MORE INFORMATION TECHNOLOGY INVESTMENTS,
MORE PERFORMANCE IN BANKING?
Department of Finance, National Formosa University,
64, Wun-Hua Road, Huwei, Yunlin 632, Taiwan ROC
This study investigates the relationship between information technology (IT) and operational performance based on a sample of Taiwan‘s banks. Using
data envelopment analysis (DEA) and stochastic frontier approach (SFA), this study finds some interesting points. First, low operating efficiencies exist in the banking Industry during the study period. These inefficiencies are in nature as-cribable to a combination of both wasteful overuse of IT resources and inappro-priate scale of IT investments. Second, operating efficiencies measured by DEA and SFA present a significant strong relationship. Third, for an inefficient indi-vidual bank, the performance can be enhanced if the amount of IT investment is enlarged. Fourth, the difference of the ownership type has a significant effect upon bank performance with respect to the contribution of IT investment to operating performance. Fifth, as a whole, both number of ATMs and diversifi-cation of IT services contribute positively to bank performance. Sixth, the op-erational efficiency of the mutually-owned banks is positively affected by number of ATMs and diversification of IT services while negatively affected by number of IT staff. Finally, number of ATMs has a positive impact on the per-formance of the privately-owned banks.
Banking, Data Envelopment Analysis (DEA), Information Technology (IT), Operating Efficiency, Performance, Stochastic Frontier Approach (SFA)
C.-F. Li / More Information Technology Investments, More Performance In Banking
Over the past few decades, modern business organizations have been in-creasingly investing substantial amounts of money in information technology (IT) with the objective of improving their operational efficiency and competitive ability in the industry. The important role IT plays in contemporary business is unquestionable. IT is regarded as a critical factor for business enterprises to sur-vive and to grow up further; however, empirical evidence to support these antic-ipated benefits has been mixed. Some researchers asserted that the IT invest-ments can really promote the enterprises' operational performance by means of reducing costs, raising profit margin, upgrading production levels, increasing service quality, advancing customer satisfaction, and improving overall opera-tions. In contrast, other researchers did not demonstrate the positive effect of IT investment and concluded that IT spending brought no significant contributions to the enterprises‘ operations, and so the ―IT productivity paradox‖ has been an
issue debated continuously for decades.
The differences among research objects, methodologies, and performance indices result in inconsistent conclusions obtained in the literature. In this re-spect, the lack of good quantitative measures for the output and value created by IT has made the studies on justifying IT investments particularly difficult. Bryn-jolfsson (1993) proposed the following four explanations for IT productivity pa-radox: (1) mismeasurement of inputs and outputs; (2) lags between cost and benefit; (3) redistribution and dissipation of profits; (4) mismeasurement of in-formation and technology. Obviously, a new research method should be able to eliminate or overcome above-mentioned defects.
To explore the effects of IT spending on organizational performance, this study provides a framework for performance evaluation using reliable quantita-tive tools, not only nonparametric data envelopment analysis (DEA) but also pa-rametric stochastic frontier approach (SFA). The data set is based on a sample of Taiwan‘s banks. The banking industry has been particularly information inten-sive. In history, banking has always been a crucial area for IT to be implemented. That is, an area where the advantage from using IT is so considerable that the state-of-the-art IT is developed almost as soon as it becomes available. A widely held belief is that IT is absolutely vital to a bank‘s survival and growth. In this
regard, it seems especially meaningful to link this issue with banking institu-tions.
To sum up, the purpose of this study is to assess the impact of IT investment on organizational performance that accomplishes the following five main objec-tives. First, we attempt to study the relevant theories and thereby to develop more complete models for performance evaluation of banks. Second, we measure operational performance for each individual bank and analyze the main sources of operating inefficiency. Third, we investigate the impact of IT investment on bank operational performance and measure its magnitude. Fourth, we compare and interpret the effects of IT investment on performance of different types of banks. Fifth, we contrast alternative approaches to the measurement of IT value.
The remainder of this paper is organized as follows. Section 2 reviews the
C.-F. Li / More Information Technology Investments, More Performance In Banking previous empirical research at the firm level. Section 3 describes the analytical techniques applied——DEA and SFA, as well as the empirical data. Section 4
reports and discusses the empirical results from three aspects: (1) evaluation of operational efficiency; (2) comparison of efficiency differences between various types of banks; and (3) impact of IT investment on operational efficiencies. Sec-tion 5 sums up the main findings and presents the conclusions.
2. LITERATURE REVIEW
Since the relevant empirical literature is very rare in the field of banking, this section also describes existing studies relating to other service organizations, manufacturing firms and hospital institutions.
Cron and Sobol (1983) examined the relationship between computerization and several measures of overall firm performance based on a sample of 138 medical wholesalers. Using correlation analysis, the results showed that compu-terization was related to overall performance. Non-users tended to be small firms with about average overall performance. On the other hand, firms owning com-puters and making extensive use of them in a variety of ways tended to be either very high or low performers.
Bender (1986) surveyed 132 life insurance companies in 1983 to investigate the financial impact of IT on firms in this industry. Organizational performance was measured in terms of the ratio of total operating expense to total premium income. The IT impact was represented by the ratio of information-processing expense to total general expense (IPE/EXP ratio). The results revealed that an appropriate level of investment in IT could have a positive impact on total ex-pense. A range in the IPE/EXP ratio of 15% to 25% seemed to produce optimum results in the life insurance industry. Contrarily, a company that had an IPE/EXP ratio of less than 15% was mostly likely not sufficiently automated to combat the escalation costs of doing business.
Alpar and Kim (1990) utilized 424-759 U. S. banks during 1979-1986 to analyze the impact of IT on economic performance. Applying cost function ap-proach they found that IT was able to reduce operating costs, increase capital expenditures of banks, save personnel costs, reduce demand deposits, and in-crease time deposits.
Strassman (1990) investigated the relationship between IT and return on investment in a sample of 38 service sector firms using correlation analysis. He found that some top performers invested heavily in IT, while some did not. He concluded that there was no correlation between spending for computers, profits and productivity.
Weill (1992) studied 33 medium and small-scaled valve manufacturing companies to explore the relationship between the IT investments and organiza-tional performance using hierarchical regression. Although transactional IT in-vestment was found to be strongly related to superior organizational performance,
C.-F. Li / More Information Technology Investments, More Performance In Banking there was no evidence that strategic IT investment, on a long-term basis, would increase or decrease organizational performance. However, the results implied that strategic IT investment was beneficial to relatively poor performing firms in the short run.
Mahmood and Mann (1993) utilized canonical correlation analysis to ex-plore the organizational impact of IT investment of 100 U. S. listed companies. The results indicated that economic performance measures such as sales by em-ployee, return on sales, sales by total assets, return on investment, and market to book value were affected by IT investment measures such as IT budget as per-centage of revenue, percentage of IT budget spent on training of employees, number of PCs per employee, and IT value as a percentage of revenue. The or-ganizational performance measure growth in revenue and IT investment measure percentage of IT budget spent on staff were not significantly related to other measures and therefore were not indicated to be useful for investigating possible effects of IT investment on organizational economic performance.
Loveman (1994) utilized OSL regression to assess the productivity impact of IT based on a sample of 60 manufacturing firms during 1978-1984. The re-sults showed that during the five-year period, the contribution of IT investment to the output of manufacturing firms was nearly zero. There existed no sufficient evidence to support the benefit of IT from productivity enhancement.
Berndt and Morrison (1995) explored relationships between industry per-formance measures and investments in high-tech office and IT capital for two-digit manufacturing industries during 1968-1986. They found limited evi-dence of a positive relationship between profitability and the share of high-tech capital in the total physical capital stock (OF/K). They also found that increases in OF/K were negatively correlated with multi-factor productivity and tended to be labor-using. Furthermore, they found some evidence that industries with a higher proportion of high-tech capital had higher measures of economic perfor-mance, although within industries increasing OF/K did not appear to improve economic performance.
Kivijarvi and Saarinen (1995) used a sample of 36 Finnish firms to probe the relationship between IT investments and of firm financial performance. Uti-lizing regression analysis, the results demonstrated that IT investments had no direct relationship with financial performance. However, IT investments were able to improve firm performance in the long term.
Brynjolfsson and Hitt (1996) used 367 large firms during 1987-1991 as a sample to study the benefits of information systems (IS) spending. The results indicated that IS spending had made a substantial and statistically significant contribution to firm output. It was also found that the gross marginal product (MP) for computer capital was at least as large as the MP of other types of capital investment and that IS labor spending generated at least as much output as spending on non-IS labor and expenses.
Hitt and Brynjolfsson (1996) applied OLS regression and iterated seemingly unrelated regression (ITSUR) to explore the business value of IT based on a
C.-F. Li / More Information Technology Investments, More Performance In Banking sample of 370 large firms. The findings indicated that IT had increased produc-tivity and created substantial value for consumers. However, they did not find evidence that these benefits had resulted in supranormal business profitability.
Mitra and Chaya (1996) used a sample of over 400 large and medium-sized U.S. corporations to analyze the performance impact of IT investment. They found that higher IT investments were associated with lower average production costs, lower average total costs, and higher average overhead costs. They also found that larger companies spent more on IT as a percentage of their revenues than smaller companies. However, they did not find any evidence that IT reduced labor costs in organizations.
Byrd and Marshall (1997) investigated the relationship between IT invest-ment and organizational performance using a sample of 350 public companies during 1989-1991. Applying correlation analysis, they found that the number of PCs and terminals as a percentage of employees was significantly and positively related to sales by employee. The value of supercomputers, mainframes, and mi-nicomputers as well as the percentage of IT budget spent on IT staff were sig-nificantly and negatively associated with the sales by employee. The IT budget as a percentage of revenue was significantly and negatively associated with sales by total assets. The percentage of IT budget spent on IT staff training was not related to any performance variable.
Rai et al. (1997) employed Cobb-Douglas cost function approach to probe the relationship between IT investment and business performance based on a sample of 497 firms during 1994. The results suggested that IT investments could make a positive contribution to firm output and labor productivity. How-ever, various measures of IT investment did not appear to have a positive rela-tionship with administrative productivity. Furthermore, IT was likely to improve organizational efficiency, its effect on administrative productivity and business performance might depend on such other factors as the quality of a firm‘s man-
agement processes and IT strategy links, which could vary significantly across organizations.
Devaraj and Kohli (2000) examined monthly data collected from 8 hospitals over a recent three-year time period to study the relationship between IT and performance. Applying correlation analysis, the results provided support for the relationship between IT and performance that is observed after certain time lags. Such a relationship may not be evident in cross-sectional data analyses. Also, results indicated support for the impact of technology contingent on business process reengineering practiced by hospitals.
Lee and Menon (2000) used DEA and Cobb-Douglas cost function approach to analyze the financial data on the hospitals during 1976-1994. They found that hospitals that were characterized by high technical efficiency also used a greater amount of IT capital than firms that exhibited low technical efficiency and that a group of hospitals exhibiting high technical efficiency also exhibited low alloca-tive efficiency, indicating that, while processes might have been efficient, re-source allocation and budgeting between various categories of capital and labor had not been efficient. Moreover, they found that IT labor had a negative con-
C.-F. Li / More Information Technology Investments, More Performance In Banking tribution to productivity and that non-IT capital had a greater contribution to productivity than IT capital.
Sircar, Turnbow and Bordoloi (2000) explored the relationship between firm performance and IT investments based on a sample of 624 firms. They used canonical correlation analyses as a research method and found that IT invest-ments had a strong positive relationship with sales, assets, and equity, but not with net income. Spending on IS staff and staff training was positively correlated with firm performance, even more so than computer capital.
Shao and Lin (2001) investigated the relationship between IT investments and technical efficiency of 370 large U.S. firms during 1988-1992. Using both Cobb-Douglas and Translog cost functions and hypothesis test, the results indi-cated that IT had a significantly positive effect on technical efficiency and, hence, contributed to the productivity growth in organizations.
Osei-Bryson and Ko (2004) employed the same data set used by Bryn-jolfsson and Hitt (1996) to explore the relationship between IT investments and firm performance using regression splines analysis. The results exhibited that depending on the conditions that applied, an unbiased observer could either con-clude that investments in IT had a positive statistically significant effect on productivity, or that there was a ‗productivity‘ paradox. This suggested that the relationship between IT investments and organizational performance is much more complex than that found in some other studies.
Ham, Kim and Joeng (2005) examined the effect of IT applications on per-formance based on a sample consisted of 13 five-star hotels and 8 four-star ho-tels in Korea. Using hypothesis test, the results supported the relationship be-tween IT usage and the performance of lodging operations. Furthermore, they found that front-office applications, restaurant and banquet management systems, and guest-related interface applications significantly and positively affected performance of lodging operations; however, guest related interface applications were not significant.
Andersen and Foss (2005) investigated the role and effects of information and communication technology in multinational enterprises. They suggested that the attendant cost–benefit tradeoff could be influenced by computer-mediated communication. Based on a sample of 88 organizations in the computer products industries, they found that multinationality in itself did not guarantee a higher level of strategic opportunity. Instead, use of information technology to facilitate communication among managers across functional and geographical boundaries enhanced coordination of multinational activities in the development of strategic opportunity, which in turn was associated with superior performance.
This paper attempts to investigate the impact of information technology on operational performance of banking firms by mean of both DEA and SFA. This
C.-F. Li / More Information Technology Investments, More Performance In Banking section introduces the methodology, where the research methods and the empir-ical data are described, respectively, as follows.
3.1 The Research Methods
In the past several years, the assessment of operational performance has re-ceived much attention in academia and in business circles due to increased competition in the market. While evaluating the performance of a decision mak-ing unit (DUM), it is indispensable to use reliable approaches. Up to the present, the three prevailing techniques developed for efficiency measurements in both the industrial and academic worlds are traditional financial ratio analysis, eco-nometric approach and linear programming approach. The financial ratio analy-sis shows the relations between two financial figures after being compared, such as return on assets, return on investments, etc. It is based on financial statements and has been widely used not only for financial and production management but also for marketing, purchasing and personnel management throughout all sectors of business and commerce. Without a doubt, these ratios do convey some finan-cial information about firm performance. It is particularly meaningful when they are compared with those ratios of prior periods or of other firms. The popularity of financial ratio analysis lies perhaps in the simplicity and ease of calculation. However, a critical limitation is that financial ratio analysis fails to consider the multiple input–output characteristics of business enterprises and cannot give an overall clear picture of organizational operations, because firm performance may exhibit considerable variation depending on the indicator chosen. In the recent banking literature, much attention mostly directed to the latter two techniques of frontier efficiency analysis??econometric approach and linear programming
approach that are able to provide comprehensive insights beyond those available from financial ratio analysis for evaluating and improving banking efficiency.
Since the seminal study by Farrell (1957), methodological development in frontier efficiency analysis has been continued at a rapid pace. To date, there are a multitude of techniques, parametric and nonparametric, stochastic and deter- ministic.The essential differences among these techniques primarily reflect dif-fering assumptions used in estimating the shape of frontier and the distributional assumptions imposed on the random error and inefficiency.
There are at least five different types of approaches in the literature that have been employed in measuring the banking efficiency. Of those, three eco-nometric approaches, such as stochastic frontier approach (SFA), distribu-tion-free approach (DFA) and thick frontier approach (TFA) are parametric, and two linear programming approaches are nonparametric, such as data envelop-ment analysis (DEA) and free disposal hull (FDH). Each of the approaches nec-essarily has weaknesses as well as strengths relative to the other. The literature has not yet come to a consensus about the preferred approach for determining the best-practice frontier against which relative efficiencies are measured.
In general, parametric approaches are stochastic, and so attempt to distin-guish the effects of inefficiency from the effects of noise. A key drawback to pa-rametric approaches is that they usually specify a particular functional form that presupposes the shape of the frontier. If the functional form is misspecified, measured efficiency may be confounded with the specification errors. In sharp
C.-F. Li / More Information Technology Investments, More Performance In Banking contrast to parametric approaches, nonparametric approaches are inherently bounding techniques, and so they impose less structure on the frontier. They are deterministic and do not allow for random error owing to luck, data problems or other measurement errors. If random errors do exist, measured efficiency may be confounded with these random deviations from the true efficiency frontier. So the former's limitations is exactly the latter's advantages and vice versa. Conse-quently, we employ both nonparametric DEA and parametric SFA as research methods at the same time. The DEA and SFA are among the most popular of frontier efficiency analysis. Thus we can not only analyze the impact of IT in-
vestment on banking operational performance but also contrast the results of both approaches. DEA and SFA are described as follows.
3.1.1 Data Envelopment Analysis
The data envelopment analysis (DEA) is basically a mathematical pro-gramming technique initially developed by Charnes, Cooper and Rhodes (1978) based on the basic concepts of relative efficiency and nonparametric frontier of Farrell (1957). DEA extends the notion of Farrell's productive efficiency from single-output case to multiple-output case. Unlike parametric frontier approaches, DEA does not require any assumptions about the functional form. The DEA frontier is formed as the piecewise linear combinations that connect a set of the best-practice DMUs, which is obtained from the observed sample, yielding a convex production possibilities set. Thus a maximal efficiency measure for each DMU relative to all other DMUs in the observed data set can be calculated only with the requirement that each DMU lies on or below the external frontier.
The most important characteristics of the DEA methodology can be pre-sented by the CCR model. For the discussions to follow, let us suppose that there are k DMUs to be evaluated. Each DMU (j=1, 2,…, k ) consumes varying j
amounts of n inputs to produce m outputs and each has at least one positive input and one positive output. The primal input-oriented CCR model is formulated as follows:
C.-F. Li / More Information Technology Investments, More Performance In Banking
osubscript () = the DMU being evaluated from the observed data;
E = the efficiency rating of the DMU being evaluated; 0
y(0Y = the observed amount of output r for DMU, rj jrj
x(0X = the observed amount of input i for DMU, ij jij
u = the weight for output r; r
w = the weight for input i; and i
ε= a non-Archimedean infinitesimal constant.
3.1.2 Stochastic Frontier Approach
The stochastic frontier approach (SFA) specifies a functional form for the cost, profit, or production relationship among inputs, outputs and environmental factors, and allows for random error. It was first proposed by Aigner, Lovell and Schmidt (1977), Battese and Corra (1977) and Meeusen and van den Broeck (1977) simultaneously on three different continents. Its residuals contain two error terms, one for inefficiency that is assumed to be either nonpositive or non-negative relying on its distributional assumption and another for noise or random error that is unrestricted to be positive or negative. The former represents factors that can be controlled by DMUs, while the latter represents those effects which cannot be controlled by the DMUs, including quality or measurement errors. The inefficiency term is supposed to follow a one-sided distribution, usually the asymmetric half-normal; whilst noise component is supposed to follow a sym-metric distribution, usually the standard normal.
Let us consider that a specific DMU (j=1, 2,…, k ) uses n inputs j
nxxxxR？；(,,,)yR； to produce scalar output , the production function 12;n;
can be expressed as follows.
Y = the observed amount of output for DMU; j j
X = the observed amount of input i (i =1, 2, …, n) for DMU; ij j
u(0u = the measure for technical inefficiency, ; ii
1 If it is assumed that u？0, the problem simplifies to one of OLS estimation of the parameters of a i
production function with no inefficiency; while if it is assumed that v？0 the problem simplifies i
to one of estimating the parameters of a deterministic production frontier with no noise. In the for-mer case there is no efficiency measurement problem to worry about; while in the latter case there is no decomposition problem to worry about.
C.-F. Li / More Information Technology Investments, More Performance In Banking
2vN~(0,)！v= a random error term indicating the usual statistical noise, . iii
3.2 The Data
The research object of this study is all banks operating in Taiwan during 1996- 2000. The definition of a bank here adopts in a broader sense; it is referred to as a financial institution able to create deposit currency, including general banks, and community financial institutions, such as credit cooperatives, credit departments of farmers' and fishermen's association. The empirical data comes
from the following two sources. The first is primary data that mainly provides information about IT investments and spending, which is taken from the results based on a questionnaire survey to current directors in charge of information unit or center of each bank. The second is secondary data that chiefly provides finan-cial information, which is taken from the financial statements of each bank an-nounced annually. For general banks, the financial statements are collected from the publication ―Financial Statistics‖ by the Bureau of Monetary Affairs of Min-
istry of Finance, R.O.C. For community financial institutions, the financial statements are collected from their annual reports. To improve survey results, a pretest is carried out before the questionnaire survey, and the questionnaire is revised accordingly.
Altogether, 74 copies of questionnaires are distributed to the participants in various ways such as mail, email, Internet, facsimile transmission, etc. Of the 74 copies, 46 are sent to general banks, 22 to credit cooperatives and 6 to credit departments of farmers' and fishermen's association. The participants of the self-administered survey are the directors in charge of the information unit or 2center. Finally, 41 responses are received after follow-ups. Among them, 11 are invalid and 30 are effective. The effective response rate is up to 41%. Thus, the total sample for this study consists of all 30 individual banks, which can be ca-tegorized into three groups based on their business ownership form——5 pub-
licly-owned banks (17%), 12 privately-owned banks (40%) and 13 mutual-ly-owned banks (43%).
According to the relevant theories and literature, six variables are used for this study, namely, number of IT staff, number of ATMs, number of PCs and terminals, number of financial cards issued, diversification of IT services and pre-tax income. Among those, the output is pre-tax income and the inputs are the other five variables. The value of each variable is defined as an annual average value during the study period. Since a strong correlation between input and out-put variables should be suggested in principle, this study conducts a Pearson correlation analysis and the result demonstrates a highly significant positive correlation between both input and output variables at the 1% significance level. The Pearson correlation coefficients are between 0.749 and 0.862, implying that the input and output variables chosen are quite suitable. In addition, this study
2 We contacted nearly 90% of the directors by telephone in advance. If they were not willing to fill out the questionnaire or unable to offer us the relevant information in the questionnaire, we deleted their banks from our sampling population..