Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

Complementarities between IT and Firm Diversification and Performance Implications Yu Liu RPI [email protected]

T. Ravichandran RPI [email protected] Abstract

This paper empirically investigates the role played by IT in diversified firms with respect to firm performance, emphasizing the complementary relationship between IT and firm diversification. First by reviewing management and finance literature, we briefly examine different types of diversification, including related diversification, unrelated diversification, and geographic diversification, particularly regarding their relationship to firm performance. Then we empirically test the interaction effects between IT and different types of diversification on firm performance and find out that the interaction between IT and related diversification has a significantly positive impact on performance. Our hypotheses and findings are consistent with prior theory-based argument that IT when used to leverage the difference in strategic resources can have a positive impact on firm performance. As a whole, this paper aims to enrich the IT business value studies by taking on a new perspective or contingency under which IT does matter to firms.

1. Introduction Information Systems (IS) researchers have long been interested in the links between Information Technology (IT) and firm performance. While early research was largely focused on the direct effects of IT on performance, there is considerable and growing evidence to suggest that these effects may be more correctly viewed as both contingent and complementary [63]. The effects of IT on firm performance have been found to depend on its complementary relationship with traditional production inputs like capital and labor [28], certain types of organization structures [7] and organizational investments [9], front-end ecommerce capability [66], and other firm characteristics. Firm diversification, as one important firm scope variable, has been extensively researched in management and finance literatures wherein studies

Shu Han RPI [email protected]

Iftekhar Hasan RPI [email protected]

have attempted to link diversification to firm performance and market valuation. Within the IS literature, the causal relationship between IT and firm scope variables (e.g. diversification and vertical integration) has been the major focus among scholars who have used transaction costs economics to study the impact of IT on firm boundaries [10, 17, 25, 27, 41]. To date there is still a salient lack of research in our field that investigates the performance implications of such a relationship. The causal link between IT and firm diversification is not unidirectional. It has been argued that the use of IT contributes to more firm diversification as well as that more diversified firms need more IT. Empirical results seem to support both arguments [17, 27], which probably implies a complementary relationship rather than causality between IT and diversification. Complementarity suggests that the effects of diversification on firm performance could be different with and without appropriate IT resources. Clemons and Row [15] discuss the complementarity between IT and firm scope when they argued that IT alone many not yield competitive advantage to a firm. Rather it is how IT is leveraged to exploit structural differences such as vertical integration and diversification across firms that are likely to be a source of sustainable competitive advantage. Such firm characteristics, when leveraged with IT, may become almost unique for the firm and thereby provide the firm with a competitive edge in the marketplace [36]. While performance effects are implicit in the notion of complementarity not much empirical work has been undertaken to address the question: does complementarities between IT and firm scope impact performance? In this paper we seek to fill this void and thereby add to our body of knowledge in this area. The rest of the paper will be organized as follows. First a review of literature that has examined the relationship between diversification and firm performance will be presented. Next, theoretical arguments about the complementarities between IT and several types of firm diversification are developed and hypotheses proposed. Then the empirical part

0-7695-2507-5/06/$20.00 (C) 2006 IEEE

1

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

follows, which consists of methodology, data, results, and discussion subsections. The conclusion section summarizes the findings and implications of this study and suggests future research directions.

2. A review of diversification studies Our review of the large body of diversificationperformance studies in the strategic management literature indicate that diversification-performance research has shifted from the relative performance of diversified and undiversified firms to a more finegrained paradigm that focuses specifically on performance differences between unrelated and related diversifiers [48]. While the unrelated-related dichotomy focuses largely on product diversity, for decades international management scholars have been exploring the competitive and performance implication of international diversification as well [30]. Conceptually diversification can be roughly positioned within a two-dimension coordinates system, one dimension for product scope, and the other for geographic scope. The remaining part of this section will examine product diversification and geographic diversification respectively, with respect to their implications on performance.

2.1 Product diversification Product diversification, the extent to which a firm operates in multiple lines of business [17], has been intensively studied by strategic management researchers, who often emphasize the benefits from diversification into related businesses. The useful distinction between related and unrelated diversification has proven to be an effective means to inspect the varied effects of different types/levels of diversification on firm performance. Firms pursing related diversification attempt to exploit economies of scope through the sharing of physical and human resources across similar lines of business while unrelated diversification pursuers seek to achieve economic benefit by being able to allocate capital and other financial resources in an internal market more efficiently than public exchanges can [17]. Although the evidence is not clear cut [26], diversification into related businesses is frequently argued to provide better performance advantage and thus should be preferred by the firm. A major reason is that different product areas can leverage knowledge gained in each other while unrelated diversification adds administrative burdens without economies of scope in developing competencies [22]. The difference in performance effects of varied types/levels of product diversification entails a curvilinear relationship between diversification and

firm performance. The curvilinear model posits that some diversification (either moderate levels or related diversification) is better than none [48]. Compared with single-business firms, firms engaging in related diversification are able to exploit synergies across product units by consolidating business activities in manufacturing, marketing, raw material purchases, R&D, etc., and thus achieve both scale and scope economies [2]. For example, at Texas Instruments, defense electronics, semi-conductors, and computer business share R&D activities and manufacturing facilities in an attempt to leverage efforts across units and gain necessary efficiencies [48]. Besides, improved learning curve efficiencies, intrafirm product/process technology diffusion [4], and enhanced market power [2] are also argued as the benefits associated with related diversification. In the long term, related or moderate diversifiers will outperform their single-business counterparts and seize a better competitive position in the market. Nonetheless as the diversification degree increases, the associated costs also escalate. One potential cost associated with unrelated or higher levels of diversification, as argued by finance researchers, is the exacerbated managerial agency problems [44]. Unrelated diversification allows firms to pool cash flows from divisions and reallocate cash to divisions in accordance with financial criteria, thereby helps to set up an internal capital market [56]. The access to an internal capital market may provide managers a greater opportunity to over invest because of excess or free cash flow, while being diversified makes it even more difficult to resolve the agency problem using equity participation [44]. In addition, coordination costs with higher levels of diversification might increase drastically when the diversified firm has tapped into businesses that have little in common. The top management in unrelated diversifiers often has little firsthand knowledge of the operating affairs of a particular division’s industry, technology, or geographic region [18], further impeding coordination efforts. As a result, the marginal cost of diversification increase rapidly as it hits high levels and one could conclude that firms experience some optimal level of diversification [48]. These arguments lead to the well-known inverted U-shaped relationship between firm diversification and performance.

2.2 Geographic diversification Product/industry is not the only dimension over which firms diversify. Both international and product diversification plays key roles in the strategic behavior of large firms [29]. Geographic diversification is defined as expansion across the

2

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

borders of global regions and countries into different geographic locations or markets [30]. Like product diversification, the core research question in geographic diversification studies is concerned with the association between geographic scope and performance. But it is noteworthy that the empirical results have been more conclusive than those concerned with product diversification in the sense that in general a consistently positive relationship has been observed between geographic scope and performance [16]. The ability to gain higher returns from exploring proprietary assets (brand equity, patents, unique processes, etc) across multiple markets endows the owner superior financial performance [16]. In particular, a firm can gain above-normal returns by exploiting its firm-specific assets, especially intangible ones, in global markets [11] and the exploitation is further reinforced by imperfections found in markets for the trade of these assets [16]. Other benefits of geographic diversification stem out of improved market power, the spread of risks across international markets, and the ability to source lower cost factor inputs [38]. However, geographic diversification does bear some significant costs as well. Escalating geographic dispersion can greatly increase transaction costs and managerial information-processing demands [29]. As these arisen costs outweigh the benefits, the effect on firm performance becomes negative. As one attempt to verify a nonlinear relationship between geographic diversification and performance, Hitt et al. [30] proposed an inverted U-shaped relationship between international diversification and firm performance, which instantly reminds us about the curvilinear model of product diversification on performance. It is interesting to observe the coincidence between these two different types of diversification, which is probably due to the fact that many of the costs associated with product diversification, such as coordination difficulties, information asymmetry, and incentive misalignment between headquarters and divisional managers, also apply to geographic diversification [40]. As Palich et al. [48] noted, “This indicates increasing advantages as both product and global diversification rise, but it also demonstrates the negative utility of these activities beyond some optimal level of diversity”.

2.3 Interaction of product diversification and geographic diversification Many internationally diversified firms also operate in multiple and disparate product markets [30]. In reality, a large portion of firms simultaneously exhibit both geographic diversification and product

diversification to some extent. Many MNCs are good examples of this kind of diversifiers. A number of strategic management studies are looking into the impact of the interaction between product diversification and geographic diversification on firm performance. Some suggested that multinationality improves the performance of low product diversified firms by providing risk diversification and a broader customer base over which to gain economies of scope to fixed resources [61]. On the contrary, Kim et al. [37] demonstrated empirically that product diversified firms do perform better when they are more geographically diversified. Based on an organization learning perspective, Hitt et al. [30] argued that experience with product diversification can build managerial capabilities that allow more effective management of international diversification and thus product diversification has a positive moderating effect on the relationship between international diversification and performance. In the meantime, some other studies reported insignificant moderating effects on performance [21,22]. To date no consensus has been reached on the interaction effect of product and geographic diversification on firm performance.

3. Complementarities between IT and firm diversification Complementarity, or the interaction effect between complementary resources, refers to the condition that the existence of one resource will significantly change the effect of another resource on the output variable, in most cases, firm performance. The strategic necessity argument of IT [14, 15] heralded studies on the contingent effects of IT on business value or the complementarity effects of IT and other resources on firm performance. Since then a substream of IS studies has undertaken the copresence-oriented perspective and examined complementarities between IT resources/capabilities and many other firm resources [see for example 33, 51, 62]. The copresence of complementary resources, however, may not necessarily lead to performance improvement. Another approach to observe complementarities focuses on how IT is utilized to reinforce the value of other resources. Clemons and Row [15] argued that IT can only provide sustained competitive advantage when used to leverage differences in complementary strategic resources like diversification and vertical integration. Ravichandran and Lertwongsatien [53] found that firm performance can be explained by the extent to which IT is used to support and enhance a firm’s core competencies. Following a similar logic we posit that complementarities arise in diversified firms when IT is used as a cost-effective coordination

3

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

mechanism across business units to support diverse business activities.

3.1 IT and product diversification Theories suggest that because of the role of IT in internal control and coordination more investment in IT is required in diversified firms while in turn more use of IT can induce firms to expand into more diversified product markets. With respect to empirical studies, Dewan et al. [17] has shown that more diversified firms have greater need for IT investment while Hitt [24] demonstrated that not only diversified firms need more IT but also the degree of firm diversification depends upon the use of IT. The causal direction is not clear cut either theoretically or empirically, which may reflect that IT and product diversification are inherently associated and tightly interdependent, leading us to speculating a complementary relationship between them. Recently a related study found that complementarities between IT and product variety have a strong effect on firm performance in terms of market evaluation [20]. In this study, we will discuss related and unrelated diversification respectively. Related diversification is characterized by the similarity among the firm’s lines of business. Management authors argue that related diversification leads to better performance because of the sharing of similar resources and capabilities across different business lines, namely synergies. These synergies can certainly be achieved without much involvement of IT, but through the use of IT, benefits from synergies can be further reinforced. IT can be leveraged to influence the performance of diversified firms by exploiting scale economies in key resources and principal activities, which are homogeneous but essential across different product units [15]. The way IT is used to support value chain activities such as inventory management, logistics, and material procurement is a good illustration. As an example, Hewlett Packard (HP) has intensively used IT to coordinate component procurement across its diverse and decentralized U.S. manufacturing operations and cut off a lot of their procurement costs by doing so [15]. The relative ease to transfer management experience and knowledge across related product units accounts for another important benefit of related diversification. The effective transfer of knowledge or best practice is crucial for achieving superior performance, especially in a highly diversified firm. For example, “Texas Instruments has avoided over $1.5 billion in unnecessary manufacturing investment from its increasing stock of 530 best practices. Xerox Europe realized $400 million in increased revenues

from a sales-related portfolio of best practices.” [3]. With respect to IT, it is expected that the Internet and other electronic means will result in more best practice transfers [32]. Nowadays most firms have their own intranets and email systems, making the knowledge transfer within the firm even more effective and efficient than before. IT can be used to develop and transfer a skill base between lines of business, thus leveraging a firm’s expertise [15]. To summarize, we hypothesize: HYPOTHESIS 1: The interaction between IT and related diversification will have a positive effect on firm performance. The motives for unrelated diversification could be quite complex and varied. Given the existence of frictions in the financial markets, such as bankruptcy costs and taxes, there may be financial motives for non-synergistic type of diversification [2]. One implication of such pure-financial diversification is that a diversified firm’s cash flow may provide a superior means of funding an internal capital market [44]. Excess cash is reallocated to divisions through the internal capital market mostly based on financial criteria, a process that IT is less capable to help. As agency problems arise when managers tend to overinvest given the existence of excess or free cash flow [35], due to the lack of knowledge about the particular industry, top management relies almost exclusively on standard financial reports, and employs incentives rather than IT-enabled monitoring as a control mechanism for agency problems [17]. Consequently the use of IT does not help much in preventing agents’ opportunistic behaviors. The less need for IT investment in unrelated diversification, as revealed by prior empirical research [17], probably exactly reflects the relatively loose coupling between IT and unrelated diversification, which might cause one to suspect that their interaction effect on performance is much weaker or more likely insignificant. Given the above arguments we hypothesize: HYPOTHESIS 2: The interaction between IT and unrelated diversification will not have any effect on firm performance.

3.2 IT and geographic diversification Global information technology (i.e. computer hardware, software, and data communications shared across country borders) can support the multinational firm as it seeks to coordinate global operations, diffuse innovation worldwide, or provide integrated service to a global corporate customer [34]. Costeffective IT has facilitated the expansion of business activities into international markets for decades. Meanwhile, escalating managerial needs for internal

4

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

coordination and control make geographically diversified firms demand more IT investment to coordinate their assets and operations across country borders. More importantly, IT can be well leveraged to fortify benefits derived from global operations, which in turn boosts the overall performance of the firm. Despite the importance of IT in geographically diversified firms little extant literature in the IS field has addressed the complementarity between geographic diversification and IT with the exception of [49], which found the relationship between geographic scope and IT scope to be complementary and have an impact on firm performance. We propose two major advantages from geographic diversification that can be facilitated by the use of global IT: achieving scale/scope economies and exploiting globally intangible assets. First, geographically diversified firms are characterized by their dispersed and heterogeneous markets throughout the globe. Multinational firms can realize cost advantages in traditional input factors such as labor and raw materials [60]. IT enables firms to realize economies of scale and scope by electronically sourcing raw materials worldwide at lower costs. For instance, Tricon Global Restaurant is making use of the company’s new proprietary global information system to permit all stores to order supplies through one global clearinghouse, a move which could substantially reduce inventory and related costs [60]. Second, internalization theory premises that intangible assets, such as technological know-how and patents, are information-intensive and unevenly distributed across countries. A firm can gain abnormal returns by exploiting its firm-specific assets, especially intangible ones, in international markets [11]. IT can effectively coordinate information-intensive activities, such as knowledge management and R&D efforts, across countries. Accenture’s KnowledgeXchange [57] and Siemens’s ShareNet [62], for instance, are both successful cases in which IT is heavily employed to facilitate the global transfer of knowledge. To summarize, the complementarity between IT and geographic diversification leads to IT-enabled synergies in firm resources and capabilities across country borders. Therefore we expect the interaction between IT and geographic diversification to have a positive impact on firm performance, which is contained in the following hypothesis: HYPOTHESIS 3: The interaction between IT and geographic diversification will have a positive effect on firm performance.

4. Methodology 4.1 Sample

Data to test the research model was compiled from several secondary sources. IT investment data was obtained from a data set published by InformationWeek. Diversification levels and other control variables were obtained from Compustat. We first matched the firms listed in InformationWeek with firms in Compustat database, and 963 observations remained after matching. After dropping observations for which diversification level or other control variables data were not available, 560 observations remained. Table 1 profiles data set in terms of the industry segments and the number of observations in each industry segment. -------------------Insert Table 1 here ------------------------

4.2 Variables The choice and measurement of dependent variables, explanatory variables and control variables are presented in the following section. Dependent variable As in prior studies [16, 30, 61], we define firm performance using accounting based measures. From Compustat, we compiled data for two performance measures: return on assets (ROA) and return on sales (ROS). The measures have a correlation of 0.92 (see Table 3). Accounting-based measures of a firm's profitability have been critiqued [1] but there is enough justification for their use [31]. Managers and external analysts frequently use ROA and ROS as a measure of management effectiveness and the various measures of profitability are typically related [54]. In addition, changes in stock prices tend to follow the announcement of such figures as ROA or ROS, indicating their importance in signaling firm performance [19]. Past research has found that both IT investments [8, 22] and strategic moves such as diversification [23] have lagged effects on firm performance. Hence, we use a one year time lag in measuring our dependent variables. Diversification level The measurement of diversification is a central issue in diversificationperformance research. Several indexes are available for measuring product diversification, including the entropy measure, the Herfindahl index, the concentric index, and counts of SIC codes. In this study we chose to use the entropy measure, not only because it is widely accepted but also because as one of its key properties, it allows the distinction between related and unrelated diversification [47]. Given our focus on related and unrelated diversification this measure is more suited than other available measures of product diversification. Following procedures used in past studies [17, 47] we calculated the entropy measures

5

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

for related, unrelated and total diversification using the business segment data from Compustat. Similar to product diversification, several measures for geographic diversification have been used in past research the most common being a unidimensional measure of international sales as a percentage of total sales [30]. Other measures for geographic diversification include the number of overseas subsidiaries, the Herfindahl index [12], the entropy measure [30, 37, 38] and the number of countries in which a firm has overseas subsidiaries [40]. In this study we used the ratio of international sales to total sales as a measure of geographic diversification which we calculated using the data from Compustat. IT investment Annual information technology spending data utilized in this study is from the annual survey conducted by InformationWeek (IW) during the period of 1991-1996. Every year since 1989, a list of 500 companies and their IT spending is published in a special issue of the magazine in September. The companies are drawn from the population of US firms with over $1 billion in sales. IT spending data of a firm includes IT hardware and software investments in multiple categories (PCs, workstations, servers, mainframes, peripheral devices, software, local and wide area networks, and telecommunications). IW data has been used previously by [5, 6, 58]. It is also found to be highly correlated with IT investment data from other sources, such as Computerworld [39]. To remove the influence of firm size on IT spending, we are using IT intensity in the model, calculated as the ratio of IT investment to total sales of the firm. Control variables [46] and [13] found evidence that performance differences among diversified firms were attributable in part to differences in market structures among the industries in which the firms competed. Hence, as in [23, 61], we introduced industry- and firm-level controls into the models. Following Grant et al. [23] and Geringer et al. [22], we include industry dummies in the model to control the industry-level effect. Year dummies are also included since the observations range from 1991 to 1996. Since within the same industry environment, firm performance varies due to the different strategies they pursue firm characteristics such as size, R&D expenditure, capital intensity, and financial leverage are included as controls in our analysis. We realize that a firm’s marketing capability, as reflected by its advertisement intensity, is an important explanatory variable of firm profitability. However, since only one fourth of the entries in Compustat had data on advertising expenditures, we were left with many missing values for this variable. Hence, we decided not to include advertisement intensity in the model.

Firm size, a commonly used control variable was measured by as the natural logarithm of firm total assets. R&D expenditure, which is related to firm innovation and technological assets, has been suggested as performance predictor [52]. Following past studies in the strategy literature [16, 42, 43], we use R&D intensity, the ratio of R&D expenditures to the firm’s total sales, as a control variable. Firm leverage has also been widely used to predict performance. It is measured as the percentage of longterm debt to total capital by [22, 61]. The same proxy is included in our model. Capital intensity is measured as the percentage of capital expenditures to sales [6, 42]. The descriptive statistics and the correlations among the variables in our model are shown in Table 2 and Table 3 respectively. --------------------------------------Insert Table 2 here Insert Table 3 here -----------------------------------------------

4.3 Econometric model Our dataset is an unbalanced panel data with 560 observations over six years of 186 firms. Since on average three observations are available for each firm, instead of employing any advanced time-series technique, we are using Ordinary Least Squares (OLS) regression for the pooled data. Using an unbalanced panel data may violate the homoskedasticity assumption of OLS, resulting in biased estimates of standard errors. To correct for this we ran the analysis with Huber-White standard errors clustering on firms, which will produce a robust estimation of standard errors given the presence of arbitrary correlations in error terms within the cluster. This method, also called Rogers standard errors [55], is robust to different specifications of the dependence in error terms within cluster. Peterson [50] compared multiple methods of calculating standard errors for panel data including OLS, Huber-White robust standard errors, Fama-MacBeth estimate, and fixed effects. He concluded that “the Rogers estimates produce correct standard errors and correctly sized confidence intervals in the presence of a firm or time effects…when clustered on the same dimension (firm or time).” Since the precise form of the dependence in the residue is unknown an estimate which is robust to different specifications is advantageous [50]. More details on this method can be found in [64] and [65].

5. Results and discussion -------------------Insert Table 4 here ------------------------

6

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

Table 4 presents the results using one year lagged ROA and ROS as the dependent variable. In both models the interaction between IT and related diversification has a positive impact on performance. Both coefficients are statistically significant (p < 0.01). Therefore Hypothesis 1 is supported, i.e. the complementarity between IT and related diversification affects firm performance positively. With regard to unrelated diversification, the interaction effect is positive but not significant in either model, which supports Hypothesis 2. Surprisingly, Hypothesis 3 receives no empirical support. Rather, the coefficient of the interaction term between IT and geographic diversification is negative and significant, implying that the investment in IT will be more profitable in firms that are more geographically focused. We suspect two reasons for this contradictory finding. First, to realize the synergy between IT and geographic diversification, the firm needs an integrated IT infrastructure across business units that span different countries. While a common IT infrastructure is more achievable in a domestic firm, it can be much more difficult to establish in a multinational firm, given the enormous difference in levels of IT application across countries. Second, our sample is drawn from the time period 1991-1996, a period when the Internet, a global IT infrastructure now was only in its embryonic state and was not utilized extensively for commercial purposes. We believe that during the time period of our study a significant proportion of IT investments of multinational firms could have been directed towards IT infrastructure development. While these investments created needed capabilities for information exchange across units operating in different countries, these capabilities may not have had a significant effect on firm performance in the short run because of time lags for infrastructure capabilities to payoff. Moreover, since any infrastructure investment only provides a platform for information management and coordination capabilities, these investments per se may not be performance enhancing and require complementary changes that leverage the infrastructure capabilities. The insignificant effect of the interaction between IT and unrelated diversification should be treated with caution. The result should not be interpreted as meaning that unrelated diversifiers do not need IT capability in their operations. Firms invest in IT for a number of reasons such as responding to competitive pressures and regulatory needs. Likewise the roles played by IT in firms are also various. In this paper we mainly focused on the coordination role of IT, based on which we reached the conclusion that the complementarity between IT and unrelated

diversification has no effect on performance. However, if one looks at other benefits derived from IT such as lower production costs [45], enhanced agility [57], improved customer orientation and knowledge codification [5], it is worthwhile to foster IT capabilities to facilitate these benefits even in unrelated diversifiers. As a whole our paper provides empirical evidence that the complementarity between IT and firm scope does exist, which is reflected by the positive relationship between the IT-diversification interaction and firm performance. This finding verifies prior arguments that IT when used to leverage the difference in structural resources can add business value to the firm and provides empirical evidence of such complementarities.

6. Conclusion Previous theoretical work suggests that a complementary relationship exist between IT and firm scope variables such as diversification, and that the complementarity can be a sustainable source of competitive advantages [15]. No empirical studies, however, have been undertaken to investigate the complementarity between IT and firm diversification and its implications on firm performance. The present study attempts to fill this gap in the literature. By employing secondary data and building econometric models, we find empirical evidence for the complementarity argument and confirm our notion that IT, when used to leverage the difference in structural resources like diversification, will have a positive effect on firm performance. Managers in highly diversified firms should view IT as a valuable asset given the performance-critical role played by IT when coupled with diversification. However, they must also be cautious that the complementary effects on performance are contingent on types of diversification. In firms predominated by unrelated diversification, IT may not contribute to performance as much as in related diversifiers. Our study suggests that managers in these firms focus not only on the coordination benefits of IT but also other enabling aspects to justify their IT spending. We used a unidimensional measure for geographic diversification which may not be rich enough to effectively tap into the coordination challenges firms face in operating in international markets. Although foreign sales ratio is the best measurement available to us to represent geographic diversification, we suggest future researchers to use more sophisticated measures for geographic diversification such as the entropy measure and the Herfindahl index. Secondly, the present study only addresses the breadth of

7

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

diversification, while diversification itself is by all means multidimensional. For example, Simmonds [59] suggested that two dimensions of diversification, breadth and mode, if combined together, explain more variance in firm performance. Therefore a viable direction for IS researchers to take is to investigate whether IT exerts different performance impacts in various breadth-mode combinations.

References [1] Aaker, D.A. and Jacobson, R., "The Role of Risk in Explaining Differences in Profitability". Academy of Management Journal, 1987. 30(2): 277-296. [2] Amit, R. and Livnat, J., "Diversification Strategies, Business Cycles and Economic Performance". Strategic Management Journal, 1988. 9(2): 99-110. [3] Ashton, C., Managing Best Practices. Business Intelligence, London, 1998. [4] Barney, J.B., Gaining and Sustaining Competitive Advantage. Addison-Wesley, Reading, MA, 1997. [5] Bharadwaj, A.S., "A Resource-Based Prospective on Information Technology Capability and Firm Performance: An Empirical Investigation". MIS Quarterly, 2000. 24(1): 169-196. [6] Bharadwaj, A.S., Bharadwaj, S.G., and Konsynski, B.R., "Information technology effects on firm performance as measured by Tobin's q". Management Science, 1999. 45(7): 1008-1024. [7] Bresnahan, T.F., Brynjolfsson, E., and Hitt, L.M., "Information technology, workplace organization, and the demand for skilled labor: Firm-level evidence". The Quarterly Journal of Economics, 2002. 117(1): 339-376. [8] Brynjolfsson, E., "The Productivity Paradox of Information Technology". Communications of the ACM, 1993. 36(12): 67-77. [9] Brynjolfsson, E., Hitt, L.M., and Yang, S., "Intangible Assets: Computers and Organizational Capital". Brookings Papers on Economic Activity: Macroeconomics, 2002(1): 137-199. [10] Brynjolfsson, E., Malone, T.W., Gurbaxani, V., and Kambil, A., "Does Information Technology Lead to Smaller Firms". Management Science, 1994. 40(12): 1628-1644. [11] Buckley, P.J., "The Limits of Explanation: Testing the Internalization Theory". Journal of International Business Journal, 1988. 19(2): 181-194. [12] Bühner, R., "Assessing International Diversification of West German Corporations". Strategic Management Journal, 1987. 8(1): 25-37. [13] Christensen, K.H. and Montgomery, C.A., "Corporate Economic Performance: Diversification Strategy versus Market Structure". Strategic Management Journal, 1981. 2(4): 327-343. [14] Clemons, E.K. and Row, M., "McKessoon Drug Company: A Case Study Economost -- A Strategic Information System". Journal of Management Information Systems, 1988. 5(1): 36-50. [15] Clemons, E.K. and Row, M.C., "Sustaining IT Advantage: The Role of Structural Differences". MIS Quarterly, 1991. 15(3): 275-292.

[16] Delios, A. and Beamish, P.W., "Geographic scope, product diversification, and the corporate performance of Japanese firms". Strategic Management Journal, 1999. 20(8): 711-727. [17] Dewan, S., Michael, S.C., and Min, C.K., "Firm characteristics and investments in information technology: Scale and scope effects". Information Systems Research, 1998. 9(3): 219-232. [18] Dundas, K.N.M. and Richardson, P.R., "Implementing the Unrelated Product Strategy". Strategic Management Journal, 1982. 3(4): 287-301. [19] Fama, E.F. and Miller, M.H., The Theory of Finance. Dryden Press, Hinsdale, IL, 1984. [20] Gao, G. and Hitt, L.M. IT and Product Variety: Evidence from Panel Data. In Proceedings of the 25th Annual International Conference on Information Systems. 2004. Washington, D.C. [21] Geringer, J.M., Beamish, P.W., and daCosta, R.C., "Diversification Strategy and Internationalization: Implications for MNE Performance". Strategic Management Journal, 1989. 10(2): 109-119. [22] Geringer, J.M. and Tallman, S., "Product and international diversification among Japanese multinational firms". Strategic Management Journal, 2000. 21(1): 51-80. [23] Grant, R.M., Jammine, A.P., and Thomas, H., "Diversity, Diversification, and Profitability among British Manufacturing Companies, 1972-84". The Academy of Management Journal, 1988. 31(4): 771-801. [24] Greene, W.H., Economic Analysis. Prentice Hall, Upper Saddle River, NJ, 2003. [25] Gurbaxani, V. and Whang, S., "The Impact of Information Systems on Organizations and Markets". Communications of the ACM, 1991. 34(1): 59-73. [26] Helfat, C.E. and Eisenhardt, K.M., "Inter-Temporal Economies of Scope, Organizational Modularity, and the Dynamics of Diversification". Strategic Management Journal, 2004. 25(13): 1217-1232. [27] Hitt, L.M., "Information Technology and Firm Boundaries: Evidence from Panel Data". Information Systems Research, 1999. 10(2): 134-149. [28] Hitt, L.M. and Snir, E.M., The Role of Information Technology in Modern Production: Complement or Substitute to Other Inputs? In Workshop on Information Systems and Economics. 1999: Wharton School. [29] Hitt, M.A., Hoskisson, R.E., and Ireland, R.D., "A Mid-Range Theory of the Interactive effects of international and product diversification on innovation and performance". Journal of Management, 1994. 20: 297-326. [30] Hitt, M.A., Hoskisson, R.E., and Kim, H., "International diversification: Effects on innovation and firm performance in product-diversified firms". Academy of Management Journal, 1997. 40(4): 767-798. [31] Hoskisson, R.E., Hitt, M.A., Johnson, R.A., and Moesel, D.D., "Construct validity of an objective (entropy) categorical measure of diversification strategy". Strategic Management Journal, 1993. 14(3): 215-235. [32] Jarrar, Y.F. and Zairi, M., "Future Trends in Benchmarking for Competitive Advantage: A Global Survey". Total Quality Management, 2001. 12(7): 906-912. [33] Jarvenpaa, S.J. and Leidner, D.E., "An Information Company in Mexico: Extending the Resource-Based View

8

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

of the Firm to a Developing Country". Information Systems Research, 1998. 9(4): 342-361. [34] Jarvenpaa, S.L. and Ives, B., "Organizing for global competition: The fit of information technology". Decision Sciences, 1993. 24(3): 547-580. [35] Jensen, M.C., "Agency costs of free cash flow, corporate finance and takeovers". American Economic Review, 1986. 76: 323-329. [36] Kettinger, W.J., Grover, V., Guha, S., and Segars, A.H., "Strategic Information Systems Revisited: A Study in Sustainability and Performance". MIS Quarterly, 1994. 18(1): 31-58. [37] Kim, W.C., Hwang, P., and Burgers, W.P., "Global Diversification Strategy and Corporate Profit Performance". Strategic Management Journal, 1989. 10(1): 45-57. [38] Kim, W.C., Hwang, P., and Burgers, W.P., "Multinational's Diversification and the Risk-Return Tradeoff". Strategic Management Journal, 1993. 14(4): 275-286. [39] Lichtenberg, F.R., "The Output Contribution of Computer Equipment and Personnel: A Firm level Analysis". Economics of Innovation and New Technology, 1995. 3: 201-217. [40] Lu, J.W. and Beamish, P.W., "International Diversification and Firm Performance: The S-Curve Hypothesis". The Academy of Management Journal, 2004. 47(4): 598-609. [41] Malone, T.W. and Yates, J.B., R.I. Benjamin, "Electronic Markets and Electronic Hierarchies". Communications of the ACM, 1987. 30(6): 484-497. [42] Markides, C.C., "Diversification Restructuring and Economic Performance". Strategic Management Journal, 1995. 16(2): 101-118. [43] Markides, C.C. and Williamson, P.J., "Related Diversification, Core Competences and Corporate Performance". Strategic Management Journal, 1994. 15(5): 149-165. [44] Martin, J.D. and Sayrak, A., "Corporate Diversification and Shareholder Value: A Survey of Recent Literature". Journal of Corporate Finance, 2003. 9(1): 37-57. [45] Mitra, S. and Chaya, A.K., "Analyzing CostEffectiveness of Organizations: The Impact of Information Technology Spending". Journal of Management Information Systems, 1996. 13(2): 29-57. [46] Montgomery, C.A., "Product-Market Diversification and Market Power". Academy of Management Journal, 1985. 28(4): 789-798. [47] Palepu, K., "Diversification Strategy, Profit Performance and the Entropy Measure". Strategic Management Journal, 1985. 6(3): 239-255. [48] Palich, L.E., Cardinal, L.B., and Miller, C.C., "Curvilinearity in the diversification-performance linkage: An examination of over three decades of research". Strategic Management Journal, 2000. 21(2): 155-174. [49] Peffers, K. and Tuunainen, V.K., "Leveraging geographic and information technology scope for superior performance: an exploratory study in international banking". Journal of Strategic Information Systems, 2001. 10(3): 175200. [50] Petersen, M.A., Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches, in Kellogg School

of Management, Northwestern University and NBER, Working Paper. 2005. [51] Powell, T.C. and Dent-Micallef, A., "Information technology as competitive advantage: The role of human, business, and technology resources". Strategic Management Journal, 1997. 18(5): 375-405. [52] Ravenscraft, D.J. and Scherer, F.M., Mergers, Sell-offs, and Economic Efficiency. Brookings Institution, Washington, DC, 1987. [53] Ravichandran, T. and Lertwongsatien, C., "Effect of Information Systems Resources and Capabilities on Firm Performance: A Resource-Based Perspective". Journal of Management Information Systems, 2005. 21(4): 237-276. [54] Robins, J. and Wiersema, M.F., "A Resource-based Approach to the Multibusiness Firm: Empirical Analysis of Portfolio Interrelationships and Corporate Financial Performance". Strategic Management Journal, 1995. 16(4): 277-299. [55] Rogers, W., "Regression Standard Errors in Clustered Samples". Stata Technical Bulletin, 1993. 13: 19-23. [56] Rowe, W.G. and Wright, P.M., "Related and Unrelated Diversification and Their Effect on Human Resource Management Controls". Strategic Management Journal, 1997. 18(4): 329-338. [57] Sambamurthy, V., Bharadwaj, A.S., and Grover, V., "Shaping Agility through Digital Options: Reconceptualizing the Role of Information Technology in Contemporary Firms". MIS Quarterly, 2003. 27(2): 237-263. [58] Santhanam, R. and Hartono, E., "Issues In Linking Information Technology Capability to Firm Performance". MIS Quarterly, 2003. 27(1): 125-153. [59] Simmonds, P.G., "The Combined Diversification Breadth and Mode Dimensions and the Performance of Large Diversified Firms". Strategic Management Journal, 1990. 11(5): 399-410. [60] Stephens, D.O., "The Globalization of Information Technology in Multinational Corporations". The Information Management Journal, 1999. 33(3): 66-70. [61] Tallman, S. and Li, J.T., "The Effects of International Diversity and Product Diversity on the Performance of Multinational Firms". The Academy of Management Journal, 1996. 39(1): 179-196. [62] Voepel, S.C., Dous, M., and Davenport, T.H., "Five Steps to Creating A Global Knowledge-Sharing System: Siemens' ShareNet". The Academy of Management Executive, 2005. 19(2): 9-23. [63] Wade, M. and Hulland, J., "Review: The ResourceBased View and Information Systems Research: Review, Extension, and Suggestions for Future Research". MIS Quarterly, 2004. 28(1): 107-142. [64] Williams, R., "A Note on Robust Variance Estimation for Cluster-Correlated Data". Biometrics, 2000. 56: 645-646. [65] Wooldridge, J., Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge, MA, 2002. [66] Zhu, K., "The Complementarity of Information Technology Infrastructure and E-Commerce Capability: A Resource-Based Assessment of Their Business Value". Journal of Management Information Systems, 2004. 21(1): 167-202.

9

Proceedings of the 39th Hawaii International Conference on System Sciences - 2006

Figures and Tables Table 1. Distribution of observations

Table 2. Descriptive statistics of the variables

by industry segments Industry

Number of Observations

Variables

N

Unit

Mean

S.D.

ROA

560

Ratio

0.0517

0.0832

ROS

560

Ratio

0.0505

0.0830

Related Diversification

560

Entropy

0.1643

0.2792

560

Entropy

0.4373

0.4584

Mining and Construction

19

Manufacturing Transportation, Communication, and Utilities

438 16

Unrelated Diversification

Wholesale and Retail Finance, Insurance, and Real Estate

45

Geographic Diversification

560

Ratio

0.2748

0.1867

IT Intensity

560

Ratio

0.0221

0.0338

Services

27

2

Others

13

Total

560

R&D Intensity

560

Ratio

0.0351

0.0399

Assets

560

Million dollars

11838.13

28184.56

Leverage

560

Ratio

0.3814

0.0630

Capital Intensity

560

Ratio

0.0634

0.0423

Table 3. Correlations matrix of the variables ROA (1) ROS (2) Related Diversification (3) Unrelated Diversification (4) Geographic Diversification (5) IT Intensity (6) R&D Intensity(7) Ln Assets (8) Leverage (9) Capital Intensity (10)

(1) 1 0.92*** -0.04 -0.13*** 0.09** 0.07 0.16*** -0.05 -0.19*** 0.07*

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

1 -0.02 -0.13*** 0.14*** 0.11** 0.27*** 0.04 -0.14*** 0.19***

1 -0.01 0.06 -0.00 -0.03 0.17*** 0.09** 0.02

1 -0.23*** -0.13*** -0.32*** 0.18*** 0.28*** -0.01

1 0.07* 0.49*** 0.16*** -0.13*** 0.10**

1 0.23*** 0.06 -0.07* 0.02

1 0.07* -0.27*** 0.14***

1 0.30*** 0.30***

1 -0.07*

(10)

1

Table 4. Results of OLS Regression Clustering on Firms using ROA and ROS as DV Independent Variables Unrelated Diversification (Centered) Related Diversification (Centered) Geographical Diversification (Centered) R&D Intensity Ln Assets Leverage Capital Intensity IT Intensity (Centered) IT Intensity × Related Diversification IT Intensity × Unrelated Diversification IT Intensity × Geographical Diversification Constant

Restricted Model ROA -0.0079 (0.0087) -0.0091 (0.0168) -0.0193 (0.0031) 0.1851 (0.1161) -0.0036 (0.0031) -0.0006* (0.0003) 0.2271* (0.1278) 0.2094* (0.1254)

ROS -0.0096 (0.0090) -0.0050 (0.0174) -0.298 (0.0281) 0.3563*** (0.1251) 0.0001 (0.0035) -0.0004* (0.0002) 0.4374*** (0.1454) 0.2355* (0.1207)

Hypothesis 1 ROA -0.0053 (0.0081) -0.0119 (0.0126) -0.0047 (0.0243) 0.1713 (0.1147) -0.0051 (0.0032) -0.0006* (0.0003) 0.2569** (0.1258) 0.4650*** (0.1061) 2.2121*** (0.5112)

ROS -0.0073 (0.0086) -0.0074 (0.0131) -0.0173 (0.0262) 0.3445*** (0.1266) -0.0012 (0.0036) -0.0004* (0.0002) 0.4629*** (0.1439) 0.4542*** (0.1179) 1.8924*** (0.5906)

Hypothesis 2 ROA -0.0063 (0.0093) -0.0095 (0.0171) -0.0188 (0.0263) 0.1765 (0.1174) -0.0038 (0.0032) -0.0006* (0.0003) 0.2332* (0.1307) 0.3213* (0.1646)

ROS -0.0091 (0.0097) -0.0051 (0.0176) -0.0297 (0.0282) 0.3535*** (0.1278) 0.0000 (0.0036) -0.0004* (0.0002) 0.4394*** (0.1483) 0.2719* (0.1637)

0.3364 (0.4246)

0.1092 (0.4192)

Hypothesis 3 ROA -0.0037 (0.0083) -0.0123 (0.0143) -0.0063 (0.0239) 0.1654 (0.1161) -0.0052 (0.0033) -0.0006* (0.0003) 0.2572** (0.1272) 0.8589*** (0.3238)

-2.1778** (0.8897) 0.0984*** 0.0531 0.1076*** 0.0610 0.1005*** 0.0538 0.1097*** (0.0355) (0.0388) (0.0358) (0.0396) (0.0352) (0.0387) (0.0353) N 560 560 560 560 560 560 560 2 R 0.1602 0.1879 0.1892 0.2094 0.1609 0.1880 0.1793 Note: 1. * p