In order to test the relationship between ownership structure, market structure and bank performance, the relative findings and hypotheses are vital for the research. Thereby, the prior studies should be introduced in this section.
Literature of international research
There are most basic and important studies on the relationship between ownership structure and performance, which provide some useful methods, results and suggestions to help doing further research. When doing research on corporate, the results appear to be quite different even opposite.
Demsetz and Villalonga (2001) hold the view that ownership structure does not associated systematically with corporate performance. Demsetz and Villalonga (2001) differences in ownership may lead to differences in company’s performance by testing ownership with some management and top five shareholdings. However, Sun et al. (2002) has demonstrated that a part of government ownership in corporate affect performance positively, but not too much or too little. The relevance between government ownership and firm performance is adverse U-shaped and nonlinear (Sun et al. 2002 p.23).
In the field of banking industry research, one study focus on the effect different ownership structures on bank economic behavior, which consequently influence bank performance, for instance, Garcia-Cestona and Surroca (2008) have considered economic behavior and regulation with public management controlled and insider controlled ownership structures that impact on saving bank performance of Spain. The results reveal that the decision making of saving banks in Spain have been indirectly influenced by legislator intervention and the mistakes of managers’ explanation in current laws impact slightly on various efficiencies of saving banks. Ownership with majority of public administrations will lead the bank to focus on services of universal finance and development of region while ownership with majority of insiders will make the bank tend to develop the growth of economy. In addition, various priorities and technologies appear in saving banks in Spain owing to different ownership structures (Garcia- Cestona and Surroca, 2008 p.597).
To estimate bank performance, Return on asset (ROA), Return on equity (ROE), Cost income (C/I), and Net interest margin (MIN) are measures to examine profitability, and non-performing loans (NPL) which is one of measures to examine quality of asset (Casu et al. 2006). By using these measures, Berger et al. (2005) has investigated Argentine banks with state, domestic and foreign ownership in 1990s combine considering the effects of static, dynamic and selection in order to test the influence of corporate governance changes on bank performance. The result indicates banks with state-owned ownership underperform than banks with foreign or domestic ownership in a long term, and domestic banks may have better perform than foreign banks. Moreover, domestic banks referring to merger and acquisitions have a bit poorer perform than that of before merger and acquisitions, and there is little relationship between performance and foreign or domestic acquisitions whereas privatization tend to be helpful for the promotion of bank performance. Banks’ portfolios allocation associated to the domestic acquisition and foreign merger (Berger et al., 2005 p.2209-2210). The survey argue other literatures “do not consider “more than one static difference at a time” and “more than one of these types of selection effects together” (Berger et al., 2005 p.2180). In order to cover this shortcoming, the static, dynamic and selection impact as three effects of various types of corporate governance are all estimated in the same test model to access bank performance. In addition, “nonrubustness” checks with different exclusive consideration have been used and display which underlying problems which appear from those exclusions are vertical (Berger et al., 2005 p.2181).
The notion and model created by Berger et al. (2005) have been made a great contribution for further research. The investigations on banks in different areas have proved Berger’s opinion. For example, Cronett et al. (2010) have investigated different performances of banks with government and private ownership form Far East in 1989-2004 periods via measure of accounting and cash flow, and noted that government-owned banks have less profitability, less main capital and high risk than private banks. Between 1997 and 2000, the performance of state-owned banks such as capital, cash flow and quality of credit is poorer than that of private banks; even it is dramatic in Asian financial crisis. Nevertheless, from 2001 to 2004, this evidence has changed and state-owned banks’ performance has been improved, even it is similar to private banks. In the courtiers where government intervention is heavy in banking industry, banks with nationalized ownership undertake the finance for government more than banks with private ownership. Bureaucratic politics is associated with poorer performance, which impact state-owned banks’ operating, and competition of financial service is beneficial for the promotion of government owned banks’ performance (Cronett et al. 2010 p.93). An useful research for 181 European banks from 1999 to 2004 by Iannotta et al. (2007) have compared risk and performance of mutual, state-owned and private banks via examine quality of asset, bank size, country, year and economic growth. The result is accordant with previous literature, which is private banks have higher profitability than government-owned and mutual banks, but much higher cost than mutual and nationalized banks owing to government guarantee protection. Public banks have less risks and profitability than other types of banks, and such banks have funding with large share, which are from capital market and wholesale interbank, moreover, these public banks have lower loan investment and higher liquidity (Iannotta et al. 2007 p.2146). Like private banks, mutual banks have a good consumer relationship that is beneficial to the quality of loans and reduces costs of operation. The reason why mutual banks have poorer performance than private banks is due to different types of asset mix, lower size and more traditional financial institution behavior (Iannotta et al. 2007 p.2147). Iannotta et al. (2007) reveal policy as an external factor affect bank performance. The limitation of this study is concentration of ownership is puzzling, and ownership with less concentration leads to lower quality of loans and higher risk of asset is not interpreted clearly by the theory (Iannotta et al. 2007 p.2147). Another survey for Egyptian banks in the period of 1996-1999 by Omran (2007) has agreed prior results that majority private ownership in banks outperforms than banks with larger government-owned shareholding; further, the study indicates how the extent of privatization and nationalization effect on bank performance in Egyptian. The findings indicates liquidity and profitability percentage of privatized banks descend dramatically but banks with majority of private ownership and private owned banks in changes of performance is faster than privatized banks. Privatized banks are more profitable than banks with majority nationalized ownership but less profitable than state-owned banks. In addition, private-owned banks and majority private-owned banks have high profitability and efficiency than majority state-owned and state-owned banks, and banks with private ownership have better performance. Better performance related to nationalized ownership with less degree, and the suggestion is that it is helpful to improve performance when reduce nationalized ownership. Nonetheless, the defects of the survey is period is a bit short and more evidence is not discovered (Omran, 2007 p. 730).
The differences in regions may give rise to different effects of ownership on bank performance. In transition countries, banking sector with is different from other banks which are located in countries of emerging market and developing nations. Such banks in transition nations have high proportion of asset and most foreign-owned ownership structure. (Bonin et al. 2005, p.32). By surveying 225 banks between 1996 and 2000 in transition countries, Bonin et al. (2005) have reported that private banks are not dramatically more efficient than state-owned banks; further, foreign ownership banks with more deposit and loans have better service than that of private banks, and banks with nationalized ownership have less efficient less service providing, which testify the hypothesis that banks which were privatized is better than banks which are not in transition states. Bonin et al. (2005) also indicate that investor of international institution play a vital role in the improvement of restruction when these investors effect on efficiency of priority positively. Compare with banks in transition countries, in developed and developing countries, the relationship between ownership structure and bank performance appear differently. In different level of economic development counties, the effects are also various. La Porta et al. (2002) illustrate that banks with government ownership will go against efficiency, development of finance and economy. One study that test world 179 nations in developed and developing countries and looks how various types of ownership effect bank performance in such countries (Micco et al., 2007). The result shows that by considering politics, in developing nations, banks with private ownership have more profitability, higher net interest margin and less cost than that of state-owned banks, while there is little significant association between ownership and bank performance in industrial nations (Micco et al., 2007, p.220). The findings are examined by looking at return on asset, net interest margin and efficiency, which display that there is no relationship between ROA and size of banks in developing nations whereas in industrial states, ROA is related negatively and statically with banks size; furthermore, margins of state-owned banks is lower than that of foreign banks in developing nations and similar to private banks, margins of private banks is higher than that of foreign banks. When looking at efficiency, there is negative correlation between banks size and costs of overhead in developing counties while size is not significantly associated with costs in industrial nations (Micco et al., 2007, p.228). Though several studies have displayed foreign banks outperform banks with state ownership, according to Lensink and Naaborg (2007), it is not the high percentage of foreign share in banks that improve bank performance. There is an adverse association between increasing foreign ownership and bank performance, such as profit of bank and net interest margin. A low proportion of foreign ownership in banks has more profitability and net interest income than banks with majority foreign ownership (Lensink and Naaborg, 2007 p.885), which is consist with the view of Berger et al. (2009), who report that low foreign ownership is benefit to promote bank efficiency. Lensink and Naaborg (2007), in addition have claimed that there is no influence on the relation between foreign ownership and bank performance in counties with various economy development levels.
Literature of research on Chinese corporate and banks
When analyzing Chinese corporate and banks, there are several researchers did the research on ownership and performance, and some scholars found relation between the change of ownership structure and corporate performance. Based on Micco et al. (2007), banks with nationalized ownership underperform than domestic private banks or foreign banks in developing nations. China is a developing country, and majority studies reveal that Chinese banks with diversification of ownership structure have better performance than banks with high state-owned share, which is consist with the view of Micco et al. (2007).
Wang (2005) has tested 747 China’s listed firms from 1994 to 1999 through ROA, OI/A (operation income to asset) and S/A (sales to asset) to estimate the performance of operation, and gained the findings that both nationalized and private ownership have little impact on the corporate performance while the centralization of ownership which are not state-owned affect changes of enterprise performance. The result has been showed that the influence of ownership on the changes of performance depends on different kinds of shareholders. The relationship between performance and ownership is curvilinear relevance as well as the relation between performance and concentration of ownership which is not nationalized (Wang 2005). In this paper, the correlation of ownership and performance has been interpreted comprehensively, nonetheless, the regional characteristic, for instance, extent of government intervention or economy level of different areas are lack of consideration.
For banking industry research, Yang and Cao (2007) have found one reason for low efficiency of banks due to large proportion of the first majority shareholder via observing five China’s listed commercial banks, and suggested that it is helpful for bank performance when weaken the controlling power of the first majority shareholder; furthermore, varied share but not too much is a better ownership structure to balance each kind of share, and prevent substantial shareholder to control the whole bank. Yang and Cao (2007) claim the share proportion of first large shareholders is adversely and significantly related to bank performance whereas the share ratio of fist five shareholders positive and significantly associated with bank performance. In addition, the rate of circulation share is dramatically negative related with the performance of listed commercial banks owing to undeveloped Chinese external market and more medium or small shareholders who support circulation share. The limitation of the study is the sample selection such as only five banks is not enough to convince in the entire banking industry system, especially increasing number of Chinese commercial banks which are going public during recent years. By comparing how ownership structure mode impact on bank performance from different countries, Li (2008) has summed up the features of two successful ownership structure modes involving external monitoring ownership from UK and US; internal monitoring ownership from Japan and Germany, and explained how these two ownership structure associates to bank outperformance. What is more, Li (2008) has selected four state Chinese state-owned commercial banks and ten joint-equity commercial banks from 1996 to 2005 as samples to examine whether ownership structure affects China’s commercial banks by ‘Chow Test’. Li (2008) has showed the result that ownership structure is a vertical factor which influences the performance of commercial banks, and the extent of market competition may increase the efficiency of banks. According to Li (2008), equity structure associates with bank performance, especially, the degree of market competition, is another considering factor has been pointed out. However, the deficiency of this study is how different kinds of ownership structures effect the performance and what is the exact relation between ownership and performance are not interpreted clearly.
According to prior literatures, geography factor are not considered. Ferri (2009) has covered this shortage and explained that the adverse correlation between ownership structure by taking over the aspect of regional economy and banking corporate governance has been indicated. In addition, more shareholders in banks would lead to less government intervention. Ferri (2009) states that ‘New Tigers’ such as city commercial banks outperform nationalized banks and it is important to handle nationalized banks when work out the problems of Chinese banking. When compare China’s commercial banks (‘New Tigers’) and state-owned banks (‘Old Mammoths’), the conclusion is “The New Tigers grow intensely and outperform the Old Mammoths” (Ferri, 2009 p.133). In order to test whether governance impact ‘New Tigers’ better performance and whether it is the reason for doing more business in developed province in China, Ferri (2009) analyzes China’s city commercial banks from the aspect of geography and corporate governance, through investigating twenty ‘city commercial banks during 2000-2003 period in Zhejiang, Sichuan and Hubei provinces which have various economic development levels. By using performance measure indicator (ROA, ROE and NPL), Ferri (2009) has displayed the result that the larger bank the better performance and the ROA and ROE of city commercial banks are higher in Zhejiang, a developed area than that of other two less developed provinces while NPL is lower; moreover, a better performance in developed province such as Zhejiang due to the lower share of state-owned enterprises, in contrast, in Sichuan and Hubei, which means bank performance is influenced by geography and policy intervention (Ferri, 2009). This study has proved negative relationship between ownership and performance. In addition, geography as a new factor is added to be considered, which is different with opinion by Li (2008). The limitation of the study is it just focuses on three provinces though these provinces represent different level of economy. Other provinces such as Dongbei, Hunan and Guangdong at different level of economy, which represent northern and southern economy of China, are not taken into account.
Another research with accordant result, has surveyed nearly all China’s banks with different types, such as state-owned banks, policy banks, joint-equity banks and city commercial banks from 1997 to 2004 (Lin and Zhang, 2009). Based on the model by Berger et al. (2005), Lin and Zhang (2009) use ROA, ROE, COI and NPL to assess bank performance with different ownership structure, and obtain the result that policy banks are more efficient than ‘Big Four’ state-owned banks whose state-ownership is high; four stated-owned banks have lower profit and efficiency than foreign banks, city commercial banks and domestic joint-equity banks dramatically. This finding indicates that banks with overseas acquisition have better performance than banks without foreign acquisition, which consist with the above literature we have mentioned (Lin and Zhang, 2009). However, according to another result, it is worth to notice that listed banks have higher ROE, ROA, COI and NPL than not listed banks, which is not similar to the early literatures. The length of the term including long and short, does not the matter performances (Li and Zhang, 2009 p. 26). From this study, the problems of China’s banking system are pointed and the resolution is suggested. Nonetheless, there is a little shortage that the data is only investigated to 2004 because after 2004, the data has been updated and some banks such as policy banks have been made a big change.
Wang (2009) notes that nationalized commercial banks shareholding reforming and going public need financial property right structure which is consistent with modern commercial banks, and the effect of property right structure on banks’ business performance depend on the impact of ownership on bank performance for the whole banking industry. Wang (2009) classify three types of enterprise’s ownership in terms of ownership concentration or dispersion, for instance, highly centralization of ownership, namely the company has only one controlling shareholder; highly decentralization ownership, namely the corporate has no substantial shareholder; and moderate centralization of ownership means there is large controlling shareholders and some other major shareholders in a company. Based on the concentration of ownership, China’s 11 listed commercial banks except city banks as sample are investigated and estimated by the indicator of ROE, ROA. The findings has showed that ROE and ROA are negatively associated with the proportion of state-owned share, and positively associated with foreign share rate; that is to say, decrease the percentage of state-owned share and bring in foreign share will help to improve the performance and competition of banks (Wang, 2009). Wang (2009) suggests that reformation and optimization of equity structure is a better way to promote Chinese nationalized commercial banks. In order to resolve the problem of concentration of state-owned banks, the share ownership should be dispersed properly and attract social security funds, private enterprises and foreign capital, which make the ownership structure be diversified (Wang, 2009, p.69). The survey also reveals the negative relativity between the concentration of ownership structure and operation performance of banks, which prove the same result from previous literatures. Moreover, the data of some banks such as nationalized banks including ‘Big Four’ commercial banks has been updated comparing with early survey by Lin and Zhang (2009).
The latest research on China’s underwriting performance of investment banks by Jia and Zhang (2010), who demonstrate that there is ‘distortion effect’ of regional government intervention that impact on investment banks which are affiliated (p.226). Underwritten by investment banks with no government participation tend to have higher quality of earnings and better performance in a long term than affiliated banks (Jia and Zhang, 2010 p.198). Jia and Zhang (2010) suggest that government ownership of financial institution may reduce the quality of underwriting for investment banks which are affiliated. This survey again reveals that too much state-owned equity ownership in banks will not facilitate bank performance.
Market structure and performance review
When measuring bank performance, market structure is another factor to take into account. There are two hypothesis concerning correlation between performance and market structure. The fist hypothesis is maret-power hypothesis, including “structure conduct performance” and “relative market power”. “structure-conduct-performance”, namely SCP, which is in the market that is concentrated, banks will offer deposit with lower rate and charge for loans with high rate, and consequently banks may monopolize rents and get high profit. When test the positive relation between market centralization and profitability, it is used to estimate SCP hypothesis (Stigler 1964, cited in Goldberg and Rai 1996 p.746). Compare with traditional hypothesis of SCP, relative-market-power hypothesis (RMP) announce that corporate will gain more profits if it is has larger market share. Lloyd et al. (1994) obtain different result of concentration-performance relation, and support SCP hypothesis that banks market behavior in Spain is related to concentration. Banking market with high centralization in Spain lead to the cost be lower and make higher profitability. The hypothesis of efficiency is not supported because higher concentration of banking market in Spain will reduce the banking competition and efficiency is not adjusted (Lloyd et al. 1994).
The second hypothesis is efficient-structure hypothesis, namely EFS that illustrate positive association between profit and market concentration, which is related to advanced management and production skills (Smirlock 1985, cited in Goldberg and Rai 1996 p.746). For EFS hypothesis, it includes X-efficiency (ESS) hypothesis namely corporate can obtain higher profits when they have advanced management and production technology with low cost, and scale efficiency hypothesis (EES) namely corporate with similar management and production skills running in diverse level of economic scale (Berger and Hannan 1993, cited in Goldberg and Rai 1996 p.749). By using performance measures such as ROA, ROE and NIM, concentration measures such as CR3 (three banks concentration ratio) and Herfindahl Index, efficiency measures of X-efficiency and scale efficiency, Goldberg and Rai (1996) note that the correlation between degree of concentration and bank performance is nonlinear in European banking industry. The concentration indicators of CR3 and Herfindahl Index have no effect on bank performance. In addition, Goldberg and Rai (1996) support hypothesis of X-efficiency for banks from counties with low concentrate market. Smirlock (1985) mainly uses market share instead of efficiency, concentration of market to establish the regression model in order to test bank performance such as indicators of ROA, ROE, and reports that market share positively impact profitability in US banking industry while concentration do not play an vital role on profit.
For Chinese banking industry, Fu and Heffernan (2009) tend to market power hypothesis instead of efficiency structure hypothesis as well as prior literature. Chinese banks with different ownership have been investigated by examine structure-conduct-performance, relative-market-power, and efficient structure hypotheses including X-efficiency and scale efficiency. The study indicates that China’s banking department is explained very well by hypothesis of relative-market-power in the first period of reformation, and efficiency does not influence market structure positively. In the second reformation period, joint-equity banks prefer promoting profitability and X-efficiency when compare with state-owned banks. Concentration does not negatively affect efficiency for state-owned banks, and the restrict regulation on rate control avoid state-owned banks to earn monopolistic profit. China’s banking with competition structure should be promoted by permitting enlarge share of joint stocks, add the entry of market, and make the interest rate keep liberalization in order to reduce concentration (Fu and Heffernan, 2009 p.50).
“Most study for banks focus on the issue concerning the measurement of bank performance, seldom examines the topic about the factors influencing the bank performance” (Chein, 2009 p.81). For this reason, recent research Chein (2009) has surveyed how the factors of loan, market and ownership structure bank performance impact bank performance, and demonstrates that there is negative relation between loan with problems and bank performance. Moreover, the association between CEO ownership and efficiency is nonlinear U-shaped, which means CEO with higher or lower ownership is related to the improvement of bank performance. For the aspect of market structure, there is positive correlation between concentration of market and efficiency. That is to say, market with higher centralization is beneficial to gain better performance of banks, expect market structure with monopolization which confirm with prior studies holding view of SCP hypothesis (Chein, 2009 p.81).