Sub-Saharan Africa because, unlike the DEA, it separates random 
error from inefficiency, and is also less sensitive to outliers.  
 
Again, there is a long standing debate in the literature about what constitutes inputs and outputs 
for financial institutions. We follow the intermediation approach proposed by Sealey and Lindley 
(1977) to determine bank inputs and outputs for this study. The intermediation approach views 
financial institutions as intermediaries between savers and borrowers, and considers total loans 
and securities (or other earning assets) as outputs, whereas deposits and labour are defined as 
inputs. Other common definitions include the operating approach, the value added approach and 
the production approach.  
 
The SFA approach, first proposed independently by Aigner, Lovell, and Schmidt (1977), and 
Meeusen and van den Broeck (1977), is based on the economics concepts of cost minimization 
and profit maximization. However, this study focuses on the cost minimization problem of banks, 
and hence, cost efficiency. A bank is considered inefficient if it incurs more cost than the best 
practice bank among its peers.   
 
University of Ghana http://ugspace.ug.edu.gh
90 
 
The SFA considers the observed inefficiency of a bank as a combination of the inefficiency 
specific to the bank and a random error, and tries to extricate the two components by making 
categorical assumptions about the underlying inefficiency process. Whilst the random error is 
usually assumed to be a normally distributed variable which can affect the overall inefficiency in 
either direction, the inefficiency term is assumed to be only one-sided (a truncated or half-normal 
distribution), and can thus affect the overall inefficiency from one direction. For a cost frontier of 
banks the inefficiency element should be non-negative, as they operate on or above the minimum 
cost. As is common in the literature, and similar to Koetter et al. (2008), we use the same 
framework (translogarithmic function) as the one used in the estimation of competition, to model 
the underlying cost structure of the banking sector.  
 
As noted by Andrieş and Căpraru (2012), restrictions concerning the function of the stochastic 
frontier are more flexible when a functional form of the translog-type production function is 
applied than when a Cobb-Douglas-type form is applied. The translogarithm form does not impose 
the assumption of constant elasticity of the production function or of elasticity of substitution 
between inputs. Moreover, the translogarithm form allows the data to indicate the real value of the 
curvature of the function rather than impose prior assumptions regarding its value. In order to take 
into account the close relationship between competition and bank efficiency, the model used to 
estimate the cost efficiency scores as stated below, is basically the same as the one used to estimate 
competition (equation 3.2). 
 
University of Ghana http://ugspace.ug.edu.gh
91 
 
  
 

3
1
3
1
,,
22
10 lnlnln)(ln2lnln k k itkitkitkkititit WTAWTATATOC 
 
 
 
  

2
1
3
1
3
1
3
1
2
,,, )(ln)2/(lnln
k
k
k
k j j
itjjitjitkij trendWWW 
+ 
ititititj
j
j trendTAtrendW u+vlnln ,
3
1



      (3.5) 
 
The variables are as previously defined for equation (3.2). However, here the standard cost 
function is modified with the assumption that the error term has two components, so that 
ititit u+v  with vit representing the random error and uit being the inefficiency part. The 
inefficiency term is derived conditional on the estimate of the composite error term. The random 
error component (vit) is symmetric and captures the random variation of the cost frontier across 
banks. The other part, uit is a one-sided variable that accounts for inefficiency relative to the cost 
frontier. The time-specific effect is implied in the trend. The maximum likelihood technique is 
used to estimate the cost frontier function, with inefficiency derived from the residuals of the 
regression, under the assumption that both components of the composite error term are 
independently distributed (Maudos & de Guevara, 2007). Equation (3.5) is estimated separately 
for each country to derive individual bank efficiency scores in the Battese and Coelli (1992) 
framework, which allows us to estimate time-varying bank cost efficiency scores. 
 
3.7 The Economic Freedom Index 
There are two major measurements of economic freedom, the Economic Freedom of the World 
Index produced by the Fraser Institute (Gwartney et al., 2014) and the Economic Freedom Index 
constructed by the Heritage Foundation (Heritage Foundation, 2015) in collaboration with the Wall 
University of Ghana http://ugspace.ug.edu.gh
92 
 
Street Journal. As noted by Chortareas et al. (2013), both indices are highly credible and their 
results are compatible in general. Even though the Economic Freedom of the World Index has 
been used extensively in the literature, we use the Heritage Foundation’s Index of Economic 
Freedom for this study because one of its components, the Financial Freedom Index captures the 
issues of interest (Chortareas et al., 2013).  
 
Financial freedom, one of Heritage Foundation’s ten measures of economic freedom, is a measure 
of banking efficiency as well as a freedom from restrictions or government control and interference 
in the financial sector. State ownership of banks and other financial institutions such as insurers 
and capital markets reduces competition and generally lowers the level of available services 
(Heritage Foundation, 2015). 
 
The Financial Freedom Index scores an economy’s financial freedom by looking into the following 
five broad areas: 
• The extent of government regulation of financial services, 
• The degree of state intervention in banks and other financial firms through direct and 
indirect ownership, 
• The extent of financial and capital market development, 
• Government influence on the allocation of credit, and 
• Openness to foreign competition. 
 
These five areas are considered to assess an economy’s overall level of financial freedom that 
ensures easy and effective access to financing opportunities for people and businesses in the 
University of Ghana http://ugspace.ug.edu.gh
93 
 
economy. An overall score on a scale of 0 to 100 is given to an economy’s financial freedom 
through deductions from the ideal score of 100, which reflects negligible government interference. 
A score of 50 indicates the existence of considerable government interference in credit allocation, 
and significant restrictions on the ability of financial institutions (especially foreign institutions) 
to offer financial services. Economies scoring below 50 are considered to have repressive policies 
with strong or extensive government control over the central bank and credit allocation. 
 
3.8 Estimating Effect of Competition and Financial Freedom on Bank Stability  
Our empirical model below, which investigates how competition and financial freedom impact 
bank stability is similar to Liu et al. (2013).   
𝑅𝑖𝑠𝑘𝑖𝑡 = 𝜆𝑅𝑖𝑠𝑘𝑖𝑡−1 + 𝛼1𝐿𝑒𝑟𝑛𝑒𝑟𝑖𝑡 + 𝛼2𝐿𝑒𝑟𝑛𝑒𝑟²𝑖𝑡 + 𝛼3𝐹𝑟𝑒𝑒𝑑𝑜𝑚𝑖𝑡 
 + 𝛼4𝐿𝑒𝑟𝑛𝑒𝑟𝑖𝑡 ∗ 𝐹𝑟𝑒𝑒𝑑𝑜𝑚𝑖𝑡 + ∑ 𝜔𝑛  Χ𝑖𝑡
𝑛4
𝑛=1 + 𝜉𝑖𝑡      ,      (3.6) 
where,  𝜉𝑖𝑡 = 𝜘𝑖 + 𝛾𝑡 + 𝜚𝑖𝑡 and the subscripts i, and t represent bank i, at year t, respectively. 𝑅𝑖𝑠𝑘𝑖𝑡 
is our measure of bank stability and 𝑅𝑖𝑠𝑘𝑖𝑡−1 is the observation on the same bank in the same 
country in the previous year. As explained earlier, the main bank stability measure we used for 
this study is the natural logarithm of Z-score (Risk). However, we also use bank asset quality as an 
alternative risk measure in view of the difficulty in measuring bank risk with a single measure. 
Lerner represents the conventional Lerner index or the funding-adjusted Lerner index, and is our 
measure of competition.  Following the findings of Martinez-Miera and Repullo (2010), we 
include a quadratic term, 𝐿𝑒𝑟𝑛𝑒𝑟² to account for a possible non-linear relationship between 
competition and bank stability. Freedom is the degree of financial freedom or economic freedom. 
  
University of Ghana http://ugspace.ug.edu.gh
94 
 
Financial Freedom is a broad indicator for the openness of a banking system, capturing whether 
foreign banks are allowed to operate freely, whether difficulties are faced when setting up domestic 
banks, and whether the government influences the allocation of credit, with larger values 
signifying more freedom (Berger et al., 2009; Demirguc-Kunt, Laeven, & Levine, 2004; 
Chortareas et al., 2011).  
 
Economic Freedom is a composite measure of a country’s overall economic freedom based on 10 
different factors. It comprises of Financial Freedom, Business Freedom, Property Rights, 
Monetary Freedom, Investment Freedom, Freedom from Corruption, Labor Freedom, Trade 
Freedom, Fiscal Freedom and Government Size. The index takes values in a scale from 0 to 100, 
with scores approaching 100 representing higher levels of freedom. Similarly, the higher the score 
on a factor, the lower the level of government interference in the economy. Lerner*Freedom is the 
interaction between competition and the freedom variables. The variable 
itX is a vector 
representing control variables. The error term, 
it  has three components: the unobserved time-
invariant bank-specific effect (𝜘𝑖), the unobserved time effects (𝛾𝑡), and the random error (𝜚𝑖𝑡). 
This is a two-way error component regression model, where 𝜘𝑖 ~ IIN (0,𝜎𝜐
2 ) and independent of 
𝜚𝑖𝑡 ~ IIN (0,𝜎𝜇
2 ). 
 
In line with the existing literature on the factors that influence bank stability, we control for a 
number of variables, including Capitalization (ratio of equity to total assets), Credit risk (ratio of 
total loans to total assets), Diversification (ratio of non-interest income to total income), and 
Financial development (domestic credit to the private sector as a percentage of GDP). Most 
previous research on bank capital and risk suggest a negative relationship, as the more capital a 
University of Ghana http://ugspace.ug.edu.gh
95 
 
bank holds, the more stable it is thought to be. Since lending is one of the most risky areas of 
banking business, Credit risk is expected to associate negatively with bank stability. The greater a 
bank’s loans exposure the higher is the probability of default, and hence the lower is the stability 
(Liu et al., 2013). It is generally expected that Diversification will enhance bank stability, yet 
recent studies suggest that more diversified banks are less stable compared to their less diversified 
counterparts (Beck et al., 2009). We anticipate that higher levels of Financial development would 
be associated with greater stability as banks gain more experience over time.  
 
3.9 Modelling Competition, Financial Freedom and Bank Profitability 
We specify a dynamic model to determine the effect of competition and financial freedom on bank 
profitability as shown below. As highlighted by Berger et al. (2000), bank profits tend to persist 
over time, as a result of impediments to market competition, informational opacity and/or 
sensitivity to regional/macroeconomic shocks, to the extent that these are serially correlated. 
Consequently, we specify a dynamic model which includes the first lag of the dependent variable 
among the regressors. This approach follows several studies on bank profitability such as 
Athanasoglou et al. (2008), Garcıa-Herrero, Gavila, and Santabarbara (2009), Trujillo-Ponce 
(2013) and Dietrich and Wanzenried (2014). 
 
𝑃𝑒𝑟𝑓𝑖𝑡 = 𝛷𝑃𝑒𝑟𝑓𝑖𝑡−1 + 𝜌1𝐿𝑒𝑟𝑛𝑒𝑟𝑖𝑡 + 𝜌2𝐹𝑟𝑒𝑒𝑑𝑜𝑚𝑖𝑡 
 + 𝜌3𝐿𝑒𝑟𝑛𝑒𝑟𝑖𝑡 ∗ 𝐹𝑟𝑒𝑒𝑑𝑜𝑚𝑖𝑡 + ∑ 𝛩𝑛  Ζ𝑖𝑡
𝑛8
𝑛=1 + 𝜖𝑖𝑡  ,      (3.7) 
where,  𝜖𝑖𝑡 = 𝛹𝑖 + 𝜁𝑡 + 𝜑𝑖𝑡 and  the subscripts i, and t represent bank i at year t, respectively. 𝑃𝑒𝑟𝑓𝑖𝑡 
is our measure of bank profitability and 𝑃𝑒𝑟𝑓𝑖𝑡−1 is the observation on the same bank in the same 
country in the previous year. We use two measures for bank profitability, Net Interest Margin 
University of Ghana http://ugspace.ug.edu.gh
96 
 
(NIM) and Return on Average Assets (ROAA). Lerner represents the conventional Lerner index 
or the funding-adjusted Lerner index, and is our measure of competition.  Freedom is the degree 
of financial freedom or economic freedom, and Lerner*Freedom is the interaction between 
competition and the freedom variables. The variable 
itX  is a vector representing control variables. 
The error term, 𝜖𝑖𝑡 has three components: the unobserved time-invariant bank-specific effect (𝛹𝑖), 
the unobserved time effects (𝜁𝑡), and the random error (𝜑𝑖𝑡). This is a two-way error component 
regression model, where 𝛹𝑖 ~ IIN (0,𝜎𝜐
2 ) and independent of 𝜑𝑖𝑡 ~ IIN (0,𝜎𝜇
2 ). 
 
We control for a number of variables that generally affect the relationship between competition 
and bank profitability. These include bank-specific and country-level characteristics. The bank-
level variables are Cost to income (ratio of non-interest operating expenses to operating income), 
Bank size (natural logarithm of total assets), Capitalization (ratio of equity to total assets), Credit 
risk (ratio of total loans to total assets), and Diversification (ratio of non-interest income to total 
income). The country-level variables are Concentration (Hirschman-Herfindahl Index-the sum of 
the squares of the market share of assets of each bank), Financial development (domestic credit to 
the private sector as a percentage of GDP), and Economic development (natural logarithm of per 
capita GDP). 
  
The operating efficiency of banks is known to be an important determinant of bank profitability. 
In this study, we use the Cost to income ratio as a proxy for efficiency. Banks with high operating 
efficiency are usually able to maintain a lower cost to income ratio, which enhances profits. Hence, 
we expect a negative relationship between Cost to income and bank profitability. Of course, if the 
benefits of improvements in efficiency are shared with customers in the form of lower loan rates 
University of Ghana http://ugspace.ug.edu.gh
97 
 
and/or higher deposit rates, then the expected increase in profitability may not materialize 
(Goddard et al., 2013). On the other hand, banks that bear higher operating costs may charge higher 
margins (if the competition allows), to balance their higher transformation costs, which implies a 
positive relationship between Cost to income and bank profitability (Maudos & de Guevara, 2004; 
Peria & Mody, 2004; Hawtrey & Liang, 2008). 
 
The relationship between Bank size and profitability is quite ambiguous. Smirlock (1985) argues 
that a growing bank size is positively related to bank profitability. Since larger banks are in a better 
position to realize economies of scale and reduce the cost of gathering and processing information, 
bank size should be positively associated with its performance (Demirguç-Kunt & Huizinga, 1999; 
Dietrich & Wanzenried, 2011). However, extremely large banks might show a negative 
relationship between size and profitability. This is due to agency costs, the overhead of 
bureaucratic processes, and other costs related to managing large firms (e.g. Stiroh & Rumble, 
2006; Pasiouras & Kosmidou, 2007). On the other hand, Athanasoglou et al. (2008) find that size 
is not significant in explaining bank profitability. They suggest that this may be because small-
sized banks usually try to grow faster, even at the expense of their profitability. In addition, newly 
established banks are not particularly profitable (if at all profitable) in their first years of operation, 
as they place greater emphasis on increasing their market share, rather than on improving 
profitability.  
 
In line with the general literature, we measure bank Capitalization with the ratio of equity to total 
assets. Highly capitalized banks have a reduced need for external funds, and might be able to 
reduce their funding costs, which should positively affect their profitability. Besides, 
University of Ghana http://ugspace.ug.edu.gh
98 
 
Athanasoglou et al. (2008) explain that the positive relationship between capital and bank 
profitability could be attributed to the ability of banks with sound capitalization to pursue business 
opportunities more effectively, and also because such banks have more time and flexibility to deal 
with problems arising from unexpected losses. Claeys and Vennet (2008) also suggest that when 
a bank holds capital in excess of the regulatory minimum, two positive effects on the interest 
margin can be distinguished. Since the bank has free capital it can increase its portfolio of risky 
assets in the form of loans or securities, and if market conditions allow the bank to make additional 
loans with a beneficial return/risk profile, this will usually increase the interest margin. Moreover, 
since capital is considered to be the most expensive form of liabilities in terms of expected return, 
holding capital above the regulatory minimum is a credible signal of creditworthiness on the part 
of the bank. And when depositors exert depositor market discipline, this may enable the bank to 
lower its deposit funding costs and, hence, increase its interest margin.  
 
However, if we consider the conventional risk-return hypothesis, we have to expect banks with 
lower capital ratios (which makes them more risky) to have higher returns in comparison to better-
capitalized financial institutions (Dietrich & Wanzenried, 2011). Hence, we have no specific 
expectation as to the direction of the relationship between Capitalization and bank profitability. 
The ratio of total loans to total assets is used as a proxy for Credit risk. An increased exposure to 
credit risk may be associated with decreased bank profitability. We thus generally expect a 
negative effect of this variable on bank returns (Dietrich & Wanzenried, 2014). As suggested by 
Dietrich and Wanzenried (2011), the negative relationship between credit risk and bank profits 
might be a reflection of exposure to high-risk loans, resulting in loan losses that lower the returns 
of the affected banks. On the other hand, banks that make risky loans may also be obliged to hold 
University of Ghana http://ugspace.ug.edu.gh
99 
 
a higher amount of provisions. And in order to compensate for the higher risk of default, they may 
charge higher margins, leading to a positive relationship (Maudos & de Guevara, 2004).  
 
In recent times, an increasing proportion of bank income have been generated from non-interest 
activities, such as payment and other commission-based services, investment banking and 
insurance underwriting, among others. We measure Diversification as the ratio of non-interest 
income to total income. Chiorazzo et al. (2008) and Elsas et al. (2010) assert that revenue 
diversification enhances bank profitability via higher margins from non-interest businesses. 
However, other previous studies (e.g. Stiroh & Rumble, 2006) show that greater diversification of 
the banking business does not necessarily translate into an improvement of the bank’s profitability. 
In fact, such diversification may be detrimental to profitability. Nguyen (2012) also found that net 
interest margin is not always inversely related to diversification. 
 
Our proxy for banking market Concentration is the Hirschman-Herfindahl Index (HHI). It is 
measured as the sum of the squares of the market share of assets of each bank in the industry. The 
Structure–Conduct–Performance (SCP) hypothesis suggests that higher concentration in banking 
markets has a positive impact on bank profitability, because collusion among banks may result in 
higher rates on loans and lower interest rates on deposits. However, if concentration is the result 
of tougher competition in the banking industry, this would suggest a negative relationship between 
performance and market concentration (Boone & Weigand, 2000). Consequently, the overall effect 
of market concentration on bank performance is uncertain. 
 
University of Ghana http://ugspace.ug.edu.gh
100 
 
We control for the level of Financial development and Economic development using domestic 
credit to the private sector as a percentage of GDP, and the natural logarithm of per capita GDP, 
respectively. The existing literature shows that GDP per capita could be expected to have a positive 
impact on bank performance (Bashir, 2003). On the other hand, we anticipate that higher levels of 
financial development would be associated with lower profits, since there may be little room for 
new business, as a result of higher competition.  
 
3.10 Modelling Effect of Financial Freedom on Bank Competition  
Two important econometric concerns might arise when modelling sources of bank competition. 
These are the dynamic nature of bank competition and the potential endogeneity of some of the 
explanatory variables (Delis, 2012). Berger et al. (2000) suggest the presence of informational 
opacity, networking, and relationship-lending even in developed banking systems, all of which 
might cause bank rents and market power to persist. We thus include the lagged dependent variable 
among the regressors in our econometric model, to account for this type of persistence. The 
empirical model follows Delis (2012) and is specified as: 
 
Competition𝑖𝑡 = 𝛳Competition𝑖𝑡−1 + 𝜏𝐹𝑟𝑒𝑒𝑑𝑜𝑚𝑖𝑡 + ∑ 𝜕𝑛  Κ𝑖𝑡
𝑛6
𝑛=1 +ώ𝑖𝑡,            (3.8) 
 
where  ώ𝑖𝑡 = 𝚠𝑖 + 𝚡𝑡 + 𝚢𝑖𝑡 and the subscripts i, and t represent bank i at year t, respectively. 
𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑡𝑖𝑜𝑛𝑖𝑡 refers to bank market power (Lerner index), our proxy for bank competition and 
𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑡𝑖𝑜𝑛𝑖𝑡−1 is the observation on the same bank in the same country in the previous year. 
We use two measures for competition, the conventional Lerner index and the funding-adjusted 
Lerner index. Freedom is the degree of financial freedom or economic freedom. The vector 
University of Ghana http://ugspace.ug.edu.gh
101 
 
variable 
it represents control variables that typically explain competitive conditions in banking 
markets. The error term,  ώ𝑖𝑡  has three components: the unobserved time-invariant bank-specific 
effect (𝚠𝑖), the unobserved time effects (𝚡𝑡), and the random error (𝚢𝑖𝑡). This is a two-way error 
component regression model, where 𝚠𝑖 ~ IIN (0,𝜎𝜐
2 ) and independent of 𝚢𝑖𝑡 ~ IIN (0,𝜎𝜇
2 ).  
 
The main drivers of market power identified in the literature include size, default risk, 
diversification, efficiency, equity, concentration, market share, ownership, monetary policy, 
inflation, and financial depth (Aboagye et al., 2008b; De Guevara and Maudos, 2007; De Guevara 
et al., 2005). We follow a similar empirical approximation, but also test for the effect of financial 
freedom and economic freedom in order to take account of the broader environment within which 
banking activity takes place in a country. An open and transparent banking environment facilitates 
access to financing and encourages competition to provide efficient financial intermediation 
between households and firms, as well as between investors and entrepreneurs (Chortareas et al., 
2013).  
 
We control for bank Capitalization (ratio of equity to total assets), Diversification (ratio of non-
interest income to total income), Bank size (natural logarithm of total assets), Market share (ratio 
of a bank’s assets to total national industry assets), Financial development (domestic credit to the 
private sector as a percentage of GDP), and Economic development (natural logarithm of per capita 
GDP). As suggested by Delis (2012), well-capitalized and larger banks may be able to set higher 
margins or gain access to cheaper sources of funds due to scale economies, informational 
asymmetries, and moral hazard issues. Also, De Guevara et al. (2005) have shown that bank size 
has a positive and significant effect on market power. Larger banks enjoy greater market power 
University of Ghana http://ugspace.ug.edu.gh
102 
 
due to either cost advantages or to their capacity to impose higher prices. While one would expect 
higher market share and activity diversification to impact market power positively, improvements 
in financial development and economic development are likely to stifle the exercise of market 
power by individual banks. 
 
3.11 Estimating Effect of Competition and Financial Freedom on Bank Efficiency  
In line with Turk-Ariss (2010) and Coccorese and Pellecchia (2010), we investigate the 
implications of the degree of financial freedom and bank market power on cost efficiency using 
the Tobit model, the most widely adopted approach in the bank efficiency literature. Bank cost 
efficiency scores are ranged between 0 and 1, hence Tobit models are more appropriate since they 
better fit models where the dependent variable is derived from a first-stage regression, as Greene 
(2005) posits. 
 
We follow a two-stage approach, where bank cost efficiency scores and Lerner index of market 
power are obtained from a first-stage regression by means of a common stochastic frontier model, 
as previously explained. Then in the second-stage, we used the freedom variables and the estimated 
market power measures, as well as a number of control variables to explain cost efficiency. We 
are thus able to test both the quite life hypothesis, and the conditioning effect of financial freedom 
on the competition-efficiency relationship, which is known to be complex (Casu & Girardone, 
2006).  
 
 
 
University of Ghana http://ugspace.ug.edu.gh
103 
 
The extended model is: 
  CostEffit = 1Lernerit +2Freedomit +3Lernerit *Freedomit +4lnHHIait +5EQTAit  
+6ROAAit +7Domcrpsit+8lngdppcit + ϧit       (3.9) 
  
where lnHHIa represents banking market concentration. EQTA is the ratio of bank equity to total 
assets while ROAA represents bank profitability. We account for the effect of financial and 
economic development with Domcrps (domestic credit to the private sector as a percentage of 
GDP) and lngdppc (natural logarithm of GDP per capita), respectively. 
  
The main independent variables of interest are the Freedom variables, representing financial 
freedom or economic freedom, and the Lerner which measures the degree of bank market power. 
Two specifications of the Lerner index are used, the conventional Lerner index and the funding-
adjusted Lerner index (to account for market power arising in the deposit market). The funding-
adjusted Lerner measure also better account for the inter-relatedness between market power and 
bank efficiency, thereby addressing endogeneity concerns (Turk-Ariss, 2010). The Lerner index is 
the mark-up of price over marginal costs, with higher values denoting higher pricing power and 
less competitive market conditions, and lower values signifying greater competition and lesser 
pricing power. We prefer to use the Lerner index as our measure of the degree of market power 
(or competition) compared to other traditional measures of market structure because it is observed 
at the bank level, just like the bank cost efficiency scores to which they are related. For example, 
unlike the Lerner index, concentration ratios and the Panza Rosse H-statistic are estimated at the 
country level. 
 
University of Ghana http://ugspace.ug.edu.gh
104 
 
Following the literature on the determinants of bank efficiency, we control for a number of 
influencing factors. According to Demirgüç-Kunt and Levine (2000), the empirical evidence on 
the links between banking market Concentration and bank efficiency is not unambiguously 
positive or negative. While some argue that concentration strengthens market power and thereby 
stifle competition and efficiency (structure-conduct-performance hypothesis), others assert that 
economies of scale result in bank mergers and acquisitions, so increased concentration goes hand-
in-hand with efficiency improvements (efficient-structure hypothesis). Hence, we could expect 
either a positive or a negative relationship between concentration and bank efficiency. 
 
According to Chortareas et al. (2012), many studies have highlighted the important role of 
Capitalization in preventing bank failure, and in protecting customers or whole economies from 
negative shocks. However, regulation may impede the efficient operation of banks. For example, 
changes in capital regulations may compel banks to engage in riskier ventures or invest in ways 
that seek to circumvent regulations, which could eventually affect their efficiency. Rules on bank 
capital requirements may not reflect the risks involved, but this could inadvertently encourage 
banks to hold either too much or too little capital. While inadequate capital increases the risk of 
bank failure, too much capital, on the other hand, inflicts needless costs on banks and their 
customers with negative repercussions for the efficiency of the banking system. 
 
Higher bank Profitability may also have varying effects on bank efficiency. Banks earning higher 
profits may be in a position to acquire more efficient technology and management capacity to 
improve efficiency, but higher profits also have the tendency to induce wastefulness. Hence, we 
could expect the relationship between bank profits and efficiency to be either positive or negative. 
University of Ghana http://ugspace.ug.edu.gh
105 
 
Since the macroeconomic environment is likely to have implications for bank efficiency levels, 
we also control for Financial development (domestic credit to the private sector as a percentage of 
GDP) and Economic development (natural logarithm of per capita GDP).  These variables should 
capture the cyclical conditions of the macroeconomic environment, and the effect of operating in 
different economic environments, since demand for financial products depends to a large extent 
on the level of economic activity (Chortareas et al., 2012). 
 
3.12 Estimating the Determinants of Bank Profit Persistence  
In line with Goddard et al. (2004) and Goddard et al. (2013), the following empirical model is used 
to estimate the determinants of bank profit persistence: 
  
πit = κi + λ1π𝑖𝑡−1 +ϐ1Economic freedomt+ Ί1xit + Ί2mt + ύit      (3.10) 
 
where πit denotes the normalized profit rate of bank i, in year t. Return on Average Equity (ROAE) 
is the profit rate measure, and is measured by the ratio of net income after tax to total equity. The 
profit rate is normalized by expressing πit as a deviation from the cross-sectional industry mean 
profit rate in year t. The main variable of interest in the above model is Economic freedom, which 
assesses the effect of general liberty to engage in business and other economic transactions on 
bank profit persistence (or intensity of competition). We could not use financial freedom as in the 
other models, because of limited variations at the country level over the study period. 
• xit denotes a vector of bank-specific variables;  
• mt denotes a vector of country-specific variables;  
• κi is an individual effect for bank i; 
University of Ghana http://ugspace.ug.edu.gh
106 
 
• λ1 is the coefficient reflecting bank profit persistence;  
• Ί1 and Ί2 are vectors of coefficients; 
• ύit is the error term, which decomposes as ύit = ϫt + zit.  
 
The model is estimated separately for each country. The coefficient of the lagged dependent 
variable, λ1 measures the speed of adjustment of short-run profits to the competitive norm. If it is 
high, then short-run profits are persistent and competition is thought to be weak. On the other hand, 
if it is small, or close to zero, then short-run profits are quickly eroded and competition is 
considered to be strong or fierce.  The motivation for the specification of a dynamic profit equation 
containing a partial adjustment mechanism (lagged dependent variable) is drawn from the 
Persistence of Profit (POP) literature. Banks usually earn excess profits either through the 
exploitation of market power of incumbency, or because incumbent banks are more efficient or 
innovative in the production or distribution of financial services. Over time, entry encourages 
competition, and eventually erodes any excess profit (Goddard et al., 2013).  
 
The other explanatory variables that typically influence banking market competition and 
comprise the vectors xit and mt are defined as follows: 
Equity = ratio of equity to total assets.  
Bank size = natural logarithm of total assets. 
Credit risk = ratio of gross loans to total assets. 
Market Share = bank share of total assets of the banking market  
 
 
University of Ghana http://ugspace.ug.edu.gh
107 
 
Concentration = sum of square of market share of bank assets. 
Financial development = ratio of credit to the private sector to GDP  
 
As suggested by Delis (2012), well-capitalized and larger banks may be able to set higher margins 
or gain access to cheaper sources of funds due to scale economies, informational asymmetries, and 
moral hazard issues. However, as with the general literature on bank profitability and competition, 
the effect of size on profit persistence may not always be in one direction. While one would expect 
higher market share to impact profit persistence positively, the effect of lending specialization (or 
credit risk) is quite uncertain. Again, the relationship between banking market concentration on 
one hand, and profitability and competition on the other hand, has been rather ambiguous in the 
literature. 
 
Goddard et al. (2011), suggest a positive relationship between GDP growth and profit persistence, 
on the back of increased business opportunities for banks, which might help banks to sustain 
positions of excess profitability. On the other hand, the availability of abundant business 
opportunities might tend to strengthen competition between banks, in which case a negative 
relationship would be expected between GDP growth and profit persistence. We expect similar 
results for the relationship between financial development and bank profit persistence. 
 
3.13 Method of Data Analysis  
We estimate most of the regressions based on the system Generalized Method of Moments (GMM) 
estimator (Arellano & Bover, 1995). This method is used to estimate the effects of Competition 
and Financial Freedom on Bank Stability, the effects of Competition and Financial Freedom on 
University of Ghana http://ugspace.ug.edu.gh
108 
 
Bank Profitability, the sources of bank Market Power, and the determinants of bank Profit 
Persistence. We used this dynamic panel model to deal with possible endogeneity of some of the 
variables used in the estimations, and to account for persistence of bank risk, profitability and 
market power. The model includes the first lag of the dependent variable as part of the covariates, 
and unobserved individual bank effects. Hence, the standard fixed effects or random effects 
estimators would be inconsistent, since by construction, the individual bank effects are correlated 
with the lagged dependent variable. To address these issues, Arellano and Bond (1991) use a 
Generalized Method of Moments (GMM) estimator for such models, popularly known as the 
difference GMM. In the difference GMM model, lagged exogenous variables in levels are used as 
instruments for the first differenced, lagged dependent variable. However, Arellano and Bover 
(1995) and Blundell and Bond (1998) have shown that these lagged variables may provide little 
information about the first differences. Consequently, Blundell and Bond (1998) expanded the 
work of Arellano and Bover (1995) to develop a system estimator that exploits additional moment 
conditions on both first-differences and levels, with lagged first-differences of the series employed 
as instruments in the levels equation. The system GMM estimator reduces potential bias in finite 
samples as well as asymptotic imprecision associated with the difference estimator (Blundell & 
Bond, 1998).  
 
Even so, the consistency of the system GMM estimator depends on two key assumptions: that the 
error term is not autocorrelated, and that the instruments used are valid. The presence of first-order 
autocorrelation in the differenced residuals does not imply that the estimates are inconsistent. But, 
the presence of second-order autocorrelation suggests that the estimates are inconsistent. We test 
the hypothesis of no autocorrelation in the error term and report the results together with the main 
University of Ghana http://ugspace.ug.edu.gh
109 
 
results. We also use the Hansen test of over-identifying restrictions to examine the validity of the 
instruments. The Hansen test analyses the sample analogue of the moment conditions used in the 
estimation procedure to determine the validity of the instruments (Liu et al., 2013). 
 
We estimate equation (3.6), which evaluates the effect of competition and financial freedom on 
bank stability, using the two-step system GMM estimator with Windmeijer-corrected standard 
errors, small-sample adjustments, and orthogonal deviations (Windmeijer, 2005). The difference 
and system GMM estimators have one-step and two-step variants. The two-step system GMM uses 
residuals from the one-step estimates and is asymptotically more efficient than the one-step. 
 
One of the main problems in assessing the determinants of bank profitability is the potentially 
endogenous character of some of the factors. For example, banks that are more profitable may 
have more resources to increase their equity, and may also find it easier to increase their customer 
base through successful advertising and, thereby, enhance profitability. Indeed, causality could 
even go in the opposite direction; that is, higher bank profitability could lead to more employees 
and less efficiency (Garcıa-Herrero et al., 2009). 
 
Besides, there are heterogeneity concerns since some features of banks that affect profitability are 
unobserved and difficult to measure or identify in an equation. If the impact of such characteristics 
are not taken into account, there could be correlations between some of the coefficients of the 
explanatory variables and the error terms that bias these coefficients (Trujillo-Ponce, 2013). To 
address these concerns, as well as the persistence of bank profitability, we use the system GMM 
estimator developed for dynamic panel models by Arellano and Bover (1995) and Blundell and 
Bond (1998), to ascertain the effect of competition and financial freedom on bank profitability. 
University of Ghana http://ugspace.ug.edu.gh
110 
 
Thus, equation (3.7) is estimated using the two-step system GMM estimator with Windmeijer-
corrected standard errors, small-sample adjustments, and orthogonal deviations (Windmeijer 
(2005). The same procedure is used to estimate equation (3.8), the sources of bank market power, 
for similar reasons. 
 
However, since bank efficiency scores are ranged between 0 and 1, a different estimation technique 
was used for equation (3.9), to assess the effect of competition and financial freedom on bank cost 
efficiency. Tobit models are more appropriate in this case since they better fit models where the 
dependent variable is derived from a first-stage regression (Greene, 2005). Again, since the cost 
efficiency scores and market power indices are obtained from the same model, we take into account 
endogeneity arising from simultaneity by estimating an instrumental variable Tobit model, in 
addition to the traditional Tobit specifications. Indeed, the efficient-structure hypothesis suggests 
a causal relationship going from efficiency to market power. Hence, in our econometric models 
the variable Lerner could be endogenous. This gives further reasons why an instrumental variables 
version of the above model is warranted. We use a Wald test of exogeneity to confirm the 
possibility of endogeneity (Coccorese & Pellecchia, 2010).  
 
Equation (3.10) is also a linear dynamic panel regression model. We estimated separate regressions 
for each country. However, due to data limitations in using the system GMM with requirements 
for instruments, it was estimated for only five of the eleven countries. This estimator is designed 
for large N-small T panels, as in our study. Accordingly, we estimate (3.10) using the two-step 
system GMM estimator, including both lagged differences and levels of the explanatory variables 
as instruments. The system GMM estimator reduces potential biases in finite samples, and 
University of Ghana http://ugspace.ug.edu.gh
111 
 
asymptotic imprecision associated with the difference estimator (Blundell & Bond, 1998). Two 
specification tests are reported. The first is a Sargan test of overidentifying restrictions, which 
examines the validity of the instruments by analyzing the sample analogue of the moment 
conditions used in the estimation procedure. The second test examines the hypothesis of no 
autocorrelation in the disturbance term. The presence of first-order autocorrelation in the 
differenced residuals does not imply that the estimates are inconsistent, but the presence of second-
order autocorrelation implies that the estimates are inconsistent. We thus report only the second-
order autocorrelation test results. 
 
3.14 Chapter Summary  
This chapter provides the methodological framework for the study. We outline the measurement 
approaches for the key variables used in the estimations. We measure bank competition using the 
Lerner index of market power to overcome the limitations of the traditional approach of using 
indirect proxies such as banking market concentration or market shares of banks to infer 
competitive conditions. We prefer to use the Lerner index because, unlike the other popular 
measure of bank competition, the Panza-Rosse framework, which is an industry level measure, the 
Lerner index measures bank market power (as a proxy for competition) at the bank level. Another 
advantage of using the Lerner index is that it allows us to measure competition and bank efficiency 
using the same framework, and thus obtain better insight into the closely related issues of bank 
competition and efficiency. Most previous studies assessing these relationships use different 
models to measure competition and efficiency and thereby fail to take into account the 
simultaneous relationship between them (Koetter et al., 2012). The Lerner index is the mark-up of 
University of Ghana http://ugspace.ug.edu.gh
112 
 
price over marginal costs, with higher values denoting higher pricing power and less competitive 
market conditions.  
 
We use a translogarithmic function to model the underlying cost structure of the banking sector. 
In addition to the conventional Lerner index, we also estimate the funding-adjusted Lerner index 
from the same model by excluding the price of funds. This is to ensure robustness of our results 
and to take into account, concerns that the conventional Lerner index estimates may be biased due 
to the exercise of market power in the deposit market (Turk-Ariss, 2010). In order to provide a 
more comprehensive assessment of bank performance, we use three main proxies that are 
commonly used in the literature to assess bank performance. These are bank stability (Z-score), 
bank profitability (NIM and ROAA), and bank cost efficiency.  
 
We use Stochastic Frontier Analysis (SFA) to measure the cost efficiency of banks because, unlike 
the DEA (Data Envelopment Analysis), which is also commonly used, it separates random error 
from inefficiency, and is also less sensitive to outliers. A bank is considered inefficient if it incurs 
more cost than the best practice bank among its peers.  The cost function is estimated separately 
for each country to derive individual bank efficiency scores in the Battese and Coelli (1992) 
framework, which allows us to estimate time-varying bank cost efficiency scores. 
 
With regards to our measure of economic freedom, we considered the Economic Freedom of the 
World Index produced by the Fraser Institute (Gwartney et al., 2014) and the Economic Freedom 
Index constructed by the Heritage Foundation (Heritage Foundation, 2015) in collaboration with 
the Wall Street Journal. As noted by Chortareas et al. (2013), both indices are highly credible and 
University of Ghana http://ugspace.ug.edu.gh
113 
 
their results are usually similar, but we settled on the Heritage Foundation’s Index of Economic 
Freedom for this study because one of its ten components, the Financial (Banking) Freedom Index 
captures the issues of interest. The financial freedom index measures the degree of government 
control and interference in the financial sector. It is measured on a scale of 0 to 100 through 
deductions from the ideal score of 100, which reflects insignificant government interference.  
 
We estimate several regressions based on the system Generalized Method of Moments (GMM) 
estimator (Arellano & Bover, 1995). This method is used to estimate the effects of Competition 
and Financial Freedom on Bank Stability, the effects of Competition and Financial Freedom on 
Bank Profitability, the sources of bank Market Power, and the determinants of bank Profit 
Persistence. This dynamic panel data model takes into account the persistence of bank stability, 
bank profitability and bank market power, as well as concerns about potential endogeneity among 
some of the variables. Following the findings of Martinez-Miera and Repullo (2010), we include 
a quadratic term, 𝐿𝑒𝑟𝑛𝑒𝑟² to account for a possible non-linear relationship between competition 
and bank stability. We also include an interaction term in the regressions to test the potential effect 
of the degree of financial freedom on the ambiguous relationship between competition and bank 
stability, as well as between competition and profitability.  
 
We estimate the regressions using the two-step system GMM estimator with Windmeijer-corrected 
standard errors, small-sample adjustments, and orthogonal deviations (Windmeijer, 2005). The 
consistency of the system GMM estimator depends on the assumptions that the error term is not 
autocorrelated, and that the instruments used are valid. We test the hypothesis of no autocorrelation 
University of Ghana http://ugspace.ug.edu.gh
114 
 
in the error term and report the results together with the main results. We also use the Hansen test 
of over-identifying restrictions to examine the validity of the instruments.  
 
In line with the existing literature on the factors that influence bank stability, we control for 
Capitalization (ratio of equity to total assets), Credit risk (ratio of total loans to total assets), 
Diversification (ratio of non-interest income to total income), and Financial development. With 
regards to the profitability model, we control for Cost to income (ratio of non-interest operating 
expenses to operating income), Bank size (natural logarithm of total assets), Capitalization, Credit 
risk, Diversification, Concentration (Hirschman-Herfindahl Index-the sum of the squares of the 
market share of assets of each bank), Financial development and Economic development. In the 
competition model, our main focus is on the effect of financial freedom on competition or market 
power, but we control for Capitalization, Diversification, Bank size, Market share (ratio of a 
bank’s assets to total national industry assets), Financial development, and Economic development.  
 
In line with Goddard et al. (2013), we estimate the determinants of bank profit persistence using 
the two-step system GMM estimator. However, unlike the previous models described above for 
bank stability, profitability and competition, which were cross-country regressions, the profit 
persistence model is estimated separately for each country. And due to data limitations in using 
the system GMM with requirements for instruments, it was estimated for only five of the eleven 
countries. Our dependent variable is the normalized profit rate of bank i, in year t. Return on 
average equity (ROAE) is the profit rate measure, which is normalized by expressing it as a 
deviation from the cross-sectional industry mean profit rate in year t. We control for other 
explanatory variables that typically influence banking market competition and profitability. These 
University of Ghana http://ugspace.ug.edu.gh
115 
 
are Capitalization, Bank size, Credit risk, Market Share, Concentration, and Financial 
development. 
 
In line with Turk-Ariss (2010) and Coccorese and Pellecchia (2010), we investigate the 
implications of the degree of financial freedom and bank market power on cost efficiency using 
the Tobit model, the most widely adopted approach in the bank efficiency literature. Bank cost 
efficiency scores are ranged between 0 and 1, hence Tobit models are more appropriate since they 
better fit models where the dependent variable is derived from a first-stage regression, as Greene 
(2005) posits. We test for both the quite life hypothesis, and the conditioning effect of financial 
freedom on the competition-efficiency relationship, which is known to be complex (Casu & 
Girardone, 2006). And in line with the bank efficiency literature, we control for banking market 
Concentration, Capitalization, Profitability (ROAA), Financial development and Economic 
development. Since the cost efficiency scores and market power indices are obtained from the same 
model, we take into account endogeneity arising from simultaneity by estimating an instrumental 
variable Tobit model, in addition to the traditional Tobit specifications.  
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
116 
 
CHAPTER FOUR 
RESULTS AND DISCUSSION  
 
4.1 Introduction 
This chapter presents results of our analysis, findings and discussions in the light of the literature. 
We begin with some descriptive statistics of key variables used in our estimations in section 4.2. 
Section 4.3 evaluates the effect of competition and economic freedom on bank risk (stability and 
bank asset quality). In section 4.4, we assess the relationship between competition, economic 
freedom and bank profitability (net interest margins and return on assets). Section 4.5 focuses on 
the competition-efficiency relationship, and the conditioning effect of economic freedom on this 
relationship. We also present the determinants of bank competition and bank cost efficiency 
estimates. Section 4.6 discusses the extent to which bank profits persist, and what drives profit 
persistence.  
 
4.2 Descriptive Statistics 
A summary of descriptive statistics for the key variables are presented in Table 4.1. The values in 
Table 4.1 represent mean country values. The financial freedom index (Financial freedom) 
constructed by the Heritage Foundation in collaboration with The Wall Street Journal measures 
the degree of independence of the financial sector from government control and interference 
(Heritage Foundation, 2015). The index has been published annually since 1995, and has been 
used in several studies, usually as a control variable or instrumental variable. Among the countries 
in Sub-Saharan Africa included in this study, Mauritius (64.3) has the highest degree of financial 
freedom while supervision of the financial sector in Ethiopia (20.0) is considered repressive.  
  
University of Ghana http://ugspace.ug.edu.gh
117 
 
Table 4.1: Summary Descriptive Statistics 
Country Financial  
freedom 
Economic 
freedom 
Z-score Asset 
Quality 
ROAA ROAE NIM 
Ethiopia 20.0 51.8 114.56 0.038 0.030 0.266 0.032 
Ghana 55.7 58.4 23.37 0.036 0.021 0.169 0.066 
Kenya 50.0 58.5 46.46 0.024 0.024 0.196 0.057 
Malawi 50.0 54.5 24.24 0.012 0.034 0.236 0.062 
Mauritius 64.3 73.3 39.08 0.023 0.011 0.106 0.026 
Mozambique 50.0 56.2 23.52 0.024 0.018 0.142 0.080 
Namibia 47.1 62.1 128.95 0.012 0.019 0.210 0.047 
South Africa 58.6 63.2 41.53 0.028 0.023 0.156 0.044 
Tanzania 52.9 57.5 64.60 0.018 0.014 0.143 0.053 
Uganda 63.3 62.7 43.47 0.013 0.022 0.108 0.073 
Zambia 50.0 57.5 16.18 0.039 0.002 0.055 0.073 
AVERAGE 51.1 59.6 47.43 0.024 0.020 0.162 0.056 
 
Country Conv. 
Lerner 
Adjusted 
Lerner 
Cost/ 
income 
Equity/ 
Assets 
Credit 
risk 
NIITI PTA OETA Financial  
Development 
Ethiopia 0.530 0.608 0.306 0.154 0.532 0.416 0.010 0.010 0.252 
Ghana 0.219 0.240 0.650 0.136 0.466 0.270 0.030 0.039 0.147 
Kenya 0.291 0.323 0.596 0.140 0.537 0.256 0.027 0.027 0.307 
Malawi 0.688 0.576 0.692 0.140 0.464 0.491 0.042 0.070 0.166 
Mauritius 0.057 N/A 0.599 0.120 0.574 0.167 0.010 0.010 0.853 
Mozambique 0.226 0.168 0.720 0.150 0.478 0.315 0.038 0.058 0.238 
Namibia 0.234 0.240 0.591 0.100 0.771 0.217 0.020 0.021 0.486 
South Africa 0.253 0.194 0.616 0.157 0.660 0.268 0.024 0.027 1.541 
Tanzania 0.263 0.214 0.657 0.119 0.461 0.313 0.027 0.033 0.159 
Uganda 0.113 0.108 0.788 0.162 0.492 0.289 0.035 0.048 0.145 
Zambia 0.058 0.080 0.895 0.105 0.490 0.376 0.050 0.050 0.138 
AVERAGE 0.267 0.275 0.646 0.135 0.539 0.307 0.029 0.036 0.403 
Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation (2015). 
They are scaled from 0 to 100 with higher values indicating greater freedom. The Z-score is defined as the return on 
assets plus the equity to total assets ratio, scaled by the standard deviation of the return on assets. Asset quality, is 
measured by the ratio of loan loss reserves to total assets. ROAA is Return on Average Assets (the ratio of net income 
to average total assets). ROAE is Return on Average Equity (ratio of net income to average total equity). Net Interest 
Margin (NIM) is defined as the ratio of the net interest income to total assets as a measure of spread. The degree of 
market power is proxied by the Lerner index or the price mark-up over marginal cost, with higher scores indicating a 
higher degree of pricing power. Two variants of the Lerner index are reported: a conventional Lerner and a funding-
adjusted Lerner. Cost to income ratio is measured by the ratio of non-interest operating expenses to operating income. 
Equity/Assets is the bank total equity to asset ratio (Capitalization). Credit risk is measured as total loans to total assets. 
NIITI represents Diversification which is measured as the ratio of non interest income to total income. PTA is defined 
by personnel to total assets while OETA is defined by other operating expenses to total assets. Financial development 
is measured as domestic credit to private sector as a percentage of GDP.  
 
University of Ghana http://ugspace.ug.edu.gh
118 
 
The average index of financial freedom (51.1) indicates that the countries in Sub-Saharan Africa 
included in this study portray significant restrictions on banking activities and the provision of 
other financial services for the study period (2006-2012). This may have adverse implications for 
the efficient operation of banks in this region.  
 
The economic freedom index (Economic freedom) measures the ability of individuals to exercise 
their fundamental right to control their own labor and property. In an economically free society, 
individuals have the freedom to work, produce, consume, and invest in any way they please. Also, 
in such societies, governments allow labor, capital, and goods to move freely, and desist from 
coercion or restriction of liberty beyond the extent needed to protect and maintain liberty itself 
(Heritage Foundation, 2015). The average level of economic freedom (59.6) signifies a reasonable 
level of liberty to own and employ capital in these countries. Again, Mauritius (73.3) and Ethiopia 
(51.8) have the highest and lowest levels of economic freedom, respectively.  
 
The results show that banks in Namibia and Ethiopia appear to be the most stable with mean 
stability scores (Z-score) of 128.95 and 114.56, respectively, while the most unstable banks are 
found in Zambia (16.18), Ghana (23.37) and Mozambique (23.52). Asset quality is poorest in 
Zambia (0.039), Ghana (0.036) and Ethiopia (0.038). This means that in general, banks in Zambia 
and Ghana face more risk than those in the other countries. Average profitability for the period are 
about 2.0% for ROAA, 16.2% for ROAE and 5.6% for NIM. Generally, the most profitable banks 
are in Ethiopia, Kenya and Malawi, while banks in Mozambique, Zambia, and Uganda earn the 
highest margins on their intermediation activities. Banks in Zambia, Mauritius and Tanzania make 
the least profit, while those in Mauritius, Ethiopia and South Africa earn the lowest margins. 
University of Ghana http://ugspace.ug.edu.gh
119 
 
Market power, represented as Conventional Lerner index and Funding-adjusted index in Table 4.1 
show a moderate level in most of the countries, averaging 0.267 and 0.275, respectively. The 
funding-adjusted Lerner is slightly higher than the conventional Lerner on average, suggesting 
little underestimation with the conventional Lerner index, but there are considerable differences in 
the two indices among the countries. Banks enjoy significantly higher monopoly power in Ethiopia 
and Malawi, with those in Mauritius, Zambia and Uganda having the least power in pricing their 
products and services. This means that banking markets in Mauritius, Zambia and Uganda are the 
most competitive, while those in Ethiopia and Malawi face little competition. 
 
Almost two-thirds of bank income is expended on cost averagely (Cost to income ratio of 64.6%). 
This gives an indication of the level of operating efficiency. Differences in operating efficiency 
are not very large, except for Ethiopian banks (30.6%) which appear to be significantly more 
efficient than the rest, with the most inefficient banks operating in Zambia, Uganda and 
Mozambique.   With an average of about 13.5% capitalization ratio, banks in Sub-Saharan Africa 
appear to be generally well-capitalized beyond the level recommended by Basel III (8%). The 
highest capitalization levels are observed in Uganda, South Africa and Ethiopia, with the least 
capitalized banks operating in Zambia and Namibia. 
 
On average, about 54% of banking assets in these Sub-Saharan African countries are devoted to 
lending activities (Credit risk), with the highest level of specialization in lending occurring in 
Namibia (77%) and South Africa (66%). On the other hand, banks in Tanzania (46.1%), Malawi 
(46.4%) and Ghana (46.6%) devote much less to lending. A significant portion of bank income is 
University of Ghana http://ugspace.ug.edu.gh
120 
 
earned from non-interest activities (NIITI) in Malawi (49.1%) and Ethiopia (41.9%). On the other 
hand, the least diversified banks are in Mauritius (16.7%) and Namibia (21.7%). 
 
Staff costs, measured as personnel costs to total assets (PTA), are lowest in Ethiopia and Mauritius, 
just about 1% of total assets, and highest in Zambia (5.0%), Malawi (4.2%) and Mozambique 
(3.8%). The average is about 2.9%. Similarly, other operating costs to total assets (OETA), are 
lowest for banks in Ethiopia and Mauritius, just about 1% of total assets, and highest in Malawi 
(7.0%), Mozambique (5.8%) and Zambia (5.0%). Domestic credit to the private sector as a 
percentage of GDP (Financial development) is significantly higher in the more developed markets 
in the region, South Africa (154.1%) and Mauritius (85.3%). Countries giving the least amount of 
credit to the private sector are Zambia (13.8%), Uganda (14.5%) and Ghana (14.7%). 
 
4.3 Competition, Economic Freedom and Bank Risk 
In this section, we analyze the effect of competition (measured as funding-adjusted and 
conventional Lerner indices of market power) and economic freedom on bank risk (stability or 
insolvency risk and bank asset quality).  
 
4.3.1 Competition, Freedom and Bank Stability 
The results on the effects of bank competition and economic freedom on bank stability are 
presented in Tables 4.2 and 4.3 (based on equation 3.6). The tables report various estimations for 
two main models. Table 4.2 shows the results using the conventional Lerner index, while Table 
4.3 show those for the funding-adjusted Lerner model. 
 
University of Ghana http://ugspace.ug.edu.gh
121 
 
Table 4.2: Effect of Competition and Freedom on Bank Stability (Conventional Lerner) 
Dependent variable: lnzscore Model 1 Model 2 Model 3 Model 4 Model 5 
lnzscoret-1 0.3707*** 
(0.0724) 
0.3853*** 
(0.0719) 
0.3956*** 
0.0755 
0.3811*** 
(0.0722) 
0.3899*** 
(0.0761) 
Lerner 2.0973*** 
(0.6644) 
2.0137*** 
(0.6580) 
2.1411*** 
0.6543 
3.6608* 
(2.0340) 
6.9220 
(4.6648) 
Lerner-squared -1.5518** 
(0.7777) 
-1.5065* 
(0.7812) 
-1.4730* 
(0.7983) 
-1.6230* 
(0.8325) 
-1.8040* 
(0.9475) 
Financial freedom  -0.0081 
(0.0083) 
 0.0022 
(0.0142) 
 
Economic freedom   0.0252* 
(0.0129) 
 0.0342** 
(0.0141) 
Conv. Lerner* Fin. Freedom    -0.0301 
(0.0335) 
 
Conv. Lerner* Econ. freedom     -0.0798 
(0.0746) 
Capitalization -0.3122 
(1.6634) 
-0.7096 
(1.7112) 
0.0807 
1.5841 
-0.5361 
(1.7124) 
0.1583 
(1.6091) 
Credit risk 0.0603 
(0.3227) 
-0.0983 
(0.3158) 
0.1381 
0.3097 
-0.1069 
(0.3212) 
0.0870 
(0.3169) 
Diversification -1.0007** 
(0.4442) 
-1.2620*** 
(0.4222) 
-0.8169* 
(0.4229) 
-1.2450*** 
(0.4198) 
-0.8094* 
(0.4197) 
Financial development 0.1754* 
(0.1035) 
0.2297* 
(0.1177) 
0.0548 
(0.1058) 
0.2109* 
(0.1153) 
0.0911 
(0.1181) 
Constant 1.8528*** 
(0.3726) 
2.4376*** 
(0.5644) 
0.1564 
(0.9173) 
1.8787** 
(0.8555) 
-0.3597 
(0.9951) 
      
Marginal effect of Lerner 
   
1.3250 1.2639 1.4081 1.2371 1.2113 
No. of observations 549 549 549 549 549 
No. of instruments  52 53 53 54 54 
F-test 8.10*** 9.00*** 8.48*** 8.46*** 7.91*** 
AR(2) 
P-value 
0.44 
(0.657) 
0.50 
(0.618) 
0.47 
(0.638) 
0.50 
(0.618) 
0.44 
(0.659) 
Hansen 
P-value 
42.88 
(0.308) 
42.56 
(0.321) 
42.5 
(0.323) 
42.22 
(0.334) 
42.14 
(0.337) 
The dependent variable bank stability is measured by the natural log of the zscore. The degree of market power is 
proxied by the Conventional Lerner Index or the price mark-up over marginal cost, with the higher scores indicating 
a higher degree of pricing power. The Marginal effect of Lerner on zscore (obtained as the partial derivatives evaluated 
at the sample means) indicates the overall impact where the quadratic term is also significant. Financial freedom and 
Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. They are scaled from 0 to 100 
with higher values, indicating greater freedom. Capitalization is the bank total equity to asset ratio. Credit risk is 
measured as total loans to total assets. Diversification, measured as the ratio of non interest income to total income, 
measures the exposure of a bank to non-interest-generating income. Financial development is measured as domestic 
credit to private sector as a percentage of GDP. All the regressions were estimated with the Windmeijer-corrected 
standard error version of the two-step system GMM model, with small-sample adjustments and orthogonal deviations. 
Robust standard errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
122 
 
Table 4.3: Effect of Competition and Freedom on Bank Stability (Adjusted Lerner) 
Dependent variable: lnzscore Model 1 Model 2 Model 3 Model 4 Model 5 
lnzscore t-1 0.4613*** 
(0.0872) 
0.4384*** 
(0.0913) 
0.4692*** 
(0.0853) 
0.4372*** 
(0.0942) 
0.4903*** 
(0.0867) 
Adjusted Lerner 1.9143*** 
(0.5221) 
1.9467*** 
(0.5296) 
1.8761*** 
(0.5176) 
1.8584 
(1.7808) 
20.9613**** 
(7.3742) 
Adjusted Lerner-squared -1.4897* 
(0.7870) 
-1.6850** 
(0.7775) 
-1.4611* 
(0.8066) 
-1.6794** 
(0.7992) 
-2.1113** 
(0.9360) 
Financial freedom  -0.0176** 
(0.0073) 
 -0.0184 
(0.0175) 
 
Economic freedom   -0.0070 
(0.0244) 
 0.1169** 
(0.0570) 
Lerner * Fin. freedom    0.0013 
(0.0322) 
 
Lerner * Econ. freedom     -0.3236*** 
(0.1239) 
Capitalization 1.6520 
(1.2174) 
1.7383 
(1.2153) 
1.5401 
(1.2545) 
1.7712 
(1.2158) 
1.8077 
(1.3621) 
Credit risk -0.2567 
(0.3070) 
-0.4504 
(0.3142) 
-0.2666 
(0.3160) 
-0.4661 
(0.3074) 
-0.2976 
(0.3186) 
Diversification -0.9286** 
(0.4013) 
-1.2954*** 
(0.4109) 
-0.9602** 
(0.4328) 
-1.2943*** 
(0.4176) 
-1.0200** 
(0.4262) 
Financial development 0.2137*** 
(0.0705) 
0.3170*** 
(0.08950 
0.2381* 
(0.1253) 
0.3227*** 
(0.1094) 
-0.0137 
(0.1768) 
Constant 1.4393*** 
(0.3804) 
2.5854*** 
(0.6161) 
1.8251 
(1.5299) 
2.6371** 
(1.0975) 
-5.3774 
(3.3637) 
Marginal effect of Lerner 1.119 1.0391 1.0891 1.0243 0.7206 
No. of observations 504 504 504 504 504 
No. of instruments  52 53 53 54 54 
F-test 14.68*** 15.43*** 14.39*** 14.85*** 15.57*** 
AR(2) 
P-value 
0.56 
(0.575) 
0.51 
(0.607) 
0.57 
(0.565) 
0.51 
(0.613) 
0.67 
(0.503) 
Hansen 
P-value 
38.74 
(0.481) 
37.92 
(0.519) 
39.8 
(0.434) 
38.41 
(0.497) 
37.99 
(0.516) 
The dependent variable bank stability is measured by the natural log of the zscore. The degree of market power is proxied by the 
Funding Adjusted Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher degree of pricing 
power. The Marginal effect of Lerner on zscore (obtained as the partial derivatives evaluated at the sample means) indicates the 
overall impact where the quadratic term is also significant. Financial freedom and Economic freedom are from the Economic 
Freedom Indicators of Heritage Foundation. They are scaled from 0 to 100 with higher values, indicating greater freedom. 
Capitalization is the bank total equity to asset ratio. Credit risk is measured as total loans to total assets. Diversification, measured 
as the ratio of non interest income to total income, measures the exposure of a bank to non-interest-generating income.  Financial 
development is measured as domestic credit to private sector as a percentage of GDP. All the regressions were estimated with the 
Windmeijer-corrected standard error version of the two-step system GMM model, with small-sample adjustments and orthogonal 
deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
123 
 
All the regressions were estimated with the Windmeijer-corrected standard error version of the 
two-step system GMM model, with small-sample adjustments and orthogonal deviations 
(Windmeijer, 2005). This dynamic panel data model takes into consideration the fact that the level 
of bank stability in one period depends to some extent on its previous level. And the significantly 
positive coefficient of the lagged dependent variable (lnzscoret-1) in all the estimations shows the 
validity of this assumption. The results of the robustness tests indicates that the model seems to fit 
the panel data reasonably well. The F-test shows overall goodness of fit, the Hansen test for the 
validity of the over-identifying restrictions in the GMM estimation is accepted for all the 
specifications, and the presence of second-order autocorrelation in the errors is also rejected by the 
test for AR (2). 
  
The results of the study show that higher bank market power (both funding-adjusted Lerner and 
conventional Lerner indices) is significantly and positively associated with greater bank stability 
(measured by the natural logarithm of z-score-lnzscore). This suggests that when banks have the 
ability to price their products in a monopolistic fashion (because of less competition), it translates 
into less likelihood of the banks becoming insolvent, and rather makes banks more sound. This is 
in line with recent studies in developing countries and emerging economies such as Amidu (2013) 
and Turk-Ariss (2010). The evidence (which confirms the first Hypothesis for our study) also 
supports the competition-fragility or competition-instability hypothesis. As postulated by Keeley 
(1990), higher market (or monopoly) power in the banking market leads to greater stability because 
the greater lending opportunities, higher profits, higher capital ratios and charter values of 
incumbent banks, put them in a better position to survive demand-side and supply-side shocks, 
which in turn provides a dis-incentive for excessive risk-taking. 
University of Ghana http://ugspace.ug.edu.gh
124 
 
However, to unravel the ambiguous relationship between competition and bank stability, some 
recent studies suggest the need to allow for a nonlinear relationship (Martinez-Miera & Repullo, 
2010). Similar to Liu et al. (2013) for European banks, we test this possibility in our model and 
find evidence that indeed, there is a non-linear relationship between competition and bank stability. 
Whereas Lerner (market power) shows a mostly positive and significant relationship with bank 
stability, the quadratic form (Lerner-squared) is consistently negatively and significantly related 
with bank stability (lnzscore). However, to determine the overall impact of market power on bank 
stability, we took partial derivatives of zscore with respect to Lerner and evaluated the resulting 
expressions at the sample means. For example, in Column (1) of Table 4.2, the marginal effect of 
1.3250 is obtained from ∂(zscore)/ ∂(Lerner) = 𝛼1 + 2𝛼2Lerner,  which is then evaluated at the 
sample means, where  𝛼1  and 𝛼2 are the estimated coefficients. This indicates an overall positive 
impact of market power on bank stability. 
 
According to Martinez-Miera and Repullo (2010) lower loan rates resulting from greater bank 
competition also reduces the interest payments from performing loans (which provide a buffer 
against loan losses) because of imperfect correlation of loan defaults.  Thus, in addition to the risk-
shifting effect in the Boyd and de Nicoló (2005) model, there is also a margin effect that goes in 
the opposite direction, so that the final effect on the risk of bank failure is in principle ambiguous. 
The risk-shifting effect tends to dominate in monopolistic markets (where concentration is high), 
whereas the margin effect dominates in competitive markets, so a non-linear relationship between 
competition and the risk of bank failure generally obtains (Martinez-Miera & Repullo, 2010).  
 
University of Ghana http://ugspace.ug.edu.gh
125 
 
The direct effect of financial freedom appears to be negative, but insignificant in the conventional 
Lerner model, showing that banks operating in environments with greater financial freedom 
generally tend to be less stable or more risky. But we are unable to obtain conclusive evidence for 
Hypothesis 2, that financial freedom has a negative effect on bank stability. Also, we did not find 
any concrete evidence of the possible conditional effect of financial freedom on the competition-
stability relationship (a rejection of Hypothesis 3, which suggests that the effect of competition on 
bank stability increases with financial freedom). This indicates that, though yet to be tested, 
concerns that ‘‘excessive’’ financial freedom may contribute to financial institutions’ propensity 
to take on greater risks, which in turn may have contributed to the recent global and European 
crises as suggested by Chortareas et al. (2013), may not be applicable in the developing country 
context. Economic freedom by itself appears not to have a clear relationship, but its interaction 
with market power (Lerner*Economic freedom) indicates a negative effect of higher market power 
on bank stability in countries with higher economic freedom, though insignificant in the 
conventional Lerner model. 
 
With respect to the control variables, we find that the relationship between bank capitalization, 
measured as total equity to total assets and stability is mostly negative in the conventional model 
and positive in the funding-adjusted model, though insignificant in both cases. A similarly 
insignificant relationship with bank stability is found for credit risk, the ratio of total loans to total 
assets, except that in this case it is mostly negative in the funding adjusted model and positive in 
the conventional Lerner model. On the other hand, diversification enters the regression with a 
strong negative and significant relationship with bank stability in all the estimations, denoting that 
highly diversified banks are more risky and unstable. This is in contrast with Kohler (2015) for 
University of Ghana http://ugspace.ug.edu.gh
126 
 
EU countries who found that banks can improve their stability by increasing their share of non-
interest income, and Amidu and Wolfe (2013) for emerging and developing countries. The results 
of the study suggest that banks tend to be more risky when they get involved in non-traditional 
business activities. Domestic credit to the private sector as a percentage of GDP (Financial 
development) has a significantly positive relationship with bank stability in most of the models, 
meaning that increases in the amount of bank credit to the private sector makes banks less risky 
and more stable.  
 
4.3.2 Competition, Freedom and Bank Asset Quality  
In view of the difficulty in assessing bank risk with a single measure, we carry out similar 
regressions on bank asset quality as an alternative to Z-score, the more popular measure of overall 
bank risk or stability. Our measure of bank asset quality is the ratio of loan loss reserves to total 
assets. Tables 4.4 and 4.5 present the results for the evaluation of bank asset quality using the 
conventional and funding-adjusted Lerner models respectively. As with the bank stability 
estimations, all the regressions were estimated with the Windmeijer-corrected standard error 
version of the two-step system GMM model, with small-sample adjustments and orthogonal 
deviations. The significantly positive coefficient of the lagged dependent variable (Asset qualityt-
1) in all the estimations shows the suitability of this dynamic panel data model. The level of quality 
of bank assets tends to persist over time. The results of the robustness tests indicates that the model 
seems to fit the panel data reasonably well. The F-test shows overall goodness of fit, the Hansen 
test for the validity of the over-identifying restrictions in the GMM estimation is accepted for all 
the specifications, and the presence of second-order autocorrelation in the errors is also rejected 
by the test for AR (2).  
University of Ghana http://ugspace.ug.edu.gh
127 
 
Table 4.4: Effects of Competition and Freedom on Asset Quality (Conventional Lerner) 
Dependent variable: 
 Asset quality 
Model 1 Model 2 Model 3 Model 4 Model 5 
Asset quality t-1 0.8956*** 
(0.0819) 
0.8884*** 
(0.0818) 
0.8953*** 
(0.0808) 
0.8881*** 
(0.0812) 
0.8895*** 
(0.0820) 
Lerner -0.0070** 
(0.0030) 
-0.0068** 
(0.0034) 
-0.0075** 
(0.0033) 
-0.0114 
(0.0135) 
-0.0246 
(0.0442) 
Lerner-squared 0.0120*** 
(0.0031) 
0.0123*** 
(0.0036) 
0.0114*** 
(0.0032) 
0.0123*** 
(0.0035) 
0.0116*** 
(0.0033) 
Financial freedom  0.0001 
(0.0001) 
 0.0000 
(0.0001) 
 
Economic freedom   -0.0001 
(0.0001) 
 -0.0001 
(0.0001) 
Conv. Lerner*Fin. freedom    0.0001 
(0.0003) 
 
Conv. Lerner* Econ. freedom     0.0003 
(0.0008) 
Capitalization 0.0012 
(0.0375) 
0.0033 
(0.0353) 
0.0025 
(0.0341) 
0.0027 
(0.0363) 
0.0019 
(0.0357) 
Credit risk 0.0151** 
(0.0068) 
0.0156** 
(0.0069) 
0.0152** 
(0.0066) 
0.0155** 
(0.0068) 
0.0151** 
(0.0063) 
Diversification -0.0063 
(0.0060) 
-0.0046 
(0.0058) 
-0.0078 
(0.0061) 
-0.0045 
(0.0059) 
-0.0076 
(0.0062) 
Financial development -0.0033** 
(0.0015) 
-0.0039** 
(0.0016) 
-0.0029* 
(0.0016) 
-0.0038** 
(0.0016) 
-0.0030* 
(0.0017) 
Constant 0.0002 
(0.0052) 
-0.0101* 
(0.0056) 
0.0067 
(0.0077) 
-0.0084 
(0.0074) 
0.0091 
(0.0092) 
Marginal effect of Lerner -0.0013 -0.0009 -0.0021 0.0520 -0.0010 
No. of observations 561 561 561 561 561 
No. of  instruments  52 53 53 54 54 
F-test 27.33*** 24.44*** 25.11*** 23.031*** 24.40*** 
AR(2) 
P-value 
-1.14 
(0.255) 
-1.15 
(0.252) 
-1.14 
(0.253) 
-1.14 
(0.256) 
-1.13 
(0.257) 
Hansen 
P-value 
47.13 
(0.174) 
47.06 
(0.176) 
48.37 
(0.144) 
46.63 
(0.187) 
48.58 
(0.140) 
The dependent variable-Asset quality, is measured by the ratio of loan loss reserves to total assets. The degree of 
market power is proxied by the Conventional Lerner Index or the price mark-up over marginal cost, with the higher 
scores indicating a higher degree of pricing power. The Marginal effect of Lerner on Asset quality (obtained as the 
partial derivatives evaluated at the sample means) indicates the overall impact where the quadratic term is also 
significant. Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage 
Foundation. They are scaled from 0 to 100 with higher values, indicating greater freedom. Capitalization is the bank 
total equity to asset ratio. Credit risk is measured as total loans to total assets. Diversification, measured as the ratio 
of non interest income to total income, measures the exposure of a bank to non-interest-generating income. Financial 
development is measured as domestic credit to private sector as a percentage of GDP. All the regressions were 
estimated with the Windmeijer-corrected standard error version of the two-step system GMM model, with small-
sample adjustments and orthogonal deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 
5% and 10% levels of significance, respectively. 
 
 
University of Ghana http://ugspace.ug.edu.gh
128 
 
Table 4.5: Effects of Competition and Freedom on Asset Quality (Adjusted Lerner) 
Dependent variable:  
Asset Quality 
Model 1 Model 2 Model 3 Model 4 Model 5 
Asset qualityt-1 0.8929*** 
(0.0619) 
0.8862*** 
(0.0598) 
0.8983*** 
(0.0553) 
0.8860*** 
(0.0599) 
0.8972*** 
(0.0544) 
Adjusted Lerner -0.0062** 
(0.0027) 
-0.0055* 
(0.0031) 
-0.0067** 
(0.0029) 
-0.0030 
(0.0207) 
-0.0518 
(0.0626) 
Adjusted Lerner-squared 0.0131*** 
(0.0037) 
0.0136*** 
(0.0042) 
0.0129*** 
(0.0036) 
0.0136*** 
(0.0042) 
0.0131*** 
(0.0028) 
Financial freedom  0.0001 
(0.0001) 
 0.0002 
(0.0002) 
 
Economic freedom   -0.0002 
(0.0003) 
 -0.0005 
(0.0005) 
Adjusted Lerner* Fin. freedom    0.0000 
(0.0004) 
 
Adjusted Lerner* Econ. freedom     0.0008 
(0.0011) 
Capitalization -0.0126 
(0.0321) 
-0.0169 
(0.0320) 
-0.0106 
(0.0324) 
-0.0174 
(0.0329) 
-0.0094 
(0.0339) 
Credit risk 0.0142** 
(0.0057) 
0.0140** 
(0.0057) 
0.0141** 
(0.0055) 
0.0140** 
(0.0057) 
0.0143** 
(0.0058) 
Diversification -0.0082 
(0.0071) 
-0.0060 
(0.0069) 
-0.0091 
(0.0073) 
-0.0060 
(0.00690 
-0.0079 
(0.0075) 
Financial development -0.0036** 
(0.0015) 
-0.0042*** 
(0.0016) 
-0.0032** 
(0.0015) 
-0.0042** 
(0.0018) 
-0.0024 
(0.0018) 
Constant 0.0030 
(0.0045) 
-0.0039 
(0.0051) 
0.0076 
(0.0155) 
-0.0050 
(0.0103) 
0.0297 
(0.0294) 
Marginal effect of Lerner 0.0005 0.0014 -0.0001 0.0039 0.0021 
 
No. of observations 516 516 516 516 516 
No. of instruments  52 53 53 54 54 
F-test 41.16*** 36.87*** 41.86*** 34.65*** 48.40*** 
AR(2) 
P-value 
-1.25 
(0.212) 
-1.25 
(0.208) 
-1.25 
(0.210) 
-1.26 
(0.206) 
-1.22 
(0.221) 
Hansen 
P-value 
44.87 
(0.239) 
43.51 
(0.285) 
44.38 
(0.255) 
43.72 
(0.278) 
41.84 
(0.349) 
The dependent variable-Asset quality, is measured by the ratio of loan loss reserves to total assets. The degree of 
market power is proxied by the Funding-adjusted Lerner Index or the price mark-up over marginal cost, with the 
higher scores indicating a higher degree of pricing power. The Marginal effect of Lerner on Asset quality (obtained as 
the partial derivatives evaluated at the sample means) indicates the overall impact where the quadratic term is also 
significant. Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage 
Foundation. They are scaled from 0 to 100 with higher values, indicating greater freedom. Capitalization is the bank 
total equity to asset ratio. Credit risk is measured as total loans to total assets. Diversification, measured as the ratio 
of non interest income to total income, measures the exposure of a bank to non-interest-generating income. Financial 
development is measured as domestic credit to private sector as a percentage of GDP. All the regressions were 
estimated with the Windmeijer-corrected standard error version of the two-step system GMM model, with small-
sample adjustments and orthogonal deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 
5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
129 
 
Generally, the results show that the coefficient of market power is negatively and significantly 
associated with the bank asset quality measure (in both funding-adjusted Lerner and conventional 
Lerner models), suggesting that as banks gain more pricing power, their asset quality improves. 
This also means that increased competition in the banking market results in greater risk-taking by 
banks. However, similar to the case of bank stability, we find evidence for a non-linear relationship 
between competition and bank risk-taking. Whereas Lerner (market power) shows a mostly 
negative and significant relationship with the bank asset quality measure, its quadratic form 
(Lerner-squared) is consistently positive and significantly related with bank asset quality. Similar 
to the bank stability analysis, we determine the overall impact of market power on bank asset 
quality, by taking the partial derivatives of Asset quality with respect to Lerner and evaluating the 
resulting expressions at the sample means. The results appear to indicate an overall negative 
relationship between market power and the asset quality measure (ratio of loan loss reserves to 
total assets), suggesting that higher market power generally improves bank asset quality. 
 
While financial freedom shows a positive but insignificant relationship in both models, economic 
freedom has a negative and insignificant relationship with the quality of bank assets. This may 
suggest that when government involvement in the banking sector is limited, banks may be tempted 
to take on greater risk, but freedom in other aspects of business improves the quality of bank loans 
somewhat. The interaction terms for financial freedom and market power and economic freedom 
and market power are positive but insignificant in both models, denoting that while greater market 
power and higher freedoms may have the propensity to lead to deteriorating bank asset quality, the 
effect is not that important. 
University of Ghana http://ugspace.ug.edu.gh
130 
 
With regards to the control variables, we find that the relationship between bank capitalization, 
measured as total equity to total assets, and bank asset quality is mostly positive in the conventional 
model and negative in the funding-adjusted model, though insignificant in both cases. This is the 
reverse of what we found with bank stability. However, here we find a statistically significant and 
positive relationship between the credit risk measure and bank asset quality in all the estimations, 
an indication of higher risk-taking and deteriorating asset quality with increased lending. Further, 
unlike the case of bank stability, diversification into non-traditional banking activities does not 
seem to be an important determinant of bank asset quality or risk-taking. The coefficient of 
diversification is negative throughout the regressions but insignificant in all cases, which shows 
that although engaging in other activities may tend to minimize bank risk-taking, its effect is not 
significant.  
 
The level of financial development appears to have a significantly positive influence on bank asset 
quality, as the coefficient of domestic credit to the private sector as a percentage of GDP shows up 
significantly negative in almost all the regressions. It appears that asset quality is likely to improve 
as banks gain more experience in the market. 
 
4.4 Competition, Economic Freedom and Bank Profitability 
This section evaluates the effect of competition and economic freedom on bank profitability. We 
discuss factors that explain net interest margins and return on assets in Sub-Saharan Africa banking 
markets, with a particular interest in the role of financial and economic freedom. As with the risk 
models, two specifications of the Lerner index are used for the estimations, the conventional 
Lerner and the funding-adjusted Lerner. 
University of Ghana http://ugspace.ug.edu.gh
131 
 
Table 4.6: Effects of Competition and Freedom on Bank NIM (Conventional Lerner) 
Dependent variable: NIM Model 1 Model 2 Model 3 Model 4 Model 5 
NIMt-1 0.5836*** 
(0.0560) 
0.5619*** 
(0.0622) 
0.5759*** 
(0.0552) 
0.5765*** 
(0.0532) 
0.5764*** 
(0.0490) 
Lerner 0.0145*** 
(0.0053) 
0.0202*** 
(0.0061) 
0.0172*** 
(0.0062) 
-0.0438* 
(0.0250) 
-0.1759*** 
(0.0612) 
Financial freedom  0.0004*** 
(0.0001) 
 0.0000 
(0.0001) 
 
Economic freedom   0.0004 
(0.0002) 
 0.0000 
(0.0002) 
Conv. Lerner* Fin. freedom    0.0012** 
(0.0005) 
 
Conv. Lerner* Econ. freedom     0.0033*** 
(0.0010) 
Concentration -0.0031* 
(0.0017) 
-0.0009 
(0.0015) 
-0.0037** 
(0.0018) 
0.0014 
(0.0017) 
-0.0011 
(0.0018) 
Cost to income -0.0028 
(0.0044) 
-0.0015 
(0.0042) 
-0.0012 
(0.0048) 
-0.0040 
(0.0039) 
-0.0034 
(0.0047) 
Bank size -0.0007 
(0.0008) 
-0.0004 
(0.0007) 
-0.0005 
(0.0007) 
-.0003 
(0.0007) 
-0.0006 
(0.0007) 
Capitalization 0.0239 
(0.0333) 
0.0172 
(0.0336) 
0.0221 
(0.0334) 
0.0123 
(0.0326) 
0.0078 
(0.0318) 
Credit risk 0.0154*** 
(0.0056) 
0.0195*** 
(0.0054) 
0.0158*** 
(0.0055) 
0.0186*** 
(0.0051) 
0.0165*** 
(0.0054) 
Diversification -0.0215*** 
(0.0070) 
-0.0197*** 
(0.0075) 
-0.0202*** 
(0.0070) 
0.0205*** 
(0.0071) 
-0.0200*** 
(0.0073) 
Financial development -0.0050* 
(0.0026) 
-0.0056** 
(0.0026) 
-0.0042 
(0.0028) 
-0.0061** 
(0.0025) 
-0.0048 
(0.0033) 
Economic development -0.0027** 
(0.0013) 
-0.0053*** 
(0.0014) 
-0.0043** 
(0.0018) 
-0.0049*** 
(0.0013) 
-0.0054*** 
(0.0019) 
Constant 0.0652*** 
(0.0194) 
0.0397** 
(0.0187) 
0.0548*** 
(0.0198) 
0.0437** 
(0.0187) 
0.0681*** 
(0.0211) 
No. of observations 561 561 561 561 561 
No. of instruments 54 55 55 56 56 
F-test 42.97*** 62.99*** 41.62*** 60.31*** 60.26*** 
AR(2) 
P-value 
1.21 
(0.225) 
1.27 
(0.203) 
1.15 
(0.251) 
1.25 
(0.212) 
1.11 
(0.269) 
Hansen 
P-value 
40.23 
(0.372) 
40.32 
(0.368) 
39.95 
(0.384) 
41.46 
(0.322) 
38.87 
(0.430) 
The dependent variable Net Interest Margin (NIM) is defined as the ratio of the net interest income to total assets as a measure of spread. The 
degree of market power is proxied by the Conventional Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a 
higher degree of pricing power.  Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. 
They are scaled from 0 to 100 with higher values, indicating greater freedom. Concentration is measured by natural logarithm of HHI, measured 
as the sum of the squares of the market share of assets of each bank. Cost to income ratio is measured by the ratio of non-interest operating expenses 
to operating income. Bank size is the natural logarithm of total assets. Capitalization is the bank total equity to asset ratio. Credit risk is total loans 
to total assets. Diversification, is measured as the ratio of non interest income to total income.  Financial development is measured as domestic 
credit to private sector as a percentage of GDP. Economic development is the natural logarithm of GDP per capita. All the regressions were estimated 
with the Windmeijer-corrected standard error version of the two-step system GMM model, with small-sample adjustments and orthogonal 
deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
132 
 
Table 4.7: Effects of Competition and Freedom on Bank NIM (Adjusted Lerner) 
Dependent variable: NIM Model 1 Model 2 Model 3 Model 4 Model 5 
NIMt-1 0.5974*** 
(0.0520) 
0.5876*** 
(0.0530) 
0.5940*** 
(0.0528) 
0.5901*** 
(0.0512) 
0.5832*** 
(0.0511) 
Funding-Adjusted Lerner 0.0113* 
(0.0066) 
0.0136** 
(0.0060) 
0.0129** 
(0.0058) 
-0.0077 
(0.0251) 
-0.0724 
(0.0825) 
Financial freedom  0.0005*** 
(0.0001) 
 0.0003 
(0.0002) 
 
Economic freedom   0.0013*** 
(0.0004) 
 0.0008 
(0.0007) 
Lerner * Fin. freedom     0.0004 
(0.0005) 
 
Lerner * Econ. freedom     0.0015 
(0.0014) 
Concentration -0.0013 
(0.0019) 
0.0025 
(0.0019) 
-0.0003 
(0.0020) 
0.0027 
(0.0020) 
0.0002 
(0.0021) 
Cost to income -0.0039 
(0.0057) 
-0.0062 
(0.0046) 
-0.0054 
(0.0048) 
-0.0066 
(0.0048) 
-0.0053 
(0.0053) 
Bank Size -0.0013* 
(0.0008) 
-0.0013* 
(0.0007) 
-0.0014* 
(0.0007) 
-0.0013* 
(0.0007) 
-0.0016** 
(0.0007) 
Capitalization 0.0152 
(0.0310) 
0.0028 
(0.0337) 
0.0042 
(0.0327) 
0.0037 
(0.0326) 
0.0015 
(0.0321) 
Credit risk 0.0146** 
(0.0057) 
0.0192*** 
(0.0052) 
0.0165*** 
(0.0059) 
0.0187*** 
(0.0052) 
0.0173*** 
(0.0060) 
Diversification -0.0238*** 
(0.0078) 
-0.0215*** 
(0.0078) 
-0.0205** 
(0.0081) 
-0.0210*** 
(0.0078) 
-0.0197** 
(0.0083) 
Financial development -0.0071** 
(0.0034) 
-0.0090*** 
(0.0031) 
-0.0076* 
(0.0039) 
-0.0080** 
(0.0035) 
-0.0066 
(0.0041) 
Economic development -0.0007 
(0.0018) 
-0.0023 
(0.0014) 
-0.0034 
(0.0021) 
-0.0026 
(0.0016) 
-0.0037* 
(0.0021) 
Constant 0.0433 
(0.0223)* 
0.0017 
(0.0204) 
-0.0197 
(0.0292) 
0.0102 
(0.0259) 
0.0087 
(0.0446) 
No. of observations 516 516 516 516 516 
No. of instruments  54 55 55 56 56 
F-test 45.84*** 63.89*** 58.00*** 61.09*** 61.47*** 
AR(2) 
P-value 
1.27 
(0.204) 
1.41 
(0.159) 
1.10 
(0.271) 
1.39 
(0.165) 
1.1 
(0.269) 
Hansen 
P-value 
40.41 
(0.364) 
43.73 
(0.241) 
43.36 
(0.253) 
44.10 
(0.229) 
42.31 
(0.290) 
The dependent variable Net Interest Margin (NIM) is defined as the ratio of the net interest income to total assets as a measure of 
spread. The degree of market power is proxied by the Funding-adjusted Lerner Index or the price mark-up over marginal cost, with 
the higher scores indicating a higher degree of pricing power.  Financial freedom and Economic freedom are from the Economic 
Freedom Indicators of Heritage Foundation. They are scaled from 0 to 100 with higher values, indicating greater freedom. 
Concentration is measured by natural logarithm of HHI, measured as the sum of the squares of the market share of assets of each 
bank. Cost to income ratio is measured by the ratio of non-interest operating expenses to operating income. Bank size is the natural 
logarithm of total assets. Capitalization is the bank total equity to asset ratio. Credit risk is total loans to total assets. Diversification, 
is measured as the ratio of non interest income to total income.  Financial development is measured as domestic credit to private 
sector as a percentage of GDP. Economic development is the natural logarithm of GDP per capita. All the regressions were estimated 
with the Windmeijer-corrected standard error version of the two-step system GMM model, with small-sample adjustments and 
orthogonal deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, 
respectively. 
 
University of Ghana http://ugspace.ug.edu.gh
133 
 
4.4.1 Competition, Freedom and Bank Net Interest Margins  
Tables 4.6 and 4.7 show the effects of competition and freedom on bank Net Interest Margins 
(NIM) (based on equation 3.7). Table 4.6 presents the results of the conventional Lerner model 
while Table 4.7 provides that of the funding-adjusted Lerner. All the regressions were estimated 
with the Windmeijer-corrected standard error version of the two-step system GMM model, with 
small-sample adjustments and orthogonal deviations. This dynamic panel data model accounts for 
the role of the previous values of the dependent variable or bank profit persistence, as reflected in 
the significantly positive coefficient of the lagged dependent variable (NIMt-1) in all the 
estimations. The results of the robustness tests indicates that the model seems to fit the panel data 
reasonably well. The F-test shows overall goodness of fit, the Hansen test for the validity of the 
over-identifying restrictions in the GMM estimation is accepted for all the specifications, and the 
presence of second-order autocorrelation in the errors is also rejected by the test for AR (2). 
 
The positive and significant coefficients of Lerner (market power) in most of the estimations (both 
conventional Lerner and funding-adjusted Lerner models), reveal that market power in the banking 
market is significantly and positively related to NIM. This indicates that banks earn higher margins 
on loans, as banks gain more market power or the market becomes less competitive. This is 
consistent with what Amidu and Wolfe (2013) reported for emerging and developing countries, as 
well as that of Hawtrey and Liang (2008) for OECD countries. The result also confirms our 
expectation that banks with market power can charge a higher loan rate and offer a lower deposit 
rate (Hawtrey and Liang, 2008). 
 
University of Ghana http://ugspace.ug.edu.gh
134 
 
Financial freedom has a positive and significant coefficient in both models, an indication that less 
intervention by government in the banking sector boosts bank margins. Similarly, economic 
freedom is positively related to NIM (but insignificant in the conventional Lerner model), 
suggesting that higher levels of freedom to carry out general business activities may enhance 
banks’ ability to generate higher margins. This confirms the findings of Sufian and Habibullah 
(2010) who provide new empirical evidence on the positive impact of economic freedom on banks’ 
performance. There is some indication that the impact of competition on bank margins is sensitive 
to the level of financial and economic freedom prevailing in an economy. This is seen in the 
positive coefficients of the interaction terms (Lerner*Financial freedom and Lerner*Economic 
freedom) in both models, though insignificant in the funding-adjusted model. We also find that the 
sign on Lerner changes to negative when interacted with the freedom variables. It appears that the 
positive effect of market power on bank margins is stronger for banks operating in countries with 
higher levels of freedom. In other words, as competition intensifies, margins of banks in freer 
countries are likely to reduce faster than those in areas with more restrictions. 
 
An analysis of the control variables show that, while not very important, banking market 
concentration is negatively related to bank margins, meaning that banks in highly concentrated 
markets may have the tendency to earn lower margins. Our measure of bank operating efficiency, 
the ratio of non-interest operating cost to operating income (Cost to income) enters the regression 
with a negative but insignificant coefficient in all the regressions. The higher the level of this ratio, 
the less efficient a bank is. The results show that banks with higher margins are usually those who 
have higher levels of operating efficiency.  
 
University of Ghana http://ugspace.ug.edu.gh
135 
 
Similarly, bank size is negatively related to NIM in all the estimations, indicating that as banks get 
larger their margins tend to dwindle. This is possibly explained by the higher cost involved in 
raising more funds. However, this is significant only in the funding-adjusted Lerner model. The 
coefficient of bank capitalization is positive but insignificant in all the estimations, a suggestion 
that highly capitalized banks may tend to have higher margins, likely because of their ability to 
engage in more intermediation activities on the back of their higher level of available funds.  
 
Credit risk (ratio of loans to total assets) appears to be an important determinant of bank margins. 
Higher credit risk associates with higher bank margins. The coefficient is significantly positive in 
all the estimations. Thus, as banks use a higher proportion of their assets to lend, they thereby 
increase their level of margins relative to assets, ostensibly as a result of additional risk premium 
charges. This is consistent with Naceur and Omran (2011), Hossain (2012), Chortareas et al. 
(2012), Amidu and Wolfe (2013) and Ahokpossi (2013). Our measure of diversification, non-
interest income to total income (Diversification), shows up negative and significant in all the 
estimations, implying that banks that engage in other activities earn lower margins, as they lose 
focus on their core activity. It is also possible that diversified banks may demand lower spreads 
for loans in order to gain higher income from non-interest activities, because they consider the two 
sources of income as substitutes for each other (Kalluci, 2010). 
 
Both domestic credit to the private sector as a percentage of GDP (Financial development), and 
per capita GDP (Economic development) have a negative and significant coefficient in most of the 
regressions, denoting that banks operating in countries with higher levels of financial and 
economic development earn reduced margins.  
University of Ghana http://ugspace.ug.edu.gh
136 
 
4.4.2 Competition, Freedom and Bank Return on Assets  
In this section, we present the results for the effects of competition and freedom on an alternative 
measure of bank profitability-Return on Average Assets (ROAA). Again, as shown by the highly 
significant coefficients of the lagged dependent variable (ROAAt-1) in Tables 4.8 and 4.9, bank 
profits tend to persist from year to year. And as with the NIM models, all the regressions were 
estimated with the Windmeijer-corrected standard error version of the two-step system GMM 
model, with small-sample adjustments and orthogonal deviations. The results of the robustness 
tests indicates that the model seems to fit the panel data reasonably well. The F-test shows overall 
goodness of fit, the Hansen test for the validity of the over-identifying restrictions in the GMM 
estimation is accepted for all the specifications, and the presence of second-order autocorrelation 
in the errors is also rejected by the test for AR (2). 
 
The coefficient of market power (Lerner) is generally positive, but significant only in the 
conventional Lerner model, suggesting that in less competitive markets (higher bank market 
power) banks generally earn higher returns on their assets. This means that as competition 
increases, bank returns on assets could be expected to go down (Hypothesis 4). Similar to the 
results for NIM, financial freedom has a positive and significant coefficient in both the 
conventional and funding-adjusted Lerner models, meaning that banks operating in countries 
where government intervention in the banking sector is minimal earn higher returns on their assets. 
Thus as expected for Hypothesis 5, financial freedom has a positive effect on bank profitability. 
Again, economic freedom is positively related to ROAA (but insignificant in the conventional 
Lerner model), denoting that higher levels of freedom to carry out general business activities may 
translate into higher bank profits. 
University of Ghana http://ugspace.ug.edu.gh
137 
 
Table 4.8: Effects of Competition and Freedom on Bank ROAA (Conventional Lerner) 
The dependent variable ROAA is the ratio of net income to average total assets. The degree of market power is proxied by the 
Conventional Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher degree of pricing 
power.  Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. They are 
scaled from 0 to 100 with higher values, indicating greater freedom. Concentration is measured by natural logarithm of HHI, 
measured as the sum of the squares of the market share of assets of each bank.  Cost to income ratio is measured by the ratio of 
non-interest operating expenses to operating income. Bank size is the natural logarithm of total assets. Capitalization is the bank 
total equity to asset ratio. Credit risk is total loans to total assets. Diversification, is measured as the ratio of non interest income to 
total income.  Financial development is measured as domestic credit to private sector as a percentage of GDP. Economic 
development is the natural logarithm of GDP per capita. All the regressions were estimated with the Windmeijer-corrected standard 
error version of the two-step system GMM model, with small-sample adjustments and orthogonal deviations. Robust standard 
errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
Dependent variable: ROAA Model 1 Model 2 Model 3 Model 4 Model 5 
ROAAt-1 0.1796*** 
(0.0561) 
0.1826*** 
(0.0572) 
0.1806*** 
(0.0586) 
0.1700*** 
(0.0548) 
0.1778*** 
(0.0534) 
Lerner 0.0341*** 
(0.0072) 
0.0354*** 
(0.0075) 
0.0353*** 
(0.0082) 
-0.0315 
(0.0232) 
-0.1881*** 
(0.0572) 
Financial freedom  0.0002** 
(0.0001) 
 -0.0001 
(0.0001) 
 
Economic freedom   0.0003 
(0.0002) 
 -0.0001 
(0.0002) 
Conv. Lerner* Fin. freedom     0.0012*** 
(0.0004) 
 
Conv. Lerner* Econ. freedom     0.0038*** 
(0.0010) 
Concentration -0.0049*** 
(0.0018) 
-0.0042** 
(0.0017) 
-0.0053*** 
(0.0019) 
-0.0015 
(0.0017) 
-0.0020 
(0.0019) 
Cost to income -0.0389*** 
(0.0074) 
-0.0392*** 
(0.0076) 
-0.0384*** 
(0.0082) 
-0.0421*** 
(0.0078) 
-0.0407*** 
(0.0079) 
Size -0.0002 
(0.0007) 
-0.0002 
(0.0006) 
-0.0002 
(0.0007) 
0.0000 
(0.0007) 
-0.0001 
(0.0007) 
Capitalization 0.0660** 
(0.0332) 
0.0621* 
(0.0315) 
0.0665** 
(0.0304) 
0.0629* 
(0.0343) 
0.0557 
(0.0343) 
Credit risk -0.0061 
(0.0065) 
-0.0033 
(0.0062) 
-0.0036 
(0.0060) 
-0.0036 
(0.0062) 
-0.0010 
(0.0061) 
Diversification 0.0267*** 
(0.0088) 
0.0289*** 
(0.0088) 
0.0289*** 
(0.0084) 
0.0281*** 
(0.0081) 
0.0307*** 
(0.0077) 
Financial development -0.0032 
(0.0025) 
-0.0036 
(0.0025) 
-0.0028 
(0.0025) 
-0.0043* 
(0.0024) 
-0.0045 
(0.0027) 
Economic development 0.0031** 
(0.0013) 
0.0021 
(0.0015) 
0.0020 
(0.0018) 
0.0019 
(0.0012) 
0.0006 
(0.0016) 
Constant 0.0390** 
(0.0175) 
0.0270 
(0.0162) 
0.0283 
(0.0180) 
0.0301* 
(0.0169) 
0.0367* 
(0.0200) 
No. of observations 561 561 561 561 561 
No. of instruments  54 55 55 56 56 
F-test 46.57*** 54.97*** 46.12*** 47.33*** 50.90*** 
AR(2) 
P-value 
-0.93 
(0.355) 
-0.87 
(0.386) 
-0.92 
(0.360) 
-0.84 
(0.400) 
-0.86 
(0.387) 
Hansen 
P-value 
37.06 
(0.513) 
36.25 
(0.550) 
36.48 
(0.540) 
34.21 
(0.646) 
33.13 
(0.694) 
University of Ghana http://ugspace.ug.edu.gh
138 
 
Table 4.9: Effects of Competition and Freedom on Bank ROAA (Adjusted Lerner) 
Dependent variable: ROAA  Model 1 Model 2 Model 3 Model 4 Model 5 
ROAAt-1 0.1547*** 
(0.0538) 
0.1621*** 
(0.0529) 
0.1658*** 
(0.0529) 
0.1501*** 
(0.0564) 
0.1528*** 
(0.0571) 
Funding-Adjusted Lerner 0.0218 
(0.0137) 
0.0217 
(0.0140) 
0.0228 
(0.0138) 
-0.0462 
(0.0284) 
-0.2176** 
(0.0996) 
Financial freedom  0.0003*** 
(0.0001) 
 -0.0003 
(0.0002) 
 
Economic freedom   0.0009** 
(0.0004) 
 -0.0005 
(0.0008) 
Lerner * Fin. freedom     0.0013** 
(0.0005) 
 
Lerner * Econ. freedom     0.0041** 
(0.0018) 
Concentration -0.0042** 
(0.0020) 
-0.0024 
(0.0019) 
-0.0041** 
(0.0019) 
-0.0012 
(0.0019) 
-0.0022 
(0.0020) 
Cost to income -0.0566*** 
(0.0124) 
-0.0577*** 
(0.0125) 
-0.0560*** 
(0.0120) 
-0.0596*** 
(0.0138) 
-0.0552*** 
(0.0124) 
Size -0.0017* 
(0.0009) 
-0.0016* 
(0.0009) 
-0.0016* 
(0.0008) 
-0.0016 
(0.0010) 
-0.0019* 
(0.0011) 
Capitalization 0.0488 
(0.0310) 
0.0441 
(0.0306) 
0.0486 
(0.0300) 
0.0483 
(0.0316) 
0.0439 
(0.0310) 
Credit risk -0.0029 
(0.0069) 
-0.0011 
(0.0065) 
-0.0023 
(0.0065) 
-0.0014 
(0.0063) 
-0.0004 
(0.0067) 
Diversification 0.0260** 
(0.0099) 
0.0285*** 
(0.0096) 
0.0277*** 
(0.0093) 
0.0293*** 
(0.0100) 
0.0310*** 
(0.0095) 
Financial development -0.0021 
(0.0027) 
-0.0035 
(0.0027) 
-0.0022 
(0.0026) 
-0.0014 
(0.0032) 
-0.0001 
(0.0034) 
Economic development 0.0040*** 
(0.0012) 
0.0033*** 
(0.0012) 
0.0022 
(0.0014) 
0.0025** 
(0.0012) 
0.0018 
(0.0013) 
Constant 0.0521** 
(0.0205) 
0.0279 
(0.0205) 
0.0105 
(0.0219) 
0.0530** 
(0.0238) 
0.0815** 
(0.0403) 
No. of observations 516 516 516 516 516 
No. of instruments  54 55 55 56 56 
F-test 54.82*** 66.54*** 63.98*** 43.08*** 42.63*** 
AR(2) 
P-value 
-0.77 
(0.439) 
-0.69 
(0.491) 
-0.75 
(0.456) 
-0.67 
(0.506) 
-0.66 
(0.507) 
Hansen 
P-value 
39.95 
(0.384) 
37.06 
(0.513) 
37.51 
(0.492) 
39.62 
(0.398) 
38.86 
(0.431) 
The dependent variable ROAA is the ratio of net income to average total assets. The degree of market power is proxied by the 
Funding-adjusted Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher degree of pricing 
power.  Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. They are 
scaled from 0 to 100 with higher values, indicating greater freedom. Concentration is measured by natural logarithm of HHI, 
measured as the sum of the squares of the market share of assets of each bank. Cost to income ratio is measured by the ratio of non-
interest operating expenses to operating income. Bank size is the natural logarithm of total assets. Capitalization is the bank total 
equity to asset ratio. Credit risk is total loans to total assets. Diversification, is measured as the ratio of non interest income to total 
income.  Financial development is measured as domestic credit to private sector as a percentage of GDP. Economic development 
is the natural logarithm of GDP per capita. All the regressions were estimated with the Windmeijer-corrected standard error version 
of the two-step system GMM model, with small-sample adjustments and orthogonal deviations. Robust standard errors are in 
parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
 
University of Ghana http://ugspace.ug.edu.gh
139 
 
The outcome for ROAA when Lerner is interacted with the freedom variables is similar to what 
we found with NIM, confirming our submission that the impact of competition on bank profits is 
sensitive to the level of financial and economic freedom prevailing in an economy. The coefficients 
of the interaction terms (Lerner*Financial freedom and Lerner*Economic freedom) are 
significantly positive in both models. Again, we find that the sign on Lerner changes to negative 
when interacted with the freedom variables, thus affirming that financial and economic freedom 
have a conditioning effect on the impact of  competition on bank profits. This suggests that banks 
with higher market power operating in countries with higher freedom for banking activities are 
more profitable than their counterparts in countries with greater restrictions on banking activities. 
Again, this is in harmony with Hypothesis 6, that the effect of competition on bank profitability 
increases with financial freedom. 
 
The results show that, in highly concentrated markets, bank profits are likely to be lower as 
indicated by the negative coefficient for Concentration, except that this is not significant is some 
cases. This is inconsistent with the Structure-Conduct-Hypothesis which suggests that 
concentration enables banks to collude and earn higher profits. Unlike the case of NIM (which 
showed an insignificant relationship), bank operating efficiency (Cost to income) shows a 
significantly negative relationship with ROAA in all the estimations, signifying that, as expected, 
banks that are able to control costs more effectively earn higher profits, and those that are 
inefficient have reduced profits. This result is also in line with previous empirical literature which 
suggests that efficiency is a more important determinant of profitability than concentration 
(Berger, 1995). There is some indication that increasing bank size is associated with lower profits 
but this is seen only in the funding-adjusted Lerner model, as was the case with NIM. In addition, 
University of Ghana http://ugspace.ug.edu.gh
140 
 
as in the NIM estimations, bank capitalization is positively related to profits (ROAA), and 
significantly so in the conventional Lerner model, an indication that highly capitalized banks have 
the tendency to earn higher returns. Similarly, higher credit risk seems to associate with lower 
bank profits, as seen in the negative coefficients for credit risk in all the regressions, except that 
unlike that of NIM, the coefficients are not statistically significant. 
 
Contrary to the NIM results (which were negative), the diversification variable has a significantly 
positive coefficient. This means that diversification is an important determinant of bank return on 
assets, and that banks that engage in other activities outside their core business earn significantly 
higher returns than those not engaging in similar activities. As suggested by Goddard et al. (2013), 
this might be attributable to synergies between core and related activities, thus enabling diversified 
banks to gain and maintain a competitive advantage over their less diversified counterparts. 
Another plausible explanation for the positive effect of diversification on bank profitability may 
be the limited losses associated with non-traditional banking activities, compared to the generally 
huge losses on loans in some of these economies. The finding is consistent with Chiorazzo et al. 
(2008) and Elsas et al. (2010) but contrary to Stiroh and Rumble (2006). 
 
The level of financial development is largely insignificant in determining bank return on assets, 
contrary to what we found for NIM, but the negative coefficients show that higher levels of 
domestic credit to the private sector have the tendency to dampen bank returns. Also, contrary to 
NIM, bank returns seems to have a positive relationship with the level of economic development. 
Increase in economic development may enhance the ability of borrowers to service loans, resulting 
in reduced loan losses and hence greater bank profitability. 
University of Ghana http://ugspace.ug.edu.gh
141 
 
4.5 Competition, Economic Freedom and Bank Efficiency 
In this section, we evaluate the relationship between competition and bank cost efficiency. We 
begin with an investigation of the factors that drive bank market power. This is followed by an 
assessment of the relationship between market power and bank cost efficiency, with particular 
interest in how financial and economic freedoms affect this relationship. 
 
4.5.1 Sources of Bank Market Power in Sub-Saharan Africa 
This section identifies factors that explain bank market power in Sub-Saharan Africa. Separate 
models were estimated for the funding-adjusted Lerner index and the conventional Lerner index, 
which are direct measures of market power (or competition) at the bank level. Again, the 
regressions were estimated with the Windmeijer-corrected standard error version of the two-step 
system GMM model, with small-sample adjustments and orthogonal deviations, and the 
statistically significant coefficients of the lagged dependent variable testifies to the suitability of 
this approach. The results of the robustness tests indicates that the model seems to fit the panel 
data reasonably well. The F-test shows overall goodness of fit, the Hansen test for the validity of 
the over-identifying restrictions in the GMM estimation is accepted for all the specifications, and 
the presence of second-order autocorrelation in the errors is also rejected by the test for AR (2). 
 
The regression results presented in Table 4.10 (obtained from estimation of equation 3.8) show 
that financial freedom has a negative relationship with market power, but this is not significant in 
the funding-adjusted Lerner model.  
 
 
University of Ghana http://ugspace.ug.edu.gh
142 
 
Table 4.10: Sources of Bank Market Power 
Dependent variable:  
Lerner 
Conventional Lerner Funding-Adjusted 
Lerner 
Lernert-1 0.2460*** 
(0.0775) 
0.2437*** 
(0.0769) 
0.2407*** 
(0.0599) 
0.2344*** 
(0.0640) 
Financial freedom -0.0023** 
(0.0010) 
 -0.0022 
(0.0013) 
 
Economic freedom  -0.0088** 
(0.0040) 
 -0.0100 
(0.0087) 
Capitalization 0.8848*** 
(0.2940) 
0.8718** 
(0.3344) 
0.8775** 
(0.3571) 
1.0036** 
(0.4043) 
Diversification 0.2830*** 
(0.0699) 
0.2665*** 
(0.0704) 
0.3871*** 
(0.1055) 
0.3679*** 
(0.1056) 
Bank Size 0.0087 
(0.0145) 
0.0038 
(0.0134) 
0.0550*** 
(0.0168) 
0.0581*** 
(0.0173) 
Financial development 0.0733 
(0.0534) 
0.0642 
(0.0461) 
-0.0314 
(0.0453) 
-0.0311 
(0.0475) 
Market share 0.0052*** 
(0.0019) 
0.0060*** 
(0.0014) 
0.0008 
(0.0010) 
0.0011 
(0.0013) 
Economic development -0.0591*** 
(0.0221) 
-0.0320 
(0.0284) 
-0.0542** 
(0.0217) 
-0.0411 
(0.0316) 
Constant 0.4038** 
(0.1733) 
0.6498*** 
(0.2004) 
0.1441 
(0.1703) 
0.5300 
(0.3891) 
No. of observations 561 561 516 516 
No. of instruments  52 52 52 52 
F-test 14.34*** 12.72*** 12.442*** 10.12*** 
AR(2) 
P-value 
-1.28 
(0.199) 
-1.29 
 (0.197) 
-1.25 
(0.211) 
-1.17 
(0.241) 
Hansen 
P-value 
43.31 
(0.255) 
39.04 
(0.423) 
41.44 
(0.323) 
40.41 
(0.365) 
The dependent variable is the Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher 
degree of pricing power. The degree of market power is proxied by the Lerner Index. Financial freedom and Economic freedom 
are from the Economic Freedom Indicators of Heritage Foundation. They are scaled from 0 to 100 with higher values indicating 
greater freedom. Capitalization is the bank total equity to total asset ratio. Diversification, measured as the ratio of non interest 
income to total income, measures the exposure of a bank to non-interest-generating income. Bank size is measured as the natural 
logarithm of total assets. Financial development is measured as domestic credit to private sector as a percentage of GDP. Economic 
development is measured as measured as the natural logarithm of GDP per capita. All the regressions were estimated with the 
Windmeijer-corrected standard error version of the two-step system GMM model, with small-sample adjustments and orthogonal 
deviations. Robust standard errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
 
This suggests that banks that operate in countries with more financial freedom (less government 
control over financial institutions) tend to have lower pricing power, likely as a result of fewer 
restrictions in establishing new banks even by foreign investors, thus leading to greater 
competition. Economic freedom also shows a negative coefficient, signifying a positive effect of 
University of Ghana http://ugspace.ug.edu.gh
143 
 
higher economic freedom on competitive conditions in the banking sector, except that the effect is 
not significant in the funding-adjusted Lerner model. This compares with Mirzaei and Moore 
(2014) who report that banks located in countries with good-quality institutional development face 
greater competition for emerging and developing economies, whereas inter-industry competition 
from insurance industries, together with financial freedom, seem to be the main drivers in 
increasing competition amongst developed economies.  
 
We find a highly significant positive coefficient for equity to total assets (Capitalization) for both 
the funding-adjusted Lerner and conventional Lerner, suggesting that banks with higher levels of 
capital exercise greater market power. A similar result is found for bank size but this may not be 
very important as indicated by the insignificant coefficient in the conventional Lerner model. At 
any rate, as noted by Delis (2012), well-capitalized and larger banks may be able to set higher 
margins or access funds from cheaper sources due to scale economies, informational asymmetries, 
and moral hazard issues. Aboagye et al. (2008b) also found that increases in bank size translate 
into greater market power for Ghana. 
 
Diversification shows up as one of the most important sources of bank market power. The 
coefficient is positive and highly significant. Clearly, increases in income from other activities put 
banks in a good position to exercise monopoly power in pricing their services, thereby reducing 
the level of competition in the market. It may be that well-diversified banks with high income from 
non-traditional banking activities are able to limit the pressure to lend at lower rates or pay higher 
rates on deposits, thereby wielding higher power in the market. Market share of banks seems to 
have a similar effect except that in this case, the coefficient of the funding-adjusted Lerner model 
University of Ghana http://ugspace.ug.edu.gh
144 
 
is rather insignificant. While the effect of per capita GDP (Economic development) is negative and 
significant in the models including financial freedom (signifying perhaps a latent effect of this 
variable on its impact), domestic credit to the private sector as a percentage of GDP (financial 
development) is rather indeterminate, with opposite signs in the two models. This suggests that 
increases in economic development diminishes the pricing power of banks as consumers also 
become more powerful. 
 
4.5.2 Competition, Freedom and Bank Efficiency 
We begin this section with a brief review of the cost efficiency estimates in Table 4.11 (based on 
equation 3.5). This is followed by an analysis of the conditioning effect of financial and economic 
freedom on the competition-efficiency relationship. The cost efficiency estimates are the observed 
level of costs per bank relative to an efficient frontier or what an efficient bank’s costs would be. 
The estimates were derived using the stochastic frontier approach. The mean efficiency estimates 
per country for the period 2006-2012 were obtained from separate frontiers for each country, and 
thus not directly comparable across countries. Even so, higher efficiency scores indicate better cost 
efficiency at the country level. 
 
The results show significant differences in cost efficiency among banks in Malawi, South Africa 
and Mauritius. On the other hand, most banks in Uganda, Ethiopia and Namibia are virtually at 
par in terms of cost efficiency. Generally, these cost efficiency estimates are in line with what has 
previously been reported in the literature, and reveal potential for considerable cost savings for 
some banks in Sub-Saharan Africa, especially among those operating in Malawi, South Africa, 
Mauritius and Ghana. 
University of Ghana http://ugspace.ug.edu.gh
145 
 
Table 4.11: Bank Cost Efficiency Estimates  
Country     Mean Std. 
Dev     
      Min      Max 
Ethiopia 0.98 0.02 0.92 1.00 
Ghana 0.86 0.07 0.66 0.98 
Kenya 0.96 0.01 0.93 0.99 
Malawi  0.73 0.24 0.24 1.00 
Mauritius 0.81 0.13 0.52 0.99 
Mozambique 0.92 0.04 0.84 0.98 
Namibia 0.97 0.01 0.95 0.99 
South Africa 0.59 0.15 0.38 0.96 
Tanzania 0.91 0.06 0.7 0.98 
Uganda 1.00 0.01 0.98 1.00 
Zambia 0.95 0.05 0.79 0.99 
 
Tables 4.12 and 4.13 present the Tobit regression results of the impact of competition and 
financial/economic freedom on bank cost efficiency using the funding-adjusted Lerner and 
conventional Lerner indices, respectively (based on equation 3.9). Bank cost efficiency scores are 
ranged between 0 and 1, hence Tobit models are more appropriate in this case since they better fit 
models where the dependent variable is derived from a first-stage regression (Greene, 2005). 
However, to take into account possible endogeneity of the market power measure and bank 
efficiency, we also use instrumental variable Tobit regression to check the robustness of our 
results. The results are reported in Tables 4.14 and 4.15. Given the general similarity in the results 
for the traditional Tobit and instrumental variable Tobit regressions, we present a common analysis 
below. We find evidence that rejects the quiet life hypothesis. The independent effect of bank 
market power on cost efficiency is consistently positive and significant in all the estimations. This 
means that banks with higher market power are the most cost efficient, contrary to the quite life 
hypothesis that banks become inefficient as they gain more market power and settle for a quiet life 
with no incentive to reduce costs. This shows that competition has a negative relationship with 
bank efficiency, as predicted by our Hypothesis 7. 
University of Ghana http://ugspace.ug.edu.gh
146 
 
Table 4.12: Effects of Competition and Freedom on Cost Efficiency (Conventional Lerner) 
Tobit regression 
Dependent variable:  
Cost Efficiency 
Model 1 Model 2 Model 3 Model 4 Model 5 
Lerner 0.0676* 
(0.0390) 
0.0183 
(0.0337) 
-0.0536* 
(0.0303) 
-0.8211*** 
(0.2016) 
-2.9205*** 
(0.8358) 
Financial freedom  -0.0067*** 
(0.0008) 
 -0.0120*** 
(0.0018) 
 
Economic freedom   -0.0298*** 
(0.0020) 
 -0.0363*** 
(0.0025) 
Lerner*Fin. freedom    0.0152*** 
(0.0041) 
 
Lerner*Econ. freedom     0.0478*** 
(0.0149) 
Concentration -0.1542*** 
(0.0155) 
-0.1754*** 
(0.0145) 
-0.1088*** 
(0.0128) 
-0.1493*** 
(0.0145) 
-0.0633*** 
(0.0147) 
Capitalization 0.1865** 
(0.0846) 
0.2233*** 
(0.0845) 
0.2631*** 
(0.0813) 
0.2184** 
(0.0858) 
0.2079** 
(0.0826) 
Profitability -0.4499 
(0.3577) 
-0.1472 
(0.3467) 
0.2706 
(0.3310) 
0.0340 
(0.4364) 
0.4686 
(0.5294) 
Financial development 0.0630* 
(0.0377) 
0.0565* 
(0.0340) 
-0.0316 
(0.0295) 
0.0453 
(0.0333) 
-0.0593** 
(0.0299) 
Economic development -0.1918*** 
(0.0192) 
-0.1630*** 
(0.0182) 
-0.0576*** 
(0.0179) 
-0.1586 
(0.0180) 
-0.0631*** 
(0.0160) 
Constant 3.2584 
(0.2130) 
3.5716 
(0.1873) 
3.8027*** 
(0.1464) 
3.6474 
(0.1750) 
3.9294*** 
(0.1307) 
F-test 133.36*** 127.93*** 245.47*** 126.19*** 270.77*** 
Log pseudolikelihood 235.29 272.86 359.34 288.55 404.37 
No. of observations 700 700 700 700 700 
The dependent variable Cost efficiency is the estimated bank cost efficiency scores.  The degree of market power is 
proxied by the Conventional Lerner Index or the price mark-up over marginal cost, with the higher scores indicating 
a higher degree of pricing power. Financial freedom and Economic freedom are from the Economic Freedom 
Indicators of Heritage Foundation. They are scaled from 0 to 100 with higher values, indicating greater freedom. 
Concentration is measured by natural logarithm of HHI, measured as the sum of the squares of the market share of 
assets of each bank. Capitalization is the bank total equity to asset ratio. Profitability is measured as the return on 
average assets. Financial development is measured as domestic credit to private sector as a percentage of GDP. 
Economic development is measured as measured as the natural logarithm of GDP per capita. Robust standard errors 
are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
147 
 
Table 4.13: Effects of Competition and Freedom on Cost Efficiency (Adjusted Lerner) 
Tobit regression 
 
Dependent variable: 
 Cost efficiency 
Model 1 Model 2 Model 3 Model 4 Model 5 
Funding-Adjusted Lerner 0.0622* 
(0.0339) 
0.0459 
(0.0303) 
0.0739** 
(0.0363) 
0.2437** 
(0.0989) 
0.8909*** 
(0.2582) 
Financial freedom  -0.0026*** 
(0.0004) 
 -0.0010 
(0.0009) 
 
Economic freedom   0.0060*** 
(0.0017) 
 0.0109*** 
(0.0024) 
Lerner*Financial freedom    -0.0035** 
(0.0017) 
 
Lerner*Economic freedom     -0.0133*** 
(0.0042) 
Concentration -0.0171 
(0.0109) 
-0.0313*** 
(0.0111) 
-0.0117 
(0.01020 
-0.0343*** 
(0.0116) 
-0.0175 
(0.0106) 
Capitalization 0.2823*** 
(0.0723) 
0.2910*** 
(0.0727) 
0.2707*** 
(0.0708) 
0.2965*** 
(0.0726) 
0.2927*** 
(0.0700) 
Profitability -0.7527** 
(0.3434) 
-0.6323** 
(0.3212) 
-0.8253** 
(0.3556) 
-0.7271** 
(0.2992) 
-1.0238*** 
(0.3011) 
Financial development -0.2317*** 
(0.0185) 
-0.2240*** 
(0.0173) 
-0.2404*** 
(0.0179) 
-0.2284*** 
(0.0176) 
-0.2481*** 
(0.0181) 
Economic development 0.0000 
(0.0085) 
0.0041 
(0.0078) 
-0.0088 
(0.0083) 
0.0065 
(0.0080) 
-0.0058 
(0.0082) 
Constant 1.0580*** 
(0.1029) 
1.2665*** 
(0.1036) 
0.7293*** 
(0.1174) 
1.1876*** 
(0.1006) 
0.4579*** 
(0.1422) 
F-test 109.11*** 105.42*** 110.87*** 94.75*** 96.50*** 
Log pseudolikelihood 555.81 569.32 561.13 571.84 568.25 
No. of observations 643 643 643 643 643 
The dependent variable Cost efficiency is the estimated bank cost efficiency scores.  The degree of market power is 
proxied by the Funding-adjusted Lerner Index or the price mark-up over marginal cost, with the higher scores 
indicating a higher degree of pricing power. Financial freedom and Economic freedom are from the Economic 
Freedom Indicators of Heritage Foundation. They are scaled from 0 to 100 with higher values, indicating greater 
freedom. Concentration is measured by natural logarithm of HHI, measured as the sum of the squares of the market 
share of assets of each bank. Capitalization is the bank total equity to asset ratio. Profitability is measured as the return 
on average assets. Financial development is measured as domestic credit to private sector as a percentage of GDP. 
Economic development is measured as measured as the natural logarithm of GDP per capita. Robust standard errors 
are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
148 
 
Table 4.14: Effects of Competition and Freedom on Cost Efficiency (IV with Conv. Lerner) 
Instrumental variable Tobit regression 
Dependent variable:  
Cost efficiency 
Model 1 Model 2 Model 3 Model 4 Model 5 
Lerner 1.4810*** 
(0.5010) 
1.4175** 
(0.6414) 
1.2894** 
(0.5984) 
3.5486** 
(1.4818) 
8.1913** 
(3.2543) 
Financial freedom  -0.0010 
(0.0029) 
 0.0152* 
(0.0090) 
 
Economic freedom   -0.0100 
(0.0094) 
 -0.0098 
(0.0078) 
Lerner*Fin. freedom    -0.0607** 
(0.0262) 
 
Lerner*Econ. freedom     -0.1360** 
(0.0550) 
Concentration -0.2237*** 
(0.0358) 
-0.2240*** 
(0.0341) 
-0.2011*** 
(0.0481) 
-0.2859*** 
(0.0489) 
-0.2446*** 
(0.0521) 
Capitalization 0.9039** 
(0.3859) 
0.8807** 
(0.4252) 
0.8528** 
(0.3813) 
0.3302** 
(0.1607) 
0.4610*** 
(0.1583) 
Profitability -12.7602** 
(4.98820 
-12.2260** 
(6.1418) 
-11.2039** 
(5.6887) 
-2.4717** 
(1.2514) 
-1.0871 
(1.2123) 
Financial development -0.0411 
(0.0542) 
-0.0379 
(0.0568) 
-0.0615 
(0.0447) 
0.0887* 
(0.0479) 
0.0451 
(0.0522) 
Economic development -0.1031*** 
(0.0364) 
-0.1024*** 
(0.0344) 
-0.0678** 
(0.0324) 
-0.1724*** 
(0.0255) 
-0.0426 
(0.0273) 
Constant 3.0086*** 
(0.2637) 
3.0645*** 
(0.3156) 
3.2172*** 
(0.3166) 
3.2023*** 
(0.2979) 
3.4021*** 
(0.2891) 
Wald-test 481.56*** 435.60*** 707.28*** 540.88*** 775.25*** 
Log pseudolikelihood 393.01 607.72 533.65 1566.03 2264.23 
No. of observations 700 700 700 700 700 
Test of Exogeneity 8.23*** 4.86** 5.16** 9.08*** 10.37*** 
The dependent variable Cost efficiency is the estimated bank cost efficiency scores.  The degree of market power is proxied by the 
Conventional Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher degree of pricing 
power. Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. They are 
scaled from 0 to 100 with higher values, indicating greater freedom. Concentration is measured by natural logarithm of HHI, 
measured as the sum of the squares of the market share of assets of each bank. Capitalization is the bank total equity to asset ratio. 
Profitability is measured as the return on average assets. Financial development is measured as domestic credit to private sector as 
a percentage of GDP. Economic development is measured as measured as the natural logarithm of GDP per capita. Robust standard 
errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
149 
 
Table 4.15: Effects of Competition and Freedom on Cost Efficiency (IV with Adj. Lerner) 
Instrumental variable Tobit regression 
Dependent variable:  
Cost efficiency 
Model 1 Model 2 Model 3 Model 4 Model 5 
Funding-Adjusted Lerner 0.2094*** 
(0.0484) 
0.1849*** 
(0.0516) 
0.2232*** 
(0.0510) 
1.3693*** 
(0.4858) 
3.7340** 
(1.7540) 
Financial freedom  -0.0017*** 
(0.0006) 
 0.0076** 
(0.0038) 
 
Economic freedom   0.0097*** 
(0.0024) 
 0.0280*** 
(0.0104) 
Lerner*Fin. freedom    -0.0226*** 
(0.0083) 
 
Lerner*Econ. freedom     -0.0593** 
(0.0290) 
Concentration -0.0224** 
(0.0102) 
-0.0315*** 
(0.0110) 
-0.0135 
(0.0099) 
-0.0506*** 
(0.0121) 
-0.0376** 
(0.0152) 
Capitalization 0.3904*** 
(0.0860) 
0.3864*** 
(0.0862) 
0.3679*** 
(0.0811) 
0.3510*** 
(0.0978) 
0.3788*** 
(0.1054) 
Profitability -2.2537*** 
(0.5879) 
-2.0346*** 
(0.6046) 
-2.3189*** 
(0.6091) 
-1.6079*** 
(0.5767) 
-1.8683*** 
(0.6229) 
Financial development -0.2279*** 
(0.0186) 
-0.2230*** 
(0.0177) 
-0.2422*** 
(0.0178) 
-0.2526*** 
(0.0227) 
-0.2749*** 
(0.0254) 
Economic development 0.0030 
(0.0088) 
0.0055 
(0.0084) 
-0.0114 
(0.0092) 
0.0205* 
(0.0114) 
0.0045 
(0.0109) 
Constant 1.0549*** 
(0.1026) 
1.1967*** 
(0.1111) 
0.5228*** 
(0.1508) 
0.7348*** 
(0.2452) 
-0.5030 
(0.5941) 
Wald-test 638.33*** 713.61*** 737.56*** 652.14*** 645.17*** 
Log pseudolikelihood 569.42 590.22 589.90 1684.96 2244.04 
No. of observations 643 643 643 643 643 
Test of Exogeneity 8.54*** 7.01*** 7.40*** 5.14** 2.41 
The dependent variable Cost efficiency is the estimated bank cost efficiency scores.  The degree of market power is proxied by the 
Funding-adjusted Lerner Index or the price mark-up over marginal cost, with the higher scores indicating a higher degree of 
pricing power. Financial freedom and Economic freedom are from the Economic Freedom Indicators of Heritage Foundation. They 
are scaled from 0 to 100 with higher values, indicating greater freedom. Concentration is measured by natural logarithm of HHI, 
measured as the sum of the squares of the market share of assets of each bank. Capitalization is the bank total equity to asset ratio. 
Profitability is measured as the return on average assets. Financial development is measured as domestic credit to private sector as 
a percentage of GDP. Economic development is measured as measured as the natural logarithm of GDP per capita. Robust standard 
errors are in parentheses. ***, **, and * show 1%, 5% and 10% levels of significance, respectively. 
University of Ghana http://ugspace.ug.edu.gh
150 
 
Our results differ from what Turk-Ariss (2010) found for developing countries, but in line with 
Casu and Girardone (2009) for some EU countries, and Koetter et al. (2012) for the U.S. This gives 
some indication that efficiency may be driving the gain of market power by banks in Sub-Saharan 
Africa, as suggested by the efficient-structure hypothesis, but we cannot reach a definite 
conclusion that the rejection of the quite life hypothesis, implies acceptance of the efficient-
structure hypothesis since we did not test for the effect of efficiency on bank market power.  
 
Also, as noted by Maudos and de Guevara (2007), market power allows banks to enjoy greater 
profits, which may motivate them to behave prudently. In turn, their prudent behavior may lead to 
the selection of less risky activities with lower costs of monitoring, thereby increasing cost 
efficiency. Again, banks that enjoy greater market power are not under much pressure to increase 
the quality of banking services (less availability of means of payment, worse attention to 
customers, etc.), thus reducing operating cost and increasing their cost efficiency. 
 
Surprisingly, the direct effect of financial freedom on bank efficiency is largely negative and 
significantly related to cost efficiency, a rejection of our Hypothesis 8. This gives an indication 
that less restrictions or government interference in the banking sector is associated with lower bank 
cost efficiency. Our results reveal that some level of restrictions and controls in the economy may 
be beneficial in enhancing bank efficiency in the developing country context. This suggests that 
policies that constrain banks’ degree of financial freedom may result in efficient allocation of 
resources. However, another plausible explanation for the negative effect of financial freedom on 
bank efficiency could be the higher costs banks may incur in an effort to obtain resources (human 
capital, technology and physical capital) in order to beat the competition, as more banks enter the 
University of Ghana http://ugspace.ug.edu.gh
151 
 
market with greater freedom. At any rate, high government borrowing from the banking sector in 
Sub-Saharan Africa may partly explain this finding, as banks are able to earn high returns from 
investment in risk-free government securities at low cost, thereby enhancing their efficiency. On 
the contrary, Chortareas et al. (2013) found that for the EU banking market, the higher the degree 
of an economy’s financial freedom, the higher the benefits for banks in terms of cost advantages 
and overall efficiency. Our results show that the direct effect of economic freedom appears 
indeterminate as the signs change between the funding-adjusted and conventional Lerner models.  
 
Given the ambiguous relationship between competition and efficiency (Casu & Girardone, 2006), 
we sought to determine if the impact of market power on efficiency depends on the institutional 
environment. When interacted with the freedom variables, we find that the coefficient of the 
interaction term is mostly negative and significant (except in the Tobit model with conventional 
Lerner where it is significantly positive). It appears that the positive effect of market power on 
bank cost efficiency is weakened for banks operating in countries with higher financial and 
economic freedom, a rejection of Hypothesis 9. Put differently, the results suggests that, a 
combination of weak competitive conditions and higher financial freedom in Sub-Saharan African 
economies lead to lower bank cost efficiency. 
 
With regards to the control variables, we find that banking market concentration has a significantly 
negative coefficient in most of the specifications, suggesting that bank efficiency reduces as the 
market becomes more concentrated. This indicates that policies that encourage mergers and 
acquisitions in Sub-Saharan African banking markets may not be helpful for efficient resource 
allocation. It appears that concentration may not be important in explaining the market power of 
University of Ghana http://ugspace.ug.edu.gh
152 
 
banks in Sub-Saharan African banking markets. Indeed, the different results obtained for the effect 
of market power and concentration on cost efficiency indicates the existence of a low relationship 
between competition and concentration, highlighting the limitations of using market concentration 
measures as proxy variables for competition (Maudos & de Guevara, 2007). 
 
On the other hand, the coefficient of the capitalization ratio is positive and significant in all the 
estimations, indicating that highly capitalized banks are more cost efficient. It may be that highly 
capitalized banks see the need to minimize cost in order to obtain reasonable returns on 
shareholders’ funds. Bank profitability (ROAA) shows a mostly negative and significant 
association with cost efficiency. This means that more profitable banks are less efficient, possibly 
because of the tendency to become wasteful with higher profits. More profitable banks may also 
hire more staff and pay higher wages leading to increased operating costs. 
 
The impact of domestic credit to the private sector as a percentage of GDP (Financial development) 
on bank cost efficiency is rather uncertain, as it is mostly negative and significant in the funding-
adjusted models, but largely positive when significant in the conventional Lerner models. Per 
capita GDP (Economic development) reveals a similarly mixed results with the conventional 
Lerner models showing a mostly negatively significant relationship with cost efficiency, and the 
funding-adjusted models having a generally positive but insignificant relationship. 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
153 
 
4.6 Bank Profit Persistence in Sub-Saharan Africa 
The estimation results for bank profit persistence by country are reported in Table 4.16 (based on 
equation 3.10). The dependent variable (eROAE) is the normalized profit rate or excess profit. 
This is measured as the yearly return on average equity per bank minus the mean return on average 
equity in the country of operation in each year. The persistence of bank profits is estimated for 
only five of the countries due to data limitations in using the system GMM with requirements for 
instrumental variables. 
 
Although there are large differences among the countries in terms of the magnitudes, the 
persistence coefficients (eROAEt-1) are significantly different from zero in most of the estimations. 
The range of persistence coefficients is from a low of 0.170 in Kenya to 0.511 in South Africa. 
This suggests that the intensity of competition in the banking sector is much stronger in Kenya and 
Tanzania (0.181) than in South Africa. However, the level of intensity in competition appear to be 
similar in Ghana (0.437) and Mauritius (0.482). The coefficients on concentration reveal that 
banking market concentration contributes positively to bank profit persistence in Ghana, Kenya 
and Tanzania, although it is not significant in Tanzania. This is consistent with what Goddard et 
al. (2011) found for some advanced and developing countries. On the other hand, market 
concentration translates into less profit persistence in South Africa significantly, and in Mauritius 
insignificantly. 
 
While economic freedom associates positively with profit persistence in Tanzania, Kenya and 
Mauritius, its effect is negative in Ghana and South Africa, although it is not very important in 
Mauritius and South Africa. The capitalization ratio coefficient is positive in all the countries 
University of Ghana http://ugspace.ug.edu.gh
154 
 
except Mauritius. However, its effect is not significant in Tanzania and South Africa. Bank size 
has a positive and significant relationship with excess profit in Mauritius and Kenya and a negative 
relationship in the other countries. 
 
Table 4.16: Determinants of Bank Profit Persistence 
Dependent variable:  
Excess Profit (eROAE) 
Tanzania South Africa Mauritius Kenya Ghana 
eROAEt-1 0.1809* 0.5111*** 0.4821*** 0.1697*** 0.4374*** 
Concentration 0.0594 -1.9368*** -0.0201 0.1646*** 0.9632*** 
Economic freedom 0.0422* -0.0001 0.0039 0.0098* -0.0426*** 
Capitalization 0.2905 0.5692 -0.9764** 0.6252*** 0.8701*** 
Bank Size -0.0215 -0.0136 0.1103* 0.0949*** -0.0756** 
Credit risk -0.0522 0.0063 -0.5351 -0.2144*** -0.1583 
Market share 0.0142** 0.0056 -0.0032 -0.0098* 0.0097 
Financial development 5.2103** -0.0539 -0.0389 0.0099 -42.3897*** 
Constant -3.7053 15.0694*** -0.3204 -2.1642*** 2.8658*** 
AR (2) 0.2745 
(0.7837) 
-0.1994 
(0.8420) 
0 .7610  
(0.4467) 
-1.1472  
(0.2513) 
.09934  
(0.9209) 
Sargan test 3.2675 
(1.0000) 
6.8210 
(0.9951) 
4.2567 
(0.9998) 
11.2795 
(0.9141) 
12.56364 
(0.8601) 
Wald 112.86*** 222.69*** 1459.15*** 308.68*** 3829.06*** 
The dependent variable eROAE is measured as the yearly return on average equity per bank minus the mean return on 
average equity. Concentration is measured by the natural logarithm of HHI, measured as the sum of the squares of the 
market share of assets of each bank. Economic freedom is from the Economic Freedom Indicators of Heritage 
Foundation. They are scaled from 0 to 100 with higher values indicating greater freedom. Capitalization is the bank 
total equity to asset ratio. Bank size is measured as the natural logarithm of total assets. Credit risk is measured as total 
loans to total assets. Market share is measured as ratio of bank assets to total market assets. Financial development is 
measured as domestic credit to private sector as a percentage of GDP. Robust standard errors are in parentheses. ***, 
**, and * show 1%, 5% and 10% levels of significance, respectively. 
 
It appears that having a higher proportion of bank assets in loans (lending specialization) has the 
tendency to translate into less profit persistence on average in all the countries except in South 
University of Ghana http://ugspace.ug.edu.gh
155 
 
Africa.  Market share is positively related to excess profit in Tanzania, South Africa and Ghana, 
and negatively related to it in Kenya and Mauritius. 
 
Domestic credit to the private sector, is an important determinant of excess profit in Tanzania and 
Ghana except that the effect is positive in Tanzania and negative in Ghana. Mauritius and South 
Africa show a negative but insignificant effect whilst the effect in Kenya is also positive but 
insignificant. 
 
The above results confirm Hypothesis 10, which refers to variations in the determinants of bank 
profit persistence across countries. 
 
4.7 Chapter Summary 
This chapter presents the results of our analysis, beginning with a summary of the descriptive 
statistics for the key variables used in the study. The mean country values show that among the 
countries in Sub-Saharan Africa included in this study, Mauritius has the highest degree of 
financial freedom while supervision of the financial sector in Ethiopia may be considered 
repressive, according to the Heritage Foundation’s index of financial freedom. Generally, 
however, countries in Sub-Saharan Africa seems to portray significant influence and interference 
of government in the financial sector (with average financial freedom of 51.1). The average level 
of economic freedom (59.6) signifies a reasonable level of liberty to own and employ capital in 
these countries, with Mauritius and Ethiopia, again, having the highest and lowest levels of 
economic freedom, respectively.  
 
University of Ghana http://ugspace.ug.edu.gh
156 
 
The results of the z-score estimates show that banks in Namibia and Ethiopia appear to be the most 
stable, while the most unstable banks are found in Zambia, Ghana and Mozambique. Asset quality 
is poorest in Zambia, Ghana and Ethiopia. This means that in general, banks in Zambia and Ghana 
face more risk than those in the other countries. Sub-Saharan African banks appear quite profitable 
with about 2.0% for Return on Average Assets (ROAA), 16.2% for Return on Average Equity 
(ROAE) and 5.6% for Net Interest Margin (NIM). This compares with what Beck and Cull (2013) 
report for the Africa region and higher than profits for banks outside Africa (ROAA of 1.5%), 
according to Beck and Cull (2013).  
 
Market power, represented as Conventional Lerner index and Funding-adjusted index show a 
moderate level in most of the countries, averaging 0.267 and 0.275, respectively. The funding-
adjusted Lerner is slightly higher than the conventional Lerner on average, suggesting little 
underestimation with the conventional Lerner index, but there are considerable differences in the 
two indices among the countries. The level of bank market power observed is consistent with what 
Fosu (2013) found for Africa using both the static and dynamic versions of the Panzar–Rosse 
(1987) model. This suggests that intermediation activities in banking markets in Sub-Saharan 
Africa are quite competitive and comparable to that of other emerging markets (Fosu, 2013). This 
indicates that the ability of banks in this region to price their products and services above the 
marginal cost is limited. This may be because of higher banking costs in Africa, generally 
reflective of the high cost of doing business, and high loan losses due to serious information 
asymmetry problems. Indeed, we found average cost to income ratio to be high (64.6%), indicating 
poor operating efficiency among banks in this region. Staff costs and other operating expenses are 
quite high, averaging about 2.9% and 3.6% of total assets, respectively. This suggests that policies 
University of Ghana http://ugspace.ug.edu.gh
157 
 
aimed at improving infrastructure and reducing information asymmetry problems, as well as 
general prices of products, such as stable macroeconomic conditions could reduce intermediation 
costs in these countries. This could in turn encourage savings, enhance financial inclusion, improve 
access to finance and promote economic growth and development.  
 
However, with an average of about 13.5% capitalization ratio, banks in Sub-Saharan Africa appear 
to be generally well-capitalized beyond the level recommended by Basel III (8%), and may partly 
explain the reasonable stability found in the banking markets in this region (average z-score of 
47.43). Nevertheless, Beck et al. (2011) warns that there is still veiled or silent fragility in several 
Central and West African countries, especially in state-owned banks, and small locally owned 
banks. On average, about 54% of banking assets in these Sub-Saharan African countries are 
devoted to lending activities with the highest level of specialization in lending occurring in the 
more developed markets in Namibia, South Africa and Mauritius. Again, this underscores the need 
for appropriate policies to improve contractual frameworks to enhance efficiency in credit 
allocation. A significant portion of bank income (about 30% on average) is earned from non-
interest activities. This may have serious negative repercussions for bank stability and may require 
policies that could limit bank risk-taking to avert potential bank failures. Diversification may be 
good for bank profitability but the uncertainty associated with income from non-traditional 
banking activities may make it quite risky especially for banks in this region who are largely 
financed by demand deposits.  
 
University of Ghana http://ugspace.ug.edu.gh
158 
 
However, domestic credit to the private sector as a percentage of GDP is generally low, at about 
40.3% average, except for the more developed markets in the region, South Africa (154.1%) and 
Mauritius (85.3%) where it was quite significant.  
 
One objective of this study is to examine the effects of competition and financial freedom on bank 
stability or insolvency risk. The regressions were estimated using system GMM. The results show 
that higher bank market power is significantly and positively associated with greater bank stability 
(measured by the natural logarithm of z-score). This suggests that when banks have the ability to 
price their products in a monopolistic fashion (because of less competition), it translates into less 
likelihood of the banks becoming insolvent, and rather makes banks more sound (Amidu, 2013; 
Turk-Ariss, 2010). The evidence also supports the competition-fragility or competition-instability 
hypothesis postulated by Keeley (1990), and indicates that policies aimed at maintaining some 
level of bank market (or monopoly) power may be required to sustain bank stability in this region. 
The study also found evidence for a non-linear relationship between competition and bank stability 
(Martinez-Miera & Repullo, 2010). This should inform competition policies in Sub-Saharan 
African countries which have generally sought to open up their banking markets to increasing 
competition. The idea that competition is good may be more naive in banking than in other 
industries (Claessens & Laeven, 2004). 
 
We did not find any concrete evidence of the possible conditional effect of financial freedom on 
the competition-stability relationship. This indicates that, though yet to be tested, concerns that too 
much financial freedom may contribute to financial institutions’ tendency to take on greater risks, 
University of Ghana http://ugspace.ug.edu.gh
159 
 
which may have contributed to the recent global and European crises as suggested by Chortareas 
et al. (2013), may not be applicable in the developing country context.  
 
With respect to the control variables, we find that diversification (non-interest income to total 
income) has a negative relationship with bank stability, denoting that highly diversified banks are 
more risky and unstable. The results suggests that banks tend to be more risky when they get 
involved in non-traditional business activities. This should be of concern to policy makers given 
the increasing trend of involvement of banks in non-traditional activities. Financial development 
also has a significantly positive relationship with bank stability, meaning that increases in the 
amount of bank credit to the private sector tend to make banks less risky and more stable.  
 
In view of the difficulty in assessing bank risk with a single measure, we carry out similar 
regressions on bank asset quality as an alternative to z-score, the more popular measure of overall 
bank risk or stability. Our measure of bank asset quality is the ratio of loan loss reserves to total 
assets. The results show that market power is negatively associated with the bank asset quality 
measure, suggesting that as banks gain more pricing power, their asset quality improves. This also 
means that increased competition in the banking market results in greater risk-taking by banks. 
However, similar to the case of bank stability, we find evidence for a non-linear relationship 
between competition and bank risk-taking. This again indicates the need for policies that allow 
banks to retain some measure of pricing power in order to reduce risk-taking. Financial freedom 
did not show any significant effect on the quality of bank assets.  
 
University of Ghana http://ugspace.ug.edu.gh
160 
 
With regards to the control variables, we find that the credit risk measure (loans to total assets) is 
positively related to bank asset quality, an indication of higher risk-taking and deteriorating asset 
quality with increased lending. The level of financial development appears to have a significantly 
positive influence on bank asset quality, suggesting that asset quality is likely to improve as banks 
gain more experience in the market. 
 
A second objective of the study is to ascertain the effects of competition and financial freedom on 
bank profitability. The regressions were estimated using system GMM. The results reveal a 
positive relationship between market power and bank Net Interest Margins (NIM), similar to 
Amidu and Wolfe (2013), confirming our expectation that banks with market power can charge a 
higher loan rate and offer a lower deposit rate, thus earning higher margins (Hawtrey & Liang, 
2008). Both financial freedom and economic freedom show a positive relationship with NIM, an 
indication that less intervention by government in the banking sector boosts bank margins. There 
is some indication that the impact of competition on bank margins is sensitive to the level of 
financial and economic freedom prevailing in an economy. This is seen in the positive coefficients 
of their interaction terms. It appears that the positive effect of market power on bank margins is 
stronger for banks operating in countries with higher levels of freedom.  
 
An analysis of the control variables show that, credit risk appears to be an important determinant 
of bank margins. Higher credit risk associates with higher bank margins, ostensibly as a result of 
additional risk premium charges (Amidu & Wolfe, 2013; Ahokpossi, 2013). Diversification shows 
a negative relationship with NIM, implying that banks that engage in other activities earn lower 
margins, as they lose focus on their core activity, or charge lower spreads for loans in order to gain 
University of Ghana http://ugspace.ug.edu.gh
161 
 
higher income from non-interest activities (Kalluci, 2010). Both financial development (domestic 
credit to the private sector as a percentage of GDP) and economic development (per capita GDP), 
appear to have a negative impact on NIM, denoting that banks operating in countries with higher 
levels of financial and economic development earn reduced margins. However, banking market 
concentration, operating efficiency, bank size, and the level of capitalization do not appear to be 
important determinants of bank net interest margins in Sub-Saharan Africa.  
 
Furthermore, similar to the results for NIM, market power is positively related to overall bank 
profitability (Return on Average Assets-ROAA), suggesting that in less competitive markets 
(higher bank market power) banks generally earn higher returns on their assets. Also, both 
financial freedom and economic freedom show a positive impact on bank profits. We find evidence 
affirming that financial and economic freedom have conditioning effects on the impact of market 
power (or competition) on bank profits. The outcome for ROAA when the market power variable 
(Lerner) is interacted with the freedom variables confirms this. This suggests that banks with 
higher market power operating in countries with higher freedom for banking activities are more 
profitable than their counterparts in countries with greater restrictions on banking activities. Thus 
policies aimed at reducing bank intermediation costs should consider measures to control bank 
market power in order to avoid abuse of market power at the expense of the larger economy. 
 
An analysis of the control variables show that bank profits are likely to be lower in highly 
concentrated markets. This is inconsistent with the Structure-Conduct-Performance hypothesis 
which suggests that concentration enables banks to collude and earn higher profits. On the other 
hand, bank operating efficiency (cost to income) shows a significantly negative relationship with 
University of Ghana http://ugspace.ug.edu.gh
162 
 
ROAA, signifying that, as expected, banks that are able to control costs more effectively earn 
higher profits. This is in line with empirical literature which suggests that efficiency is a more 
important determinant of profitability than concentration (Berger, 1995). Contrary to the NIM 
results (which were negative), the diversification variable has a significantly positive coefficient. 
This means that diversification is an important determinant of bank return on assets, and that banks 
that engage in other activities outside their core business earn significantly higher returns than 
those not engaging in similar activities (Goddard et al., 2013). One plausible explanation for the 
positive effect of diversification on bank profitability may be the limited losses associated with 
non-traditional banking activities, compared to the generally huge losses on loans in some of these 
economies. However, bank size, capitalization, credit risk, financial development and economic 
development do not appear to be important determinants of overall bank profitability. 
 
A third objective of this study is to assess the effects of competition and financial freedom on bank 
efficiency. We first considered the relationship between financial freedom and competition, and 
found that higher freedom generally increases competition (and reduces bank market power). This 
suggests that banks that operate in countries with more financial freedom (less government control 
over financial institutions) tend to have lower pricing power, likely as a result of fewer restrictions 
in establishing new banks even by foreign investors, thus leading to greater competition (Mirzaei 
& Moore, 2014). Other important determinants of market power are capitalization, diversification 
and market share, suggesting that well-diversified and highly capitalized banks, with higher market 
shares exercise greater market power. This gives an indication that policies targeted at controlling 
bank market power should aim at limited government involvement in the banking sector, and some 
restrictions on non-traditional banking activities.  
University of Ghana http://ugspace.ug.edu.gh
163 
 
We estimated bank cost efficiency scores by constructing separate frontiers for each country, and 
thus they are not directly comparable across countries. The results reveal potential for considerable 
cost savings for some banks in Sub-Saharan Africa, especially among those operating in Malawi, 
South Africa, Mauritius and Ghana. Subsequently, we used Tobit regression to assess the impact 
of competition and financial freedom on bank cost efficiency. However, to take into account 
possible endogeneity of the market power measure and bank efficiency, we also used instrumental 
variable Tobit regression to check the robustness of our results. We present a common analysis for 
the two approaches given the general similarity in the results. The results show that banks with 
higher market power are the most cost efficient, contrary to the quite life hypothesis which suggests 
that banks become inefficient as they gain more market power and settle for a quiet life with no 
incentive to reduce costs. Banks with market power may not be under pressure to improve the 
quality of their services, and may also be motivated to behave prudently as they enjoy greater 
profits (Maudos & de Guevara, 2007). 
 
Surprisingly, the direct effect of financial freedom on bank efficiency is largely negative and 
significantly related to cost efficiency. Our results reveal that some level of restrictions and 
controls in the economy may be beneficial in enhancing bank efficiency in the developing country 
context. This suggests that policies that constrain banks’ degree of financial freedom may result in 
efficient allocation of resources. On the contrary, Chortareas et al. (2013) found that for the EU 
banking market, the higher the degree of an economy’s financial freedom, the higher the benefits 
for banks in terms of cost advantages and overall efficiency. Given the ambiguous relationship 
between competition and efficiency (Casu & Girardone, 2006), we sought to determine if the 
impact of market power on efficiency depends on the institutional environment. The results 
University of Ghana http://ugspace.ug.edu.gh
164 
 
indicate that the positive effect of market power on bank cost efficiency is weakened for banks 
operating in countries with higher financial and economic freedom.  
 
With regards to the control variables, we find that higher banking market concentration and higher 
profitability associate with lower bank cost efficiency. This indicates that policies that encourage 
mergers and acquisitions in Sub-Saharan African banking markets may not be helpful for efficient 
resource allocation. Also, limiting bank profitability may motivate banks to avoid wastage and 
improve efficiency. On the other hand, highly capitalized banks appear to be more cost efficient, 
perhaps due to the need to minimize cost in order to obtain reasonable returns on shareholders’ 
funds.  
 
A fourth objective of this study is to examine the determinants of bank profit persistence. The 
persistence of bank profits is estimated for only five of the countries due to data limitations in 
using the system GMM with requirements for instrumental variables. The dependent variable is 
the normalized profit rate or excess profit. This is measured as the yearly return on average equity 
per bank minus the mean return on average equity in the country of operation in each year. The 
range of persistence coefficients is from a low of 0.170 in Kenya to 0.511 in South Africa. This 
suggests that the intensity of competition in the banking sector is much stronger in Kenya and 
Tanzania (0.181) than in South Africa. However, the level of intensity in competition appear to be 
similar in Ghana (0.437) and Mauritius (0.482). Banking market concentration contributes 
positively to bank profit persistence in Ghana and Kenya, but negatively to profit persistence in 
South Africa. Also, while economic freedom associates positively with profit persistence in 
Tanzania, and Kenya, its effect is negative in Ghana and insignificant in the other countries. 
University of Ghana http://ugspace.ug.edu.gh
165 
 
 The level of capitalization has a positive effect on profit persistent in Kenya and Ghana and a 
negative effect in Mauritius. But bank size has a positive relationship with excess profit in 
Mauritius and Kenya and a negative relationship in Ghana.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
166 
 
CHAPTER FIVE 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS  
 
5.1 Introduction 
This thesis analyzed the effects of competition and financial freedom on bank stability, the effects 
of competition and financial freedom on bank profitability (net interest margins and return on 
assets), and the effects of competition and financial freedom on bank efficiency. We also assessed 
the determinants of bank profit persistence, and the sources of bank market power. In this final 
chapter, we present a summary of findings, contributions of the study, conclusions, policy 
recommendations, as well as limitations of the study and suggestions for further research. 
 
5.2 Summary of Findings  
The results of our analysis show that Mauritius and Ethiopia have the highest and lowest degree 
of financial freedom, respectively, among the countries in Sub-Saharan Africa included in this 
study. Generally, we find a significant influence and interference of government in the financial 
sector, but a reasonable level of liberty to own and employ capital in these countries. The results 
show that, generally, banks in this region are very profitable, well-capitalized and quite stable. The 
levels of bank market power observed also indicate that banking markets in Sub-Saharan Africa 
are quite competitive and comparable to those of other emerging markets. This indicates that the 
ability of banks in this region to price their products and services above the marginal cost is limited. 
This may be because of higher banking costs in Africa, generally reflective of the high cost of 
doing business, and high loan losses due to serious information asymmetry problems. Indeed, we 
find poor operating efficiency among banks in this region, with high staff costs and other operating 
University of Ghana http://ugspace.ug.edu.gh
167 
 
expenses. This suggests that policies aimed at improving infrastructure and reducing information 
asymmetry problems, as well as general prices of products, such as stable macroeconomic 
conditions could reduce intermediation costs in these countries. This could in turn encourage 
savings, enhance financial inclusion, improve access to finance and promote economic growth and 
development.  
 
Furthermore, the percentage of banking assets devoted to lending activities is relatively low 
compared to other regions of the world. This underscores the need for appropriate policies to 
improve contractual frameworks to enhance efficiency in credit allocation. On the other hand, a 
significant portion of bank income is earned from non-interest activities. This may have serious 
negative repercussions for bank stability and may require policies that could limit bank risk-taking 
to avert potential bank failures. Domestic credit to the private sector as a percentage of GDP is 
generally low, and calls for appropriate policies to engender bank lending activities, such as 
improvements in measures to address information asymmetry and moral hazard problems. 
 
The regression results show that higher bank market power (or less competition) is significantly 
and positively associated with greater bank stability. This suggests that when banks have the ability 
to price their products in a monopolistic fashion (because of less competition), it translates into 
less likelihood of the banks becoming insolvent, and rather makes banks more sound. We also find 
a non-linear relationship between competition and bank stability, but we did not find any concrete 
evidence of the possible conditional effect of financial freedom on the competition-stability 
relationship. This indicates that, though yet to be tested, concerns that excessive financial freedom 
may contribute to banks’ propensity to take on greater risks, which in turn may have contributed 
University of Ghana http://ugspace.ug.edu.gh
168 
 
to the recent global and European crises may not be applicable in the developing country context. 
With regards to the control variables, diversification shows a strong negative and significant 
relationship with bank stability in all the estimations, denoting that highly diversified banks are 
more risky and unstable. On the other hand, financial development has a significantly positive 
relationship with bank stability, implying that increases in the amount of bank credit to the private 
sector makes banks less risky and more stable.  
 
In view of the difficulty in assessing bank risk with a single measure, we carry out similar 
regressions on bank asset quality, as an alternative to z-score, the more popular measure of overall 
bank risk or stability. Generally, the results show that the coefficient of market power is negatively 
and significantly associated with the bank asset quality measure, suggesting that as banks gain 
more pricing power, their asset quality improves. This also indicates that increased competition in 
the banking market results in greater risk-taking by banks. However, similar to the case of bank 
stability, we find evidence for a non-linear relationship between competition and bank risk-taking, 
but financial freedom did not show any significant effect on the quality of bank assets. With 
regards to the control variables, we find that the credit risk measure is positively related to bank 
asset quality, an indication of higher risk-taking and deteriorating asset quality with increased 
lending. The level of financial development appears to have a significantly positive influence on 
bank asset quality, suggesting that asset quality is likely to improve as banks gain more experience 
in the market. 
 
A second objective of the study is to ascertain the effects of competition and financial freedom on 
bank profitability. The results reveal a positive relationship between market power and bank net 
University of Ghana http://ugspace.ug.edu.gh
169 
 
interest margins, indicating that banks with market power charge a higher loan rate and offer a 
lower deposit rate. Both financial freedom and economic freedom show a positive relationship 
with NIM, suggesting that less intervention by government in the banking sector boosts bank 
margins. It also appears that the positive effect of market power on bank margins is stronger for 
banks operating in countries with higher levels of freedom. An analysis of the control variables 
show that higher credit risk associates with higher bank margins, ostensibly as a result of additional 
risk premium charges. Also, diversification shows a negative relationship with NIM, implying that 
banks that engage in other activities earn lower margins, as they lose focus on their core activity. 
Both financial development and economic development appear to have a negative impact on NIM, 
denoting that banks operating in countries with higher levels of financial and economic 
development earn reduced margins. However, banking market concentration, operating efficiency, 
bank size, and the level of capitalization do not appear to be important determinants of bank net 
interest margins in Sub-Saharan Africa.  
 
Furthermore, similar to the results for NIM, market power is positively related to overall bank 
profitability (ROAA), suggesting that in less competitive markets banks generally earn higher 
returns on their assets. Also, both financial freedom and economic freedom show a positive impact 
on bank profits. We find evidence affirming that financial and economic freedom have 
conditioning effects on the impact of market power (or competition) on bank profits. It appears 
that banks with higher market power operating in countries with higher freedom for banking 
activities are more profitable than their counterparts in countries with greater restrictions on 
banking activities. An analysis of the control variables show that bank profits are likely to be lower 
in highly concentrated markets. This is inconsistent with the Structure-Conduct-Performance 
University of Ghana http://ugspace.ug.edu.gh
170 
 
hypothesis which suggests that concentration enables banks to collude and earn higher profits. On 
the other hand, the bank operating efficiency measure shows a significantly negative relationship 
with ROAA, signifying that, as expected, banks that are able to control costs more effectively earn 
higher profits. Contrary to the NIM results, the diversification variable has a significantly positive 
coefficient. This means that diversification is an important determinant of bank return on assets, 
and that banks that engage in other activities outside their core business earn significantly higher 
returns than those not engaging in similar activities. However, bank size, capitalization, credit risk, 
financial development and economic development do not appear to be important determinants of 
overall bank profitability. 
 
The study identified the sources of bank market power in Sub-Saharan Africa to be capitalization, 
diversification, and market share. On the other hand, financial freedom, economic freedom and 
economic development appear to dampen bank market power. This suggests that banks that 
operate in countries with greater freedom tend to have lower pricing power, likely as a result of 
fewer restrictions in establishing new banks even by foreign investors, thus leading to greater 
competition.  
 
With regards to the competition and bank efficiency relationship, we find evidence that rejects the 
quiet life hypothesis. The independent effect of bank market power on cost efficiency is 
consistently positive and significant in all the estimations. This means that banks with higher 
market power are the most cost efficient, contrary to the quite life hypothesis that banks become 
inefficient as they gain more market power and settle for a quiet life with no incentive to reduce 
costs. Surprisingly, the direct effect of financial freedom on bank efficiency is largely negative 
University of Ghana http://ugspace.ug.edu.gh
171 
 
and significantly related to cost efficiency. This gives an indication that less government 
interference in the banking sector is associated with lower bank cost efficiency. However, another 
plausible explanation for the negative effect of financial freedom on bank efficiency could be the 
higher costs banks may incur in an effort to obtain resources (human capital, technology and 
physical capital) in order to beat the competition, as more banks enter the market with greater 
freedom. Given the ambiguous relationship between competition and efficiency, we sought to 
determine if the impact of market power on efficiency depends on the institutional environment, 
by interacting the Lerner index with the freedom variables. It appears that the positive effect of 
market power on bank cost efficiency is weakened for banks operating in countries with higher 
financial and economic freedom.  
 
With regards to the control variables, we find that generally, banking market concentration has a 
negative effect on cost efficiency, suggesting that bank efficiency reduces as the market becomes 
more concentrated. The different results obtained for the effect of market power and concentration 
on cost efficiency indicates the existence of a low relationship between competition and 
concentration, highlighting the limitations of using market concentration measures as proxy 
variables for competition. On the other hand, the coefficient of the capitalization ratio is positive 
and significant in all the estimations, indicating that highly capitalized banks are more cost 
efficient. Bank profitability (ROAA) shows a mostly negative and significant association with cost 
efficiency. This means that more profitable banks are less efficient, possibly because of the 
tendency to become wasteful with higher profits.  
The persistence of bank profits is estimated for only five of the countries due to data limitations in 
using the system GMM with requirements for instrumental variables. The level of profit 
University of Ghana http://ugspace.ug.edu.gh
172 
 
persistence observed suggests that the intensity of competition in the banking sector is much 
stronger in Kenya and Tanzania than in South Africa. Banking market concentration contributes 
positively to bank profit persistence in Ghana and Kenya, but translates into less profit persistence 
in South Africa. Also, economic freedom associates positively with profit persistence in Tanzania, 
and Kenya, but its effect is negative in Ghana. The level of capitalization has a positive effect on 
profit persistent in Kenya and Ghana and a negative effect in Mauritius. But bank size has a 
positive relationship with excess profit in Mauritius and Kenya and a negative relationship in 
Ghana.  
 
5.3 Contributions of the Study 
This study has made important contributions to the competition and bank performance literature. 
First, our analysis of the effect of financial freedom on the relationship between competition and 
bank efficiency has shown that higher degrees of freedom has a detrimental impact on bank 
efficiency. This is new in the literature, and should assist policy makers in designing appropriate 
policies for banking markets in Sub-Saharan Africa. For example, policies that ensure some level 
of restrictions in the banking sector may still be required to enhance bank efficiency.  
 
Second, our use of the same model in estimating bank market power and cost efficiency is another 
significant contribution to the literature. Although very necessary, given the close relationship 
between competition and efficiency, only a handful of studies have jointly estimated competition 
and bank efficiency using the same framework. Most previous studies estimate competition and 
efficiency separately, and then assess their relationship. Third, our test results which rejects the 
quiet life hypothesis which has received very little attention in banking shows that for banks in 
University of Ghana http://ugspace.ug.edu.gh
173 
 
Sub-Saharan Africa, the gain of market power is not a disincentive to controlling costs. Again, this 
has implications for policy on consumer welfare.  
 
Fourth, our confirmation of the existence of a non-linear relationship between market power and 
bank stability adds to the scant literature on this relatively new development in the literature. Most 
previous studies assume a linear relationship between competition and bank stability which may 
have resulted in poor policy prescriptions in the past. And by testing the effect of financial freedom 
on the relationship between competition and bank stability (which has not received any attention 
in the literature), we obviate concerns that excessive financial freedom may contribute to banks’ 
propensity to take on greater risks. 
 
Fifth, whereas other studies use mostly market structure indicators (concentration indices and 
market share) as a proxy for market power in testing the Structure-Conduct-Performance (SCP) 
hypothesis, we used a direct measure of market power at the bank level, the Lerner index and find 
support for the SCP hypothesis. This provides a more accurate assessment of the effect of bank 
market power. Sixth, we assess the effect of financial freedom on the relationship between 
competition and bank profitability, and show that higher freedom for banks enhances their margins 
and overall profitability. This has not been the focus of other studies and should help determine 
appropriate policy measures to ensure that banks do not earn higher profits at the expense of the 
rest of the economy. We also assess the determinants of bank profit persistence and variations 
across countries, which is largely ignored in the literature. And finally, we provide a more 
comprehensive assessment of bank performance by considering bank efficiency, profitability and 
University of Ghana http://ugspace.ug.edu.gh
174 
 
stability together, adding to the scant literature on competition and bank performance in the 
developing countries context.  
 
5.4 Conclusions  
Based on the results of the study, we can draw some conclusions. There is evidence of the 
competition-fragility hypothesis in Sub-Saharan Africa banking. Greater competition may 
generally result in bank insolvency. However, the relationship between competition and bank 
stability in Sub-Saharan Africa is found to be non-linear, implying that beyond a setting threshold, 
increases in market power may also be damaging to bank stability. It is clear from the results that 
higher financial freedom for banks improves their profitability, but is harmful for cost efficiency, 
especially for banks with higher market power. On the other hand, financial freedom does not 
affect bank stability.  
 
Higher level of banking system concentration is clearly harmful to bank profitability and efficiency 
in Sub-Saharan Africa. Again, higher bank equity seems to improve bank efficiency. It may be 
that bank managers are more concerned about controlling costs when bank capital is high, in order 
to earn reasonable returns for shareholders. While helpful for improving bank profitability, the 
results show that high diversification exposes banks to greater risk. Banking costs in Sub-Saharan 
Africa also appear high, which may be attributable to information asymmetry problems, and poor 
infrastructure such as lack of access to adequate electricity, technology and roads. On the 
determinants, we can conclude that there are significant variations in the factors influencing bank 
profit persistence in different countries, which calls for different policy responses.  
University of Ghana http://ugspace.ug.edu.gh
175 
 
5.5 Policy Recommendations 
The results of the study have important policy repercussions. Policy makers usually take the view 
that opening up banking markets to greater competition may lead to higher efficiency, as it reduces 
the monopoly power of banks and challenge them to be innovate. However, our results have shown 
that allowing banks to maintain some level of market power may be necessary to ensure banking 
system efficiency and overall stability. Hence, caution is needed in implementing policies that will 
flood the market and eliminate pricing power of banks. Of course, the fact that higher financial 
freedom associates with greater bank margins and overall profitability suggests that banks may be 
earning part of their income at the expense of the rest of the economy which needs to be controlled. 
It would appear that policy measures to minimize bank market power moderately may be well 
placed. Also, given that higher financial freedom improves bank profitability but harms cost 
efficiency, especially for banks with higher market power, policies that ensure some level of 
restrictions in the banking sector may improve bank efficiency.  
 
Higher level of banking system concentration is clearly harmful to bank profitability and efficiency 
in Sub-Saharan Africa. Hence, measures that have been taken to reduce the level of concentration 
in banking markets in the region should be continued to enhance bank performance. Again, higher 
bank equity seems to improve bank efficiency. This indicates that policies that require banks to 
hold higher capital may be helpful in this region. It may be that bank managers are more concerned 
about controlling costs when bank capital is high, in order to earn reasonable returns for 
shareholders. However, policies that allow banks to engage in non-traditional activities need to be 
carefully crafted. While helpful for improving bank profitability, the results show that high 
diversification exposes banks to greater risk. Banking costs in Sub-Saharan Africa appear high, 
University of Ghana http://ugspace.ug.edu.gh
176 
 
which may be attributable to information asymmetry problems, and poor infrastructure such as 
lack of access to adequate electricity, technology and roads. Thus, policy measures need to be put 
in place to address lack of transparency in the financial behavior of borrowers, and banking 
operations to reduce costs and improve efficiency. Investments in infrastructure also needs to be 
accelerated to provide adequate resources at reasonable cost to improve consumer welfare.  
 
5.6 Limitations and Suggestions for Further Research 
While providing useful results, this study is subject to some limitations. Inadequate data to cover 
the whole Sub-Saharan Africa region, particularly some of the larger banking markets such as 
Nigeria and Angola is one of the limitations of the study. Future studies should explore the 
possibility of obtaining complete data to ensure general applicability of the results. Of course, most 
of the important banking markets in the region, in terms of assets, were covered in the study. 
Another limitation is the use of a single measure of bank competition. It would be useful for future 
studies to consider the use of other measures of competition such as the newly introduced Boone 
indicator, or the Panza-Rosse H-statistic to confirm our findings. 
 
In addition, future studies could consider the feedback effect of efficiency on bank competition. 
The close relationship between competition and efficiency suggest possible reverse causality, yet 
most studies assess the effect of competition on bank efficiency, without considering whether 
efficiency in turn engenders competitive conditions in banking markets. And for effective and 
specific policy prescriptions, it may be more helpful if the relations of interest in this study could 
be replicated at the country level, instead of the cross-country approach used for this research. 
University of Ghana http://ugspace.ug.edu.gh
177 
 
REFERENCES 
Aboagye, A. Q. Q., Akoena, S. K., Antwi‐asare, T., & Gockel, A. F. (2008). Explaining interest 
rate spreads in Ghana. African Development Review, 20(3), 378-399. 
Aboagye, A. Q. Q., Akoena, S. K., Antwi‐asare, T., & Gockel, A. F. (2008). Explaining the market 
power of Ghanaian banks. South African Journal of Economics, 76(4), 569-585. 
Agoraki, M. E. K., Delis, M. D., & Pasiouras, F. (2011). Regulations, competition and bank risk-
taking in transition countries. Journal of Financial Stability, 7(1), 38-48. 
Ahokpossi, M. C. (2013). Determinants of bank interest margins in sub-Saharan Africa. (No. 13-
34). International Monetary Fund. 
Aigner, D.J., Lovell, C.A.K., & Schmidt, P. (1977). Formulation and estimation of stochastic 
frontier production function models. Journal of Econometrics, 6, 21-37. 
Allen, F., & Gale, D. (2000). Financial contagion. Journal of Political Economy, 108, 1–33. 
Allen, F., & Gale, D. (2004). Competition and financial stability. Journal of Money, Credit, and 
Banking, 36, 453–480. 
Allen, L. (1988). The determinants of bank interest margins: A Note. Journal of Financial and 
Quantitative Analysis, 23, 231-235. 
 Allen, F., Otchere, I., & Senbet, L. W. (2011). African financial systems : A review. Journal of 
Advanced Research, 1(2), 79–113. 
Almarzoqi, R., & Naceur, S. B. (2015). Determinants of bank interest margins in the Caucasus and 
central Asia. International Monetary Fund WP/15/87. 
Al-Muharrami, S., & Matthews, K. (2009). Market power versus efficient-structure in Arab GCC 
banking. Applied Financial Economics, 19(18), 1487-1496. 
University of Ghana http://ugspace.ug.edu.gh
178 
 
Amidu, M. (2013). The effects of the structure of banking market and funding strategy on risk and 
return. International Review of Financial Analysis, 28, 143–155.  
Amidu, M., & Wolfe, S. (2013). Does bank competition and diversification lead to greater 
stability? Evidence from emerging markets. Review of Development Finance, 3, 152–166. 
Amidu, M., & Wolfe, S. (2013). The impact of market power and funding strategy on bank-interest 
margins. The European Journal of Finance, 19(9), 888-908.  
Andries, A. M., & Căpraru, B. (2012). Competition and efficiency in EU27 banking systems. 
Baltic Journal of Economics, 12(1), 41-60.  
Angbazo, L. (1997). Commercial bank net-interest margin, default risk, interest rate risk and off-
balance sheet banking. Journal of Banking and Finance, 21, 55-87. 
Arellano, M. & Bover, O. (1995). Another look at the instrumental variable estimation of error 
component models. Journal of Econometrics, 68, 29–51. 
Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence 
and an application to employment equations. Review of Economic Studies, 58, 277–297. 
Athanasoglou, P. P., Brissimis, S. N., & Delis, M. D. (2008). Bank-specific, industry-specific and 
macroeconomic determinants of bank profitability. Journal of International Financial 
Markets, Institutions and Money, 18(2), 121-136. 
Bain, J. S. (1951). Relation of profit rate to industry concentration: American manufacturing, 
1936-1940. The Quarterly Journal of Economics, 65(3), 293-324. 
Barth, J., Caprio, G., & Levine, R. (2013). Bank regulation and supervision in 180 countries from 
1999 to 2011. Journal of Financial Economic Policy, 5(2), 111-219. 
Bashir, A-H. M. (2003). Determinants of profitability in Islamic banks: some evidence from the 
Middle East. Islamic Economic Studies, 11(1), 31–57. 
University of Ghana http://ugspace.ug.edu.gh
179 
 
Battese, G., & Coelli, T. (1992). Frontier production functions, technical efficiency and panel data: 
with application to paddy farmers in India. Journal of Productivity Analysis, 3, 153–169. 
Beck, T., & Cull, R. (2013). Banking in Africa. Policy Research Working Paper 6684. The World 
Bank Development Research Group, Finance and Private Sector Development. 
Beck, T., & Hesse, H. (2009). Why are interest spreads so high in Uganda? Journal of Development 
Economics, 88(2), 192-204. 
Beck, T., De Jonghe, O., & Schepens, G. (2013). Bank competition and stability: cross-country 
heterogeneity. Journal of Financial Intermediation, 22, 218–244.  
Beck, T., Demirgüc-Kunt, A., & Levine, R. (2006). Bank concentration, competition, and crises: 
first results. Journal of Banking & Finance, 30(5), 1581–1603. 
Beck, T., Demirgüc-Kunt, A., & Maksimovic, V. (2004). Bank competition and access to finance: 
international evidence. Journal of Money, Credit and Banking, 36, 627–648. 
Beck, T., Fuchs, M., & Uy, M. (2009). Making finance work for Africa – the role of governments 
in a global world in crisis. Regional economic outlook: sub-Saharan Africa. International 
Monetary Fund.  
Beck, T., Maimbo, S. M., Faye, I., & Triki, T. (2011). Making finance work for Africa-financing 
Africa: through the crisis and beyond. ADB WP.  
Berger, A. N. (1995). The profit-structure relationship in banking-tests of market-power and 
efficient-structure hypotheses. Journal of Money, Credit and Banking, 27(2), 404-431. 
Berger, A. N., Bonime, S. D., Covitz, D. M., & Hancock, D. (2000). Why are bank profits so 
persistent? The roles of product market competition, informational opacity, and 
regional/macroeconomic shocks. Journal of Banking & Finance, 24(7), 1203-1235. 
University of Ghana http://ugspace.ug.edu.gh
180 
 
Berger, A.N., & Hannan, T.H. (1998). The efficiency cost of market power in the banking industry: 
a test of the ‘quiet life’ and related hypotheses. Review of Economics and Statistics, 8(3), 
454–465. 
Berger, A. N., Klapper, L. F., & Turk-Ariss, R. (2009). Bank competition and financial stability. 
Journal of Financial Services Research, 35(2), 99-118. 
Berger, A.N., & Mester, L.J. (1997). Inside the black box: what explains differences in the 
efficiencies of financial institutions? Journal of Banking & Finance 21(7), 895–947. 
Besanko, D., & Thakor, A. (1993). Relationship banking, deposit insurance and bank portfolio. In 
C. Mayer & X. Vives (Eds.), Capital markets and financial intermediation (292–318). 
Cambridge University Press, Cambridge, UK. 
Blundell, R. W., & Bond, S. R. (1998). Initial conditions and moment restrictions in dynamic panel 
data models. Journal of Econometrics, 87(1), 115–143. 
Boone, J. (2008). A new way to measure competition. The Economic Journal, 118(531), 1245-
1261. 
Boone, J., & Weigand, J. (2000). Measuring competition in the Dutch manufacturing sector: how 
are cost differentials mapped into profit differentials? CPB Working Paper No. 131, Den 
Haag.  
Boubakri, N., Cosset, J. C., Fischer, K., & Guedhami, O. (2005). Privatization and bank 
performance in developing countries. Journal of Banking & Finance, 29(8), 2015-2041.  
Boyd, J., & De Nicolo, G. (2005). The theory of bank risk taking and competition revisited. 
Journal of Finance, 60(3), 1329–1343. 
University of Ghana http://ugspace.ug.edu.gh
181 
 
Brissimis, S. N., Delis, M. D., & Papanikolaou, N. I. (2008). Exploring the nexus between banking 
sector reform and performance: evidence from newly acceded EU countries. Journal of 
Banking & Finance, 32(12), 2674-2683. 
Caminal, R., & Matutes, C. (2002). Market power and bank failures. International Journal of 
Industrial Organization 20, 1341–1361. 
Carbo V. S., & Rodriguez F. F. (2007). The determinants of bank margins in European 
banking. Journal of Banking & Finance, 31(7), 2043-2063. 
Carletti, E. (2008). Competition and regulation in banking. In A.W.A Boot & A. Thakor (Eds.), 
Handbook of Financial Intermediation and Banking. Elsevier, Amsterdam. 
Carow, K. A., & Kane, E. J. (2002). Event-study evidence of the value of relaxing long-standing 
regulatory restraints on banks, 1970–2000. The Quarterly Review of Economics and 
Finance, 42(3), 439-463. 
Casu, B., & Girardone, C. (2006). Bank competition, concentration and efficiency in the single 
European market. The Manchester School, 74(4), 441-468. 
Casu, B., & Girardone, C. (2009). Testing the relationship between competition and efficiency in 
banking: a panel data analysis. Economics Letters, 105(1), 134-137. 
Chiorazzo, V., Milani, C., & Salvini, F. (2008). Income diversification and bank performance: 
evidence from Italian banks. Journal of Financial Services Research, 33, 181–203.  
Chirwa, E. W., & Mlachila, M. (2004). Financial reforms and interest rate spreads in the 
commercial banking system in Malawi, IMF staff papers 51(1). 
Chortareas, G. E., Garza‐Garcia, J. G., & Girardone, C. (2011). Banking sector performance in 
Latin America: market power versus efficiency. Review of Development Economics, 15(2), 
307-325. 
University of Ghana http://ugspace.ug.edu.gh
182 
 
Chortareas, G. E., Garza-García, J. G., & Girardone, C. (2012). Competition, efficiency and 
interest rate margins in Latin American banking. International Review of Financial 
Analysis, 24, 93-103. 
Chortareas, G. E., Girardone, C., & Ventouri, A. (2012). Bank supervision, regulation, and 
efficiency: evidence from the European Union. Journal of Financial Stability, 8(4), 292-
302. 
Chortareas, G. E., Girardone, C., & Ventouri, A. (2013). Financial freedom and bank efficiency: 
evidence from the European Union. Journal of Banking & Finance, 37(4), 1223-1231. 
Chronopoulos, D. K., Liu, H., McMillan, F. J., & Wilson, J. O. (2013). The dynamics of U.S. bank 
profitability. The European Journal of Finance, 21(5), 426-443. 
Claessens, S., & Laeven, L. (2004). What drives bank competition? Some international 
evidence. Journal of Money, Credit and Banking, 36(3), 563-583. 
Claeys, S., & Vennet, V. R. (2008). Determinants of bank interest margins in Central and Eastern 
Europe: a comparison with the west. Economic Systems, 32 (2), 197-216. 
Coccorese, P., & Pellecchia, A. (2010). Testing the ‘quiet life’ hypothesis in the Italian banking 
industry. Economic Notes, 39(3), 173-202. 
Da Silva, G. J. C., Oreiro, J. L., de Paula, L. F., & Sobreira, R. (2007). Macroeconomic 
determinants of banking spread in Brazil: An empirical evaluation. In Proceedings of the 
35th Brazilian Economics Meeting. 
De Guevara, J. F., & Maudos, J. (2007). Explanatory factors of market power in the banking 
system. The Manchester School, 75(3), 275-296. 
De Guevara, J. F., Maudos, J. & Pérez, F. (2005). Market power in European banking. Journal of 
Financial Services Research, 27(2), 109–137. 
University of Ghana http://ugspace.ug.edu.gh
183 
 
Delis, M. D. (2012). Bank competition, financial reform, and institutions: the importance of being 
developed. Journal of Development Economics, 97(2), 450-465. 
Delis, M. D., Molyneux, P., & Pasiouras, F. (2011). Regulations and productivity growth in 
banking: evidence from transition economies. Journal of Money, Credit and 
Banking, 43(4), 735-764. 
Delis, M. D., & Tsionas, E. G. (2009). The joint estimation of bank-level market power and 
efficiency. Journal of Banking & Finance, 33(10), 1842-1850. 
Demirgüc-Kunt, A., & Huizinga, H. (1999). Determinants of commercial bank interest margins 
and profitability: Some international evidence. World Bank Economic Review, 13(2), 379–
408.  
Demirgüc-Kunt, A., & Levine, R. (2000). Bank concentration: cross-country evidence. World 
Bank, Mimeo, October. 
Demirgüc-Kunt, A., Laeven, L., & Levine, R. (2004). Regulations, market structure, institutions 
and the cost of financial intermediation. Journal of Money, Credit, and Banking, 36, 593–
622. 
Demsetz, H. (1973). Industry structure, market rivalry and public policy, Journal of Law and 
Economics, 16(1), 1-9. 
Diallo, B. (2015). Bank competition and crises revisited: new results. Economics Letters, 129, 81–
86. 
Dietrich, A., & Wanzenried, G. (2011). Determinants of bank profitability before and during the 
crisis: evidence from Switzerland. Journal of International Financial Markets, Institutions 
and Money, 21(3), 307–327. 
University of Ghana http://ugspace.ug.edu.gh
184 
 
Dietrich, A., & Wanzenried, G. (2014). The determinants of commercial banking profitability in 
low-, middle-, and high-income countries. The Quarterly Review of Economics and 
Finance, 54(3), 337–354. 
Ding, C. G., Wu, C. H., & Chang, P. L. (2013). The influence of government intervention on the 
trajectory of bank performance during the global financial crisis: a comparative study 
among Asian economies. Journal of Financial Stability, 9(4), 556-564. 
Eldomiaty, T.I. (2007), Determinants of corporate capital structure: evidence from an emerging 
economy, International Journal of Commerce and Management, 17(1/2), 25-43. 
Elsas, R., Hackethal, A., & Holzhauser, M. (2010). The anatomy of bank diversification. Journal 
of Banking and Finance, 34(6), 1274–1287. 
Estrada, D., Gomez, E., & Orozco, I. (2006). Determinants of interest margins in Columbia. Banco 
Dela Republica, Series Borradores de Economia No. 393. 
European Central Bank (2010). Beyond ROE – how to measure bank performance. Appendix to 
the report on EU banking structures. ECB. 
Flamini, V., Schumacher, L., & McDonald, C. A. (2009). The determinants of commercial bank 
profitability in sub-Saharan Africa. International Monetary Fund wp/09/15. 
 Forssbæck, J., & Shehzad, C. T. (2014). The conditional effects of market power on bank risk—
cross-country evidence. Review of Finance, 1-42. 
Fosu, S. (2013). Banking competition in Africa: sub-regional comparative studies. Emerging 
Markets Review, 15, 233-254. 
Fu, X. M., & Heffernan, S. (2009). The effects of reform on China’s bank structure and 
performance. Journal of Banking & Finance, 33(1), 39-52. 
University of Ghana http://ugspace.ug.edu.gh
185 
 
Fu, X. M., Lin, Y. R., & Molyneux, P. (2014). Bank competition and financial stability in Asia 
Pacific. Journal of Banking & Finance, 38(1), 64-77.  
Fungáčová, Z., & Poghosyan, T. (2011). Determinants of bank interest margins in Russia: Does 
bank ownership matter? Economic systems, 35(4), 481-495. 
Fungacova, Z., & Weill, L. (2013). Does competition influence bank failures? Evidence from 
Russia. Economics of Transition, 21(2), 301–322.  
Gaganis, C., & Pasiouras, F. (2013). Financial supervision regimes and bank efficiency: 
international evidence. Journal of Banking & Finance, 37(12), 5463-5475. 
Garcia-Herrero, A., Gavila, S., & Santabarbara, D. (2009). What explains the low profitability of 
Chinese banks? Journal of Banking and Finance, 33(11), 2080–2092. 
Gischer, H., & Juttner, D.J. (2003). Global competition, fee income and interest rate margins of 
banks. Kredit and Kapital 36(3), 368–394. 
Goddard, J., Liu, H., Molyneux, P., & Wilson, J. O. S. (2011). The persistence of bank 
profit. Journal of Banking & Finance, 35(11), 2881-2890. 
Goddard, J., Liu, H., Molyneux, P., & Wilson, J. O. S. (2013). Do bank profits 
converge? European Financial Management, 19(2), 345-365. 
Goddard, J., Molyneux, P., & Wilson, J. O. S. (2004). Dynamics of growth and profitability in 
banking. Journal of Money, Credit and Banking, 36(6) 1069-1090. 
Greene, W. (2005). Reconsidering heterogeneity in panel data estimators of the stochastic frontier 
model. Journal of Econometrics, 126, 269–303. 
Gschwandtner, A. (2005). Profit persistence in the ‘very’ long run: evidence from survivors and 
exiters. Applied Economics, 37, 793–806.  
University of Ghana http://ugspace.ug.edu.gh
186 
 
Gschwandtner, A. (2012). Evolution of profit persistence in the USA: evidence from three 
periods. The Manchester School, 80(2), 172-209. 
Gwartney, J., Lawson, R. & Hall, J. (2014). Economic freedom of the world: 2014 annual report, 
Vancouver: The Fraser Institute. 
Hafer, R. W. (2013). Economic freedom and financial development: international evidence. Cato 
Journal, 33(1), 111-126. 
Hakenes, H., & Schnabel, I. (2011). Capital regulation, bank competition, and financial 
stability. Economics Letters, 113(3), 256-258. 
Hassan, M. K., Sanchez, B., Ngene, G. M., & Ashraf, A. (2012). Financial liberalization and 
foreign bank entry on the domestic banking performance in MENA countries. African 
Development Review, 24(3), 195-207. 
Hawtrey, K., & Liang, H. (2008). Bank interest margins in OECD countries. The North American 
Journal of Economics and Finance, 19(3), 249-260.  
Hellmann, T., Murdock, K., & Stiglitz, J. (2000). Liberalization, moral hazard in banking, and 
prudential regulation: Are capital requirements enough? American Economic Review, 90, 
147–165. 
Heritage Foundation (2015). Index of economic freedom. The Heritage Foundation. Retrieved 
from http://www.heritage.org/index. 
Hicks, J. (1935). The theory of monopoly. Econometrica, 3, 1–20. 
Ho, T.S., & Saunders, A. (1981). The determinants of bank interest margins: theory and empirical 
evidence. Journal of Financial and Quantitative Analysis, 16, 581–600. 
Honohan, P., & Beck, T. (2007). Making finance work for Africa. The World Bank. 
University of Ghana http://ugspace.ug.edu.gh
187 
 
Hossain, M. (2012). Financial reforms and persistently high bank interest spreads in Bangladesh: 
pitfalls in institutional development? Journal of Asian Economics, 23(4), 395-408. 
Jeon, Y. and Miller, S. M., (2002). Bank concentration and performance. Economics Working 
Papers, Paper 200225, University of Connecticut. 
Jiang, N., & Kattuman, P. A. (2010). Intensity of competition in china: profitability dynamics of 
Chinese listed companies. Asia Pacific Business Review, 16(3), 461-481. 
Jimenez, G., Lopez, J. A., & Saurina, J. (2013). How does competition affect bank risk-taking? 
Journal of Financial Stability, 9, 185– 195. 
Kablan, A. S. (2010). Banking efficiency and financial development in sub-Saharan Africa (No. 
10/136). International Monetary Fund. 
Kalluci, I. (2010). Determinants of net interest margin in the Albanian banking system. Working 
Paper Series, April, Bank of Albania. 
Kasman, A., Tunc, G., Vardar, G., & Okan, B. (2010). Consolidation and commercial bank net 
interest margins: evidence from the old and new European Union members and candidate 
countries. Economic Modelling, 27, 648–655. 
Keeley, M. (1990). Deposit insurance, risk, and market power in banking. American Economic 
Review, 80, 1183–200. 
Klein, M.A. (1971). A theory of the banking firm. Journal of Money, Credit, and Banking, 3, 205-
218. 
Koetter, M., Kolari, J., & Spierdijk, L. (2008). Efficient competition? Testing the ‘‘quiet life” of 
U.S.  banks with adjusted Lerner indices. Working Paper, Groningen University. 
University of Ghana http://ugspace.ug.edu.gh
188 
 
Koetter, M., Kolari, J., & Spierdijk, L. (2012). Enjoying the quiet life under deregulation? 
Evidence from adjusted Lerner indices for U.S. Banks. Review of Economics and Statistics, 
94 (2), 462–480. 
Koetter, M., & Poghosyan, T. (2009). The identification of technology regimes in banking: 
Implications for the market power-fragility nexus, Journal of Banking & Finance 33(8), 
1413–1422. 
Kohler, M., (2015). Which banks are more risky? The impact of business models on bank stability. 
Journal of Financial Stability, 16, 195–212. 
Kouki, I., & Al-Nasser, A. (2014). The implication of banking competition: evidence from African 
countries. Research in International Business & Finance. Retrieved from 
http://dx.doi.org/10.1016/j.ribaf.2014.09.009.  
Laeven, M. L., & Valencia, F. (2008). The use of blanket guarantees in banking crises (No. 8-250). 
International Monetary Fund. 
Laeven, L., & Valencia, F. (2012). The use of blanket guarantees in banking crises. Journal of 
International Money and Finance, 31(5), 1220-1248. 
Lee, C. C., Hsieh, M. F., & Dai, H. W. (2012). How does foreign bank ownership in the banking 
sector affect domestic bank behavior? A dynamic panel data analysis. Bulletin of Economic 
Research, 64(1), 86-108. 
Lin, K., Doan, A. T., & Doong, S.  (2016). Changes in ownership structure and bank efficiency 
in Asian developing countries: The role of financial freedom. International Review of 
Economics & Finance, 43, 19–34. 
Liu, H., & Wilson, J. O. (2010). The profitability of banks in Japan. Applied Financial 
Economics, 20(24), 1851-1866. 
University of Ghana http://ugspace.ug.edu.gh
189 
 
Liu, H., Molyneux, P., & Wilson, J. O. (2013). Competition and stability in European banking: a 
regional analysis. The Manchester School, 81(2), 176–201.  
Mamatzakis, E., Kalyvas, A. N., & Piesse, J. (2013). Does regulation in credit, labour and business 
matter for bank performance in the EU-10 Economies? International Journal of the 
Economics of Business, 20(3), 341-385. 
Martinez-Miera, D., & Repullo, R. (2010). Does competition reduce the risk of bank failure? 
Review of Financial Studies, 23, 3638–3664. 
Maudos, J., & de Guevara, J. F. (2007). The cost of market power in banking: Social welfare loss 
vs. Cost inefficiency. Journal of Banking & Finance, 31(7), 2103-2125. 
Maudos, J., & de Guevara, J. F. (2004). Factors explaining the interest margin in the banking 
sectors of the European Union. Journal of Banking & Finance, 28(9), 2259-2281.  
Maudos, J., & Solis, L. (2009). The determinants of net interest income in the Mexican banking 
system: an integrated model. Journal of Banking & Finance, 33(8), 1920–1931.  
McShane, R.W., & Sharpe, I.G. (1985). A time series-cross-section analysis of the determinants 
of Australian trading bank loan-deposit interest margins: 1962-1981. Journal of Banking 
and Finance 9, 115-136. 
Meeusen, W., van den Broeck, J. (1977). Efficiency estimation from Cobb-Douglas production 
functions with composed error. International Economic Review, 18, 435-444. 
Mirzaei, A., Moore, T., & Liu, G. (2013). Does market structure matter on banks’ profitability and 
stability? Emerging vs. Advanced economies. Journal of Banking & Finance, 37(8), 2920-
2937. 
University of Ghana http://ugspace.ug.edu.gh
190 
 
Mirzaei, A., & Moore, T. (2014). What are the driving forces of bank competition across different 
income groups of countries ? Journal of International Financial Markets, Institutions & 
Money, 32, 38–71.  
Mlambo, K., & Ncube, M. (2011). Competition and efficiency in the banking sector in South 
Africa. African Development Review, 23(1), 4-15. 
Moyo, J., Nandwa, B., Oduor, J., & Simpasa, A. (2014). Financial sector reforms, competition and 
banking system stability in sub-Saharan Africa. New Perspectives.  
Mu, Y., Phelps, P., & Stotsky, J. G. (2013). Bond markets in Africa. Review of Development 
Finance, 3, 121–135. 
Mueller, D.C. (1977). The persistence of profits above the norm. Economica, 44, 369–80. 
Naceur, S. B., & Omran, M. (2011). The effects of bank regulations, competition, and financial 
reforms on banks' performance. Emerging Markets Review, 12(1), 1-20. 
Nguyen, J. (2012). The relationship between net interest margin and noninterest income using a 
system estimation approach. Journal of Banking & Finance, 36, 2429–2437. 
Oreiro, J. L., & de Paula, L. F. (2010). Macroeconomic determinants of bank spread in Latin 
America: a recent analysis with special focus on Brazil. International Review of Applied 
Economics, 24(5), 573-590.  
Panzar, J. C., & Rosse, J. N. (1987). Testing for" monopoly" equilibrium. The Journal of Industrial 
Economics, 443-456. 
Pasiouras, F., & Kosmidou, K. (2007). Factors influencing the profitability of domestic and foreign 
commercial banks in the European Union. Research in International Business and Finance, 
21 222–237. 
University of Ghana http://ugspace.ug.edu.gh
191 
 
Pasiouras, F., Tanna, S., & Zopounidis, C. (2009). The impact of banking regulations on banks' 
cost and profit efficiency: cross-country evidence. International Review of Financial 
Analysis, 18(5), 294-302. 
Perera, S., Skully, M., & Chaudrey, Z. (2013). Determinants of commercial bank profitability: 
South Asian evidence. Asian Journal of Finance & Accounting, 5(1), 365-380. 
Peria, M. S. M., & Mody, A. (2004). How foreign participation and market concentration impact 
bank spreads: evidence from Latin America. Journal of Money, Credit and Banking, 36(3), 
511-537. 
Poghosyan, T. (2013). Financial intermediation costs in low income countries: the role of 
regulatory, institutional, and macroeconomic factors. Economic Systems, 37(1), 92-110.  
Poshakwale, S. S., & Qian, B. (2011). Competitiveness and efficiency of the banking sector and 
economic growth in Egypt. African Development Review, 23(1), 99-120. 
Pruteanu-Podpiera, A., Weill, L., & Schobert, F. (2008). Banking competition and efficiency: a 
micro-data analysis on the Czech banking industry. Comparative Economic Studies, 50(2), 
253-273. 
Saunders, A., & Schumacher, L. (2000). The determinants of bank interest rate margins: an 
international study. Journal of International Money and Finance, 19(6), 813-832. 
Schaeck, K., & Čihák, M. (2008). How does competition affect efficiency and soundness in 
banking? New empirical evidence. ECB working paper no. 932. 
Schaeck, K., Čihák, M., & Wolfe, S. (2009). Are competitive banking systems more stable? 
Journal of Money, Credit and Banking, 41(4), 711-734. 
Schaeck, K., & Čihák, M. (2014). Competition, efficiency, and stability in banking. Financial 
Management, 43(1), 215 – 241. 
University of Ghana http://ugspace.ug.edu.gh
192 
 
Sealey, C., & Lindley, J. T. (1977). Inputs, outputs and a theory of production and cost at 
depository financial institutions. Journal of Finance, 32(4), 1251–1266. 
Seelanatha, L. (2010). Market structure, efficiency and performance of banking industry in Sri-
Lanka. Banks and Bank Systems, 5(1), 20-31. 
Senbet, L. W., & Otchere, I. (2006). Financial sector reforms in Africa: Perspectives on issues and 
policies, Annual World Bank Conference on Development Economics, St. Petersburg, 
Russia. 
Shepherd, W. G. (1982). Economies of scale and monopoly profits. In John V. Craven (Eds), 
Industrial organization, antitrust, public policy, Boston, Mass.: Kluwer Nijhoff. 
Smimou, K., & Karabegovic, A. (2010). On the relationship between economic freedom and equity 
returns in the emerging markets: evidence from the Middle East and North Africa (MENA) 
stock markets. Emerging Markets Review, 11(2), 119-151. 
Smirlock, M. (1985). Evidence on the (non) relationship between concentration and profitability 
in banking. Journal of Money, Credit, and Banking, 17 (1), 69–83. 
Stephan, A., & Tsapin, A. (2008). Persistence and determinants of firm profit in emerging 
markets. Applied Economics Quarterly, 54(4), 231-253.  
Stern, J. M., Stewart, G. B., & Chew, D. H. (1995). The EVA® financial management system. 
Journal of Applied Corporate Finance, 8(2), 32-46. 
Stiroh, K. J., & Rumble, A. (2006). The dark side of diversification: the case of U. S. financial 
holding companies. Journal of Banking and Finance, 30(8), 2131–2161. 
Sufian, F., & Habibullah, M. S. (2010). Does economic freedom foster banks’ performance? Panel 
evidence from Malaysia. Journal of Contemporary Accounting & Economics, 6(2), 77-91.  
University of Ghana http://ugspace.ug.edu.gh
193 
 
Tabak, B. M., Fazio, D. M., & Cajueiro, D. O. (2012). The relationship between banking market 
competition and risk-taking: do size and capitalization matter? Journal of Banking & 
Finance 36(12), 3366–3381. 
Tregenna, F. (2009). The fat years: the structure and profitability of the U. S banking sector in the 
pre-crisis period. Cambridge Journal of Economics, 33(4), 609-632. 
Trujillo-Ponce, A. (2013). What determines the profitability of banks? Evidence from Spain. 
Journal of Accounting and Finance, 53, 561–586. 
Turk-Ariss, R. (2010). On the implications of market power in banking: evidence from developing 
countries. Journal of Banking & Finance, 34(4), 765-775. 
Uhde, A., & Heimeshoff, U. (2009). Consolidation in banking and financial stability in Europe: 
empirical evidence. Journal of Banking & Finance, 33(7), 1299-1311. 
Van Leuvensteijn, M., Sørensen, C. K., Bikker, J. A., & Van Rixtel, A. A. (2013). Impact of bank 
competition on the interest rate pass-through in the Euro area. Applied Economics, 45(11), 
1359-1380. 
Weill, L. (2004). On the relationship between competition and efficiency in the EU banking 
sectors. Kredit und Kapital, 37 (3) 329-352. 
Weill, L. (2013). Bank competition in the EU: How has it evolved? International Financial 
Markets Institutions and Money, 26, 100– 112. 
Williams, J. (2012). Efficiency and market power in Latin American Banking. Journal of 
Financial Stability, 8(4), 263-276. 
Windmeijer, F. (2005). A finite sample correction for the variance of linear efficient two step 
GMM estimators. Journal of Econometrics, 126(1), 25–51. 
University of Ghana http://ugspace.ug.edu.gh
194 
 
Wong, K. (1997). On the determinants of bank interest margins under credit and interest rate risks. 
Journal of Banking and Finance, 21, 251-71. 
 
 
 
University of Ghana http://ugspace.ug.edu.gh