Information and communications technologies (ICTs) have changed the way in which firms do business and create value. However, empirical data for the linkage between ICT impacts and business process re-engineering (BPR) is rare and cross-country comparison is also missing. The objective of our study is to improve our understanding of the contribution of ICTs to firm performance and explore the linkage between ICT impacts and BPR by applying multiple-group structural equations modeling to test the Cross-Nation-ICT adoption Model. To test the model, a survey data form the United State, Taiwan, Korea, and Chile was conducted.
This paper contributes to IS research by offering some insights and exploring BPR generalization by cross national comparison. First, it provides empirical evidence to examine impacts of ICTs adoption on business process reengineering. Second, we emphasize two major ICT infrastructures: resource planning infrastructure and e-commerce infrastructure that firms adopt for advancing business processes in order to compete real business environments. Finally, we explore the national difference on ICTs adoption and its impacts on financial performance. More precisely, the model encapsulates the effects of both ICTs impacts for business process re-engineering and ICTs support for firm assets on firm performance.
Keywords: Information and communications technologies, Resource-based view, Cross-cultural, Structural equation modeling
Information and communications technologies (ICTs) have changed the way in which firms do business and create value. Many researchers focus on ICT systems and interaction with firms, such as information capability and business process re-engineering (BPR), and claim that ICTs increase firm's competitiveness. For example, Sarker and Jagjit (2006) and Ziaul, Faizul, and Ken (2006) report that ICTs may change business practices to re-optimize business processes, which leads to increased efficiency and improved performance. Neirotti, Cantamessa, and Paolucci (2008) find that ICTs help firms to improve their knowledge about customers and performance in the manufacturing industries in Italy. Rhim, Park, and Kim (2007) show that adoption of ICTs may improve general financial performance, reduce functional area costs through business process outsourcing, customer relationship management (CRM), supply chain management (SCM), etc.
Despite the importance of ICT, most research focus on one or two enterprise applications, such as electronic commerce or enterprise resource planning (Lee, Chu, & Tseng, 2009). We argue that examination of one kind of technologies is not a good approach to studying business reengineering. It ignores the mixed effects of various ICTs on BPR and misses to offer a complete picture of the relationship between ICT adoption and BPR. contributions of ICTs at country level are rarely examined and compared. In addition, ICT activities in firms are complicated, and may be different significantly across countries. The impact of ICT uses may depend on national culture and in the specific environment in which they are embedded (Limaye & Victor, 1999). Generalizations about the outcomes of related to ICT adoption cannot be made until research accumulates evidence of similar effects across different countries (Spanos, Prastacos, & Poulymenakou, 2002).
Based on the rich body of research on BPR and the adoption and diffusion of ICTs, we first propose an initial ICT-BPR model (Lee, Chu, & Tseng, 2009). The survey results indicate that organizational innovation, market pressure and competitive intensity positively affect information technology adoption, which in turn trigger changes or business process in terms of workplace, workforce and business structure (Lee, Chu, & Tseng, 2009). We extend our initial ICT-enabled BPR model to evaluate ICT-enabled BPR and business performance by investigating 377 senior information systems managers from top 2,000 Taiwan corporations, the results show that ICTs technologies facilitate firms to optimize business processes, and business process re-engineering has significant positive impacts on internal process performance and corporate financial performance (Lee, Chu, & Tseng, 2010). In this study, we further revise our ICT-BPR-Performance model and apply multiple-group structural equation modeling to test the revised model on data derived from cross-national research. The data are collected from the Business and Information Technologies (BIT) project, which lead by researchers at UCLA Anderson School of Management. The main purpose of the BIT project is to study the impacts of ICTs on business practices in various countries over an extended time horizon. The findings would be valuable in understanding the complexity of ICT adoption and predicting outcome of business process reengineering.
The paper is organized in six sections. The second section provides an overview of past studies on ICT adoption and possible impacts on business process and profit, which will help us situate the present study. The third section presents an analytical framework for examining the relationship among ICT adoption, business process re-engineering, and performance, drawing from business, economics and organizational development literature. The data and methodology used in the study are discussed in section four, and the empirical results are presented in section five. The final section discusses the results, implications and limitations of the study.
ICT adoption and BPR
. Business process reengineering is an approach that firms take to re-optimize business processes for the purpose of obtaining competitive advantages and enhancing business performance, such as costs saving, quality breakthrough, better customer services, time reduction, and revenue increases (Morris & Brandon 1993). In most BPR cases, ICTs play a critical role in reshaping business practices. For example, Morton (1996) finds that ICT is an important enabler of BPR because ICTs permit the distribution of power, function, and control to wherever they are most effective, and further change the ways in which production, coordination activities, and data processing are carried out. Many experts (Attaran, 2004; Akhavan, Jafari & Ali-Ahmadi, 2006; Freeman, 2000; Venkatraman, 1991) agree that ICTs could facilitate firms to gather and analyze information, develop strategic visions, allow collaborative teamwork, and find the best approach for business process redesign. Firms sometimes need to change the organizational structure to align with the adopted ICTs to obtain positive business performance and better operational efficiency (Kang, Park & Yang, 2008).
ICT application and its influence on BPR are related to many interconnected systems, such as organizational power, cultural fit, etc. (Akhavan et al., 2006; Bartis and Mitev, 2008). Lee et al. (2009) suggest that ICTs change BPR in three dimensions. The first one is changes of workplace. In many cases, ICTs enhance control and coordination and permit changes the economics and functionality of the coordination process (Morton, 1996). For example, ICTs collapse space and time constraints and allow people to work outside the office, engage in non-face-to-face coordination activities at different locations and even in different time zones (Morton, 1996; Turban, King, Lee, Warkentin, & Chung, 2002; Sarkar & Jagjit, 2006; Ziaul, Faizul & Ken, 2006; Dub, Liu, Wynter, & Xia, 2007). ICTs enable the staff to telecommunicate from home or other locations, send and receive multimedia data, exchange information, and accomplish tasks by using electronic mail, teleconference and various Internet-based cooperation tools. Communication and connectivity within an organization becomes more direct and cross-unit collaboration among geographically dispersed business units is increased (Sarkar & Jagjit, 2006).
The second one is changes of workforce. ICTs activate series of automation in many business practices, such as supply chain management, order management, customer service management, which leads to manpower reduction, e.g., office staff and production workers reduction (Wymer & Regan, 2005; Sarkar & Jagjit, 2006; Ziaul, Faizul & Ken, 2006; Lee et al., 2009). Contract out programming, network management, customer services and market research allowing firms to make alliances with organizations that are more professional or labor-intensive (Kearney, 2000; Dube Liu, Wynter & Xia, 2007). Responding to automation and outsourcing, ICT skills or knowledge become necessary requirements for employees and are the greatest problems when firms engage in BPR (Ahadi, 2006).
The third one is changes of organizational structure. What ICTs could do is be flexible enough to handle changes of environment and business processes, which at last leads organizational structure reform (Bhatt, 2000; Chanopas et al., 2006; Venkatraman, 1991). By sharing and exchanging information through ICTs, employees can directly report to managers and business units diminish the process of mediation and increase cross-unit collaboration (Teng, Grover, & Fiedler, 1994; Sarkar & Jagjit, 2006; Ziaul, Faizul, & Ken, 2006). As a result, the organizational hierarchies become flatter, and the degree of centralization of decision making has been changed (Orman, 1998; Pan & Jang, 2008). In addition, ICTs also enhance managers' information-processing capacity (Orman, 1998) and increase managers' control (Karmarkar & Vandanal, 2004).
ICT, BPR and performance (This part - please check the logic and order of the article)
Many researchers believe that new ICTs bring better performance (Akhavan et al., 2006; Dube, Liu, Wynter & Xia, 2007; Kearney, 2000; Sarkar & Jagjit, 2006). Other researchers argue that ICT is not a driver of performance per se. It should be associated with higher performance if accompanied by organizational change (Francalanci & Morabito, 2008). ICT plays a useful role in BPR, and BPR is expected to re-optimize business processes and improve business performance (Hammer, 1990; Goll, 1992; Hammer & Champy, 1993; Al-Qirim, 2007). Industrial economists (e.g., Hempell, et al., 2004; Premkumar & Roberts, 1999; Sarkar & Jagjit, 2006; Shin, 2006) assume that ICTs have direct impacts on productivity. They employ the structure-conduct-performance framework, follow traditional input and output mathematical calculation and use regression to predict how much capital and employment of ICTs are accounted for the productivity. However, they often ignore the moderate effect of ICTs on organizational reform processes.
To evaluate ICT-driven performance, it is common to use objective profit data on stock returns and accounting metrics. Santhanam and Hartono (2003) find out that firms with superior IT capability indeed exhibit superior firm performance when compared to average industry performance in financial indicators, such as return on assets, operating income to assets, etc. Francalanci and Morabito (2008) use ROE and ROI index to test organizational absorptive capacity and analyzes its effects on the relationship between IT and business performance in small and medium enterprises. Hendricks et al. (2007) document the effect of investments in ERP, SCM, and CRM systems on a firm's stock price performance and profitability measures such as return on assets and return on sales. However, the financial benefits of these implementations yields mixed results. Many multi-years, multi-million dollar ERP projects, do not reap benefits from ICTs (Attewell, 1996; Dryden, 1998). Even the gains are realized, given the high levels of expenditure and efforts for ICTs, the benefits for business in these investments are relatively small in comparison (Bird, 1994; Lyytinen & Robey, 1999; Carr, 2003; Na, et al., 2004). Thus, how ICTs affect businesses re-engineering process, and how ICT-enabled changes influence business performance remain a puzzle and need further clarification.
Based on resource-based theory, Melville, Kraemer, and Gurbaxani (2004) propose an integrative model of IT business value that comprises three domains: value generation process (e.g., IT & organizational resources, process & organizational performance, etc.), competitive environment (e.g., industrial and trading partner characteristics), and macro environment (e.g. country characteristics). The conceptual model describes how phenomena resident within each domain shape the relationship between ICT and organizational performance. Only partial models are tested on empirical data and the results show no persistent evidence of performance associated with ICT investments (Santhanam & Hartono, 2003; Jin, 2006; Hendricks et al., 2007; Francalanci & Morabito, 2008). They further classify performance into two parts: (1) business process performance-a range of measures associated with operational efficiency enhancement, e.g., inventory turnover (Barua, et al., 1995), customer satisfaction (Devaraj & Kohli, 2000), and on-time shipping (McAfee, 2002); (2) organizational performance-IT-enabled performance impacts across all firm activities, e.g., operations measures (cost reduction, productivity enhancement, etc.), market-based measures (stock market value, Tobin's q, etc.) (Dehning & Richardson, 2002), and perceptual measures (usage metrics, balanced scored card, etc.) (Tallon et al., 2000; Lee et al., 2010). This resource-based theoretical approach informs understanding of the linkage between the type of ICTs and the nature of business process and organizational performance impacts. However, such approach rarely links to BPR and researchers could not reach consensus on the performance measurement standards.
We therefore propose to investigate the moderate effect of ICTs on performance and examine how ICTs re-engineer business processes. As firms emphasize more on profitability than productivity (Dewan & Draemer, 2000), we assess the performance from the perspective of profitability.
Hypotheses and Research Questions
ICT-enabled BPR often contributes to higher performance. ICTs re-engineer the organizational structure, workplace and manpower (Lee, et al., 2009; Sarkar & Jagjit, 2006; Ziaul et al., 2006). By employing ICTs, the organization structure become flatter with efficient decision making and the staff can work and cooperate by telecommunicating each other outside the office (Lee, et al., 2009; Teng, Grover & Fiedler, 1994; Sarkar & Jagjit, 2006; Ziaul, et al., 2006). ICTs also allow organizations to function efficiently with less manpower (Wymer & Regan, 2005; Sarkar & Jagjit, 2006; Ziaul et al., 2006). We thus develop the following hypotheses:
Hypotheses 1a-1c: Workplace reform (WP) (1a), workforce reform(WF) (1b), and organizational structure reform (OS) (1c) all have positive and significant impacts on business profit (profit).
Hypotheses 2a-2b: Resource planning infrastructure (RPI) (2a) and e-commerce infrastructure (ECI) (2b) both have positive and significant impacts on workplace reform (WP).
Hypotheses 3a-3c: Workplace reform (WP) (3a), resource planning infrastructure (RPI) (3b), and e-commerce infrastructure (ECI) (3c) all have positive and significant impacts on workforce reform (WF).
Hypotheses 4a-4c: Workforce reform (WF) (4a), resource planning infrastructure (RPI) (4b), and e-commerce infrastructure (ECI) (4c) all have positive and significant impacts on organizational structure (OS).
Shall we Add these 2 research questions?????
Research Question 1: Are there differences of ICT adoption impacts on BPR among countries?
Research Question 2: Are there differences of BPR influences on profit among countries?
Figure 1 presents the Cross-Nation ICT-enabled BPR model proposed by this study.
[Insert Figure 1. here] The Cross-Nation ICT-enabled BPR Model
The preceding theoretical arguments and our proposed model can be formally stated by the following system of simultaneous regression models.
Based on the preceding theories and hypotheses, this section details the subsequent constructs and questionnaire items, and the survey settings to collect the empirical data from the senior information systems managers of organization units that make independent decisions on information technology systems.
The survey instrument for this study was developed using validated items from past research. ICT adoption refers to the perception of respondents to accept an ICT. We distinguish among two types of ICT, resource planning infrastructure (including enterprise resource planning, business intelligence, content management, and supply chain management) and e-commerce infrastructure (including storage area networks and network attached storage, digital receipts, collaboration & portal tools, website and e-commerce, business process modeling, and enterprise Instant messaging). Respondents are asked to indicate their acceptance of the above mentioned ICTs. BPR refers to dramatic and distinct management reforms and redesign of workflow as a result of ICT adoption. It was assessed from three dimensions: Organizational structure, workplace, and workforce. Organizational structure reform refers to changes of organizational hierarchy. The measurement comprised of a combination of the instruments used in Sarkar and Singh (2006), Karmarkar and Mangal (2006), asking respondents if ICT adoption decreases middle level managers, makes the organization more flatter, and increases managers' control (Karmarkar & Mangal, 2006). Workplace reform describes changes of where people work and engage in non-face-to-face coordination activities. A scale for workplace was developed to ask individuals to indicate 2 items if they telecommute and teleconference. Workforce reform is defined as manpower reduction. construct was a revision of the construct used in the survey instruments of (Karmarkar & Mangal, 2004), asking if the companies reduce manpower because of engaging in automation, outsourcing, and automated monitoring. Respondent also rated 2 items of "The need for IT skills at lower levels is going up," and "The IT function is shifting from staff to line" for this construct. Performance in this study is evaluated from the perspective of profits. The survey instruments used by Karmarkar and Mangal (2004) are used to ask respondents to rate 2 items about how they felt technology impacts on the organizations in terms of financial profits and margins. The items of WP, WF, OS, and profit were measured on a five-point Likert scale, where '1' represented 'strongly disagree', and '5' represented 'strongly agree'. The questionnaires were pre-tested on 15 respondents in order to determine the length of the instrument, the format, and the clarity and appropriateness of the wording of the scales, which resulted in the revision of some questions regarded as ambiguous.
The final questionnaire, developed partially in accordance with the BIT project, includes two major parts. The first part is composed of twenty two questions as shown in Table 1. The second part contains information about the surveyed company, including size, industry type. The twenty two items in Table 1 stand for the variables that measure the theoretical constructs serving as predictors of the Cross-Nation ICT-enabled BPR Model in Figure 1.
[Insert Table 1 here]: Detailed Constructs and Questionnaire Items
The research was conducted by obtaining responses from Chief Information Officers or senior information systems managers that make independent decisions on information and communications technologies systems in firms of Taiwan, the United States, Korea, and Chile.  Table 2 summarizes the sample sizes and demographics of each country. We selected the four countries for three reasons: (1) generalizability-that is, the countries broadly represent the international clusters; (2) construct equivalence-that is, the adoption of ICTs provides relatively similar experiences in terms of basic want/need fulfillment, for the business organizations in the four countries (and arguably many others); and (3) data availability-that is, the available BIT network that provides quality data for the four countries.
[Insert Table 2 here]: The sample sizes and demographics of each country
In Taiwan, questionnaires were mailed to 2,000 companies by systematically sampling the database of The Largest 5,000 Corporations. In Korea, the survey was sent by fax or delivered by interviewers depending on the location of the respondent. Collection process was managed by a reliable market research company. More than 50% of the respondents were manager level. In Chile, The data has a weighting factor because 300 case samples were split equally among big, medium, and small-sized firms (100 cases each, randomly selected in each category). To ensure proper representativeness of these 300 cases, they had to be weighed up/down according to the real distribution of these three sizes- there are smaller than big companies. In the United State, the survey was sent to a database of over 25,000 individuals across all industry sectors in the United State. The data was acquired from an independent entity that collects corporate data. The CIO ware requested to complete the survey either by mail or on-line, where the survey questionnaire was also made available.
The sample consists of 382 respondents from Taiwan, 248 from the United State, 262 from Korea, and 301 from Chile. We exclude 40 samples from public sectors and media sectors from Taiwan, those data are low ICT adoption due to industry characteristic. To deserve to be mentioned, this low response rate is common in the management area of obtaining responses from top management (Ferratt, Agarwal, Moore, & Brown, 1999).
Among these organizations, the largest industry in Taiwan is "Manufacturing (38.3%)", followed by "Service (28.1%)", among these organizations, 35.5% organizations with fewer than 100 employees are categorized as small organizations, and 18.7% organizations with more than 1,000 employees are large ones. In the United State, the largest industry is "Service (56.5%)", followed by "Manufacturing (31.5%)", they employ from fewer than 50 to more than 1,000 workers. Among these organizations, 7.2% organizations with fewer than 100 employees are categorized as small organizations, and 39.7% organizations with more than 1,000 employees are large ones. In Korea is "Service (63.7%)", followed by "Manufacturing (27.5%)", they employ from fewer than 50 to more than 1,000 workers. Among these organizations, 27.7% organizations with fewer than 100 employees are categorized as small organizations, and 26.2% organizations with more than 1,000 employees are large ones. In Chile, the largest industry is "Manufacturing (39.3%)", followed by "Service (28.7%)", among these organizations, 78.4% organizations with fewer than 100 employees are categorized as small organizations, and 8.6% organizations with more than 200 employees are large ones.
As a more rigorous method, structural equation modeling (SEM) may offer a more holistic picture of ICT adoption, as relationships between business process reform and business profit can be examined simultaneously in one unified model. In addition to descriptive statistical summaries, multiple-group structural equation modeling was utilized to test the model by software packages using SPSS 12.0 and AMOS 16.0 software packages.
Evaluations of the Measurements
Among four countries, enterprise resource planning is the most popular information system in the resource planning infrastructure and supply chain management system is the last (Table 1). Website and e-Commerce are the most widely adopted ICTs in the e-commerce Infrastructure. Digital receipt is the least adopted ICTs in Taiwan, Korea, and the United State. Business process modeling is the least in Chile. With regard to the workplace reform, the result of this study shows that most firms experience the increase usage of telecommuting and teleconferencing, except Korea. The results of mean of these two questions in Korea are 2.25 and 2.71. ICT-enabled workforce reforms include hiring more staff of lower level that possess IT skills, increased automated monitoring of workforce productivity, and increased IT function to work-front line from staff for all four countries. Taiwan is the only country that reform workforce by reduction because of automation and outsourcing. Firms in four countries engage in organizational structure reforms by making the structure more flat so that the business has a wider span of control and employees can report directly to managers. The surveys in this study (Table 1) demonstrate that ICTs increase profits and margins for firms of the four countries.
Quality of the Proposed Model
Scale reliability and construct validity were tested with confirmatory factor analysis (CFA) (Bentler, 1995) before assessing the hypothesized relationships shown in [Figure 1]. It is also necessary to identify equivalent phenomena when conducting cross-national research (Calantone et al., 1996). A separate CFA that involved raw data as input was performed for each of the four countries (Anderson & Gerbing, 1988). All constructs in the four samples were composed of identical items. Based on CFA results, we eliminated items that loaded on multiple constructs or had low item-to-construct loadings (Anderson & Gerbing, 1988). Then, the fit of the measurement model for each of the four countries (Bagozzi & Yi, 1988) was assessed. As Table 3 shows, the loadings of items on their respective factors were highly significant ( p < 0.01). The goodness of fit index (GFI), comparative fit index (CFI), and Bollen's fit index (IFI) range between 0.82 and 0.91 for each of the four countries. The root mean square of error approximation (RMSEA) values was found to be 0.061, 0.067, 0.064, and 0.061 for Taiwan, US, Korea, and Chile, respectively. The results suggest that the data converge and the CFA model for all four countries fits the data adequately (Bentler, 1995; Bollen, 1989).
[Insert Table 3 here]: Summary of reliability and validity index
Measurement invariance addresses the question whether the same models hold across different countries. More specifically, instruments can only be compared when showing cross-national invariance. The underlying reasoning is that cross-national differences in scale means or structural relationships between scale scores then indicate true differences across countries, instead of being caused by systematic response biases or problems with scale artifacts, reliability, or nonequivalence (Mullen, 1995; Steenkamp & Baumgartner, 1998).
We tested for measurement invariance across the four countries following the method suggested by Mullen (1995) and Steenkamp and Baumgartner (1998). A test of unconstrained model (Model 1) produced X2 =1768.62 with 788 degrees of freedom (df) as shown in Table 4. CFI was found to be 0.859 and RMSEA was 0.033. On the basis of these fit indices, the six-factor model is supported in all groups. These findings reveal that the four countries exhibit the same simple factor structure, and the same baseline model can be used for each sample population. The second model (Model 2) examines whether or not the factor covariance are equal across the four countries. This was accomplished by constraining the Î» to be equal across groups. The results in Table 4 show that the chi-square difference between the simple structure model (Model 1) and the equal variance model (Model 2) was 135 with 48 df. This is significant at p<0.05, although the chi-square tests is significant, indicative of a poor fit, the other goodness-of-fit indices indicated a good fit. Steenkamp and Baumgartner (1998) commented chi-square tests that "it has been observed that one should not rely exclusively on the chi-square difference test as it suffers from the same well-known problems as the chi-square test for evaluating overall model fit (see, e.g., Anderson and Gerbing 1988; Marsh and Grayson 1990)". Therefore, Steenkamp and Baumgartner (1998) recommend using the following four alternative fit indices: the root mean square error of approximation (RMSEA); the consistent Akaike information criterion (CAIC), the Comparative Fit Index (CFI); and the Tucker-Lewis Index (TLI) or nonnormed fit index (NNFI). The increase in chi-square between model 2 and model 3 is significant, Î”X2 =239.05 with Î”df=57. However, the remaining goodness-of-fit indices (TLI=0.815, CFI=0.821, and RMSEA=0.035), which are less sensitive to sample size, show a less marked decrease in fit. Overall, we find support for configural invariance and partial support for measurement invariance. Our findings suggest that the factor loadings, factor correlations and factor variance are invariant across countries. The error variances are not invariant across countries. However, the invariance of error variances is typically regarded as the least important constraint (Marsh, 1994).
[Insert Table 4 here]: Measurement invariance tests through a CFA constrained at several levels
Results for Hypotheses Testing
Table 5 shows path coefficients of four countries. In Taiwan, the resource planning infrastructure has significant and positive impacts on the workforce reform, with the standardized path coefficients 0.15. The e-commerce infrastructure has significant and positive impacts on the workplace reform and workforce reform, with the standardized path coefficients 0.31 and 0.17 respectively. The workplace reform positively affects the workforce reform; and, the workforce reform positively influences the organizational structure reform, with the standardized path coefficients 0.38 and 0.94 respectively. The workforce reform strongly influences company profits with the standardized path coefficients 0.05. The results in Taiwan support hypotheses 1b, 2b, 3a, 3b, 3c, and 4a.
In the United State, the resource planning infrastructure significantly affects BPR (path coefficients are 0.19, 0.34, and -0.19 for workplace reforms, workforce reforms, and organizational structure reforms respectively), supporting hypotheses 2a, 3b, and 4b. The results support hypothese 2b and 4c, confirming that the electronic commerce infrastructure significantly changes workplace (path coefficient = 0.39) and organizational structure (path coefficient = 0.24). Profit is directly affected by workplace reforms (path coefficient = 0.16), workforce reforms (path coefficient = 0.13), and negatively by organizational structure reforms (path coefficient = -0.09). Profit is also indirectly influenced by workplace reforms (path coefficient = 0.37) through workforce reforms. However, it is negatively influenced by workforce reforms (path coefficient = 0.63) through organizational structure reforms. The results therefore support hypotheses 1a, 1b, 3a, and 4a.
The ICT adoption in Korea tells a similar story. The resource planning infrastructure has significant and positive impacts on organizational structure reforms with the standardized path coefficients 0.19, and the workplace reform, with the standardized path coefficients 0.10. Nevertheless, it negatively influenced workforce reforms. The e-commerce infrastructure has significant and positive impact on the workforce reform (path coefficients 0.17). Results of ICT-enabled BPR support hypotheses 2a, 3c, and 4b. The workplace reform positively affects the workforce reform; and, the workforce reform positively influences the organizational structure reform, with the standardized path coefficients 0.56 and 0.75 respectively. Profit is positively and directly affected by workplace reforms (path coefficient = 0.27) and organizational structure reforms (path coefficient = 0.12). Profit is also negatively affected by workforce reforms (path coefficient = -0.28), the results therefore support hypotheses 1a, 1c, 2a, 3a, 3b, 3c, and 4a.
In Chile, the resource planning infrastructure significantly affects workplace reforms (path coefficient = 0.09. The electronic commerce infrastructure has significant and positive impacts on workplace (path coefficient = 0.26). Profit is directly affected by workplace reforms (path coefficient = 0.10). Profit is also indirectly influenced by workplace reforms (path coefficient = 0.34) through workforce reform and by workforce reforms (path coefficient = 0.43) through organizational structure reforms. The results therefore support hypotheses 1a, 2a, 2b, 3a, and 3b.
[Insert Table 5 here]: Results of structural equation model analysis of individual country models
Research Question 1 asks if there are differences of ICT adoption impacts on BPR among countries. The resource planning infrastructure significantly makes firms in USA to engage in workforce reforms than firms in Korea and Chile. The resource planning infrastructure also encourages more USA firms to telecommunicate for tasks accomplishment than Korea and Chile firms. The infrastructure reforms organizational structure in Korea only. The e-commerce infrastructure changes more USA firms to telecommunicate than Taiwan and Chile. Firms in Taiwan exhibits more ECI enabled workplace re-optimization than Chile. It also improves workforce efficiency for firms in Taiwan and Korea, and, reforms organizational structure of America enterprises. (add explanation)
Research Question 2 tries to figure out the differences of BPR impacts on profit among countries. Workplace reforms positively influence profits of firms in Korea the most, firms in the USA the second, and firms in Chile the third. Workforce reforms positively and significantly influence profits of firms in the USA the most and firms in Taiwan the second. Re-optimization of organizational structure brings better profits for Korean firms. Though some BPR dimensions do not excert impacts on profits directly, they do affect profits indirectly (need references) in this study: For firms of all four countries, profits can be indirectly influenced by workplace reforms through workforce reforms, and by workforce reforms through organizational structure reforms. (add explanation)
Conclusion and discussion
ICTs play an important role in business practices and management. Yet, many implementations of ICTs end with poor results (Dryden, 1998; Gallagher, 1998). Enterprises may successfully grasp computerization payoffs only if they engage in re-optimization of business processes to align with ICTs (Al-Qirim, 2007; Buckby et al., 2008; Francalanci & Morabito, 2008). Past studies rarely investigate what ICTs that firms adopt, how ICTs affect businesses, and how ICT-enabled changes influence business performance at international level. This research applies multiple-group structural equations modeling to test the Cross-Nation ICT-enabled BPR Model on data derived from cross-national research.
By investigating 1193 senior information systems managers from four countries, the results show that the resource planning infrastructure and the e-commerce infrastructure positively affect some dimensions of business process engineering. They are consistent with the literature (Ahadi, 2006; Lee et al., 2009; Sarkar & Jagjit, 2006; Wymer & Regan, 2005; Ziaul, Faizul & Ken, 2006). The results also discover ICT-enabled BPR affects enterprise performance (Hammer & Champy, 1993; Al-Qirim, 2007). In Taiwan, ICT adoption makes firms to engage changes in terms of accepting automation in business practices, outsourcing business functions, and recruiting people that have ICT skills or knowledge. Firms tend to use the e-commerce infrastructure to overcome space and time constraints for communicating, exchanging information, and finishing tasks. Taiwan used to be a major manufacturer country that exports various products. As China economy has grown, many Taiwanese firms shift their production lines to China by either contracting out labor intensive operation or establishing subsidiaries in China (China Review News, 2009). By telecommunicating the parent company in Taiwan using ICTs seems logical.
However, for firms in the United State, two negative significant path coefficients were from RPI to OS and from OS to Profit, with the standardized path coefficients -0.19 and -0.09, respectively. In Korea, two negative significant path coefficients were from RPI to WF and from WF to Profit, with the standardized path coefficients -0.15 and -0.28, respectively. The reasons of why RPI doesn't have positive impacts on OS in US might be that because firm size in US is much bigger than other country in our survey. Further, it also impacts the profit, because firm may need more capital to maintain operation.
In Korea, WF has negative and significant influence on profit which means workforce such as automation of function did not leads to human resource released. ICT implement is huge and long term investment, Thouin et al. (2009) suggest that IT activities that have become commodities (having 'low specificity') should be outsourced to improve the firm's financial performance.
This paper contributes to IS research by offering some insights and exploring BPR generalization by cross national comparison. First, it provides empirical evidence to examine impacts of ICTs contributions to business process reengineering and performance. Prior research on this topic usually focused on BPR theoretical discussion and implementation strategies based on case study. What has been largely missing is development of empirical evidence of generalization. Second, research usually focuses on one or two enterprise application, such as KM or ERP and lack in offering a complete picture of the relationships between IT adoption and BPR (Lee, Chu, & Tseng, 2009). The findings will be valuable in understanding the complexity of various ICT adoptions and predicting outcome of BPR. Finally, we exploded the national difference on ICTs adoption and its impacts on financial performance. It should be noted that BPR instruments is not a significant determinant of performance. This inconsistency with prior literature may be because ICT investment may have time delay effect on it and financial index is not the only indicator for performance estimation. For example, Santhanam & Hartono (2003) remind that prior financial performance of firms must be taken into consideration in future tests of IT capability.
Despite all the care given to this study, there are several limitations of our research should be considered in further research. First of all, we do have Europe data form BIT research. However due to model converge and some missing value problems, we eliminated European country from this paper. Secondly, we use five-point Likert-type response format to ask change of financial performance in firms instead of using ROA or ROI data, because it is difficult to get financial index of each sample in every country. Thirdly, as a national comparison, there might be survey biases related to the subjective nature of the data. CIO is our only informant about ICT adoption and performance evaluation may cause cognitive biases; however, James & Hatten (1995) have demonstrated this type of measurement is valid. Notwithstanding these limitations, this study has demonstrated that integrating the ICT impacts and business process re-engineering could provide further understanding of ICTs contribution to firm performance. We recommend two patterns for further research: (1) add more country in study; (2) use economic location category to tell the difference on each region, such as: Asia, American, and European, in order to gain knowledge and understanding on the processes by which firms use ICTs to support their performance and culture difference.