Relationship Between Competition and Financial Stability

CHAPTER I. INTRODUCTION

The economy is increasingly competitive, which is considered as an essential driver for promoting productivity (Harris, 1988). Smith (1776) describes this move as an exercise of distribute different resources and foster efficiency. According to Case et al., (2012), competition would lead to more progressive products and services. As a consequence, consumers can capitalise on being given a better selection of products and services and greater quality at good prices. When it comes to the competition in banking industry, there are a wide range of different views. Particularly, in terms of bank management’s perspectives, Besanko and Thakor (1992) and Guzman (2000) maintain that competition can be considered to utilise detrimental influences on both banking and other industries as fiercer competition would result in a cost reduction and smoother approach to finance. Lawrence (1989) suggests that the cost reduction in banking context refers to a decrease in commission fees and lending interest rate, and an increase in deposit interest rate. However, Claessens (2009) argued that the degree of competition in the financial sector could impact the quality of financial products and the efficiency of financial services not to relate to the level of banking innovation. Therefore, according to Mirzaei and Moore (2014), anticompetitive behaviours can result in a decrease in efficiency or even a seriously downward trend in banking industry. Moreover, as mentioned by Boot and Thakor (2000), this issue is also concerned by both academics and policy makers when it comes to the industry’s soundness and health.  Indeed, there have been debated on the influences of competition on financial stability in banking context for years, but this issue has been seriously taken into account since the financial crisis in 2008 (Caletti, 2010; Carletti, 2008; Acharya and Richarson, 2009, Beck., et al, 2010; Beck, 2008; and OECD, 2011).

When competition which is preferred by bank management is still highly contentious, there are two different perspectives connecting to it, including ‘competition-fragility’ perspectives introduced by Marcus (1984); and ‘competition-stability’ hypothesis developed by Boyd and De Nicolo (2005). According to the former, Marcus (1984) suggests that it is unlikely that banks can acquire monopoly rent in the competitive market and he also maintains that exaggerated competition is found to erode the market power. Thus, it would cause a reduction in capital ratios and profits, which then would negatively lead to franchise value (is also charter value that indicate the ability of a bank to carry on their business in long term and it is mirrored as a part of a bank’s share value). According to Marcus (1984) and Keeley (1990), it negatively influences banks as they have to deal with demand-supply shocks and to recover the financial losses, they are likely to take part in more risk projects. More specifically, Keeley (1990) examines a harder competition in the banking industry and deregulation in the United States in 1980s when their restrictions in state branches make the monopoly rent collapsed and result in a dramatic rise in bank failures (check it).  Furthermore, an empirical research conducted by Jimenez et al. (2007) indicates that a higher level of competition is found to be connected with more portfolios of extremely risky loans which refer to be non-performing loans in Spain. In addition, charter value would be eroded and moral hazard behaviours in banks would rise relatively because of the retraction of interest rate ceiling on deposits and savings.

By contrast, ‘competition – stability’ is a contrary view has developed by Boyd and De Nicolo (2005) in which they argue that the competition in banking industry can improve bank’s stability. It is believed that the harder competition is the lower interest rate for loans because when banks provide less interest rates, borrowers are encouraged to repay loans to banks and demotivated to take risky projects. It then can lead to reduce moral hazard behaviours. Moreover, they also suggest that a drop in interest rate may lead to a careful selection of less risky borrowers as regarding to adverse selection theory (need in-text reference..,.). According to it, in a more competitive market, interest rate refers to be lower, net subsidies appear to be safer, and Too-Big-To-Fail problem of bank failures are less likely to occur as it would indirectly be sheltered by the Government. Similar to Boyd and De Nicolo (2005), Fiordelisi and Mare (2014) maintain that a competitive market environment in the banking industry can positively influence bank financial stability as the market power can negatively result in soundness in banks.

Surprisingly, both views are supported by some empirical evidence. Specifically, a study conducted by Matinez-Miera and Repullo (2010) generated a mixed result of the association between competition and stability in which they concluded that the competition is non-linearly associated with stability and the relationship is U-shape. Furthermore, Berger et al. (2009) maintain that when it comes to the influences of bank competition on bank stability, there is no need for two individual views because of the fact that they can totally coexist and bring both fragility and stability.

It has been favourable for both academics and policy makers to explore the association between competition and financial stability in banking industry (….in-text ref). Particularly, the empirical studies on this research area were normally undertaken in the same groups of countries in the same continent, for instance, developing countries in South East Asia. By contrast, these academic studies are still under researched in a multi-continent group of countries, for instance, G7-the seven largest advanced economies throughout three different continents in the world. Thus, this research paper is to examine the connection between competition and financial stability in this group of countries with further evidence from 2010 to 2015.

It is important for all literature relating to this topic to be critically analysed. Besides, research data will be collecting from SNL financial and then they are put in Excel before getting ready for running regression. Nevertheless, there are some main limitations in this research, including data collection and time frame.  Furthermore, some helpful sources of data gathering are not available or have not been updated to the latest versions, for instance, World Bank database has not updated some of their data as such Entry restriction, Deposit Insurance, and Capital Requirement, and so on. As a consequence, these drawbacks may cause unexpected influences on the result of the employed model.

The research paper will be organised in a logic manner. First and foremost, the first chapter will be an introduction, followed by the theoretical and empirical literature review which will be presented in chapter 2 in order to the association between the bank competition and financial stability in both main points of view and in different contexts. The following chapter is about data collection and the establishment of models, which helps to measure the degree of each element. At the same chapter, the regression will be established and a number of tests will be conducted. After that, chapter 4 aims to analyse the results which are generated in chapter 3. Last but not least, chapter 5 will be discussions and an overall conclusion of the whole research.

CHAPTER 2: LITERATURE REVIEW

This chapter aims to discuss a wide range of both theoretical and empirical literature relating to the association between competitions and financial stability in the sphere of banking services. The relationship is found to be complex as it has been broadly researched and even strongly debated in the academic sphere and between policy makers, particularly since the financial crisis in 2008 ( Beck, 2008; Acharya and Richardson, 2009; Beck et al., 2010; Carletti, 2008; Careletti, 2010; OECD, 2011). Specifically, many academic research papers have shown a broad range of models, measurement instruments and variables in order to examine this connection. In this paper, two contradicting views of this association will be critically analysed, consisting of ‘competition-fragility’ perspective which has been supported by Marcus (1984), Keeley (1990), Allen and Gale (2000; 2004) for instance, and ‘competition-stability’ hypothesis which was maintained by a number of modern literatures such as Boyd and De Nicolo (2005), Fu et al. (2014); Fiordelisi and Mare (2014) and so on. Nevertheless, most of the literature was studied and has been developed in the same group of countries in the same region as such an emerging market in South East Asia (…in-text references..) or developed markets in the European countries (in-text ref). Hence, it seems that critically studying the relationship between bank competition and stability needs a diversity of contexts in order to trace the differences in both theoretical and empirical perspectives. Furthermore, the level of competition in banking services is closely linked to banking structure, Government regulations, institutions, and other factors which fundamentally characterise a country or a context since they are crucial for welfare-related public policy set in the banking industry (Mirzael and Moore, 2014). Those authors also suggest that the differences in regulation and economic orientation in different contexts can create a diversity in competition shapes since regulations play an important role in forming competition.

2.1. ‘Competition-fragility’ hypothesis

According to Markus (1984), ‘competition-fragility’ perspective maintains that competition in the banking industry have a negative association with financial stability. In other words, the less competitive market in banking services is the less fragile financial stability. This statement is critically underpinned by ‘charter/franchise value’ perspective which is empirically developed by Marcus (1984) and Keeley (1990). It is indicated that the competition in banks is found to erode the franchise value, further be a driver for taking at-risk projects due to a reduction in bank’s charter value, and eventually disrupt the financial stability (Keeley, 1990). According to Ren and Schmit (2006), franchise value or charter value is defined as an intangible asset which is differentiated from the value of tangible assets and it is particularly identified from a bank’s market power, distribution networks, goodwill, renewal rights and expertise knowledge shown in the financial risks generating from business book, in an extent to which the bank continues running their business. Furthermore, Demsetz et al. (1996) suggest that the more franchise value refers to the less risk-taking projects as banks who have greater charter value are appear to be reluctant to involve in risky activities that can cause bank’s indebtedness. Through a study in U.S banks, these authors also indicate that the stronger competition will result in a rise in insolvency ratios and asset risk level.  Similar to it, a higher franchise/charter value, as is stated by Marcus (1984) and Keeley (1990), can be generated from more market power, which may demotivate bank’s managers and shareholders to decide on taking risky projects. This perspective is widely supported by a range of studies, for instance, Saunders and Wilson (1996) generated a same result as Keeley (1990) over a sample of US banks in the 20^th century. In addition, supporting to the model used by Marcus (1984) and Keely (1990), a study conducted by Brewer and Saidenberg (1996) through a number of publicly traded thrifts in the US demonstrates that franchise/charter value has totally negative association with risks which are identified by measuring the unpredictability of bank’s market stock prices. Another study undertaken by Bofondi and Gobbli (2004) in Italy’s banking industry points out that higher market power or weak competitive market would be linked to more opportunities of bank solvency and less credit risks. Similarity, Hellmann et al. (2000) add that the liberalisation in Japan in the 1990s fosters the bank competition in this country and it weakens domestic bank’s franchise/charter value and these banks’ profit relatively. As a consequence, it leads to an exacerbation in both financial system in Japan and financial crisis in the East Asia. The Marcus (1984)’s and Keeley (1990)’s view is also supported by a study in Spain in which Salas and Saurina (2003) found an obvious association between Tobin’s q, solvency and non-performing loans ratio when applying Keeley (1990)’s framework in Spain’s banking context. In summary, it is found that higher bank concentration/ bank power is maintained to have positive relationship with greater opportunities of solvency and lower levels of credit risk.

Fu et al. (2014) suggest that stability appears to be influenced by competition in the market through contagion effects. Moreover, inconstant banking condition is intensified by competition towards two main ways (Vive, 2010). One of them suggests that competition can stimulate risk-taking activities in order to push bank’s franchise value and as a result, it may lead to a higher possibility of bank collapse. The second path points out that tension and panic in banks can be caused by an extent to which there are serious problems in the coordination with depositors and investors on the accountability side. Martutues and Vives (1996) also suggest that these stakeholders are found to be sensitive and totally conscious to perceive any problems that can be arisen such as the probability of bank failure. Therefore, when a serious problem is perceived, they can easily withdraw deposits or discount banknotes, which can seriously cause bank runs or even the breakdown of the whole system. Furthermore, it is believed that financial crisis can absolutely be an issue when the fragility exists in the financial market (Alen and Gale, 2004). Additionally, Kiyotaki and Moore (1997) add that the enormous effects of a financial shock are impossible to be controlled and it can lead to the whole system collapse. The perfect competition can result in liquidity problems because banks are considered as price takers who are not motivated to help illiquid peers. Eventually, those illiquid banks would easily bankruptcy as a result of the whole banking industry. However, Saez and Shi (2004) argue that banks can strategically cooperate to overcome the liquidity issues, but only in an extent to which the financial market is imperfectly competitive. Alen and Gale (2000) point out that an environment to which the market power is higher and competition is lower, can be favourable for banks to control and supervise; thus, they would be more confident to face with financial shocks. Furthermore, Fu et al. (2014) maintain that speaking of a provision of credit monitoring, big-sized financial institutions appears to acquire more competitive advantages. When it comes to an involvement in ‘credit rating’ procedure, if they only pay attention on high-creditworthiness investments, they can generate profit through each investment and further improve banking system’s soundness (Boot and Thakor, 2000).

Specifically, Beck et al. (2006), who conducted a research using data collected from 69 nations from 1980 to 1997, indicate 47 stages of crisis. Here, the term ‘systemic crisis’ is introduced as a stage in which the whole system is financially underwent, and the non-performing loan ratio is above 10% of total asset if it is without any intervention by the Government and other authorities. The research outcome indicates a banking system with higher market power can rapidly response to and positively impact on each stage of the financial shock. Additionally, these authors demonstrate that it is less likely that financial crises can occur in the less competitive banking environment since the government and authorities strictly control institutional, micro and macroeconomic factors through their regulations and policies. Similarity, Allen and Gale (2004) also state that banking crisis is more likely to happen in the environment with higher level of competition. It is due to the inaccessibility of financial services given by influential issuers in which banks may take this advantage to avoid asset losses. Supporting to this, Kasman and Carvallo (2014) argue that bigger-sized financial institutions can harvest financial benefits and weaken the influences of banking fragility. Furthermore, the gained profit from the financial advantages can be considered as an effective protector against shocking financial events or liquidity issues.

The traditional view of ‘competition-fragility’ explains moral hazard issues through the existence of deposit insurance systems which help to reduce the fragility in banks on the one hand, motivate banks to engage in risky projects on the other hand (Fu et al., 2014). Specifically, Craig and Dinger (2013) undertook an investigation of 589 banks from not only retail deposit market but also wholesale capital funding market in the U.S. Here, risks are measured by a number of variables, including stock price volatility, non-performing loans, and ROA while deposit interest rate was the only measurement unit for competition among banks. Eventually, the research outcome concludes that banks refer to involve in at-risk activities when operating in the environment with less market concentration and more competition. For instance, the depositors are offered the higher cost in the retail deposit market, compared with the wholesale funding market, which motivates banks to involve in riskier investments. According to Diamond and Dybvig (1983), when it comes to a more competitive market, financial stability is weakened by the deposit insurance regulation given by the Government. Moreover, Matutes and Vives (1996) maintain that the deposit insurance can be a motive for moral hazards since banks appear to take risks and have weaken monitoring activities for their own borrowers. To deal with this issue, Hellmann et al. (2000) suggest that banks must set the interest rate ceiling for deposits in order to promptly avoid involvements in at-risk projects. It is also suggested that if this interest rate ceiling is put in a competitive market, bank’s charter/franchise value would be eroded and banks would have more motives to become risk-takers (Hellmann et al., 2000). These authors carried out their research which used data collected from the US savings and loans shocks during the period of time of 1980s and 1990s, and the financial crunch in Japan in the 1990s in order to examine the association among three factors, excessive risk-taking cost, social cost, and financial liberalization.

Bank’s size is a factor that should be taken into account when deciding whether bank competition relates to financial stability. Rmakrisman and Thakor (1984), Diamond (1984), and Boyd and Prescott (1986) and Williamson (1986) all maintain that big-sized financial institutions have their intensive attendance in the financial market with higher market power and lower competition; therefore, they are able to reap on benefits of economies of scale and scope. These authors also point out that big-sized banks have a strength of a greater capital compared with smaller peers so that they could have more diversified investing portfolio. What’s more, when these banks are set in high financial positions in the market, they can obtain a much greater credit rationing procession. It can be explained by two paths, (1) larger banks own advanced technological applications which help them to screen their loans and monitor their services, (2) big-sized banks are more likely to invest in huge-value lending and have relatively lower number of loans that need to be monitored and thus take greater care of their lending activities. Additionally, Boot and Greenbaum (1993) suggest that the lower competitive environment is favourable for big-sized banks to gather information from their borrowers, further decrease their information rent, improve the effectiveness of monitoring leading activities, and eventually foster financial stability. Similarity, Allen and Gale (2004) suggest that when regulations and guiding practises are applied properly and effectively, it is easier to expect and control the financial crisis and its probability.

It is found that the number of research studying ‘competition-fragility’ perspective in emerging markets has increased. An investigation in MENA region of 18 countries in the East Asia and North Africa from 2000 to 2008 conducted by Labidi and Mensi (2015) demonstrate the negative association between bank competition and financial stability. In those countries, it is found that the banking markets are highly concentrated and it can sustain the market stability. In the research, interest rate ceiling for deposits for several years is picked to measure its impact on the connection between competition and stability in banking services in those countries. After conducting a research in the same region, Troug and Sbia (2015) also agree that less competitive banking environment would contribute to a more stable financial market. They also add that regulation and supervision play an important role in maintaining and improving the level of stability for the whole banking systems. In terms of regulatory and supervisory practices, Agoraki et al. (2011), who carried out a case-study research in Central and Eastern European countries from 1994 to 2005, state that more competitive market would cause at-risk decisions and the stability in banking services is partially generated by strictly regulatory policies established for banks. According to ‘franchise/charter value’ hypothesis, the authors also point out that risk-taking strategies are motivated in strong competitive environments. It is believed that competition is one of the main factors that specify the charter/franchise value and an increase in this value can make bank’s managers and shareholders indisposed to be risk-takers.  In terms of risk-taking behaviours, Marchrouh and Tarazi (2011), who investigated 12 countries in Asia during the period from 2001 to 2007, empirically point out that it is easier for banks that have more market power in less competitive operating market to obtain quick capital mobilization and hence enhance their stability level in the market.

2.2. ‘Competition – stability’ hypothesis

The alternative ‘competition – stability’ perspective suggests that higher concentrated or less competitive markets negatively influence financial stability in banking industry (Fu et al., 2014). Similarity, Boyd and De Nicolo (2005), the very first authors created this hypothesis, maintain that for banks, competition is a great chance for them to be stable in the strong competitive market. Particularly, a small number of banks makes the market concentrated and here, big-sized banks have strong market power so that they can influence other smaller peers and the whole banking system. It, therefore, is more likely to cause instability in the entire system if a large is supposed to collapse. To deal with it, the bigger banks appears to be given subsidies from the Government and this event is called as ‘too big to fail’ dogma. Similar to it, Miskin (1999, 2006) and Barth et al. (2012b) explain that the Government and authorities have certain tensions about bank downfall when there are few banks in the market with less competition. Thus, the Government tends to provide banks with superior guarantees, which may easily cause moral hazard issues. In other words, as the Government’s concerns that if bank disasters happen, they have more incentives to get involve in at-risk decisions and further worsen the financial volatility (Kane, 2010; Rosenblum, 2011). Additionally, Fu et al. (2014), who state that contagion effects appear to be generated through the influences of competition on banking stability, also indicate that bank collapses are likely to happen in the market with weak competition and a small number of big-sized banks operating in.

Furthermore, the hypothesis is supported by Caminal and Matutues (2002), stating that weak competition in banking environment can reduce credit rationing and make a rise in loan’ size. In other words, banks tend to both increase their willingness to offer credit to their debtors regardless to the creditworthiness and offer excess loans, which may further be a reason for bank collapse. However, the traditional ‘competition-fragility’ hypothesis did not take the borrower’s side into account. On this vein, Boyd and De Nicolo (2005) suggest that lending interest rate is set by big-sized financial institutions in the concentrated market with relative high market power. A rise in interest rate for loans can cause a relative drop in profit for borrowers due to a higher interest cost, which may motivate the borrowers to take riskier activities. In conclusion, moral hazard issues among borrowers are more likely to be fostered in the weak competitive market, which results in a growth of non-performing credits and loan default.

In terms of the association between bank’s size and the complexity of its organisation, Beck et al. (2006a, b) suggest that larger financial institutions tend to be more organisationally complex, and more challenging to control as opposed to smaller peers in the industry. Specifically, Beck et al. (2006a, b) also found that the expansion strategies to penetrate more potential markets, expanding their services or opening  more subsidiaries to achieve their structural expansion process and diversification  are generally erode the bank’s transparency. As a result, it is believed to cause operational risks due to less effective management and weak internal control as a result of bank’s expansion strategies (Fu et al., 2014). Furthermore, Cetorelli et al. (2007) suggest that as the complexity of a bank’s organisation is increased, market discipline would obtain more incentives and regulatory initiatives tend to be less constructive about controlling risk disclosure. On this vein, Beck (2008) also maintain that big-sized banks, that appear to have more complex organisations, are found to be more involved in risky larger financial projects or even international investments. De Nocole et al. (2004), who conducted a study in developed countries including the U.S, Japan, and some European nations, found the positive association between the size of banks and its failures. However, a study of more than 100 holding organisations in the U.S over two decades from 1971, carried out by Boyd and Runkle (1993) shows the negative connection between bank’s size and their instability, even though they could not make a conclusion about the association between competition and bank collapse.

As mentioned above, the traditional view of ‘competitive-fragility’ shows that consolidation in banking services have positive impacts on the stability of the whole market. However, it causes some negative influences, particularly in the occurrence of financial conglomerate phenomena. Here, it is argued that larger banks in the concentrated market tend to expand their financial services, which stimulates banks to control and monitor their activities relating to their new services (Fiordelisi and Mare, 2014). Supporting the idea of ‘competition – stability’ perspective, Berger et al. (2009) who carried out a large research in 23 developed markets confirm that the less competitive banking market banks are running in brings the riskier projects banks tend to take.

When it comes to regulation and institution, the probability of financial shocks in the whole scheme can be diminished through features that can only offered in a competitive banking environment (Beck et al., 2006). On this vein, Schaeck et al. (2010) made a crucial improvement in an advanced study that based on Beck et al. (2006)’s research outcome. Here, they analysed data gathered from 38 developed nations from 1980 to 2003 and apply Panzar and Rosse H-statistic to measure the market competition. As a result, the outcome of the research contribute to the ‘competition – stability’ hypothesis in which it suggests that the banking market with higher concentration and lower competition can lead to higher possibility of systemic disaster.

There is a wide range of scholars holding up for the ‘competition – stability’ perspective. As referring to Amidu and Wolfe (2013), who conducted an investigation on 55 developing countries for 7 years from 2000 to 2007 just before the financial shock in 2008, empirically conclude that stronger competitive banking market could increase bank soundness and significantly enhance bank’s performance measured by ROA and ROE and revenues can be generated from diversified sources. Williams (2012) conducted a research of commercial banks in Latin America, stating that banking yield efficiency reap from the financial reorganisation during the late 1990s. The study shows the influences of the reorganisation on competition in banking services from 1985 to 2010 and particularly concludes that the reform results in a reduction of market concentration and an increase in monopolistic competition and further stimulate financial stability. Turning to Demirguc-Kunt et al. (2010)’s study in MENA region, particularly in Jordan, the authors empirically indicate that the banking market has weak competition and is highly concentrated as there is only a small number of large banks in this country. With very high market power, those banks push the interest rate, which causes inefficiency in bank’s performances. To deal with this issue, the Government make interventions to diminish the interest rate in order to recover bank efficiency; however, this rate is still bigger than that in Jordan’s neighbour countries like Israel or Lebanon. Another study was undertaken in Malaysian banking market over the period of time from 2001 to 2005 shows that the less competitive market can negatively impact larger bank’s efficiency rather than smaller counterparts and it can be seen that this country is facing with diseconomies of scale (Majid et al., 2007). Soedarmono et al. (2013) develop their previous work by an extension in the research time from 1994 to 2009, confirming that banks operate in less competitive market can take advantages on capital mobilization activities, but the acquired advantages are not enough to cover negative influences caused by at-risk behaviours and indebtedness risk. Consequently, the study is supposed to support the ‘competition-stability’ perspective.

2.3. Other opinions for the association between competition and financial stability

As referring to Berger et al. (2009) who state that the connection between competition and bank stability can have both positive and negative sides, it is also argued that competition and concentration can coexist, which can simultaneously generate both stability and fragility. Moreover, Martinez – Miera and Repullo (2010) suggest that there are two paths of the literature, including risk-shifting impact and margin impact. The latter suggests that interest rates tend to be decrease due to a competitive market, which can lead to a drop in bank’s profit gained by loans. As mentioned by Boyd et al. (2004), that profits can improve the bank’s financial solidity through giving them a great ‘capital buffer’ in order to avoid macroeconomics downturn and liquidity issues. Hence, it can be seen that if the profit gained from loans is dropped, banks are more likely to face more risks. The former ‘risk-shifting’ effect suggests that stronger competitive market can result in lower interest rate, which can lead to an improvement in non-performing loan issues (Martinez – Miera and Repullo, 2010). They also point out that risk-shifting impact can be considered as a consequence of a lower competitive market with high concentration, while margin impact is supposed to happen in higher competitive environment. They also figure out that the association between competition and financial stability is U-shaped, which is measured by the probability of bank’s collapse.

As mentioned above, it is argued that although there are more banks that tend to take at-risk projects in the concentrated market, the banking crisis is less likely to occur. It is clarified that banks are likely to offer diversified approaches to balance possible risk exposure in order to maintain their franchise/charter value. Here, a number of methods are announced, for instance, promoting bank’s equity capital, avoiding interest-related risks, and settling credit derivatives.

Competition is not found as a significant factor that influence financial solidity (Berger et al., 2004; Beck, 2008). On this vein, they suggest that financial stability is mostly impacted by the market concentration through different channels. The authors also maintain that market structure is not always an ideal measure for probing competition in the banking market. There is a wide range of works that examine the connection between banking competition and stability in single countries and a group of nations. It is found that cross-country studies have been undertaken more in recent times due to the insufficiency of database for studying in the past, whereas studies in single countries have available database for a long period of time. Zhao et al. (2010) identify some initiatives for deregulations delivered by Indian Government to stimulate competition and the authors also figure out the ones that negatively influence bank’s behaviours on risk-taking situations. Eventually, they confirm that stronger competition can foster risk-taking behaviours, supporting for ‘competition-fragility’ perspective. When it comes to Turkish banking environment, Ak Kocabay (2009) and Bazzana (2010) figure out extents to which the degree of competition can relate to non-performing loan risk. Ak Kocabay (2009)’s study outcome varies depending on the changes in model specifications, while Bazzana (2010)’s study suggests that an increase in a bank’s market power can lead to a reduction in credit risks. The linkage between bank competition and its solidity in Chine was investigated by Pino and Araya (2013), showing that heterogeneity tends to principally influence the correlation between competition and financially risky decisions. Moreover, Pak and Nurmakhanova (2013, who study extents to which market concentration refers to impact stability and ways that banks are found to be engaged in credit risks, suggest that the higher market power is, the more credit risky decisions banks appear to take. Similarity, as suggested by Jimenez et al. (2003) in Spanish banking environment, the correlation between competition and stability in banking services is non-linear when applying average measurements for market power.

When it comes to cross-country studies, the research outcome varied at different degrees. For instance, Berger et al. (2009) suggest that banks running in a less competitive environment are less likely to get involved in risk disclosure, which provides evidence for the hypothesis of ‘competitive-stability’. Here, the authors collected data from a large sample of 8235 banks throughout 23 nations for 6 years from 1999 to 2004; in which the level of competition and risk level was measured through Lerner index and Z-scores respectively. Furthermore, it is found that on the one hand, powerful banks are more likely to take non-performing loans and riskier loan portfolios, on the other hand they are also remain their charter/franchise value by taking more equity capital. Similarity, Aniger et al. (2012), who gathered data form 1872 banks across 63 nations from 1997 to 2009, capture the degree of bank competition by Lerner index and systemic stability by credit risk. The research result shows that competition has positive influence on systemic solidity and the result is consistent even after significant check running or taking new indicator for bank competition. Nevertheless, the measurement for bank stability is argued by Merton’s (1974) contingent pricing model. Moreover, Schaeck and Cihak (2008) also found the evidence for that association by a study of over 3600 banks across 10 nations in Europe and about 8900 banks throughout the U.S from 1995 to 2005. Here, competition was identified by Boone Index and found to improve banking solidity due to a rise in efficiency and revenues and simultaneously banking soundness can stimulate banking power. On the vein, Schaeck (2009) carried out a similar investigation in 45 nations from 1980 to 2005, particularly in 31 crises linking to systemic shocks. Here, competition was captured by Panzar Hosse H-statistic and found to contribute to a reduction of financial shocks even in the case that market concentration is significantly intervened by the Governments. In short, the study is supporting to the hypothesis ‘competition-stability’. Generating the similar research results, Yeyati and Micco (2007) gathered data from 8 emerging countries in the South America for nearly a decade from 1993 to 2002 and they indicate that competition, measured by Panzar and Rosse (1987)’s H-statistic, has negative impacts bank’s risks that identified by Z-scores.  Moreover, the research of the relationship between competition and stability is also undertaken across 4 developing countries in the South East Asia by Liu et al. (2011) for a decade from 1998. Particularly, risks are inspected through a number of variables, including instability of tax-adjusted ROA, loan loss reserves which are identified by total loan ratio and natural logarithm of Z-index, and loan damage which is total loan ratio. Here, it is found that most of the variables show significantly negative association with competition and from that the study is found to hold up for ‘competition-stability’ perspective. The study also further asserts that Government’s regulations and interventions negatively influence the stability in banking services in those countries.

In summary, it can be concluded that studies in individual countries show clear results, whereas it tends to be mixed results in the cross-country studies of the linkage between competition and financial solidity in banking services. Particularly, similar to Berger et al. (2009)’s study outcome, many research papers maintain that market concentration and competition absolutely have certain possibilities to coexist to not only enhance the stability of fragility in one way but also towards a wide range of channels.

CHAPTER III: DATA AND METHODOLOGY

The research aims to empirically explore the association between competition and financial stability; therefore, the most appropriate measures for these two main variables will be chosen depending on some certain requirements. Importantly, each measurement method will be critically evaluated and explained its significance in terms of measuring bank competition and stability. As discussed in the literature review, these two perspectives ‘competition-fragility’ and ‘competition-stability’ also link to the association between market concentration and bank stability and it is suggested that the high market concentration significantly contributes to bank’s soundless. Hence, the proposed model used for exploring the two perspectives will be considered as a supporting model to examine the influences of concentration on bank stability.

3.1. Competition measurements

In this section, there are some common measures of competition and they will be then discussed to provide the background information of measuring the variable before a selection of the most appropriate measure –the Lerner Index used in the employed model.

3.1.1. Competition measurement’s background

Bank competition is mainly measured towards two main methods, including structural approaches and non-structural ones. In structural methods, there are two distinct hypotheses mentioned, ‘structural performance’ and ‘efficient structure’. Regarding the former, it is suggested that  (confused)…. According to Stigler (1968) and Demsetz (1973), the market is considered by SCP as imperfect in structure and it, therefore, requires a market power checking. However, the lower market power tends to increase competition in the market, which in turn, results in a reduction in price and further benefits for customers (Gilbert, 1984; Molyneux, 1996). Nevertheless, Bikker and Haaf (2002) argue that concentration ratio may not be an exact measurement for competition. Regarding the latter ‘efficient structure’ hypothesis, it is maintained that the performance of a bank positively relates to the bank’s efficiency since in this case, banks with efficiency are likely to generate more revenue, obtain more market share, and further improve their market concentration (Demsetz, 1973). In other words, when revenue rises as a result of an expansion in market share, the efficiency is relatively stimulated without the presence of collusive behaviours based on the view of traditional SCP model (Molymeux and Forbes, 1995). On this vein, Berger (1995) also points out that banks should reach their market power through their pricing strategies, particularly in financial services of providing interest rates, and gain high profits with larger market share in the market.

Having realised the drawbacks of the traditional SCP model, New Empirical Industrial Organisation (NEIO) has developed non-structural methods for competition measurement. The term ‘structure’ seems to be a crucial element to access market power; nevertheless, many progressive studies maintain that competition is more likely to be identified by ‘conduct’ performance. The non-structural approaches include three main types: the Lerner index, Boone index, and H-statistic.

• H-statistic type

This measurement for measuring the degree of competition is developed by Panzar and Rosse (1987). As stated in the report of Global Financial Development published by World Bank, H-statistic is generated by conducting the regression of reduced form of a company’s revenue (maybe log of gross total revenues or interest revenues)….. The input prices are calculated by a number of elements such as labour price, credit price, and equipment and fixed capital price. Panzar and Rosse (1987) also suggest that stronger competition in the banking industry is caused by the higher rate of H-statistics. Furthermore, as being suggested by Claessens and Laeven (2004), the domination banking environment is where marginal cost tend to increase, but profit appears to go down as a consequence of a rise in input prices, which may cause a downwards demand curve. As a result, H-statistic is likely to be negative or equal to 0. Nevertheless, this figure can be 1 in a perfect competitive environment due to an equal rise in both revenues and marginal cost as a result of increase in input expenses.

• The Boone index

The Boone index, created by Boone (2008), is also an important indicator for measuring the level of competition. The measurement is the elasticity between marginal cost and revenues, and it is found that the index negatively relates to the level of competition. Particularly, the Boone index is more negative, banks tend to experience stronger competition and this situation is clarified by the influences of more significant reallocation. This model is considered for the efficiency hypothesis in which banks appears to reach larger market share if they have greater efficiency and it also means that they are supposed to make more profits as opposed to less efficiency counterparts at the same spending. Additionally, competition, in some certain situations, can significantly promote the more efficient banks in one hand, weaken other banks that are less efficient on the other hand (Boone, 2008). It is also found that the greater competitive environment can relatively make the impact more significant.

3.1.2. Lerner index

According to Fu et al. (2014), H-statistic has its drawbacks in which particularly this calculation can be only delivered in long-run equilibrium. On this vein, the Lerner Model is created to address that issue; therefore, it is widely used in a number of studies in the area of bank competition, such as Claessens and Laeven (2004), Maudos and Fernandez de Guevara (2004), Fernandez de Guevara et al. (2005), Berger et al. (2009); Maudos and Solis (2009); and Fu et al. (2014). The power of price is calculated by the Lerner model through the difference between marginal cost and price. Lerner rate can be between 0 and 1, depending on the level of monopoly market. Particularly, the index is 0 when the market is perfectly competitive and 1 under a pure monopoly. Especially, the figure is negative, which means that marginal cost is higher than the price. Demirguc-Kunt and Martinez Peria (2010) calculate Lerner index through data gathered from FDED (shorts for Federal Reserve Economic Data) and they then develop a formulas for calculating Lerner index as below:

Lernerit=Pit-MCitPit

In which,

Pit:

The value of assets and is identified by the ratio of total revenues (including interest and non-interest income) to total assets of bank i at time t.

MCit:

The marginal cost for bank i at time t

nCit= a0i+b0ln⁡Qit+ b10.5[ln⁡Qit]2+ a1ln⁡W1it+ a2ln⁡W2it+a3ln⁡W3it+b20.5ln⁡Qit*ln⁡W1it+b30.5ln⁡Qit*ln W2it+b40.5ln⁡Qit*ln⁡W3it+a4ln⁡W1it*ln⁡W2it+a5ln⁡W1it*ln⁡W3it+a6ln⁡W2it*ln⁡W3it+ a70.5ln⁡W1it2+a80.5ln⁡W3it2+d1Trend+ d2Trend2+d3Trend*ln⁡Qit+ d4Trend*ln⁡W1it+ d5Trend*ln⁡W2it+ d6Trend*ln⁡W3it+ uit

• i: represents banks and t is represents years
• C: operating and financial costs
• Q: total assets
• W1: proxy for input price of deposits, equalling to ratio of interest expenses to total deposits and money market funding
• W2: Proxy for input price of labour, equalling to the ratio of personal expenses to total assets.
• W3: proxy for input price of equipment or fixed capital, equalling the ratio of other operating and administrative expenses to total asset

It is found that the Lerner index can equal to 0 if the value of assets balance the marginal cost. In other words, it also means that bank (i) is found to have no market power. Additionally, Fu et al. (2014) suggest that if the asset price is lower than the incremental cost, the value of Lerner index would be negative and it may further result in non-optimal bank activities. The total assets are valued at the rate of total profits (interest and non-interests) to total assets, whereas the incremental cost is log shaped. One point should be taken into account is that Lerner index is calculated regardless to the breakthrough of technology (Demirguc-Kunt and Martinez Peria, 2010).

3.2. Stability measurement

Financial stability measurement is not an easy stage, particularly due to the ways to define ‘financial stability’ context (Achinasi, 2004). This context should be favourable for banks to reach a range of financial elements, for example, the efficiency in allocating economic resources; the assessment of price for financial risks; fostering the capability of running key functions and being capable of well withstanding under financial shocks from both internal and external factors. Furthermore, a bank is seemed as a financially stable company when they have capabilities of stimulating economy performances, eliminating imbalances generating from financial turbulences or unforeseen events.

Compared to other measures, Z-score is found as the most applicable indicator as it has been applied in a wide range of studies, for instance, Roy (1952) – the first scholar applied Z-score to figure out financial stability; Boyd and Runkle (1993); Lepetit et al. (2008) Cihak and Hesse (2007) and Liu and Wilsonm (2011). According to Beck et al. (2009), Z-score has less default risks, which suggests the ways in which banks with strong capital face the instability in revenues and the authors also suggest Z-score is able to reveal the ability of a bank in absorbing loss with their power of capital. Furthermore, Boyd and Runkle (1993) suggest that Z-score is identified as a measure for standard deviations in which the bank’s revenue should be reduce from the average to take more for equity capital. Moreover, Fu et al. (2014) also point out that this score has a negative association with bankruptcy probability and they also maintain that bank are only found as insolvent when the asset value is less than debt value. Z-score equation is shown as below:

Zit=ROAit+Eit/AitσROAit

Where:

ROAit

: Return on bank i’s assets at time t

Eit/Ait

: Ratio of bank equity capital to total assets for bank i at time t

σ(ROA)it

: Standard deviation of return on bank’s assets for bank i at time t

According to the result of the equation, it can be concluded that higher value Z-score means that banks are less likely to face the insolvency. In other words, banks are more likely to be asked for lesser capital capability and the stable environment is less required. It is claimed to be a straight calculation for bank’s stability, as opposed to the other indicators for risks in banking services (Kasman, 2015).

3.3. Concentration measurement

In the model examining the linkage between stability and competition, concentration is identified as a national-level structural indicator which is recognised by the concentration of assets in the biggest financial institutions in every country (also called World Bank). The level of concentration is calculated in the formula below:

Concentration

=

∑i=15Assetsit∑i=1mAssetsit

Where,

m: the number of banks in one country

i, t: referring to each bank and each year

Concentration rate tends to be the same across all banks in each nation for one-year period. The higher value of concentration implies about a more concentrated banking market, meaning that five biggest banks have higher majority of the assets in the whole system.

Herfindahl-Hirschman index (is also called HHI index) is applied to reveal the bank-level competition in the second model. This index is calculated as follows:

HHI

=

assetsit2∑i=1massetsit2

In which,

i, t: referring to each bank and each year

m: the number of banks in one country

Compared to the first model, this model suggests that each bank has its own HHI rate. It is found that banks with higher value of HHI are maintained to reach higher market concentration.

3.4. Data and model

3.3.1. Data

In this research paper, the data is primarily collected from 109 commercial banks across 7 developed countries in G7 from 2010 to 2015 and secondarily gathered from SNL financial. To achieve the purpose of examining the association between competition and financial stability in banking industries, the data is targeted as such ROA, total loans, net interest margin, total assets, equity, loan loss and some other sorts. Additionally, information on regulations and principles is gathered from Financial Freedom index 2010 and Property Rights index issued by the Heritage. The data used for measuring competition and Lerner index is collected from the SNL financial and targeted to G7 countries before being calculated. More importantly, missing, negative and invalid value of cost information for calculating Lerner index, Z-score and loan loss ratio are omitted in order to reach a balanced panel data.

3.3.2. Model

The data gathered for the investigation of the association between competition, market concentration and financial stability is panel data. Hence, GMM model will be used to examine that linkage since it is found as a more efficient model than 2SLS when heteroscedasticity issue is taken into account (Hall, 2005). The panel data designed to explore the association between competition and stability, and the connection between financial stability and market concentration is created as follows (as developed by Fu et al. (2014)).

Financial stability = f (Concentration, Competition, Bank controls, Regulatory and Institutional Controls)

And,

Financial stability = f (Concentration, Bank controls, Regulatory and Institutional Controls)

Or:

Financial stabilityit= α+ β1concentration+β2competition+ β3 Bank controls  +β4Regulatory and Institutional Controls +εit

Financial stabilityit= α+ β1concentration+ β2 Bank controls  +β3Regulatory and Institutional Controls +εit

Where:

i, t: referring to each bank, country and year respectively

The dependent variable:

The dependent variable is bank stability which is calculated by Z-score. It is found that higher Z-score value means that banks has higher overall financial stability and lower risks (Fu et al., 2014). Here, data used for the equation is gathered from SNL financial.

The independent variables:

• Concentration (CR5): The first independent variable is concentration that is a country-level structural measure. The data will be directly collected from World Bank data record. This figure is computed by taking total assets of 3 biggest banks in every nation, implying that if CR5 has higher value, there would be higher market concentration.
• Concentration (HHI): as mentioned above, HHI index is also a structural indicator, but at organisational level. The figure is computed by data gathered from SNL Financial. It is also suggested that when HHI has its higher value, the bank experiences more concentration.
• Lerner Index (LERNER): as opposed to two last independent variables, this index is a non-structural indicator for competition measurement at organisational level. The figure is computed by multiplying national-level LERNER and HHI value. It is maintained that higher value of LERNER implies about less competitive banking market. Additionally, the national-level LERNER is directly collected from World Bank records.
• There is a range of bank controls variables which are selected based on a variety of literature, such as Schaeck and Cihak (2008), Laeven and Levine (2009) and Uhde and Heimeshoff (2009).

• Bank size: identified by the total assets of a bank in USD and gathered from SNL Financial.
• The net interest margin (NIM) is net income of a bank from interest to interest-bearing assets. According to Fu et al. (2014), this figure is to assess bank’s profitability in investment and lending activities. The data is also collected from SNL Financial.
• Loan-loss provisions ratio (LLP): Fu et al. (2014) suggest that this figure is to specify bank’s output and appraise the ways in which banks invest their output in non-performing loans. LLP is available on SNL Financial.

• Regulation variables:

• The first variable in terms of regulation is Financial Freedom. This term is identified by the level of openness in banking system and the regulations and interventions given by the government and other authorities into financial market. It is found that high degree of financial freedom fosters performance efficiency, weakens financial limitations, and in turn, helps to promote financial solidity (Fu et al., 2014).
• Property Rights: The Property right index is published by the Heritage, which is a fundamental pre-requisite for the whole banking system to assess to what extent their system is working well or not. It is suggested that the higher value of this index is, the inferior property protection is likely to be (Fu et al., 2014).