Since the latest Chinese banking reform started from 2001 focus on the ownership transformation via foreign participation and stock listing, there is a great progress for the Chinese 'Big Four' state-owned commercial banks (SOCBs) in their operational efficiency and profitable results. The paper reviews the process of Chinese banking reforms and compares the banking efficiency of nine typical Chinese commercial banks before and after their IPOs empirically (through DEA approach). Using data from 2005 to 2009 of nine banks and analysis them, the objective of this paper is to find out whether the Big Four state-owned commercial banks (SOCBs) are more efficient after their ownership reform than before, and to conclude what is the implication of this ownership reform in Chinese banking system. Key Words: Chinese SOCBs; Ownership reform; Banking efficiency; Data envelopment analysis (DEA) Since Chinese economic reform and open up started in the year 1978, Chinese modern banking system went through a great transformation and achieved a great improvement. After the year 1978, Chinese economic reforms have been focused on Chinese economic transformation from a planning economy to a market-based economy (Chen, Skully and Brown, 2005). As for Chinese banking system, the reform point is the ownership transformation via foreign participation and stock listing. In the year 2006, Bank of China (BOC) and Industrial and Commercial Bank of China (ICBC) became the first two listed banks in China, which also created a history-the largest scale of IPO around the world at that time. By the end of the year 2009, there were 14 listed banks in China, including Chinese 'Big Four' state-owned commercial banks (SOCBs). Among the 14 listed banks, six of the top seven commercial banks listed in Shanghai and Hong Kong stock exchange. When China participated in the World Trade Organization (WTO) at December of 2001, Chinese government promised to totally open up its national banking market for all the qualified foreign banks and other foreign financial institutions. In order to cope with the immediate challenges from international financial institutions, the recent banking reform in China was aim to transform the Chinese banking system from a monopolistic and centralized structure to a multi-ownership and competitive banking system (Wu, 2006). With the faster development of banking reform, the problem of banking efficiency has become a focus in modern finance. The banking efficiency indicates the comparison relationship of inputs and outputs, or cost and income during the banking operational business. In other words, banking efficiency refers to the effective allocation of banking resources, which is the combination of banks' competitive ability, input-output ability and sustainable development ability. In addition, the banking efficiency is an essential index to measure the performance of banks, the usage of resources, standard of banking management, and ability of sustainable development. On the background of recent banking reform and worldwide financial crisis, the objective of this paper is to compare the bank efficiency of nine typically Chinese commercial banks before and after their IPOs empirically (using DEA approach), to find out whether the Big Four SOCBs are more efficient after their ownership reformation, and to conclude what is the implications of this banking reform. The paper will mainly use Data Envelopment Analysis (DEA) to conclude the efficiency lever of nine mainly Chinese commercial banks over the period 2005-2009, especially focus on the Big Four SOCBs. Among them, Bank of China (BOC, established in 1912), China Construction Bank (CCB, established in 1954) and Industrial and Commercial Bank of China (ICBC) were the biggest winner after world financial crisis, and replaced the western top financial institutions to be the world's first three most profitable banks since the year 2008. The aim of this paper is twofold. First of all, it reviews the process of Chinese banking reforms and the methods people used to analysis the efficiency of commercial banks around the world. Secondly, we examine the efficiency of Chinese commercial banks, especially the four state-owned commercial banks (SOCBs), to find out whether the four SOCBs are more efficient after their IPOs than before, and to conclude what is the implication of Chinese recent banking reform. The structure of this paper is organized as follows: Chapter 1 gives a brief introduction of the background and objective of this paper; Chapter 2 reviews the literature on the studies of banking efficiency; Chapter 3 discusses the process of Chinese banking ownership reform especially the latest one; Chapter 4 describes empirical model we use to analysis the banking efficiency- the Data Envelopment Analysis (DEA); Chapter 5 describes the data we choose and empirical modeling results; Chapter 6 will discusses conclusion and implications of the whole paper.
Chapter 2 Literature Review 2.1 About banking efficiency Banking efficiency can be viewed as the ability of social financing that commercial banks could possess under the condition of security, profitability and liquidity. Banking efficiency includes micro efficiency and macro efficiency, and the former is the object of study in this paper. According to western banking management theory, micro efficiency in banking indicates the comparison relationship of inputs and outputs of commercial banks. In most of studies, banking efficiency is briefly showed as: efficiency = outputs/ inputs. Throughout those studies, there are about three efficiencies in the banking efficiency theory analysis: scale efficiency, scope efficiency and X efficiency (John, Iftekhar and Paul, 2005). Specifically, scale efficiency indicates the improvement of labor productivity or the decline of average product cost, which all stem from the same proportion addition of whole production factors. The scale efficiency can measure the effects of inputs of banks on cost saving, while scope efficiency stands for the efficiency of the diversity on products and services, which measures the effects of diversity business of banks on cost saving. The scale efficiency and scope efficiency are the definitions from neoclassical economics, which are all used depends on their external features. According to the neoclassical economics, the direct reason of scale efficiency and scope efficiency is that the large scale and diversity business create more market power, which could decrease the products prices and raise service prices. In the mean time, it is convenient to adopt advanced technology. The X efficiency was put forward aim at the internal situation in banks itself. Harvey Leeibenstein, one of the most famous American economists, put forward the X efficiency theory in 1966 after he studied the factory efficiency in developing countries. Harvey Leeibenstein suggested that firms may own allocative economic efficiency at any circumstances. Since it does not have a clear characteristic, it is called X efficiency. The effect of X efficiency theory is based on the information asymmetry and incomplete of labor contract. Under this assumption, the economic efficiency is not a result of the enterprises as a whole decision-maker, but a result of their internal strategy (Fu and Heffernan, 2007). Recently, X efficiency has became a popular topic in efficiency research (Fu and Heffernan, 2007). In this paper, we define the banking efficiency as the comparison relationship of inputs and outputs, or of cost and income during the banking operation business. In other words, banking efficiency refers to the effective allocation of banking resources, which is the combination of competitive ability, input-output ability and sustainable development ability. 2.2 Review on banking efficiency When it comes to measuring banking efficiency, most of the studies before were mainly included three parts: the consequence of mergers and acquisitions around banks (Berger and Humphery, 1992; Rhoads, 1993); the contrast of private banking ownership, public and foreign banking ownership (Hasan and Marton, 2003); and the effects of access and deregulation over foreign institution (Chen, 2001; Claessens, 2001). Recently, however, researches on banking efficiency have transferred to compare efficiencies among different commercial banks (including both SOCBs and JSCBs) or to test the coherence among different frontier methods such as SFA and DEA approach (Weill, 2004; Bos and Kolari, 2005). Since the year 1980, with more and more intense competition in banking system, all banks around the world aimed to improve their competition ability and enhance their banking management. Those studies on banking efficiency have transferred to the internal resource allocation in banks, and the X efficiency was paid attention to in theory (Leeibenstein, 1966). At the same time, the relative researches were more depend on the frontier efficiency analysis. Farrell (1957) firstly studied the firm efficiency from micro aspect, and firstly introduced the definition of frontier efficiency function. The main idea of frontier efficiency is that defining all the possible outputs external frontier based on a series of inputs observations. The frontier efficiency is a relatively valid definition, which is now become the most general approach in the micro efficiency research area. The frontier efficiency analysis on banking efficiency comprises the parametric methods and non-parametric methods (Ariff and Luo, 2008). The most popular parametric methods are Stochastic Frontier Approach (SFA), Distribution Free Approach (DFA) and Thick Frontier Approach (TFA), while the most familiar non-parametric methods are Data Envelopment Analysis (DEA) and Free Disposal Hull (FDH). Among these methods, the methods used in this paper-DEA is viewed as the most frequently used approach (Wei and Wang, 2000; Zheng and Zhang, 2004; Chen et al., 2005; Fu and Heffernan, 2007; Ariff and Can, 2008; Lin and Zhang, 2009). The most controversy problems in the studies of financial institution efficiency are two aspects: the one is how to define and compute the inputs and outputs of banks; the other is how to make sure the best business boundary to evaluate banking performance. There are three methods for measuring inputs and outputs of banks: production method, intermediate method and profit-output method. Berger and Humphrey (1997) indicated that neither production method nor intermediate method is perfect, they should supply each other: production method is better on analysis the overall standard of banking efficiency and intermediate method is suitable for measure the efficiency of branches. Ferrier and Lovell (1990) demonstrated that considering the operational cost of banks, production method is good to research the banks' cost efficiency, while intermediate method can be used to analysis the economic difference of different banks. The early studies in banking efficiency mostly focus on scale efficiency and scope efficiency. In European, Dietsch (1993), Zardkoohi and Kolari (1994) indicated that economy of scale plays an important role in banking sector. And in US, studies proved that the average cost of banks is relatively associate with some evidence of scale efficiency (Berger, Hunter and Timme, 1993; Gilbert, 1994; Mester, 1987). Apart from studies on the European and American banks, researches of banks in developing economies, such as in China and India, have become increasingly popular after 1990s. A wealth of information is to be found in the statistics literature, they generally point out that the whole efficiency of Chinese commercial banks has been improved generally after a series of banking reforms since the open up reform in 1978, and the joint-stock commercial banks (JSCBs) were more efficient than those state-owned commercial banks (SOCBs) by about 10 to 20 percent (Wei and Wang, 2000; Zhang, 2003; Zheng and Zhang, 2004; Chen, 2005; Fu and Heffernan, 2007; Ariff and Can, 2008; Lin and Zhang, 2009). According to one of the popular classification on banking efficiency, the banking efficiency can be divided into scale efficiency (SE) and technical efficiency (TE). According to Laurenceson Â¼Ë†2001Â¼â€° and Zhao (2008), the inefficiency of the joint-stock commercial banks (JSCBs) mainly because of their unsuccessful on scale efficiency. As for the reason of the inefficiency of state-owned commercial banks (SOCBs), pure technical inefÂ¬Âciency (PTE) should took responsible for it (Wei and Wang, 2000; Kumbhakar and Wang, 2005). When it comes to the influence of banking reform, apart from improved efficiency concluded by Chen (2005), Berger, Iftekhar and Zhou (2009) indicated the effectiveness of bringing in foreign institutions for banking operational performance enhancement. In addition, the WTO participation of China was found to be related to a decline in overall efficiency, scale efficiency (SE) and pure technical efficiency (PTE) (Kumbhakar and Wang, 2005; Fu and Heffernan, 2007; Hu et al., 2008). In the meantime, factors that affect banking efficiency had also been explored (Zhang, 2003; Chang and Chiu, 2006; Wang and Tan, 2007; Yao et al., 2007; Ariff and Can, 2008; Fu and Heffernan, 2007; Hu et al., 2008). As for the relationship between stock listing and enhancement of banking efficiency, Liu and Song (2004) proved that among those joint-stock commercial banks (JSCBs), listed banks, such as China Merchants Bank of China (CMBC) and Pudong Development Bank (PDB), had a higher efficiency score than average efficiency score. In addition, Lin and Zhang (2009) observed that listed banks had better performance than those non-listed banks. Since most of the Chinese commercial banks were listed on the Hong Kong and Shanghai stock exchanges over the year 2006 and 2007, the post-IPO data were available. In recent years, a large amount of studies have paid attention to Chinese banking efficiency to obtain an understanding of the process of Chinese banking reform since the year 1978. Sehrt and Park (2001) demonstrated that Chinese banks are inefficient, and the banking ownership reform started from 1990s has not substantially raise the operational performance and banking efficiency of Chinese banking. In terms of Data Envelopment Analysis (DEA), Wei and Wang (2000) tried to empirically conclude the scale efficiency (SE), pure efficiency (PE) and technical efficiency (TE) of Chinese mainly commercial banks within the year 1997, their results proved that the state-owned commercial banks (SOCBs) do suffer technical inefficiency. According to their work, the average technical efficiency of the SOCBs was 62.39% and was much lower than the newly established commercial banks, which the average efficiency of 84.59%. Similarly, the empirical results of Chi, Skuly and Brown (2005) concluded that the scale economies of joint-equity commercial banks (JSCBs) were better than that of state-owned commercial banks (SOCBs). Using DEA approach, Zhao (2000), Qin and Qu (2001), Chen (2002), Zhao et al. (2002), Zhang (2003), and Liu (2004) also confirmed that Chinese SOCBs do suffer inefficiency. To conclude, studies on banking efficiency always focus on the relationship between inputs and outputs. There are three ways to measure the indicators: production method, intermediate method and profit-output method, the first two are usually adopted. Stochastic Frontier Approach (SFA) and Data Envelopment Analysis (DEA) are the most popular approaches in computing banking efficiency.
Chapter 3 Process of Chinese banking reforms Chinese financial system has a long history, but the modern Chinese banking system was created after the year 1948, when People's Bank of China (PBOC) established. There were two banking reforms since Chinese modern banking system set up: the first banking transformation started from the economic reform and open up in 1978, and the latest banking ownership reform began after China entered World Trade Organization (WTO) at the end of 2001. The Chinese modern banking system was initially created following the socialist banking system in the former Soviet Union. At that time, Chinese owned a mono banking system, which concentrate all the roles of central banking and commercial banking to the People's Bank of China (PBOC). So PBOC, established in 1948, acted as a central bank and the only commercial bank in China for a long period (Dai, 2003). Following the economic reform and open up in 1978, there were several state-owned commercial banks (SOCBs) founded. People's Bank of China (PBOC) was divided into two parts: the central bank part and the bank with commercial operation part. The central bank part was still named People's Bank of China (PBOC), but the commercial operation part was changed into state-owned banks. Among the state-owned commercial banks, the Bank of China (BOC, established in 1912), China Construction Bank (CCB, established in 1954), Agriculture Bank of China (ABC, established in1979), and Industrial and Commercial Bank of China (ICBC) made up the Chinese 'Big Four' state-owned commercial banks Â¼Ë†SOCBsÂ¼â€°, which initially owned some lending function separated from PBOC and eventually became the core part of modern Chinese banking system. Big Four were focus on different aspects of Chinese social hierarchy: the Bank of China (BOC) was mainly focus on foreign exchange business, the China Construction Bank (CCB) was mainly provide services to the large constructions in urban area, the Agriculture Bank of China (ABC) was mostly concentrate on rural finance, and the Industrial and Commercial Bank of China (ICBC) was responsible for industrial and commercial activities in urban areas. In the year 1985, the Big Four were allowed to compete in all sectors. Until the middle of 1990s, Chinese banking system was dominated by the PBOC and Big Four. At that stage, some non-state commercial banks and joint-stock commercial banks Â¼Ë†JSCBsÂ¼â€° were gradually allowed to run business throughout China. Before the year of 1986, all Chinese banks were under the direct leading of the Ministry of Finance and People's Bank of China (PBOC). The state-owned enterprises (SOEs) like Big Four were not set up for profit purpose, but played essential social roles in the whole Chinese economy (Subal C.K and Dan W, 2007). After 1978, banking reforms turned its point into Chinese economic transformation from a planning economy to a market-based economy (Chen, Skully and Brown, 2005). The structure of Chinese modern banking system could be illustrated as Figure 1, which is a pyramid structure (Enrico G and Rubens P, 2009). The top of pyramid is represented by People's Bank of China (PBOC), the central bank of China. Banks on the second layer are the state-owned banks, which make up the core power of Chinese banking system. It can be divided into state-owned commercial banks (SOCBs) and state-owned policy banks. In addition, joint-stock commercial banks (JSCBs) and city commercial banks are also the significant bank types in the Chinese banking system. Figure 1 Structure of Chinese Banking System In order to raise competition ability after joining WTO and with the foreign financial institution access, Chinese second banking reform started and enhanced after the end of 2001. During this reform, the Chinese government unloaded more than 2.6 trillion yuan of non-performing loans (NPLs) from the three biggest SOCBs (CCB, BOC and ICBC) during the year 1999 to the year 2005 to restructure Chinese financial system and then pushed the ownership banking reform to the last phase, ownership restructuring via stock listing and foreign merger & acquisition. In the year 2003, the Chinese Banking Regulatory Commission (CBRC) changed its guidelines to encourage external investment, and this encouragement worked. As a result, all the four biggest SOCBs attracted a large amount of foreign investment from world top financial institutions, such as Royal Bank of Scotland (RBS), Merrill Lynch and Bank of America (BOA). In October 2005, China Construction Bank (CCB) became the first listed SOCB, and this listing trend was followed by Bank of China (June 2006) and Industrial and Commercial Bank of China (October 2006). The initial public offering (IPO) of ICBC (almost $22 billion) was larger than any other IPOs issued in New York, London and Tokyo. By the year of 2009, the last SOCB, Agriculture Bank of China (ABC) finished its listing process. This time also witnessed the listing of other joint-stock commercial banks (JSCBs) such as China Mingsheng Bank (CMSB) and China Merchants Bank of China (CMBC), which also had significant impact upon the financial markets (Yao et al., 2008). During the process of listing of Chinese commercial banks, how to deal with the non-performing loans (NPLs) was the most important step. Over this banking reform, the Chinese government has unloaded over 2.6 trillion yuan of NPLs from the three biggest SOCBs (CCB, BOC and ICBC) during the period 1999 to 2005. Table 1 below indicates the amount and proportion of NPLs in major Chinese commercial banks from the year 2005 to 2009. We can see during this period, the whole amount of NPLs decreased from 1313 billion yuan (2005) to 497.3 billion yuan (2009), which achieved a great progress for dealing with NPLs. Specifically, the percentage of NPLs in state-owned banks dropped from 10.49% (2005) to 1.80% (2009), which made it possible for the listing in stock exchange of state-owned banks. Table 1: NPLs in Major Chinese Commercial Banks (Billion yuan and %) Years 2005 2006 2007 2008 2009 All banks 1313 8.61% 1255 7.09% 1268 6.17% 560.3 2.42% 497.3 1.58% State-owned commercial banks (SOCBs) 1072 10.49% 1053 9.22% 1115 8.05% 421 2.81% 362.7 1.80% Joint-stock commercial banks (JSCBs) 147 4.22% 117 2.81% 86 2.15% 65.7 1.35% 63.7 0.95% City commercial banks 84.2 7.73% 65.5 4.78% 51.2 3.04% 48.5 2.33% 37.7 1.30% Sources: China Banking Regulatory Commission (CBRC) 2005-2009, Bank scope (2005-2009) and author's calculation. Influenced by the world financial crisis started in 2008, share prices of Chinese commercial banks dropped at first. However, as the national economy developed, Chinese banking stocks recovered rapidly (Luo et al. 2011). In the early of 2009, China's three largest SOCBs had replaced the American and European financial giants and become the three largest commercial banks in the world, which measured by the market after they were listed on the Hong Kong Stock Exchange (HKSE) and Shanghai Stock Exchange (SSE) only two years later. Thanks to their nature of risk-averse, Chinese listed SOCBs were not influenced too much by financial crisis and turned out to be the least affected financial institutions by the current financial crisis among all other large banks in the world. However, in the process of continued step of globalization, the commercial banks need to adjust corporate structure, accelerate their business innovation, cut operational expenditures and improve their risk management ability to make them healthy and powerful in the future (Luo et al. 2011).
Chapter 4 Empirical model 4.1. Data Envelopment Analysis (DEA) Data Envelopment Analysis (DEA) is the main pattern of non-parametric method originally put forward by Charnes, Cooper and Rhodes in 1978, while Sherma and Gold(1985) first applied DEA to the efficiency evaluation of a bank branch. In 1993, Yue and Berger used DEA to analysis the efficiency among banks again. And in the year 1997, Berger and Mester did the research on the scale of bank, the form of organization and standard of capitalization using DEA approach. DEA is a system analysis method, which is aim to synthesize different indicators and make it an overall indicator, and then use this overall indicator to evaluate the overall situation. As a non-parametric linear programming approach, DEA needs no estimation parameters, and could have an objective result, which makes it a perfect method to analysis the efficiency of banking system. DEA does not need the clear relationship between inputs and outputs; it just uses the linear programming to find the relative valid sample, and uses the plane combined by the relative valid decision-making unit (DMU) combination as the efficiency frontier. If the DMU at last hit the efficiency frontier boundary, it is efficiency with the value of 1; and if the DMU eventually lands inside the efficiency frontier boundary, it is relative inefficiency with the value between 0 and 1. DMU stands for the decision-making unit, which is estimated by the ratio of the weighted outputs over the weighted inputs. Every DMU is an entity responsible for converting multiple inputs into multiple outputs (Wu. 2006). Charnes, Cooper and Rhodes (CCR) model is an input-orientated DEA model, which means measuring the ratio of reduced inputs under a certain output. When it comes to commercial banks, it is easier to control their inputs rather than outputs. So it is better to use an input-orientated DEA model to measure banking efficiency. However, there also exist some drawbacks on DEA approach. When data is incomplete, the results from DEA will not the reliable explanation. According to the returns to scale, Data Envelopment Analysis (DEA) includes two models: the Charnes, Cooper and Rhodes (CCR) model and the Banker, Charnes and Cooper (BCC) model. CCR model refers to the DEA approach under the assumption of constant returns to scale (CRS), while BCC model stands for the DEA approach with variable returns to scale (VRS). The paper will use both CCR and BCC methods to analysis the banking efficiency and then compares their results. If the two results under different assumption are similar, it stands for the banking efficiency has no relationship to bank scale, otherwise, it is related to returns to scale. 4.2 The Charnes, Cooper and Rhodes (CCR) model CCR model is one of the DEA approachs estimated under the assumption of constant returns to scale (CRS), which developed by Charnes, Cooper and Rhodes in 1978. It uses the linear programming to obtain the efficiency frontier and computes the relative efficiency in each decision-making unit (DMU). If the DMU at last hit the efficiency frontier boundary, it is efficiency with the value of 1; while the other DMU at last inside the efficiency frontier boundary is relative inefficiency with the value between 0 and 1. CCR can be expressed as following: Minise ÃŽÂ¸Â¼Å’ Subject to Ã¢Ë†â€˜XijÃŽÂ·j Ã¢â€°Â¤ ÃŽÂ¸Xik i=1, Ã¢â‚¬Â¦, m Ã¢Ë†â€˜yrjÃŽÂ·j Ã¢â€°Â¥ yrk r=1, Ã¢â‚¬Â¦, s; ÃŽÂ·j=1, Ã¢â‚¬Â¦,n Where xik and yrk are the ith input and rth output of the kth decision-making unit (DMU), and ÃŽÂ· is a non-negative vector of variables. The simple express of Cooper and Rhodes (CCR) Model is based on the assumption that decision-making units (DMUs) were acted under constant returns to scale (CRS). 4.3. The Banker, Charnes and Cooper (BCC) model Considering the variable returns to scale (VRS), Banker, Charnes and Cooper (1984) gave up the constant returns to scale (CRS) assumption to measure the efficiency under different returns to scale. The value of scale efficiency in BCC model can compare the technical efficiency of a bank under constant returns to scale and variable returns to scale. If there is no difference between these two results, it stands for the bank inefficiency does not come from the scale factor. Otherwise, the inefficiency of that bank comes from the scale inefficiency. The relationship can be showed as following: TECRS= TEVRS* SE Or SE= TECRS/ TEVRS As shown above, TECRS stands for the technical efficiency value under constant returns to scale (CRS); TEVRS stands for the technical efficiency value under variable returns to scale (VRS). Under the BCC model, the technical efficiency can be classified as pure technical efficiency (PTE) and scale efficiency (SE). It is convenient to tell how much inefficiency of commercial banks comes from pure technical inefficiency, and how much from scale inefficiency. 4.4 Variables of Inputs and Outputs under DEA approach In general, there are three methods for measuring inputs and outputs of banks: production method, intermediate method and profit-output method. Under production method, banks are viewed as the service provider to account holders. This assumption need to measure the output under the amount of dealing transaction, which is private and difficult to obtain. The production method is more suitable for study the efficiency of branch of banks. When it comes to intermediate method, financial institutions are treated as the media for savers and investors. Under this assumption, the deposits of savers, salaries and assets are banks inputs, and loans and currency amount are the outputs. The main default of intermediate method is the neglect of risk. The third approach is the profit-output method makes the operating income as output and the deposits as input. The problem is that the operating income stands for the real cash input, but cannot completely reflect the operating performance. This paper chooses to combine the production method and intermediate method to select the input and output indicators. Specifically speaking, the input variables include deposits, number of employees and fixed assets of commercial banks. And the output indicators include loans and net incomes of those banks. All of three inputs are strongly correlated to the two outputs, and the two output variables are correlating each other. Table 2 below demonstrates the variables of inputs and outputs in this paper. Table 2 Variables of Inputs and Outputs X1 X2 X3 Y1 Y2 Deposits Employees Fixed assets Loans Net incomes
Chapter 5 Data and Empirical results 5.1. Data Description All the data for the study in this paper are obtained from the Almanac of China's Finance and Banking (2005-2009), Chinese Statistical Yearbook (2005-2009) and Bankscope (2005-2009). The sample was composed of nine listed commercial banks: Industry and Commercial Bank of China (ICBC), Agricultural Bank of China (ABC), Bank of China (BOC), China Construction Bank (CCB), China Investments and Trust Bank (CITIC), China Bank of Communications (CBC), China Mingsheng Bank (CMSB), Shenzhen Development Bank (SDB) and China Merchants Bank of China (CMBC). As showed above, there are three methods for measuring inputs and outputs of banks: production method, intermediate method and profit-output method. In this paper, we choose the combination of production method and intermediate method to select the input and output indicators. The input variables include deposits, number of employees and fixed assets of commercial banks. And the output variables choose loans and net incomes of these 9 commercial banks. The software for the computing the result is DEA Excel Solver developed by Zhu in 2003. 5.2 Empirical results Table 3 describes the inputs and outputs of the 9 Chinese mainly commercial banks in the year 2009. The data for efficiency analysis compose of all 9 banks over the period 2005-2009. According to the table, in the year 2009, Industry and Commercial Bank of China (ICBC) owned the largest scale both in fixed assets (117850.53 billion yuan) and deposits (97713 billion yuan), while Agricultural Bank of China (ABC) had the largest number of employees (389827). As for the output variables, ICBC also owned the most amount of loans (40614.44 billion yuan) and net incomes (1672.48 billion yuan). In general, the 4 state-owned commercial banks had the larger scale than those 5 joint-stock commercial banks both in inputs and outputs. Table 3 Inputs and Outputs of Chinese Commercial Banks: 2009 (billion yuan) Bank Deposits Employees Fixed Assets Loans Net Incomes ICBC 97713 389827 117850.53 40614.44 1672.48 ABC 74974 440830 88811.55 30149.84 738.09 BOC 58878 262566 77711.53 19680.85 929.65 CCB 80013 301537 96233.55 35113.15 1387.26 CITIC 13419 24180 17750.31 6649.24 192.66 CBC 23728 79122 32949.08 13285.90 373.32 CMSB 11254 12301 14040.87 6584.10 166.00 Sources: Almanac of China's Finance and Banking (2005-2009); Chinese Statistical Yearbook (2005-2009) Table 4 presents the estimated CCR efficiency scores for the 9 commercial banks in 2005-2009, while Figure 2 presents the efficiency of the Big Four SOCBs in this period. The results of DEA-CCR model are shown below in Table 5, while Table 6 indicates the efficiency of Chinese Listed Banks before and after IPOs under DEA-BBC model. Table 4 Efficiency Scores of Chinese Commercial Banks: 2005-2009 (CCR) DMU 2005 2006 2007 2008 2009 Average ICBC 0.80 0.80 0.87 0.94 0.97 0.88 ABC 0.73 0.81 0.87 0.91 0.99 0.87 BOC 0.82 0.90 0.90 0.97 1.04 0.92 CCB 0.73 0.74 0.80 0.97 0.98 0.85 CITIC 0.91 0.95 0.91 0.98 1.01 0.95 CBC 0.82 0.83 0.88 0.87 0.93 0.87 CMSB 0.90 0.78 0.81 0.88 0.90 0.86 SDB 0.82 0.83 0.88 0.88 0.90 0.87 CMBC 0.81 0.83 0.83 0.81 0.84 0.83 Sources: Almanac of China's Finance and Banking (2005-2009); Chinese Statistical Yearbook (2005-2009). Figure 2 Efficiencies of the Big Four SOCBs (2005-2009) It can be concluded from Table 4 that during the year 2005-2009, the average efficiency scores of Big Four SOCBs were lower than that of the other commercial banks. However, it is clearly that in the year 2008 and 2009, the efficiencies of state-owned commercial banks were higher than that of joint-stock commercial banks. According to Figure 2, during the period of 2005 to 2009, the efficiency of the Big Four SOCBs all witnessed a huge increase. Agricultural Bank of China (ABC) is the last SOCB listed on stock exchange. After its ownership reform finished in 2009, the efficiency of ABC rushed to 0.99, about 25% higher than that in 2005. Table 5 Efficiency of Chinese Listed Banks before and after IPO - CCR DMU One year before IPO IPO year One year after IPO ICBC 0.80 0.85 0.86 ABC 0.81 0.85 0.87 BOC 0.90 0.93 0.90 CCB 0.74 0.78 0.80 CITIC 0.95 0.98 0.91 CBC 0.83 0.90 0.88 CMSB 0.78 0.80 0.81 SDB
- CMBC 0.81 0.84 0.83 Sources: Almanac of China's Finance and Banking (2005-2009); Chinese Statistical Yearbook (2005-2009). Table 6 Efficiency of Chinese Listed Banks before and after IPO - BCC DMU One year before IPO IPO year One year after IPO ICBC 0.94 0.99 1.01 ABC 0.91 0.95 0.97 BOC 0.97 1.04 1.01 CCB 0.97 0.98 1.01 CITIC 0.98 1.01 0.99 CBC 0.87 0.93 0.90 CMSB 0.88 0.90 0.92 SDB
- CMBC 0.81 0.84 0.85 Sources: Almanac of China's Finance and Banking (2005-2009), Chinese Statistical Yearbook (2005-2009). It is obviously to be seen from tables above is that, after the recent banking reform, there is a great progress of banking efficiency for Chinese state-owned commercial banks (SOCBs). Taking Bank of China (BOC) as an example, before its IPO year (2006), the banking efficiency was 0.82, which was a medium level among all the nine commercial banks. However, after it was listed in 2006, the efficiency of BOC increased rapidly over 0.9, and even reached 1.04 in 2009. Compared with the other commercial banks, BOC is now an outstanding international bank with high operational efficiency. As for the other three SOCBs, there is a same trend as BOC. By the end 0f 2009, ABC, the last listed SOCB owned the efficiency of 0.99, which get rid of the fortune of low efficiency and low competition ability. According to table 5 and table 6, the DMUs under CCR and BCC approach are different, which stands for that the efficiency of these commercial banks comes from the scale efficiency. Since the preparation of listing for the SOCBs, they did a lot of works including expanded their scale. The banking efficiency of the Big Four was clearly increased as the result of IPOs. The empirical results above indicate that stock listing is an effective way for commercial banks to improve banking efficiency. It can be seen that the average efficiency of SOCBs after IPOs year was about 6 percent more than that before IPOs. So we could confirm that the recent banking reform in China had achieved its expected objective. Since the banking reform focus on the ownership transformation via foreign participation and stock listing, there is a great progress for the 'Big Four' state-owned commercial banks (SOCBs) both in their operational efficiency and in profitable results.
Chapter 6 Conclusion and Implications The paper uses DEA approach to analysis the banking efficiency of nine commercial banks, and compares the efficiency scores before and after their IPOs. Since the process of Chinese banking reform started in 2005, the Chinese banking system has developed impressively, especially the Big Four SOCBs. After their IPOs, the Big Four achieved a great progress in the operational efficiency. Nowadays, the SOCBs still dominate the Chinese banking market. Since the world financial crisis happened in 2008, Chinese SOCBs became the top profitable commercial banks around the world, which turned out to be the least affected financial institutions by the current financial crisis. In the year 2009, CCB, BOC and ICBC replaced the American and European giants to become the first three largest commercial banks in the world measured by market after they were listed on the Hong Kong Stock Exchange (HKSE) and Shanghai Stock Exchange (SSE). In this paper, we evaluate the efficiency of 9 listed commercial banks over the period 2005-2009, to compare the operational banking efficiency of SOCBs before and after their IPOs. We use the Data Envelopment Analysis (DEA) to measure the banking efficiency, and both CCR model and BCC model are adopted. As the empirical results shown above, we can find out that Chinese SOCBs have benefited a lot from the latest banking reform. After this transition, they have developed a range of competitive strategies in both domestic banking market and overseas market. It is overwhelming evident that a well-functioning financial system is important for a country's economic growth. What's more, the Â¬Ânancial intermediation determines the efficient allocation of savings as well as the return of savings and investment. China's banking sector is the most important component of the Chinese Â¬Ânancial system (with 66% of total Â¬Ânancial assets in 2009) Although there is a great progress in banking efficiency of SOCBs since the ownership reform, we can still find problems in their ways of operate. Compared with the other commercial banks, SOCBs still have the historical disadvantage which will block the step of their fast development. It is important for SOCBs to solve their historical problems and make them more efficient. Only in this way, Chinese banks could compete with other international financial institutions around the world.
REFERENCE Allen L. and A Rai (1996) 'Operational Efficiency in Banking: An International Comparison'. Journal of Banking and Finance. 20, 4, 655-72. Allen N. Berger, Iftekhar H and Ming Zhou (2006) 'Bank ownership and efficiency in China: What will happen in the world's largest nation?' Journal of Banking & Finance. 33, 113-130. Almanac of China's Finance and Banking. Beijing, 1999-2008. Ariff M. and L Can (2008) 'Cost and ProÂ¬Ât EfÂ¬Âciency of Chinese Banks: A Non-Parametric Analysis'. China Economic Review. 19, 2, 260-73. Chen X M, Skuly and K Brown (2005) 'Banking Efficiency in China: Application of DEA to Pre- and Post-Deregulation Eras: 1993-2000'. China Economic Review. 16, 3, 229-45 Dan Luo, Shujie Yao, Jian Chen and Jianling Wang (2011) 'World Financial Crisis and Efficiency of Chinese Commercial Banks'. The World Economy. 1467-9701. Desheng Wu (2006) 'A note on DEA efficiency assessment using ideal point: An improvement of Wang and Luo's model'. Applied Mathematics and Computation. 183, 819-830. Enrico G and Rubens P (2009) 'The Chinese Banking System: Economic Performance and Prospect for Future Development'. Transition Finance, Banking and Currency Research. 16, 92-113 Fu X Q. and S Heffernan (2007) 'Cost-X-Efficiency in China's Banking Sector'. China Economic Review. 18, 1, 35-53. Hsiu-Ling Wu and Chien-Hsun Chen (2010) 'Operational performance of commercial banks in the Chinese transitional economy'. The Journal of Developing Areas. 44, 1, 383-396. John P. B, Iftekhar H and Paul W (2005) 'Bank Performance, efficiency and ownership in transition countries'. Journal of Banking & Finance. 29, 31-53. Lin X C and Y Zhang (2009) 'Bank Ownership Reform and Bank Performance in China'. Journal of Banking and Finance. 33, 1, 20-29. Paul B, McGuinness and Kevin K (2010) 'The Listing of Chinese State-Owned Banks and their Path to Banking and Ownership Reform'. The China Quarterly. 201, 125-155. Shujie Yao, Zhongwei Han and Genfu Feng (2008) 'Ownership reform, Foreign competition and efficiency of Chinese commercial banks: A non-parametric approach'. The World Economy. 1467-9701. Shujie Yao, Chunxia Jiang, Genfu Fen and Dirk Willenbockel (2007) 'WTO challenges and efficiency of Chinese banks'. Applied Economics. 39, 629-643 Wei Y and L Wang (2000) 'Zhongguo shangye yinhang xiaolv yanjiu: yizhong fei canshu de fenxi' ('EfÂ¬Âciency Study of Chinese Commercial Banks, A Non-Parametric Approach'). Jingrong Yanjiu (Journal of Financial Research). 3, 88-96. Y. Wang and Y. Luo (2006) 'DEA efficiency assessment using ideal and anti-ideal decision-making units'. Applied Mathmatics and Computation. 173(2), 902-915. Zhang J. H. (2003) 'Woguo shangyeyinhang de X-xiaolv fenxi' ('X-EfÂ¬Âciency Study ofChinese Commercial Banks'). Jinrong Yanjiu (Journal of Financial Research). 6, 46-57. Zheng M and Y Zhang (2004) 'Zhongguo yinhang xiaoyi de shizheng fenxi' ('Empirical Study on the EfÂ¬Âciency of Chinese Banks'). Xiamen Daxue Xebio (Journal of Xiamen University). 6Â¼Å’106-14.