You are on page 1of 40

Balanced Performance Index and Its Implications: Evidence from Taiwans Commercial Banks

Dar-Yeh Hwang a,* ,Chi-Chun Liu b , Lishu Ouyang c


b

------------------------------------------------------------------------------------------------------Abstract Taking into account only financial factors does not provide complete information on performance, and could possibly lead to less profitable policies or strategies. This paper takes into consideration of both financial and non-financial performances when evaluating 35 sampled publicly traded commercial banks in Taiwan. The performance of banks is measured using an indexing method consisting of financial and non-financial measures. Banks are classified into two categories according either to the year when founded, (i.e., old and new banks), or to the type of major stockholders of a bank when founded, namely, privatized government-owned and private banks. The results show that privatized government-owned/old banks are larger than private/new banks, respectively. Moreover, privatized government-owned banks have significantly higher financial performance index than private banks but both types of banks are not significantly different from each other in non-financial performance index. New and old banks are not significantly different from each other in both financial and non-financial performance indexes. With relatively large scale, higher profitability and better management, banks will perform relatively better among competitors in the following year. Furthermore, non-financial factors are important predictors of future financial and total performance indexes, though individual factor may not be consistently significant. More branch offices, better capital structure and solvency, and higher rates of growth in deposits and loans all result in more profits, and lead to higher customer satisfaction and more efficient management. Providing better technology to customers is an efficient way in promoting customer services, which in turn produces more profits and results in efficient management. CEOs, on average, have plans for better management and more profits. Among the factors that have direct and positive impacts on profitability, increasing the efficiency of management is the most efficient way; on the contrary, adding more branch offices contributes the least profits. Therefore, to increase bank profits, CEOs should aim to improve bank management and capital structure and solvency rather than to add more branch offices.
JEL classification: G21, G28, C12 Keywords: Balanced Performance Measures; Performance Indexing Approach; Privatized Government-Owned Bank, CEO Leadership *Correspondence author: Tel:011-8862-2363-0987; fax:011-8862-2365-7095. E-mail addresses: hwangdar@handel.mba.ntu.edu.tw (Dar-yeh Hwang), ccliu@mba.ntu.edu.tw (Chi-Chun Liu), oyls@faculty.pccu.edu.tw (Lishu Ouyang).

Department of Finance, College of Business, National Taiwan University, Taipei, Taiwan. Department of Accounting, College of Management, National Taiwan University, Taipei, Taiwan. c Department of Economics, College of Management, Chinese Culture University, Taipei, Taiwan.

We thank the Taiwan Academy of Banking and Finance (TABF) for their financial supports. We thank the followings for helpful discussions, comments, and insights: Edward Altman, Gang Shyy, Paul Chiu, Ming-Teh Yu, Conference participants at the 2000 Chung-hwa Banking Association Meetings, the 2001/2002 Enhancing Banking Competitivity Conference at TABF and 2003 Taiwan Conference on Economics, Finance and Accounting,. 1

1. Introduction The first step is to measure whatever can be easily measured. This is OK as far as it goes. The second step is to disregard that which cant be easily measured or to give it an arbitrary quantitative value. This is artificial and misleading. The third step is to presume that what cant be measured easily really isnt important. This is blindness. The fourth step is to say that what cant be easily measured really doesnt exit. This is suicide. --The McNamara Fallacy1 Performance measurement systems play a critical role in evaluating the achievement of firms goals, compensating managers, and developing strategies. With increasing global competition and technology changes, designing a balanced performance measure is critical to the survival and success of companies. We develop a balanced performance measurement as a management tool for enhancing decision-making and accountability, not for evaluating stock or bond performance. As a strategic process, the balanced performance index can be used to assess accomplishment of organizational strategic goals and objectives. Existing financial measures are insufficient at expressing corporate value. Managers depending wholly on financial performance only get an incomplete view of the companies. Thus, there is a pressing need for a set of widely accepted metrics by which managers and investors can rely on to measure the value creation in firms (Kaplan and Norton 1992, 1996). How financial and non-financial performance measures can be integrated into one measure is a necessary ingredient. The performance index should include outcome measures, the performance drivers of those outcomes, short-term and long-term objectives, hard objective measures and more subjective measures. By articulating them clearly, managers can channel the energies, the abilities, and knowledge towards achieving the firms long-term goals. In addition, a balanced performance index can serve as the focal point for the organizations efforts, defining and communicating priorities to managers, employees, investors, even customers, and can be used as a communication, information, and learning system.

The McNamara Fallacy was tagged to Robert S. McNamara, a strikingly successful executive who sought to quantify virtually everything. McNamara was elected as a director of the Ford Motor Company in 1957, and president of the company in 1960. At the request of President John F. Kennedy, McNamara served as Secretary of Defense of the United States, a position he held from 1961 until 1968. He became president of the World Bank Group of Institutions in April of 1968, retiring in 1981. 2

Depository financial institutions in Taiwan include commercial banks, credit cooperatives, credit departments of farmers and fishermen associations, the Postal Remittances, Savings Bank and local branches of foreign banks. At the end of 2001, 53 commercial banks in Taiwan, relatively larger compared with other depository financial institutions, accounted for 71.33% of total deposits accepted and 89.38% of total loans extended. Out of 53 commercial banks, 35 banks are publicly traded in 2001, and some of them are ranked among top 500 banks in the world. Facing the trend of worldwide financial deregulation, commercial banks in Taiwan make all efforts to enhance their performance. After the Commercial Bank Establishment Promotion Decree being approved in 1991, a number of domestic private and foreign banks join the highly competitive banking industry, and result in lower profitability for most banks. Furthermore, the wave of consolidations and globalization has been transforming the financial services landscape. Thus, how to maximize the shareholders value is always the most dominant variable in bank managers decisions. In response to the question of what drives the shareholders value, there are numerous competing measures being developed both in theory and in practice. Some use the economics-based approach or financial information metrics. Since performance measures strongly affect the behavior of managers, employees, and investors (Handy, 1994; Kaplan and Norton, 1992), a more balanced approach, a combination of financial and non-financial measures, has been introduced in economics, strategy, finance and accounting (Porter, 1992; Liebowitz, 1999; Lev, 2001; Kaplan and Norton, 2001a, 2001b; Stewart, 1991a, 1991b). The performance indexing approach is proposed in this paper to measure the performance of Taiwans commercial banks. This paper takes into consideration of both financial and non-financial performances when evaluating 35 sampled publicly-traded commercial banks in Taiwan. Performance measurement systems play a critical role in evaluating the achievement of firms goals, compensating managers, and developing strategies. The performance index takes out the fuzziness and subjectivity. It offers a yardstick by which to compute the impact of various factors. It allows managers and investors a more complete view of the wealth creating potential of their companies, eliminating the partial and restricted view of a strictly financial perspective. Banks are classified into two categories according either to the year a bank was founded, i.e., old and new banks, or to the type of major sponsors of a bank when founded, namely, privatized government-owned and private banks. The categories and weights of the performance index in this paper are selected according to their relative impact based upon the surveys of diverse experts from accounting, finance, strategy, and management. This study
3

seeks to offer a performance metrics and implement this measure to evaluate the Taiwan commercial banks. The major contribution of this paper is the consideration of both financial and non-financial factors in constructing the performance indexes. This provides a complete picture of banking performance both from quantitative (objective) and from qualitative (subjective) perspectives. The results show that privatized government-owned/old banks are larger than private/new banks, respectively. Moreover, privatized government-owned banks have significantly higher financial performance index than private banks in 2001 but both types of banks are not significantly different from each other in non-financial performance index. New and old banks are not significantly different from each other in both financial and non-financial performance indexes. With relatively large scale, higher profitability and better management, banks will perform relatively better among competitors in the following year. Furthermore, non-financial factors are important predictors of future financial and total performance indexes, though individual factor may not be consistently significant. In addition, more branch offices, better capital structure and solvency, and higher rates of growth in deposits and loans all result in more profits, and lead to higher customer satisfaction and more efficient management. Providing better technology to customers is an efficient way in promoting customer services, which in turn produces more profits and results in efficient management. CEOs, on average, have plans for better management and more profits. Finally, increases in the size of a bank in terms of total assets cause inefficient management, and reduce profits or impair customer services. Among the factors that have direct and positive impacts on profitability, increasing the efficiency of management is the most efficient way; on the contrary, adding more branch offices contributes the least profits. Therefore, to increase bank profits, CEOs should aim to improve bank management and capital structure and solvency rather than to add more branch offices. The remainder of the paper is organized as follows. Section 2 briefly reviews some of performance measures. The methodology used in this paper is discussed in Section 3. The

data and the empirical results are presented in Section 4 and Section 5, respectively. Finally, the conclusion is shown in Section 6. 2. Prior Performance Measures Previous studies have measured bank performance from different aspects.
4

Numerous

studies estimated X-efficiency.

Others construct performance indexes based on financial

and/or non-financial data. There have been many measures being proposed over the last two decades to complement the current financial information. The most cited measures include the value creation index (VCI) (Cap Gemini Ernst & Yong), the invisible balance-sheet (Annell et al., 1989), the intangible assets monitor (Sveiby, 1997), the balanced scorecard (Kaplan and Norton, 1996), economic value added (Stewart, 1991a, 1991b), IC-index (Roos et al., 1997), technology broker (Brooking, 1998), Skandia AFS business navigoator, and the financial method of intangible assets measurement (Rodov and Leliaert, 2002). Balanced Scorecard is introduced by Kaplan and Norton (1992) to motivate and measure business performance. The Scorecard with financial and non-financial (i.e., customer, internal business process, and learning and growth) provides a balanced picture of current operating performance as well as the drivers of future performance. Cap Gemini Ernst & Youngs Center for Business Innovation (CBI) develops a value creation index (VCI), a list of the nine most critical categories of non-financial performance that determine corporate value creation: innovation, quality, customer relations, management capabilities, alliances, technology, brand value, employee relations, and environmental and community issues. Economic value added (EVA) is introduced by Stern Stewart and Co., as a comprehensive performance measure to explain corporate value added or lost. The IC-index combines strategy, non-financial measurements, finance, and management value added, and consolidates those factors into a single index. Numerous prior studies adopt frontier approaches to measure bank X-efficiency. Two popular techniques are the nonparametric linear programming approach, often referred to as data envelopment analysis (DEA), and the parametric econometric approaches, specifically, the stochastic frontier approach (SFA). On SFA approach, Kraft and Tirgiroglu (1998) build that during 1994 and 1995 in Croatia, new banks were more X-inefficient and scale-inefficient than old banks and profitability was negatively correlated to X-efficiency. Berger and DeYoung (1997) analyze the relationship between loan quality and cost efficiency in commercial banks and found that cost efficiency was a good indicator of future problem loans or problem banks. By controlling for scale, Kwan and Eisenbeis (1996) find that small banking firms in U.S. were, on average, less X-efficient, and the degrees of X-inefficiency varied a lot among small banks than large banks. In addition, banks with more capital are more efficient than those with less

capital; less efficient banks are higher risk-taking than more efficient banks.2 On DEA approach, Grabowski, Rangan, and Rezvanian (1993) find that branch banking is more efficient than the bank holding company. Grabowski et al. (1994) observe that banks with deposits in excess of $1 billion have the highest technical efficiency. Miller and Noulas (1996) measure technical inefficiency of 201 large U.S. banks during the years of 1984 to 1990 and concluded that banks with relatively lager size and more profits are more technical efficient. Chen, and Yeh (2000), on the other hand, finds that in Taiwan privatized government-owned banks are less technically efficient than private banks in 1996. Aly et al. (1990) suggest that, based on a sample of 322 U.S. independent banks in 1986, bank efficiency is positively correlated with bank size and is negatively related to product diversity. Using a sample of 580 branches of a commercial bank in the UK, Athanassopoulos (1998) find technical inefficiency and diseconomies of scale existed at the branch level. The empirical evidence in Avkirans

study (1999) indicates that bank efficiency rises slowly and steadily in Australia from 1986 to 1995.3 Bauer, Berger, and Humphrey (1998) investigate the consistency and differences of measured operation efficiency obtained using different approaches. The evidence indicates

that nonparametric DEA method yields much lower average efficiency than the SFA parametric approaches do. 3. Methodology Bank performance in this study is measured using indexing procedure. Performance measure is decomposed into financial and non-financial perspectives. Banks are further classified by (1) the year of the establishment (new or old banks), or (2) the government being the major stockholder (privatized government-owned or private banks). The performance indexes of individual bank are first constructed; those of four types of banks (i.e., new, old, privatized government-owned as well as private banks) are derived from individual bank performance indexes accordingly. The performance indexes of different types of banks are examined. The Spearmans correlations of performance indexes of financial and non-financial perspectives as well as overall performance indexes of banks are analyzed.
Other studies focusing on different issues are available. See Huang and Wang (2001), DeYoung et al. (1998), Mitchell and Onvural (1996), Atkinson and Cornwell (1994), Berger, Hancock and Humphrey (1993) Huang (2000) and Kumbhakar (1996). 3 More studies on banking issues using DEA are also available. See Sathye (2001), Chen(2001), Chiu et al. (2000), Chen and Yeh (2000), Camanho and Dyson (1999), Chen and Yeh (1998), Chang (1997), Resti (1997), Schaffnit, Rosen and Paradi (1997), Fukuyama (1993), Elyasiani and Mehdian (1992), Oral and Yolalan (1990), Sherman and Gold (1985). 6
2

Finally, the impacts on commercial banks profitability of several different factors concerning financial and non-financial performance indexes are studied. 3.1. Constructing Performance Index 3.1.1. Selection of Performance Measures and Weights Kaplan and Norton (1996) emphasize that balanced scorecards should reflect four types of measures: (1) financial and nonfinancial; (2) external and internal; (3) input/drivers and outcomes/results; (4) objective and subjective. However, different types of measures are not mutually exclusive. For example, financial measures (such as return on assets) could be external, outcomes/results, or objective. Our performance index includes outcome measures, the performance drivers of those outcomes, short-term and long-term objectives, hard objective measures and more subjective measures. This study initially reviews prior literature on performance measures, workers compensation and CEO incentive plans, and selects four types of possible measures for our performance index. The preliminary list of performance measures (about 60 different measures) is provided to NTU EMBA students who rank each of the measures by its importance to the success of commercial banks. Then, we reduce the list to about 30 measures. The final list of performance measures, categories and weights is determined according to their relative impact based upon the surveys of diverse experts from accounting, finance, strategy, and management. The performance index is evaluated from two aspects, namely, financial and non-financial aspects. Financial and non-financial performance components are measured based on five and four characteristics, respectively. characteristic is composed of several important factors, as listed in Table 1. Each

3.1.2. Financial performance index Financial performance of a bank is dependent on its capital structure and solvency, management efficiency, profitability, scale and growth, all of which are evaluated based on different financial ratios derived from financial statements. Capital structure and solvency is determined by three ratios, i.e., liability ratio, risk-based capital ratio, and the current ratio. Management efficiency is measured by NPL ratio, asset turnover and operating revenue per employee. The size of a bank (i.e., bank scale) is defined by total assets. The growth rates of both deposits and loans are taken into account when evaluating the growth of a bank. For the purpose of intertemporal and cross-sectional comparisons, the values of all considered factors of individual bank are standardized. Let Xijt be the value of jth factor of ith
7

bank at time t. The standardized value is calculated as


X ijt jt

Z ijt =

jt

(1)

where jt and jt are, respectively, the sample mean and standard deviation of the jth factor at t4. Bank i is doing relatively well at time t than the average in terms of jth factor if the standardized value (Zijt) is greater than zero, and is doing relatively worse if the value is less than zero. -Insert Table 1 about hereThe performance index of each financial characteristic is constructed by averaging standardized values through relevant factors with predetermined weights. The performance index represents the relative importance of each category: the more important a factor is in determining a banks value, the greater its weighting in the index. The weights, as indicated in the second column of Table 1, are selected according to their relative impact based upon the surveys of diverse experts from accounting, finance, strategy, and management.5 For example, the performance index of capital structure and solvency for Bank i was calculated as follows: Capital Structure and Solvency Index = Ei1t = 0.45*(-1)*Zi1t+0.5* Zi2t+0.05* Zi3t where Zi1t: the standardized ratio of liability to total assets for the ith bank at time t; Zi2t: the standardized risk-based capital ratio for the ith bank at time t; Zi3t: the standardized current ratio for the ith bank at time t. Since the higher is the liability ratio, the more likely does the bank have troubles of paying its customers, and hence, the more risky is the bank in terms of capital structure and solvency. The negative multiplier (-1) associates with the ratio of liability to total assets takes into account the negative influence of liability ratio on bank performance, as was indicated in the third column of Table 1. Bank performance indexes for management (Ei2t), profitability (Ei3t), scale (Ei4t), and growth (Ei5t) were calculated similarly. Finally, financial performance index of each bank was the weighted average of performance indexes of capital structure and solvency, management,
jt (jt)) are the sample mean and standard deviation excluding outliers. The standardized values of outliers are replaced by 3 and 3, depending upon whether they are three standard deviations above or below the mean value. 5 The actual impact of a factor may differ from the common perception of experts. 8
4

(2)

profitability, scale, and growth (with weights listed in Table 1). That is, Financial Performance Index of the ith bank at time t = FEit = W j Eijt
j =1 5

(3)

where W1=0.15, W2=0.35, W3=0.20, W4=0.25, W5=0.05, respectively. To evaluate the performances of banks of different types, banks were grouped by the selected criteria. Based on the year when a bank was founded, banks are divided into new banks (founded after the announcement of Commercial Bank Establishment Promotion Decree in 1991) and old banks. Banks are also classified into private banks and privatized government-owned banks depending upon whether the bank was first founded mainly with private funds or with public funds. The performance indexes of banks of different types are calculated by averaging performance indexes through banks of the same type. For new banks, Performance index of jth factor at time t = ZNjt = i new banks Z ijt /n N ,

(4)

Performance index of jth characteristic at time t = ENjt = i new banks E ijt /n N (5) Financial Performance index at time t = FENt = i new banks FE it /n N where nN is the number of new banks. (6)

3.1.3. Non-financial performance index Non-financial performance of each bank is evaluated based on peers and customers ratings. Questionnaires using five-point Likert scale are specifically designed for this research. questions regarding non-financial performance are summarized in Table 2. The

Each bank is

evaluated by its customers from two main categories: (1) customer services, and (2) technology. Questions concerning customer services include six factors, i.e., employees knowledge about their work, employees attitudes toward customers, fees and rates, the diversities of financial information and services provided to customers, security and reliability, and lobby and other facilities. Questions regarding technology concern the services of ATM, Tele-banking as well as e-banking.6 On the other hand, CEOs leadership of each bank is evaluated by peers.

Tele-banking (telephone-banking) and e-banking (electronic-banking) refers to provision of banking products and services through telephone or electronic platforms. These products and services include deposit-taking, lending, 9

Several questions are designed to measure customers satisfactions in services and technology provided by banks and CEOs leadership among peers. The rating of customers satisfactions in any particular service provided by each bank, for example, Reasonable Service Charges, is constructed as the mean value of sample ratings with respect to that service. The rating of any specific factor, for example, Fees and Rates, is the average of the ratings corresponding to the questions (services) listed under that factor. - Insert Table 2 about here -To construct non-financial performance index, the ratings of all factors are first converted into standardized values, respectively, as is in equation (1). The standardized values are averaging through relevant factors to calculate the performance indexes of CEO leadership, customer services, as well as technology. Equal weights are used in all cases. Finally, non-financial performance indexes of banks of different types are constructed by averaging performance indexes through banks of the same type. For new banks, for example, Performance index of jth non-financial factor at time t = ZNjt =
i new banks

NZ

it

/n N ,

(7)

Performance index of jth non-financial characteristic at time t = NENjt =


i new banks

NE

ijt

/n N

(8)

Non-financial performance index at time t = NFENt =

i new banks

NFE

it

/n N

(9)

3.1.4. The composite index of total performance index The composite indexes of total performance index of each individual bank (TEit) and banks of different types are constructed by averaging through financial and non-financial performance indexes with weights 2/3 and 1/3, correspondingly. That is, Total performance index of bank i at time t = TE it = (2 FEit + NFEit ) / 3
electronic bill payment, account management, and other financial services. In Taiwan, stand-alone virtual banks are not allowed. That is, e-banking services are offered only as an extension and complement to existing other services. 10

Total performance index of new banks at time t = TE Nt = (2 FE Nt + NFE Nt ) / 3 Likewise, performance indexes of banks of other types are calculated. Based upon the surveys of diverse experts from accounting, finance, strategy, and management, we assign relatively heavier weight to objective evaluation (financial performance index) than to subjective evaluation (non-financial performance index).

3.2. Hypotheses Testing Government-owned banks used to command the Taiwan banking industry until the early 1990s, when Taiwan started financial liberalization by granting new banking licenses and encouraging foreign banks to join the domestic market. With increasing competition and the governments privatization policy, 1991 marked the turning point for market dominance to shift from government-owned banks to private banks. Thus, we examine (1) whether banks established before or after 1991 (Old vs. New banks) perform differently because of timing of their entry into the market and (2) whether privatized government-owned or private banks perform differently because of government support and/or favor. Performance index of banks of different types are compared. The following hypotheses are tested in all categories, including capital structure and solvency, management, profitability, scale, growth, customer services, technology, CEO leadership as well as financial and non-financial performance indexes. This study examines how the performance varies between banks of different types. Our sample includes 35 banks (i.e., 8 privatized government-owned banks and 27 private banks, and 16 new banks and 19 old banks). In sum, our first hypothesis is as follows (in alternative form):

Hypothesis 1: Old (privatized government-owned) banks perform better than new (private) banks. That is, old (privatized government-owned) banks have higher performance indexes than new (private) banks.

Our performance index includes outcome measures and the performance drivers of those outcomes, hard objective measures and softer, more subjective measures. Our performance measure also provides a balance of short-term and long-term objectives. As a result, the
11

balanced performance index can serve as the focal point for the organizations efforts, defining and communicating priorities to managers, employees, investors, even customers, and can be used as a communication, information, and learning system. By articulating them clearly, managers can channel the energies, the abilities, and knowledge towards achieving the firms long-term goals. Non-financial performance measures in our study can be interpreted as leading indicators that provide information on future performance that is not contained in current financial measures. Thus, we expect that our financial and non-financial performance indexes can be used to predict bank future performance. Our second hypothesis is stated in alternative form as follows:

Hypothesis 2: The current performance indexes can be used to predict future performance. That is, current performance indexes are positively associated with future performance indexes.

The other purpose in this study is to examine whether a tailor-made financial and non-financial performance measures, unique to banks, can provide significant insights into bank value creation. Thus, we further examine the impacts of financial and non-financial factors on Management, Profitability and Customer Services The interactions of profitability, management and customer services performance indexes are investigated by using a system of three equations in order to take their endogeneity into account. Our third hypothesis is stated as follows (in alternative form): Hypothesis 3: Other financial and non-financial factors affect profitability, management and customer services performance indexes. Furthermore, profitability, management and customer services performance indexes are interrelated.

4. The Data The data used in this paper consists of panel data of 35 sampled publicly-traded

12

commercial banks in Taiwan for the years of 2000 and 20017.

Nineteen out of 35 sampled

banks are old banks founded before 1992 and the other sixteen of them are new banks founded after 1992. Eight out of 35 sampled banks are privatized government-owned banks and 27 of them are private banks. Table 3 lists the members of banks of all types. -Insert Table 3 about here-

Financial data of each bank are collected mainly from ROC Securities and Futures Institute database and bank annual reports. Non-financial data are collected by surveys based upon peers rating (CEO questionnaires) and customers rating (customer surveys). CEO questionnaires are self-administered fax-delivered to CEOs of the 35 sampled commercial banks, 23 and 21 of which responded for the years of 2000 and 2001, respectively. On the other hand, both personal-interviewing and self-administered methods are employed on customer surveys8. 2583 and 2792 response data are collected for the years of 2000 and 2001, respectively.

5.

The Empirical Results

5.1.

The performance index of individual banks To examine how performance of each bank is relative to that of the peers in certain

category, banks are ranked according to performance index of interest. Table 4 reports the ranking results of individual banks based on the performances of financial and non-financial perspectives as well as total performance index. It is interesting to see that banks continuously being on the top-ten list of the best financial performance are mostly old banks. These banks are Chang Hwa Commercial Bank, First Commercial Bank, Hua Nan Commercial Bank, International Commercial Bank of China, Central Trust of China, Chiao Tung Bank, United World Chinese Commercial Bank and Taipei Bank. On the contrary, banks continuously being among the worst tens are all private banks.
7

Several banks were merged or were converted into holding companies in 2002 after the Taiwan congress passed the bill of Financial Holding Company Act. Financial and non-financial data of these banks in 2002 were not available and hence not compatible to those before 2002, and thus year 2002 and later are excluded from our sample. 8 A personal interview and a self-administered questionnaire seeking the same data generally provided sufficient similarity of answers to enable them to be combined (Cooper and Schindler, 1998). 13

These banks are Hsinchu International Bank, Tainan Business Bank, Kaohsiung Business Bank, Taitung Business Bank, Taichung Commercial Bank, Chung Shing Bank, Pan Asia Bank and Bank of Overseas Chinese. Similarly, banks continuously being the worst non-financial performance are private banks. Do privatized government-owned banks on average have better performance than private banks? Do old banks perform better than new banks? Are they true in what aspect? In order to

find out the answers, we compare the performance indexes of banks of different types. -Insert Table 4 about here-

5.2.

The performance indexes of banks of different types The descriptive statistics and test results for the comparisons of financial performance

indexes between old and new banks and between private and privatized government-owned banks are summarized in Tables 5. Likewise, the descriptive statistics and test results for the comparisons of non-financial performance indexes between banks of different types are reported in Tables 6. -Insert Table 5 about here-

5.2.1 Privatized Government-Owned Banks vs. Private Banks Table 5 shows that private banks have, on average, higher performance scores than privatized government-owned banks in terms of current ratio; their performance in liability ratio is also better than privatized government-owned ones in 2001. Privatized government-owned banks are larger than private banks in terms of scale; they have better performance than private banks in profitability only in 2000. The two types of banks are

always not significantly different from each other in management and growth. Overall, privatized government-owned banks are significantly better performers than private banks in financial performance of both 2000 and 2001. Table 6 shows that private banks have higher performance scores than privatized government-owned banks in fees and rates, but the two types of banks are no difference in
14

terms of customer services.

The CEOs of privatized government-owned banks are considered

by the peers to have better visions in 2001. Overall, privatized government-owned banks and private banks are not significantly different from each other in the non-financial performance category. -- Insert Table 6 about here -5.2.2 New Banks vs. Old Banks On financial performance indexes, Table 5 shows that new banks have, on average, higher performance indexes than old banks in NPL ratio in 2001. Old banks are always larger in scale. In general, old banks and private banks are not significantly different from each other in the category of financial performance index during 2000 and 2001. On non-financial performance indexes, Table 6 shows that new and old banks are no different except in 2001, in which new banks perform better in customer services due to better attitudes toward customers (i.e., service quality) and relatively more comfortable lobby and facilities. Overall, new and old banks are not significantly different from each other in the non-financial performance category.

5.3. Prediction of Future Performances 5.3.1. Spearmans rank correlation coefficient To examine the correlations among the rankings, Spearmans rank correlation coefficients for financial and non-financial performance indexes as well as total performance indexes of banks are computed and reported in Table 7. -- Insert Table 7 about here -Table 7 shows that the rankings of banks based on total performance indexes are highly and positively correlated with those based on financial performance indexes. In addition, the rankings of banks based on non-financial performance indexes are positively correlated with those based on financial performance indexes. This result supports that our non-financial performance captures the drivers of outcomes and provides a balance of short-term and long-term objectives. As a result, the balanced performance index can serve as the focal point for the organizations efforts.
15

The evidence also suggests that lagged performance in financial and non-financial categories affects future total performance indexes. In other words, financial and non-financial performance indexes of this period will have impacts on banks future performance with different degree. This result supports our second hypothesis that our financial and non-financial performance indexes can be used to predict bank future performance. Thus, by articulating performance measures clearly, bank managers can channel the energies, the abilities, and knowledge towards achieving the firms long-term goals. Our results indicate that non-financial performance measure is highly related to both current and future financial performance indexes. Our findings imply that non-financial performance measures in our study are highly value-relevant for banks, and leading indicators that provide information on future performance are not contained in current financial measures. Consistent with the literature, predictive ability is one of the primary benefits of non-financial measures.

5.3.2. Prediction Results High correlations among lagged and current performance ranks denote that lagged performance indexes can be used to predict current performances. Two new variables measuring the size of a bank are introduced in the prediction regression analyses. They are BRANCH (the number of branch offices) and EPLEBR (the number of employees per branch office, in hundred). Eight different models are considered here. Models A1 through A4 examine the effects on total performance indexes of lagged financial and non-financial performance factors. Model A1 : TP2001 = A0 + A1CSS 2000 + A 2 MANGMT2000 + A3 PROFIT2000 + A4 GROWTH 2000 + A5 SCALE 2000 + A6 CEO2000 + A7 SRVCE 2000 + A8TECH 2000 + A, 2001 Model A2 : TP2001 = A0 + A1CSS 2000 + A 2 MANGMT2000 + A3 PROFIT2000 + A 4 GROWTH 2000 + A5 BRANCH 2000 + A6 CEO2000 + A7 SRVCE 2000 + A8TECH 2000 + u A, 2001 Model A3 : TP2001 = A0 + A1CSS 2000 + A 2 MANGMT2000 + A3 PROFIT2000 + A 4 GROWTH 2000 + A5 EPLEBR2000 + A6 CEO2000 + A7 SRVCE 2000 + A8TECH 2000 + v A, 2001 Model A4 : TP2001 = A0 + A1 FP + A 2 NFP2000 + A, 2001

where TP2001: total performance score in 2001 CSS2000 : the performance score of Capital Structure and Solvency in 2000;
16

MANGMT2000 : the performance score of Management in 2000; PROFIT2000 : the performance score of Profitability in 2000; GROWTH2000 : the performance score of Growth in 2000; CEO2000 : the performance score of CEO Leadership in 2000; SRVCE2000 : the performance score of Customer Services in 2000; TECH2000 : the performance score of Technology in 2000. BRANCH2000 : the number of branch offices in 2000; EPLEBR2000 : the number of employees per branch office in 2000. Models B1 through B4 inspect the effects on financial performance indexes of lagged financial and non-financial performance factors. Model B1 : FP2001 = B 0 + 1CSS 2000 + B 2 MANGMT2000 + B 3 PROFIT2000 + B 4 GROWTH 2000 + B 5 SCALE 2000 + B 6 CEO2000 + B 7 SRVCE 2000 + B8TECH 2000 + B , 2001 Model B 2 : FP2001 = B 0 + B1CSS 2000 + B 2 MANGMT2000 + B 3 PROFIT2000 + B 4 GROWTH 2000 + B 5 BRANCH 2000 + B 6 CEO2000 + B 7 SRVCE 2000 + B8TECH 2000 + u B , 2001 Model B3 : FP2001 = B 0 + 1CSS 2000 + 2 MANGMT2000 + 3 PROFIT2000 + 4 GROWTH 2000 + B 5 EPLEBR2000 + B 6 CEO2000 + B 7 SRVCE 2000 + B 8TECH 2000 + v B , 2001 Model B 4 : FP2001 = B 0 + B1 FP + B 2 NFP2000 + B , 2001

where FP2001: financial performance score in 2001.

Table 8 reports the prediction results of current total performance index on lagged financial and non-financial performance indexes. The prediction results of current financial performance on lagged financial and non-financial performances are reported in Table 9. Unless stated otherwise, we say that a coefficient is significant if it exceeds the 90% confidence level in one-tailed test. -- Insert Table 8 about here --- Insert Table 9 about here -Table 8 and 9 indicate that customer services and technology are significantly positively related to future total performance (financial performance) in Models A1 (B1) and A3 (B3)
17

respectively. In addition, non-financial factors have a significant and positive impact on total performance and financial performance in the following year. Among five financial factors, profitability, management and scale in terms either of the number of branch offices or of assets all significantly and positively affect total performance and financial performance in the following year. As a whole, current financial performance also has a significant and positive impact on total performance in the following year. Similarly, current financial performance also leads to better financial performance in the following period. The above results suggest that with relatively large scale, higher profitability and better management, banks will perform relatively better among competitors in the following year. Furthermore, non-financial factors are important predictors of future financial and total performance indexes, though individual factor may not be consistently significant. 5.4. Simultaneous Equations In addition to BRANCH (the number of branch offices) and EPLEBR (the number of employees per branch office), EPLEE (the number of employees) is also created and tested for its influence on performances of different factors. The standardized values of all variables relating to financial and non-financial performance indexes are relatively too small compared to the numbers of employees, branch offices and employees per branch, and are adjusted by 100.

5.4.1. Correlation Analysis The correlation coefficients of factors appeared in regression analysis are calculated and reported in Table 10. -- Insert Table 10 about here -Bank scale (SCALE), the number of branch offices (BRANCH), and the number of employees (EPLEE) are highly correlated. These variables would not be included in the same regression. They are used interchangeably in three equations to measure size effects.

5.4.2. Regression Results Based on the correlations among variables, the single regression model for each of
18

profitability, management, and customer services is first developed. To control for endogenous effects, the three equations are estimated simultaneously using three-stage least square method. Several systems of equations with different scale variables are studied. Table 11 reports, among several possible models, the best regression results for analyzing the impacts on profitability, management, and customer services of variables of interest. The system of three equations is as follows: PROFITt = 0 + 1CSS t + 2 MANGMTt + 3 GROWTH t + 4 SRVCEt + 5 BRANCH t + t1 MANGMTt = 0 + 1 PROFITt + 2 GROWTH t + 3SCALE t + 4 CEOt + 5 SRVCEt + t 2 SRVCEt = 0 + 1 MANGMTt + 2 CEO t + 3TECH t + 4 EPLEBRt + t 3

All variables are the same as those defined for Models A1 through A4. Subscript t denotes the time. It is a panel data model with 33 observations (banks) in each of the two years (2000 and 2001). -- Insert Table 11 about here -Regression results suggest that capital structure and solvency, management, growth, and the number of branch offices all have significant and positive impacts on profitability. Profitability and the quality of customer services as well as CEO leadership all have significant and positive impacts on a banks performance in management. The size of bank, on the other

hand, has a significant and negative impact on management suggesting that larger banks in Taiwan, on average, probably do not take advantage of their competitive edge. Consequently,

the larger is the banks scale; the worse is a banks performance in management. Finally, financial performance had a positive impact on customer services directly through management and indirectly through profitability. On non-financial performance, technology provides to customers, as expected, has a significant and positive effect on customer satisfactions. On the contrary, CEOs have strategies planned for long-term profitability, which may contradict short-term customers expectations. As a result, CEO leadership has a significant and negative impact on customer satisfactions. Figure 1 illustrates the relationship among variables. Solid lines indicate significantly positive effect while dash lines indicate the opposite.
19

The arrows show the cause to effect.

-- Insert Figure 1 about here -Figure 1 shows that 1. More branch offices, better capital structure and solvency and higher rates of growth in deposits and loans all result in more profits, which in turn lead to efficient management and better customer services. 2. 3. CEO leadership, on average, has plans for better management and more profits. The investment in technology is an efficient way in promoting customer services, which in turn results in efficient management and more profits. In the following, the coefficients in the profitability regression are converted to coefficients to compare the effects on profitability9: Profitability = -0.0915 + 0.0558*Asset Quality + 0.1792*Management + 0.002295*Branch office + 0.0455*Growth - 0.0473*Customer Services. The coefficients of regressions for profitability suggest that, for an increase in one standard deviation of the explanatory variables, efficient management brings in the greatest profits, followed by better asset quality and the growth in deposits and loans; an increase in the number of branch offices brings in the least profits. Therefore, to increase bank profits, CEOs should aim to improve bank management and capital structure and solvency rather than to add more branch offices. 6. Conclusions Conventional performance measures are mainly based on current financial data, which are comparable and well accepted. However, such traditional financial information paradigms do not fully reflect performance in the new economy. Non-financial factors have become increasingly more significant. Internal and external needs would be served by appropriate performance measures that capture value creation activities linked to long-term strategies. This paper evaluates the performance index of 35 publicly-traded commercial banks in Taiwan for the years of 2000 and 2001. The performance indexes of financial and non-financial aspects as

Each regression coefficient is adjusted by multiplying the original coefficient by the ratio of SXi/SY, in which SXi is the standard error of Xi and SY is the standard error of Y. The new coefficient represents the change in the units of standard errors of profitability due to the increase in one unit of standard errors of the independent variable of interest. 20

well as total performance index are constructed. Thirty-five banks are divided into new and old banks based on the year a bank founded or into public and private banks according to the type of major sponsors of a bank when founded. The results show that privatized government-owned/old banks are larger than private/new banks, respectively. Moreover, privatized government-owned banks have significantly higher financial performance index than private banks but both types of banks are not significantly different from each other in non-financial performance index. New and old banks are not significantly different from each other in both financial and non-financial performance indexes. With relatively large scale, profitability and better management, banks will perform relatively better among competitors in the following year. In addition, non-financial factors, as a whole, can be used to predict future total and financial performance indexes. More branch offices, better capital structure and solvency, and higher rates of growth in deposits and loans all result in more profits, and lead to higher customer satisfaction and more efficient management. Providing better technology to customers is an efficient way in promoting customer services, which in turn produces more profits and results in efficient management. CEOs, on average, have plans for better management and more profits. Finally, increases in the size of a bank in terms of total assets cause inefficient management, and reduce profits or impair customer services. Among the factors that have direct and positive impacts on profitability, increasing the efficiency of management is the most efficient way; on the contrary, adding more branch offices contributes the least profits. Therefore, to increase bank profits, CEOs should aim to improve bank management and capital structure and solvency rather than to add more branch offices. Our results indicate that non-financial performance measure is highly related to current and future financial performance indexes. Our findings imply that non-financial performance measure in our study is highly value-relevant for banks, and our performance measure can serve as the focal point for the banks efforts, and be interpreted as leading indicators of future performance. Consistent with the literature, predictive ability is one of the primary benefits of non-financial measures.

21

References Aly, H.Y., Grabowski, R., Pasurka, C., Rangan, N., 1990. Technical , scale, and allocative efficiencies in banking: an empirical investigation. Review of Economics and Statistics 23, 211-218. Annell, E. et al. (Konrad Group), 1989. The Invisible Balance Sheet: Key Indicators for Accounting, Control and Valuation of Know-How Companies, Karl-Erik Sveiby, Sweden. Athanassopoulos, A. D., 1998. Nonparametric frontier models for assessing the market and cost efficiency of large-scale bank branch networks. Journal of Money, Credit and Banking 30, 172-192. Atkinson, S.E., Cornwell, C., 1994. Parametric estimation of technical and allocative inefficiencies with panel data. International Economic Review 35, 231-243. Avkiran, N. K., 1999. The evidence on efficiency gains: the role of merger and the benefits to the public. Journal of Banking and Finance 23, 991-1013 Bauer, P.W., Berger, A. N., Ferrier, G..D., Humphrey, D.B., 1998. Consistency conditions for regulatory analysis of financial institutions: A comparison of Frontier efficiency method. Journal of Economics and Business 50, 85-114. Berger, A. N., DeYoung, R., 1997. Problem loans and cost efficiency in commercial banks. Journal of Banking and Finance 21, 849-870. Berger, A. N., Hancock, D., Humphrey, D.B., 1993. Bank efficiency derived from the profit function. Journal of Banking and Finance 17, 317-347. Brooking, A., 1998. Intellectual Capital, International Thomson Publishing, London. Camanho, A.S., Dyson, R.G.., 1999. Efficiency, size, benchmarks and targets for bank branches: An application of data envelopment analysis. Journal of Operational Research Society 50, 903-915. Chang, C.C., 1997. The nonparametric risk-adjusted efficiency measurement: An application of Taiwans credit departments of farmers associations. Presented at Taipei International Conference on Efficiency and Productivity Growth, June 20-21, Taipei, Taiwan, R.O.C. Chen, T.Y., 2001. An estimation of X-inefficiency in Taiwans banks. Applied Financial Economics 11(3), 237-242. Chen, T.Y. and Yeh, T.L., 1998. A study of efficiency evaluation in Taiwans banks. International Journal of Service Industry Management 9(5), 402-415. Chen, T. Y., Yeh, T.L., 2000. A measurement of bank efficiency, ownership and productivity changes in Taiwan. Service Industries Journal .20, 95-109. Chiu, Y.H., Hu , J.L., Yang, L., 2000. Ownership and production efficiency: evidence from Taiwans mixed banks. Presented at Taiwan Economic Conference 2000, December 17, Taipei, Taiwan, R.O.C. Cooper, D. R., Schindler, P. S., 1998. Business Research Methods, 6th edition, Irwin McGraw-Hill companies, Inc.
22

DeYoung, R., Iftekhar, H., 1998. The performance of de novo commercial banks: a profit efficiency approach. Journal of Banking and Finance 22, 565-587. Elyasiani, E., Mehdian, S., 1992. Productive efficiency performance of minority and nonminority-owned banks: a nonparametric approach. Journal of Banking and Finance 19, 933-948. Fukuyama, H., 1993. Technical and scale efficiency of commercial banks: a non-parametric approach. Applied Economics 25, 1101-1112. Grabowski, R., Rangan, N., Rezvanian, R., 1993. Organizational forms in banking : an empirical investigation of cost efficiency. Journal of Banking and Finance 17, 531-538. Grabowski, R, Rangan, N., Rezvanian,R., 1994. The effect of deregulation on the efficiency of U.S. banking firms. Journal of Economics and Business 46, 39-54. Handy, C., 1994. The Age of Paradox, Harvard Business School Press. Huang, T.H., Wang, M.H., 2001. Estimating scale and scope economies with Fourier flexible functional form--evidence from Taiwan's banking industry. Australian Economic Papers 40 (2), 213-31. Huang, T.H., 2000. Estimating X-efficiency in Taiwanese banking using a translog shadow profit function. Journal of Productivity Analysis 14, 225-245. Kaplan, R. S., Norton, D.P., 1992. The Balanced Scorecard Measures that Drive Performance. Harvard Business Review (Jan-Feb), 71-79. Kaplan, R. S., Norton,D.P., 1996. The Balanced Scorecard: Translating Strategy into Action, Harvard Business School Publishing, Boston, MA. Kaplan, R. S.Norton, D.P., 2001a. Transforming the balanced scorecard form performance measurement to strategic management: Part I. Accounting Horizons 15 (1), 87-104. Kaplan, R.S., Norton, D.P., 2001b. Transforming the balanced scorecard form performance measurement to strategic management: Part I. Accounting Horizons 15 (2), 147-160. Kraft, E., Tirgiroglu, D.,1998. Bank efficiency in croatia: a stochastic-frontier analysis. Journal of Comparative Economics 26, 282-300. Kumbhakar, S.C., 1996. A parametric approach to efficiency measurement using a flexible profit function. Southern Economic Journal, 473-487. Kwan ,S., Eisenbeis, R.A., 1996. An analysis of inefficiencies in banking: a stochastic cost frontier approach. FRBSF Economic Review, 16-26. Lev, B., 2001. Intangibles: Management, Measurement and Reproting, Brookings Institution Press, Washington, D. C. Liebowitz, J., 1999. Knowledge Management Handbook, CRC Press LLC. Miller, S.M., Noulas, A.G., 1996. The technical efficiency of large bank production. Journal of Banking & Finance 20, 495-509. Mitchell, K., Onvural, N.M., 1996. Economies of scale and scope at large commercial banks: Evidence from the Fourier flexible functional form. Journal of Money, Credit, and Banking 28(2), 178-199.
23

Oral, M., Yolalan, R., 1990. An empirical study on measuring operating efficiency and profitability of bank branches. European Journal of Operational Research 46, 282-294. Porter, M. E. 1992. Capital disadvantage: Americas failing capital investment system. Harvard Business Review (September/October), pp. 65-82. Resti, A., 1997. Evaluating the cost-efficiency of the Italian banking system: what can be learned from the joint application of parametric and non-parametric techniques. Journal of Banking and Finance 21, 221-250. Rodov, I., Leliaert, P., 2002. FiMIAM: financial method of intangible assets measurement. Journal of Intellectual Capital 3 (3), 323-336. Roos, J., G. R., Dragonetti, N.C., Edvinsson, L., 1997. Intellectual Capital: Navigating in the New Business Landscape, Macmillan, Houndmills. Sathye, M., 2001. X-efficiency in Australian banking: an empirical investigation. Journal of Banking and Finance 25, 613-630. Schaffnit, C., D. R., Paradi, J.C., 1997. Bestp practice analysis of bank branches: an application of DEA in a large Canadian bank. European Journal of Operational Research 98, 269-289. Sherman, H.D., Gold, F., 1985. Bank branch operating efficiency. Journal of Banking and Finance 9, 297-316. Stewart, G.. B., 1991a. The Quest for Value: A Guide for Senior Managers, Harper Business, New York. Stewart, G..B., 1991b. The Quest for Value: the EVATM Management Guide, HarperCollins, Philadelphia, PA. Sveiby, K. E., 1997. The Intangible Assets Monitor, Karl-Erik Sveiby, Sweden.

24

Table 1: Factors Considered in Financial and Non-Financial Performance Indexes


Item Financial Performance 1. Capital Structure and Solvency (1) Liability Ratio (2) Risk-Based Capital Ratio (3) Current Ratio 2. Management (1) NPL Ratio (2) Asset Turnover (3) Operating Revenue per employee (in Billions, NT$) 3. Profitability* (1) Return on Assets (2) Return on Stockholders Equity (3) Net Profit Margin 4. Scale (1) Assets (in Trillions, NT$) 5. Growth (1) Deposits Growth Rate (2) Loans Growth Rate II. Non-financial Performance 1. Customer Services (1) Quality of Employees (2) Services (3) Fees and Rates (4) Information and Convenience (5) Security and Reliability (6) Lobby and Facilities 2. CEO Leadership (1) CEO vision (2) CEO strategy 3. Technology (1) ATM (2) Tele-Bank (3) E-Bank 3.78 3.61 3.50 0.29 0.36 0.36 1/3 1/3 1/3 2.92 2.90 0.61 0.57 1/2 1/2 1/3 + + + 3.78 3.79 3.36 3.67 3.76 3.76 0.20 0.21 0.20 0.17 0.17 0.17 1/6 1/6 1/6 1/6 1/6 1/6 1/3 + + 0.05 0.01 0.18 0.09 0.40 0.35 1 0.05 0.5 0.5 1/3 1/3 + + + + + + Customer Surveys 1 question 3 questions 2 questions 6 questions 3 questions 5 questions CEO surveys Peer rating Peer rating Customer Surveys 2 questions 2 questions 2 questions + + (Deposits t Deposits t-1) / Deposits t-1 (Loans t Loans t-1) / Loans t-1 0.07 0.06 0.01 0.05 0.02 0.008 0.92 0.09 5.10 0.03 0.02 4.06 Mean 2001 Std dev. Weight Impact 2001 2/3 0.15 0.45 0.5 0.05 0.35 0.5 0.3 0.2 0.20 0.002 0.01 0.05 0.01 0.10 0.19 0.4 0.4 0.2 0.25 + Total Assets + + + Net Income/ Average Assets Net Income / Average Net Stockholders Equity Net Income /Operating Revenue + + Non-Performing Loans/Total Loans Operating Revenue / Average Assets Operating Revenue /Employees Explanation

+ +

Liability/Total Assets Capital/Risk-Based Assets Current Asset/Current Liability

*. The means and standard deviations of Return on Assets and Net Profit Margin exclude outliers such as Taitung Business Bank and Dah An Commercial Bank.

25

Table 2 Questions concerning customer services and CEOs leadership Non-Financial Performance Questions
1. Customer Services (1) Quality of Employees (2) Services (3) Fees and Rates Professional and Familiar with his/her works Attitude toward customers Patience with customers questions and/or complaints Awareness of customers needs Satisfactory deposit rates Reasonable service charges Multi-financial products available Convenient location Easy and simple applications Prompt response to any requests Relevant information updated accordingly Diversified services Security Privacy in customers personal data Banks reputation and reliability Welcoming facilities Enough ATM machines Low down rate of ATM Easy-to-follow screen directions in ATM Friendly designed lobby

(4) Information and Convenience


(5) Security and Reliability

(6) Lobby and Facilities

2. CEO Leadership (1) CEO vision (1) CEO strategy 3. Technology (1) ATM (2) Tele-Bank (3) E-Bank

Peer Ratings in both vision and strategy


Waiting time Satisfaction

26

Table 3 The Classifications of Sampled Commercial Banks in Taiwan


Bank Chang Hwa Commercial Bank First Commercial Bank Hua Nan Commercial Bank International Commercial Bank of China Hsinchu International Bank International Bank of Taipei Tainan Business Bank Taitung Business Bank Taichung Commercial Bank Central Trust of China The Farmers Bank of China Chiao Tung Bank United World Chinese Commercial Bank Grand Commercial Bank Dah An Commercial Bank Taipei Bank The Chinese Bank Taiwan Business Bank Cathay United Bank Bank of Kaohsiung Cosmos Bank Union Bank of Taiwan Bank Sino Pac E.Sun Commercial Bank Fubon Commercial Bank Asia Pacific Bank Tai Shin International Bank Far Eastern International Bank Ta Chong Commercial Bank Entie Commercial Bank Pan Asia Bank Baodao Commercial Bank Bank of Overseas Chinese Total 8 15 Privatized Government-owned New

27

Table 4 Rankings of Banks Based on Performance Indexes


Bank Type Financial Non-financial Total operation

Banks
Private New 2000 2001 2000

2001 16 10 7 19 5 17 20 31 32 4 33 13 6 22 3 15 8 25 29 12 18 24 2 1 9 28 14 11 21 23 26 30 27

2000 2001 14 4 6 3 27 13 29 33 30 1 24 2 8 16 20 10 12 15 26 21 28 18 9 11 7 17 5 19 22 25 31 23 32 12 8 3 7 16 15 25 33 31 2 30 1 6 27 28 9 17 23 22 13 19 21 5 4 10 24 11 14 18 20 29 26 32

Chang Hwa Commercial Bank First Commercial Bank Hua Nan Commercial Bank International Commercial Bank of China Hsinchu International Bank International Bank of Taipei Tainan Business Bank Taitung Business Bank Taichung Commercial Bank Central Trust of China The Farmers Bank of China Chiao Tung Bank United World Chinese Commercial Bank Grand Commercial Bank Dah An Commercial Bank Taipei Bank The Chinese Bank Taiwan Business Bank Cathay United Bank Bank of Kaohsiung Cosmos Bank UNION BANK OF TAIWAN Bank Sino Pac E.Sun Commercial Bank Fubon Commercial Bank Asia Pacific Bank Tai Shin International Bank Far Eastern International Bank Ta Chong Commercial Bank Entie Commercial Bank Pan Asia Bank Baodao Commercial Bank Bank of Overseas Chinese

10 3 5 4 27 15 31 33 30 1 21 2 6 18 20 8 16 13 28 19 29 17 11 12 9 22 7 14 24 23 26 25 32

8 6 3 4 25 13 28 33 29 2 24 1 7 27 32 5 26 21 15 14 23 20 12 11 10 17 9 16 19 22 30 18 31

27 7 10 5 26 14 21 33 28 1 30 2 12 15 16 13 9 18 22 25 24 17 6 8 4 11 3 23 19 31 32 20 29

28

Table 5 Descriptive statistics and tests for financial performance indexes Panel A: The means and standard deviations of financial performance indexes for privatized government-owned and private Banks Privatized Government-owned Private Item 2000 2001 2000 2001
1. Capital Structure and Solvency (1) Liability Ratio (2) Risk-Based Capital Ratio (3) Current Ratio 2. Management (1) NPL Ratio (2) Asset Turnover (3) Operating Revenue per employee (in millions) 3. Profitability

0.12(0.80) -0.36(0.73) 0.62(0.98) -0.64(0.46) 0.28(0.75) 0.37(0.51) -0.46(0.68) 1.18(1.86)

-0.01(0.62) -0.43(0.56) 0.43(0.84) -0.58(0.30) 0.18(1.02) -0.04(0.68) 0.18(1.54) 0.71(1.83)

0.02(1.00) 0.11(1.08) -0.08(1.03) 0.23(1.01) -0.03(0.83) -0.05(1.11) 0.14(1.07) -0.26(0.65) -0.11(1.07) -0.12(1.11) -0.11(1.12) -0.12(1.11) -0.36(0.58) -0.01(1.04) 0.02(1.11) -0.03(1.10) -0.12(0.68)

0.00(0.78) 0.14(1.08) -0.14(1.02) 0.18(1.08) -0.06(0.67) 0.01(1.09) -0.06(0.79) -0.23(0.36) -0.06(1.06) -0.07(1.11) -0.07(1.10) -0.02(1.14) -0.37(0.61) -0.01(0.61) 0.08(0.40) -0.10(0.97) -0.12(0.56)

0.39(0.31) 0.18(0.51) (1) Return on Assets 0.41(0.37) 0.22(0.51) (2) Return on Stockholders Equity 0.36(0.23) 0.21(0.58) (3) Net Profit Margin 0.42(0.33) 0.06(0.38) 4. Scale 0.48(0.77) 1.16(1.14) 5. Growth -0.22(0.39) 0.03(1.28) (1) Deposits Growth Rate -0.28(0.49) -0.25(1.98) (2) Loans Growth Rate -0.16(0.41) 0.31(1.11) 0.30(0.38) 0.39(0.45) Financial Performance Notes: the values in the parentheses are standard deviations.

29

Table 5 (Contd) Panel B: The means and standard deviations of financial performance indexes for new and old Banks New Old Item 2000 2001 2000 2001
1. Capital Structure and Solvency (1) Liability Ratio (2) Risk-Based Capital Ratio (3) Current Ratio 2. Management (1) NPL Ratio (2) Asset Turnover (3) Operating Revenue per employee(in millions) 3. Profitability

0.02(0.81) 0.20(0.87) -0.14(0.84) 0.08(0.76) 0.13(0.63) 0.22(0.98) 0.26(0.75) -0.26(0.32)

0.10(0.78) 0.30(1.30) -0.03(0.61) -0.26(0.49) 0.11(0.33) 0.41(0.41) -0.11(0.87) -0.30(0.28)

-0.02(1.04) -0.16(1.10) 0.11(1.13) -0.07(1.18) -0.11(0.90) -0.18(1.01) -0.22(1.14) 0.22(1.30) 0.00(0.89) 0.00(0.96) 0.02(0.93) -0.01(0.96) 0.44(1.19) -0.05(0.79) -0.16(0.85) 0.07(0.85) 0.07(0.71)

-0.09(0.71) -0.25(0.59) 0.03(1.26) 0.22(1.26) -0.09(0.98) -0.34(1.22) 0.09(1.11) 0.25(1.29) 0.07(0.83) 0.09(0.96) 0.12(0.74) -0.06(0.81) 0.46(1.16) 0.02(0.90) 0.04(1.33) 0.00(0.99) 0.08(0.70)

0.00(1.09) -0.09(1.11) (1) Return on Assets 0.00(1.08) -0.10(1.06) (2) Return on Stockholders Equity -0.02(1.11) -0.15(1.25) (3) Net Profit Margin 0.01(1.08) 0.07(1.22) 4. Scale -0.53(0.11) -0.55(0.22) 5. Growth 0.05(1.11) -0.02(0.68) (1) Deposits Growth Rate 0.19(1.15) -0.05(0.37) (2) Loans Growth Rate -0.08(1.18) 0.00(1.05) -0.08(0.60) -0.10(0.37) Financial Performance Notes: the values in the parentheses are standard deviations.

30

Table 5 (Contd) Panel C: Test results for financial performance indexes Privatized vs. Private Item 2000 2001
1. Capital Structure and Solvency (1) Liability Ratio (2) Risk-Based Capital Ratio (3) Current Ratio 2. Management (1) NPL Ratio (2) Asset Turnover (3) Operating Revenue per Employee (in millions) 3. Profitability (1) Return on Assets (2) Return on Stockholders Equity (3) Net Profit Margin 4. Scale 5. Growth

New vs. Old 2000 Old 2001 New Old Old

Private Private Privatized Privatized Privatized Privatized Privatized Privatized

Private Private Privatized

(1) Deposits Growth Rate (2) Loans Growth Rate Privatized Privatized Financial Performance (1) denotes no significant difference between the two types, otherwise, the significantly better one was reported. (2) The significance level is 5%.

31

Table 6 Descriptive statistics and tests for financial performance indexes Panel A: The means and standard deviations of non-financial performance indexes for privatized government-owned and private Banks Privatized Private Government-owned Non-financial Aspect 2000 2001 2000 2001 1. Customer Services (1) Employees Quality (2) Service Quality (3) Fees and rates -0.22(0.89) -0.22(0.77) 0.09(0.76) 0.09(0.89) -0.39(1.13) -0.17(0.90) 0.03(0.91) 0.06(1.06) -0.53(0.94) -0.33(1.06) -0.36(0.96) -0.40(0.71) -0.11(0.88) -0.08(0.91) -0.22(1.08) 0.17(0.51) 0.22(0.50) 0.12(0.53) -0.13(1.04) -0.36(1.32) 0.02(1.22) -0.04(1.04) 0.12(1.00) 0.30(0.81) 0.07(0.89) -0.11(0.94) 0.09(0.97) -0.22(1.05) -0.23(1.03) -0.21(1.07) -0.08(0.70) -0.01(1.01) -0.01(0.95) -0.21(0.74) -0.07(0.61) 0.13(1.02) 0.19(0.99) 0.12(0.93) -0.06(1.07) 0.09(0.91) -0.25(1.03) -0.27(1.03) -0.22(1.04) 0.09(0.70) 0.14(0.80) 0.04(0.82) 0.09(0.92) -0.04(0.56)

(4) Information and Convenience 0.05(1.03) -0.01(0.91) (5) Security and Reliability (6) Lobby and Facilities 2. CEO Leadership (1) CEO vision (2) CEO strategy 3. Technology (1) ATM (2) Tele-Bank (3) E-Bank -0.14(1.12) 0.07(0.40) 0.08(0.43) 0.07(0.40) -0.07(0.82) -0.03(0.94) -0.22(0.84) 0.03(1.05)

-0.08(0.59) -0.08(0.50) Non-financial Performance Notes: the values in the parentheses are standard deviations.

32

Table 6 (Contd) Panel B: The means and standard deviations of non-financial performance indexes for new and old Banks Non-financial Aspect 2000 1. Customer Services (1) Employees Quality (2) Service Quality (3) Fees and rates (5) Security and Reliability (6) Lobby and Facilities 2. CEO Leadership (1) CEO vision (2) CEO strategy 3. Technology (1) ATM (2) Tele-Bank (3) E-Bank Non-financial Performance -0.10(0.75) -0.27(0.95) -0.03(0.98) 0.21(0.80) -0.26(0.94) -0.11(0.89) -0.01(0.96) -0.02(0.98) 0.01(0.95) 0.04(0.58) 0.12(0.84) 0.18(0.87) -0.19(0.80) -0.02(0.51) New 2001 0.31(0.88) 0.30(0.88) 0.36(1.01) 0.32(0.93) 0.37(1.03) 0.15(1.06) 0.35(0.85) -0.10(0.94) -0.12(0.95) -0.08(0.94) 0.14(0.71) 0.38(0.78) 0.13(0.92) -0.08(0.84) 0.11(0.49) 2000 0.03(1.00) -0.08(1.06) 0.07(1.00) 0.20(0.94) 0.03(0.91) 0.10(1.08) Old 2001 -0.20(1.06) -0.24(0.96) -0.16(0.93) -0.14(0.79) -0.21(0.99) -0.21(0.98)

0.06(0.83) -0.19(0.81)

(4) Information and Convenience -0.15(0.88)

-0.22(0.90) -0.15(0.94) -0.22(0.87) -0.14(0.93) -0.22(0.93) -0.16(0.95) -0.15(0.81) -0.04(0.86) -0.11(1.06) -0.23(1.04) -0.23(0.92) -0.02(0.96) -0.12(0.87) 0.14(1.01) -0.11(0.66) -0.16(0.55)

Notes: the values in the parentheses are standard deviations.

33

Table 6 (Contd) Panel C: Test results for non-financial performance indexes Non-financial Aspect 1. Customer Services (1) Employees Quality (2) Service Quality (3) Fees and rates (5) Security and Reliability (6) Lobby and Facilities 2. CEO Leadership (1) CEO vision (2) CEO strategy 3. Technology (1) ATM (2) Tele-Bank (3) E-Bank Privatized vs private 2000 Private Private 2001 Private Privatized Privatized New vs 2000 Old

2001 New New New New

(4) Information and Convenience

Non-financial Performance (1) denotes no significant difference between the two types, otherwise, the significantly better one was reported. (2) The significance level is 5%.

34

Table 7 Spearmans Rank Correlation Coefficients Rank correlation 2000 2001 2000 2001 2000 2001 Total performance 2000 2001 1 0.89 0.96 0.89 0.90 0.64 1 0.89 0.88 0.76 0.85 Financial performance 2000 2001 Non-financial performance 2000 2001

Total performance Financial performance Non-financial performance

1 0.91 0.76 0.62 1 0.70 0.55 1 0.63 1

35

Table 8 Prediction Results of Total Performance on lagged Financial and Non-financial Factors Independent Variables Intercept Financial performance Non-Financial performance Capital Structure and Solvency Management Profitability Growth Scale CEO Leadership Customer Services Technology Branch per hundred Hundred Employees per branch Adjusted R2 + + + + + + + + + + + +
0.693 0.708 0.057 0.405*** 0.298** 0.076 0.154** -0.097 0.096* -0.025 0.11275 0.07316 -0.01546 0.454*** -0.585 0.641 0.690 0.048 0.076 0.144* 0.07659 0.536*** 0.272** 0.06970 -0.024 0.306** 0.268** 0.052

Predicted sign

Model A1
-0.139

Model A2
-0.415

Model A3
0.147

Model A4
-0.072 0.607*** 0.236**

24 24 Degree of freedom 24 30 (1) Dependent variable is 2001 total performance index, and independent variables are components of 2000 performance indexes. (2) *, ** and *** indicate significance at the 10%, 5% and 1% (one-tail) respectively.

36

Table 9 Prediction Results of Financial Performance on lagged Financial and Non-financial Factors Independent Variables Intercept Financial performance Non-Financial performance Capital Structure and Solvency Management Profitability Growth Scale CEO Leadership Customer Services Technology Branch per hundred Hundred Employees per branch Adjusted R2 + + + + + + + + + + + +
0.7066 0.7360 0.046 0.608*** 0.260** -0.103 0.264*** -0.053 0.117* -0.075 -0.080 0.079 -0.058 0.778*** -0.645 0.5779 0.6678 0.177 0.097 0.186* 0.079 0.831*** 0.216* -0.114 -0.089 0.390** 0.228 -0.130

Predicted sign

Model B1
-0.129*

Model B2
-0.601***

Model B3
0.213

Model B4
-0.097* 0.841*** 0.209*

Degree of freedom 24 24 24 30 (1) Dependent variable is 2001 financial performance index, and independent variables are components of 2000 performance indexes. (2) *, ** and *** indicate significance at the 10%, 5% and 1% (one-tail) respectively.

37

Table 10 The Correlation Coefficients of Variables Appeared in Regression Analysis


2000-2001 EPLEBRBRANCH EPLEE MANGMT CSS PROFIT SCALE GROWTH CEO SRVCE TECH EPLEBR BRANCH EPLEE MANGMT CSS PROFIT SCALE GROWTH CEO SRVCE TECH 1 -0.10 0.40 0.58 0.22 0.30 0.28 0.23 0.59 0.00 0.15 1 0.86 -0.18 -0.16 0.17 0.86 -0.07 0.09 -0.03 0.13 1 0.09 -0.05 0.26 0.94 0.02 0.36 -0.06 0.18 1 0.46 0.60 0.14 0.54 0.67 0.24 0.24 1 0.53 0.00 0.28 0.43 1 0.29 0.53 0.52 1 0.05 0.40 -0.05 0.18 1 0.40 0.12 0.16 1 0.01 1 1

-0.01 -0.05 0.06 0.07

0.19 0.59

Variable Definitions: EPLEBR: the number of employees per branch office; BRANCH: the number of branch offices; EPLEE: the number of employees; MANGMT: the performance score of Management; CSS: the performance score of Capital Structure and Solvency; PROFIT: the performance score of Profitability; GROWTH: the performance score of Growth; CEO: the performance score of CEO Leadership; SRVCE: the performance score of Customer Services; TECH: the performance score of Technology.

38

Table 11 3SLS Regression Results of Profitability, Management and Customer Services Variables Intercept Capital Structure and Solvency Management Profitability Growth Scale CEO Leadership Customer Services Information & Technology Branch Employees per branch System adjusted R
2

Predicted sign Profitability Management -40.089 + + + + + + + + + + 0.5635 181 0.688*** -0.190 0.245** 0.319*** 0.610*** 0.324** 0.100 -13.621 0.387*** 0.243** 3.622

Services 11.999

0.428**

-0.282 0.562** -0.375

Degree of freedom

*, ** and *** indicate significance at the 10%, 5% and 1% (one-tail) respectively.

39

Growth

Profitability

Capital Structure & Solvency

Branch Offices

Managemnent

CEO Leadership

Technology

Customer Services

Figure1 Impacts of Financial and Non-Financial Factors on Management, Profitability and Customer Services

40