International Journal of Accounting Information Systems 13 (2012) 21–40
Contents lists available at ScienceDirect
International Journal of Accounting
Information Systems
Role of IT executives in the firm's ability to achieve
competitive advantage through IT capability
Jee-Hae Lim, Theophanis C. Stratopoulos ⁎, Tony S. Wirjanto
University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
a r t i c l e
i n f o
Article history:
Received 5 April 2011
Received in revised form 28 June 2011
Accepted 1 July 2011
Keywords:
Business value of IT
CIO
IT capability
Structural power
Tobin's q
a b s t r a c t
Contrary to prior studies that have tried to examine the role of IT
capabilities (ITC) on firm performance in isolation from the role of senior
IT executives, we propose that the two are linked. More specifically we
argue that there is a positive relationship between the structural power
of senior IT executives and the likelihood that the firm will develop
superior ITC. Furthermore, the contribution of ITC to a firm's competitive
advantage is much stronger in firms with powerful senior IT executives
as they are the driving force that may ensure the continuous renewal of
ITC. We develop a two-stage econometric model designed to test this
chain hypothesis that the structural power of senior IT executives will
affect a firm's ability to achieve superior ITC, in turn driving firm
performance. Empirical evidence based on a sample of large US firms
strongly supports both of our hypotheses.
© 2011 Elsevier Inc. All rights reserved.
1. Introduction
Anecdotal evidence based on companies that have been known for their superior organizational IT capability
indicates that powerful senior IT executives1 play a pivotal role in orchestrating their company's success. The
objective of this study is twofold. A two-stage approach is used in order to examine the contribution of senior IT
executives on their firm's ability to achieve superior IT capability, and the impact of co-presence of powerful
senior IT executives in firms with superior IT capability on their firm's competitive position.2
⁎ Corresponding author.
E-mail addresses: jh2lim@uwaterloo.ca (J.-H. Lim), tstratop@uwaterloo.ca (T.C. Stratopoulos), twirjant@uwaterloo.ca
(T.S. Wirjanto).
1
Salient examples include Shack at PNC, Kepler at Dow Chemical, Webb at Equifax, Stanley at Harrah's, and Mott at Wal-Mart.
2
Our study concurs with prior research that firms use many different titles for their senior IT executives (e.g., CIO, CTO, VP,
Director, Managing Director). We are speculating that for brevity the majority of studies tend to use the blanket label ‘CIO’ to refer to a
firm's senior IT executive. While this is convenient, it is an oversimplification. It ignores the fact that a senior executive's title and
number of titles is a proxy of their power within the organization, i.e., a proxy for their structural power (Finkelstein, 1992).
1467-0895/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.accinf.2011.07.001
22
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
Our study builds on and integrates two streams of research that fall under the broader IT business
value literature: (1) Studies that examine role of organizational IT capability on firm performance. One of
the first studies in this area by Bharadwaj (2000) has shown strong support for the hypothesis that firms
with superior IT capability perform better than their competitors. However, a follow up study by
Santhanam and Hartono (2003) found that the role of IT capability on firm performance is weaker once
you account for the firm's prior performance. In general the results from IT capability studies remain
mixed. (2) Studies that examine the role of senior IT executives on their firm's market and financial
performance. According to Chatterjee et al. (2001), markets react positively to the announcement of
the creation of a new CIO position. Khallaf and Skantz (2011) find that CIO appointments improve
firm financial performance but the performance improvement is largely limited to ‘firms appointing a CIO
for the first time’ and first movers. Notwithstanding the contribution of both of these studies (Chatterjee
et al., 2001; Khallaf and Skantz, 2011), a question with significant practical implications for senior
business executives and senior IT executives remains unanswered. Do incumbent senior IT executives
add value? The importance of evaluating the value adding contribution of incumbent senior IT executives
has been suggested by Chatterjee et al. (2001; p. 58): “…evidence suggests that CIOs experience a high
turnover rate, relative to other senior executives. Is this solely a function of poor performance by
incumbents?”
In this study we propose integrating these two streams of research in order to better understand the
value adding contribution of senior IT executives and the role of IT capability on their firm's performance.
Prior studies have either tried to examine the role of IT capabilities on firm performance (e.g., Bharadwaj,
2000; Santhanam and Hartono, 2003) or the role of senior IT executives on firm performance (Khallaf and
Skantz, 2011) in isolation. We propose that one should consider the role of senior IT executives in a
firm's ability to develop superior IT capability, and then the value added by senior IT executives
through superior IT capabilities. Building on existing literature (e.g., Raghunathan and Raghunathan,
1989; Grover et al., 1993; Armstrong and Sambamurthy, 1999) this study argues that there is a positive
relationship between the structural power of senior IT executives and the likelihood that the
firm will develop superior IT capability. Structural power is ‘perhaps the most commonly cited type
of power; it is based on formal organizational structure and hierarchical authority' (Finkelstein, 1992;
508). Furthermore, given the dynamic nature of IT capabilities (Bhatt and Grover, 2005; Lim et al.,
forthcoming) the contribution of IT capability to a firm's competitive advantage is much stronger in firms
with powerful senior IT executives as they are the driving force that may ensure the continuous renewal
of IT capability.
The data set for testing our hypotheses was developed by integrating data from InformationWeek 500
(proxy for firms with superior IT capability), Lexis-Nexis and a variety of online sources (search for official
title and number of titles to proxy for structural power of senior IT executives) and Compustat (firm
performance data) for the period 1997–2004. A two-stage econometric model is used to test the chain
hypothesis that the power of senior IT executives will affect a firm's ability to achieve superior IT capability,
and that this in turn drives firm performance. Empirical evidence based on a sample of large US firms
strongly supports our hypotheses.
This study contributes to prior literature on IT capability and firm performance by demonstrating
that the impact of superior IT capability on a firm's competitive performance is contingent upon the
structural power of the firm's senior IT executive. The results show that in firms where senior IT
executives have high power, the effect of superior IT capability on firm performance remains strong,
even when accounting for the firm's prior performance, time, and industry effect. Furthermore, our
study contributes to the growing literature on the role of senior IT executives (Chatterjee et al., 2001;
Dehning and Stratopoulos, 2003; Khallaf and Skantz, 2011; Masli et al., 2011) by illustrating that the
impact of a firm's senior IT executive on its performance is channelled through superior IT capability.
Identifying antecedents of IT capability, i.e., factors that lead a firm to achieve and sustain superior IT
capability, as well as recognizing the role that these factors play on the rent yielding capacity of IT
capability, helps to illuminate the proverbial black box of IT business value research. Finally, this study
has important practical implications for senior business executives who want to compete through IT
enabled strategies. The message is that bestowing more power on senior IT executives increases the
likelihood that the firm will develop and sustain superior IT capability, the sine qua non for an IT
enabled strategy.
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
23
2. Theoretical foundation and hypotheses development
2.1. IT capability and firm performance
Studies exploring the effect of organizational IT capabilities on firm performance figure prominently in the
body of IT business value literature. The pioneering empirical study by Bharadwaj (2000) defines organizational
IT capability as a firm's ability to mobilize and deploy IT based resources (i.e., flexible IT infrastructure, human
IT resources, and intangible IT enabled resources such as knowledge, synergy, and customer orientation) in
combination with other resources and capabilities.3 Bharadwaj (2000) and all subsequent studies employed
what Newbert (2007, p. 127) calls a ‘resource heterogeneity approach.’ That is, they argue on theoretical
grounds that IT capability is valuable, rare, inimitable, and/or non-substitutable. They use InformationWeek 500
(IW500), Computerworld's Premier 100 (CW100), or CIO Magazine IT excellence awards (CIO100 Award) to proxy
the ‘amount’ of IT capability possessed by a firm (i.e., these firms posses superior IT capability), and correlate
this IT capability superiority to some measure of competitive advantage or performance.
Using ‘IT leaders’ drawn from the annual special issue of InformationWeek 500 (IW500) over the period
1991–94 to identify firms with superior IT capability, Bharadwaj (2000) finds that these firms perform better
than their matched competitors based on profit- and cost-based performance measures. Using the same data
set but with industry performance as the benchmark, a subsequent study by Santhanam and Hartono (2003)
finds that firms with superior IT capability perform above the industry average, but the results are weaker once
they account for the firm's prior financial performance. Identifying firms with superior IT capability based on
the 1993 Computerworld's Premier 100 list, Dehning and Stratopoulos (2002) use DuPont analysis to show that
firms with superior IT capability achieve profitability as well as efficiency payoffs while high-performing nonIT capable firms have profitability but no-efficiency advantage.
More recent studies have explored and contrasted the effect of superior IT capability on firm performance
pre Y2K (eighties and nineties) versus post Y2K with mixed results. Wang and Alam (2007) use Information
Week 500 ranking, for the years 1991, 1992, and 1995–2002, as a proxy for superior IT capability and they find
that IT capability is value-relevant and provides incremental explanatory power for firm valuation beyond
traditional accounting information. Muhanna and Stoel (2010) use the top 40 to 60 firms from the annual
Information Week 500 ranking to evaluate and contrast the effect of superior IT capability on stock market
valuation, and they find strong results in the pre- (1992–1994) and post-Internet commercialization era
(1999–2006). However, the results of the Masli et al. (2011) study contradict these findings. Using CIO
Magazine IT excellence awards (CIO 100 Award) over the 1988–2007 period, Masli et al. (2011) find that while
firms with superior IT capability performed above the industry benchmark before 2000, this performance
advantage disappeared in the post-1999 period. The authors argue that this is due to a structural shift
associated with the dot com crash. Furthermore they observe that only firms achieving IT capability
superiority in multiple years outperform the industry benchmark before as well as after the dot com crash.
In general the findings of prior studies seem to reflect the perceptions and frustrations of senior
business executives with IT. While executives agree on the value adding contribution potential of IT
resources and capabilities, they also realize that there are significant barriers that prevent firms from
exploiting the full benefit of IT investments (Masli et al., 2011). Hence, an exploration of the antecedents
that lead to the continuous development of IT capability and their reinforcing effect on firm performance
would be the next logical step for studies on the effect of organizational IT capabilities on firm performance.
The reification of ‘firm’ in the definition of IT capability (a firm's ability mobilize and deploy IT based
resources in combination with other resources and capabilities) renders the firm as a purposeful entity and
detaches IT capability development from the people involved. However, according to the upper echelon theory,
organizational outcomes are predicted by managerial characteristics (Hambrick and Mason, 1984). Therefore it
seems reasonable to consider the role of senior IT executives as an antecedent in the firm's ability to develop
superior IT capability, as well as a factor that could enhance a firm's ability to exploit its superior IT capability.
3
Numerous other studies have argued or demonstrated the benefits of integration of IT with improvements in organizational
structure and business processes (complementary resources). A representative sample of these studies includes: Bhatt and Grover
(2005), Brynjolfsson and Hitt (1998, 2000, 2003), Milgrom and Roberts (1990), Powell and Dent-Micallef (1997), Ravichandran and
Lertwongsatien (2005), and Zhu, 2004. In a similar fashion Armstrong and Sambamurthy (1999) refer to the effective rutinization/
infusion of IT on the firm's value chain and business strategy as IT assimilation.
24
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
2.2. Senior IT executives and firm performance
As top management teams recognized the strategic potential of IT during the mid-1980s, the
management of IS functions naturally evolved. Data processing managers were gradually replaced with IT
executives and focus shifted from delivery of basic IT services to the exploitation of IT for competitive
advantage (Galliers and Leidner, 2003). In order to achieve more effective management of IT resources and
capabilities, during the late eighties and nineties an increasing number of firms elevated the structural
power of their senior IT executives by announcing the creation of a new position: Chief Information Officer
(Chatterjee et al., 2001). This trend is consistent with the view that assets of higher importance will be
entrusted to managers of higher rank (Raghunathan and Raghunathan, 1989).
Given that higher ranking IT executives have more power to affect organizational actions (Raghunathan
and Raghunathan, 1989), the explicit presence of more powerful senior IT executives in a firm highlights the
role and importance of IT capability to enhance strategic advantage (Applegate and Elam, 1992; Armstrong
and Sambamurthy, 1999). For example, such managers are more likely to concentrate on the following issues:
the creation of shared knowledge between business and IT managers (Chatterjee et al., 2001), the alignment
between business and IT objectives (Sambamurthy and Zmud, 1994; Bassellier et al., 2003), and the creation of
a positive shared view towards IT across the upper echelon of the organization (Chen et al., 2010).
Arguing that the creation of a new CIO position reflected changes in top management thinking and policy
regarding the role of IT and firms' approaches to IT governance, Chatterjee et al. (2001) use event study
methodology to find shareholders' reaction to these announcements. Analyzing a sample of 96
announcements of newly created CIO positions, from the period 198 to 1998, they find that markets react
positively to a firm's decision to assign higher rank to its senior IT executive. A follow up study by Khallaf and
Skantz (2007) contrasted market reaction to new versus existing CIO positions. Based on evidence regarding
461 CIO appointments during the period 1987–2002, they conclude that there is no significant difference in
the market's reaction to the two types of announcements. A more recent study by the same authors (Khallaf
and Skantz, 2011) looked at the change in long-term accounting performance relative to the corresponding
industry benchmark in the years following the appointment of a new senior IT executive. Analyzing 359
announcements from 1987 to 2002 they report that CIO appointments improve firm financial performance
but the performance improvement is largely limited to ‘firms appointing a CIO for the first time’ and first
movers.
Another study by Dehning and Stratopoulos (2003) approaches the issue of the value adding contribution
of senior IT executives from a different angle. Instead of looking at market reaction or change in financial
performance, Dehning and Stratopoulos (2003) evaluate factors that can contribute to a company's duration
of competitive advantage. They use ‘management rating,’ one of the criteria used to select the companies in the
1993 Computerworld's Premier 100 list to proxy the ‘amount’ of superior managerial IT skills possessed by a
firm, and find a positive association with the firm's ability to sustain its competitive advantage.
In general these studies have drawn on the Resource Based View (RBV) to provide the theoretical
justification for the value adding contribution of senior IT executives on the firms' stock market valuation,
financial performance, or duration of competitive advantage. They have used RBV to argue that a senior IT
executive is an asset or strategic resource that delivers value through IT resources and capabilities. Mata et al.
(1995) explained that ‘managerial IT skills’ are a likely source of competitive advantage because they are
valuable and heterogeneously distributed across competing firms. According to Dehning and Stratopoulos
(2003) superior managerial IT skills are believed to be a source of sustainable competitive advantage because
they enable companies to manage the technical and market risks of IT investment, they are developed over
time through the accumulation of experience, they are tacit and causally ambiguous, and they are the result of
socially complex processes.
Two gaps were illuminated through a review of literature on the role of senior IT executives on firm
performances. First, with the exception of Dehning and Stratopoulos (2003), the main emphasis has been on
new appointments of senior IT executives. As Chatterjee et al. (2001) have pointed out, relatively little is
known about the value adding contribution of incumbent senior IT executives. Second, the path through
which senior IT executives add value to their firms has been treated as a black box. Empirical studies that we
have reviewed on the role of IT executives on firm performance either ignore the role that senior IT
executives have played in developing a firm's IT capability or they treat their firm's IT capability as a control
variable.
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
25
2.3. Senior IT executives and IT capability
As data processing managers were gradually replaced with IT executives and their focus shifted from
day-to-day operations and IT spending to IT strategic approaches, the scope and responsibilities of senior IT
executives expanded, leading to promotion and increased power (Chatterjee et al., 2001; Galliers and
Leidner, 2003). This combination of power and expanded roles suggests a greater influence on the
leadership team (Mintzberg, 1983). Several studies looked at the role and relationships of IT executives
within the firm and their efforts to infuse IT on the firm's value chain and business strategy (e.g.,
Raghunathan and Raghunathan, 1989; Grover et al., 1993; Armstrong and Sambamurthy, 1999).
Analyzing managerial roles of IT managers, Grover et al. (1993) argue that the hierarchical level of
management influences the roles of senior IT executives. In general, higher-level senior IT executives tend to
focus more on such roles as: (1) ‘liaison’ between the organization and suppliers or customer; (2) ‘monitor’ of
the external environment for technical changes, competition and identification of new ideas from sources
outside of the organization; and (3) ‘entrepreneur’ who identifies business needs and develops solutions that
change business situations. As an entrepreneur the senior IT executive ensures that rapidly evolving technical
opportunities are understood, planned, implemented, and strategically exploited in the organization. Through
entrepreneurship they play a greater role in influencing overall strategy and planning in the organization and
they translate these organizational strategies and plans into definite IS actions (Grover et al., 1993).
Raghunathan and Raghunathan (1989) explore whether the organizational rank of the senior IT executive
influences the way in which IS tasks are undertaken in organizations. They conclude that rank has significant
linkages to the organizational role of IS and to the emphasis placed on various dimensions of IS planning.
Building on organizational theory Raghunathan and Raghunathan (1989) argue that hierarchical organization
structure confers formal authority and constitutes a base of power to affect and influence organizational
actions. To this point Karimi et al. (1996) have shown that senior IT executives holding the title CIO actively
spend more time interacting with stakeholders outside the firm and on more general corporate strategic
planning. Building on prior literature that has proposed, among other things, that with lower rank of
responsible IT executives the likelihood of a successful IT effort declines, Raghunathan and Raghunathan
(1989) conclude that the rank of senior IT executives in the organizational hierarchy could be an important
factor affecting the success of IS and its use as a strategic, competitive weapon.
In addition to this, research has shown that participating in the firm's top management team and interacting
with senior business executives enhances the business knowledge of senior IT executives. This chain-effect is very
important as it affects a firm's ability to successfully integrate IT with improvements in organizational structure
and business processes (firm's value chain) and business strategies (Armstrong and Sambamurthy, 1999).
Raising the profile and rank of senior IT executives in the executive team provides economic justification to the
executive team for IT projects. This in turn supports the argument for increased attention and allocation of
resources for strategic IT projects. In contrast, if the senior IT executive is not part of the executive team, IT
prioritization is less likely to have the executive team's support and commitment (Applegate and Elam, 1992;
Armstrong and Sambamurthy, 1999). In summary this stream of research provides the theoretical foundation for
the following position: The structural power of senior IT executives (i.e., their power within the firm due to their
formal position) is an antecedent to their firm's ability to develop superior organizational IT capability.
2.4. Synthesis
Examining these three streams of research sequentially and in isolation from each other serves two purposes.
First, it reveals potential gaps in each stream. Second, it clarifies the ease with which these gaps can be addressed
by integrating these three streams. Fig. 1 summarizes the main points of our review and the proposed synthesis.
More specifically, Fig. 1 — Panel A shows that ‘antecedents of IT Capability’ are missing from the stream of
research ‘IT Capability and firm performance.’ Panel B shows that the impact of IT executives on firm performance
is going through a ‘black box.’ Panel C shows that both of the above gaps can be filled by incorporating the stream
of research on the senior IT executives and IT capabilities. The resulting synthesis is shown in Panel D.
The proposed synthesis of these three streams of research is well supported by anecdotal evidence. For
example, Intel senior IT executives played a critical role in the firm's ability to align IT with business goals,
eventually developing superior IT capability. In the late-nineties, Intel's IT organization was known for its
inefficiency (high cost) and ineffectiveness (failing IT projects). In an attempt to change this, D. Busch was named
26
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
A: IT Capability and firm performance
B: IT executives and firm performance
C: Senior IT executives and IT Capabilities
D: Synthesis
Fig. 1. Theoretical foundation and synthesis.
VP of IT in 1999. He created a vision for the IT organization to become a value-adding contributor to Intel's internal
and external success, and a mission to enhance the company's success with outstanding strategic leadership and
IT services (Westerman and Curley, 2008). He planned to achieve this by managing IT like a business. Busch set a
goal to create $100 million in net business value by 2002, without incurring any additional costs. In his role, Busch
not only achieved this goal, but the IT organization he redesigned won the coveted Intel Quality Award (Curley,
2006). In 2002, D. Busch was appointed as the first VP and co-CIO in the company's history, with a mandate to
help the company in its efforts to become a total e-corporation (Rosencrance, 2002).
Clearly, the case of Intel as well as the cases of several other firms known for their superior
organizational IT capability indicates the pivotal role of powerful senior IT executives in orchestrating their
company's success. This study argues that senior IT executives contribute to their firm's competitive
position through the development of such superior IT capability. Thus we postulate that:
H1. The structural power of senior IT executives will be positively associated with the development of
superior organizational IT capability.
H2. Impact of senior IT executives on firm's competitive position is channelled through superior IT capability.
3. Research methodology
3.1. Sample selection
The data set for testing these hypotheses was developed by integrating data from several sources.
Consistent with prior studies (Bharadwaj, 2000; Santhanam and Hartono, 2003; Wang and Alam, 2007;
Muhanna and Stoel, 2010; Stratopoulos and Lim, 2010; Lim et al., forthcoming) a firm's appearance in the
annual special issue of the InformationWeek 500 is considered a proxy for IT capability superiority. We use
Lexis-Nexis and a variety of online sources in order to search for official title and number of titles to proxy for
structural power of senior IT executives. Finally, we use Compustat for financial data. To account for the
proposed structural shift associated with the dot com crash (Masli et al., 2011), hypotheses were tested based
on data from the period 1997 to 2004.
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
27
Given that for RBV, the de facto theoretical foundation of this study, understanding how a firm
compares to its competitor is critical; and in order to make the results more interesting to professionals,
Hoover's Handbook of American Business was used to identify and annually match each IT capable firm with
its top three competitors. Hoover's compiles the profiles of major US firms each year (i.e., firms with sales
of $2 billion or more). Each profile includes a list of the firm's competitors, and this list is updated each year.
On average there are approximately 20 competitors listed for each firm. Hoover's was chosen for the
matching process for two reasons. First, Hoover's was involved in the preparation of the IW500 report as a
provider of financial data (Weston, 1998). Therefore it is likely that the sample of firms used by IW overlaps
with the population of firms tracked by Hoover's. Second, Hoover's is a source of firm-specific information
geared towards the professional community, and hence is more likely to reflect a practitioner-oriented list
of close competitors of IW500 firms. The final sample for firms with superior IT capability and their
competitors (non-IT capable firms) contains 960 publicly traded companies. The data set covers the same
spectrum of industries and maintains a similar composition (622 IT leader firms and 338 non-IT leader
firms) as in prior studies (Stratopoulos and Lim, 2010; Chen et al., forthcoming; Lim et al., forthcoming).
3.2. Measurement of senior IT executives' structural power
Titles that are stated in annual reports clearly relate to authority and responsibility. In fact, a higher
number of official titles brings greater power and expanded responsibility (Finkelstein, 1992), as well as
more practical benefits such as higher compensation (Marlin, 2004). 4 Accordingly, the official title and the
number of secondary titles of senior IT executives indicate a wider spectrum of expertise, greater influence
outside the IT organization, as well as more structural power. For each IT capable and non-IT capable firm in
the sample, we searched proxy statements, such as Form 10-Ks and DEF-14A from the U.S. Securities and
Exchange Commission (SEC), to identify each senior IT executive's name, official title, and number of titles.
Due to a lack of publicly-available data regarding senior IT executives, the combination of data sources was
expanded to include proxy statements, Lexis-Nexis, and several online sources. In order to verify and
enhance the completeness and accuracy of our data, a subsequent manual review of each senior IT
manger's bio data was conducted via Lexis-Nexis as well as seven online search engines (LinkedIn, Company
Biographies, Factiva, Forbes People Tracker, Marketwatch.com, Zoominfo People, and BusinessWeek People).
This search produced 317 different official titles for senior IT executives. 5 Consistent with the extant
literature (Finkelstein, 1992), we use the official title as well as the number of titles as a proxy for the
structural power of senior IT executives. In other words we assume that a senior IT executive with the
formal title of CIO or CIO with additional official titles (e.g., CIO and Executive Vice President) possesses
greater power than senior IT executives without the CIO title (e.g., Managing director). In the remainder of
this study the group of high power senior IT executives are referred to as CIO and the rest as Non-CIO.
3.3. Market performance measure: Tobin's q
Firm performance in this study is measured by Tobin's q. Tobin's q is defined as the market value of the firm
divided by the replacement cost of its assets. As a performance measure it is forward-looking, risk-adjusted, and
less vulnerable to changes in accounting practices. It has also been used as a proxy for the firm's intangible value.
More specifically, q values that are greater than one have been interpreted as an indicator of an unmeasured
source of value attributed to the firm's intangible value (Bharadwaj et al., 1999). Given that typical IT benefits are
related to such intangibles as improved customer satisfaction, and improved service quality and agility, Tobin's q
has been used as a performance proxy for examining how firm performance relates to IT investment (Bharadwaj
et al., 1999; Chari et al., 2008; Ravichandran et al., 2009), IT synergies (Tanriverdi, 2006), and superior IT
capability (Masli et al., 2011). Overall, Tobin's q is the kind of performance measure more likely to capture and
4
The result of a survey by the IT research firm Alinean shows that senior IT managers with multiple titles are among the highest
paid IT managers and receive compensation packages of $1 million or more (Marlin, 2004).
5
A sample of these titles included Chief Information Officer (CIO), Chief Technology Officer (CTO), Sr. V.P. of Information Systems
(IS)/Information Technology (IT)/Computing Information Systems (CIS)/Management Information Systems (MIS); V.P. of IS/IT/CIS/
MIS; Dir. of IS/IT/CIS/MIS; Exec. Dir. of IS/IT/CIS/MIS; Managing Director of IS/IT/CIS/MIS; Pres. Dir. of IS/IT/CIS/MIS.
28
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
reflect co-presence of such intangibles as good management skills (Adams et al., 2010) and superior IT capability
(Masli et al., 2011).
Consistent with the previous AIS literature, Tobin's q is computed as a ratio of market value [(fiscal yearend market value of equity)+ (liquidating value of the firms' outstanding preferred stock) + (current
liabilities)− (current assets) + (book value of inventories)+ (long-term debt)] to book value of total assets
(Chung and Pruitt, 1994).
3.4. Control variables
Two sets of control variables were selected to evaluate the role of IT executives on firm performance.
These control variables were based on a review of prior studies on IT investment and firm performance.
The first set controls for the effects of known factors that might be associated with IT capability, such as
size, prior performance, and reputation. Firm size (Lev, 1983; Cheng, 2005), natural log of total assets
(SIZE), represents a firm's ability to sustain a competitive advantage from its market power or positional
advantages, as well as superior financial and human resources endowments (Brynjolfsson and Hitt, 1996;
Roberts and Dowling, 2002; Morrow et al., 2007). The need to control for the past financial performance of
a firm (ROA) is twofold: (1) a firm's strategic IT capability choice could be a function of its past performance (Santhanam and Hartono, 2003), and (2) it is likely that the selection of firms with superior IT
capability by industry experts might be influenced by the firm's past performance (Bharadwaj, 2000). A
firm's reputation might be another factor that may influence the selection of firms with superior IT
capability by industry experts. Market-to-book-value (MV) has been suggested as a proxy for reputation
because it captures tangible and intangible assets (Roberts and Dowling, 2002), as well as the market's
expectations of future economic returns (Mueller, 1990).
The second set controls for the effects of known factors that might be associated with firm performance,
such as R&D, advertising and capital expenditures. R&D intensity is viewed as a proxy for a firm's intangible
capital, which may enhance technology pursuit and firm performance (Scott and Pascoe, 1987; Miller, 2006).
Advertising (AD) and capital expenditure (CAP) are incorporated as well, as they were identified by Hitt and
Brynjolfsson (1996) and Masli et al. (2011) as having a significant impact on firm value. Finally, we control for
the fixed industry and year effects. See Appendix A for a list of variables and their description.
3.5. Econometric model
This study argues that IT capable firms perform better than their competitors and executives with more
power are more likely to contribute to the firm's ability to achieve superior IT capability. Ideally this
argument would be tested by contrasting the performance of the same firm given two different scenarios.
In the first scenario a powerful IT executive in a firm helps develop superior IT capability, whereas in the
second scenario the IT function of the same firm has been managed by a less powerful IT executive and the
firm had chosen not to develop the IT capability. Since we cannot make this type of comparison with
archival data, the second best approach is to contrast performance of firms with superior IT capability
versus the performance of firms with no IT capability. However, this approach is likely to lead to biased
estimates of the effect of strategic choice (IT capability versus non IT capability) on firm performance
unless we allow for endogeneity of strategic choices.
The underlying assumption behind a firm's strategic choices (e.g., whether or not to empower IT executives,
develop superior IT capability) is that such choices are not random, but compatible with firm attributes and
industry conditions. That is strategic choices are endogenous and self-selected (Hamilton and Nickerson, 2003;
Shaver, 1998). Ignoring the self-selection issue is equivalent to assuming that firm performance is independent of
the firm's choice regarding IT executives and IT capability. But such an assumption is unrealistic and would
introduce bias in the estimated coefficients. Furthermore, there is no way of predicting ex ante the direction and
size of bias (Hamilton and Nickerson 2003). In other words, we don't know whether firms that chose to assign
higher power to their senior IT executives in the hope that this will lead to superior IT capability and better
performance could have performed better or worse had they chosen not to make these strategic choices. Failure
to control for endogeneity may therefore lead to faulty conclusions about theoretical propositions.
This study attempts to alleviate endogeneity bias by employing a two-stage estimation with correlated
error terms. Formally, we specify a system of two equations (Eq. (1) and (2)) for testing the following chain
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
29
hypothesis: IT executives affect a firm's ability to achieve superior IT capability (Eq. (1)), and this in turn
drives their firm's performance (Eq. (2)).
yit = 1
ρyit−1 + δ1 CIOit−1 + δ2 SIZEit−1 + δ3 ROAit−1 + δ4 MVit−1 +
+ ηi + ϕt + uit > 0
qit = μqit−1 + β1 R&Dit + β2 ADit + β3 CAPit + γŷit + αi + ψt + εit
ð1Þ
ð2Þ
where t = 1,..,T, i = 1,…,N, and
1 for firms with superior IT capability, 0 otherwise
1 for Senior IT executive with the formal title of CIO or CIO with additional official titles,
0 otherwise
SIZEit − 1 natural log of total assets
ROAit − 1 return on assets
MVit − 1 market-to-book value
Tobin's q,
qit
research and development
R & Dit
advertising expenditures
ADit
capital expenditures
CAPit
ŷit
predicted value from the first-stage regression
ηi and αi fixed effects in Eqs. (1) and (2) respectively
ϕt and ψt time effects in Eqs. (1) and (2) respectively
uit and εit other time-variant unobserved variables assumed to be uncorrelated with vector of observable
firm characteristics in Eqs. (1) and (2) respectively.
yit
CIOit − 1
1[.] in Eq. (1) is an indicator function based on Lim et al. (forthcoming) and models firm i's ability to
achieve superior IT capability (yit). The expression in square brackets represents the ability of firm i to
achieve superior IT capability. If the ability is sufficiently high, firm i has superior IT capability, in which
case yit takes the value of 1. The parameter ρ and the parameter vector δ′ represent respectively the effects
of past IT capability superiority and firm characteristics, such as role of IT executives, on current ability to
achieve IT capability superiority. Eq. (2) models current firm performance (qit). The parameter γ and the
parameter vector β′ represent the effects of past firm's performance and firm characteristics on its current
performance respectively. Ceteris paribus, the impact of a senior IT executive's contribution on the firm's
competitive position is channeled through his/her success in helping the company develop superior IT
ϑqit
ϑqit ϑ ŷit
capability ϑCIO
= ϑy
. The dynamic binary choice model of stage one is estimated using the
ˆ it ϑCIOit−1
it−1
random-effect (RE) approach proposed by Wooldridge (2005). The second stage is estimated using OLS
with correction suggested by Murphy and Topel (2002). See Appendix B for a more detailed econometric
discussion of the estimation methodology.
The predicted value of IT capability (γ) in Eq. (2) will not have the desired properties, unless we account for
the fact that the residual in Eqs. (1) and (2) are correlated, Cov(uit, εit) ≠ 0 (Shaver 1998), and omitted
variables that affect both a firm's strategic choices and performance (Hamilton and Nickerson, 2003). To
account for the former, i.e., to allow for correlation between the processes governing the ability of the firm to
develop superior IT capability and the ensuing improved firm performance, we specify the two-stage model as
a system of two equations with correlated error terms Cov(uit, εit) ≠ 0. To account for the latter, fixed firm
effects are included in the specification to control for unobservable firm characteristics, which affect the firm's
strategic choices such as power of IT executive and IT capability. Firm specific fixed effects (ηi and αi
respectively) are proxied by firm dummies, and time effects (ϕt and ψt respectively) are proxied by year
dummies. Finally, to mitigate the possibility of endogeneity bias in the estimation of (1) due to the
contemporaneous inclusion of IT capability and firm-specific time-varying exogenous variables, we add the
time-varying exogenous variables (Sizeit − 1, ROAit − 1, and MVit − 1) with a one-period lag.
30
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
4. Empirical results
4.1. Descriptive statistics
During the sample time period 1997–2004 the majority of senior IT executives, 72% to 82%, were NonCIO. The average percentage of high power senior IT executives (CIO) was in the range of 18% to 28%. A
similar pattern was observed when we looked at the breakdown between CIO and Non-CIO within specific
industry sectors. On average, in the machinery, electronic, and other equipment industry the percentage of
CIO has been around 23%. In the service, transportation/utility, and financial industries the percentages
were 21%, 20%, and 19% respectively. In manufacturing the percentage of CIO was around 18%. Overall there
seems to be a gradual increase in the percentage of CIO.
Descriptive statistics (Panel A of Table 1) seem to indicate that the power of senior IT executives does
indeed matter for the firm's ability to achieve superior IT capability. More specifically, there is more than a
61% probability that firms with high power senior IT executives (CIO) this year will develop superior IT
capability next year. This result is about 40% higher than the probability that a firm with a less empowered
senior IT executive (Non-CIO) this year will develop superior IT capability next year.
Previous studies have shown that the probability that a firm will develop superior IT capability in the next year
is strongly dependent upon its current IT capability status (Lim et al., forthcoming). To account for this persistence
in IT capability superiority, we investigate the probability that the title of the senior IT executive in the current
year will affect the firm's ability to achieve superior IT capability in the following year. Concentrating on non-IT
capable firms and those that had a CIO leading their IT organization in the current year, the probability of these
firms developing superior IT capability in the following year is over 30% (untabulated results). The corresponding
probability for firms with Non-CIO was less than 10%. This is consistent with our prediction that the higher the
power of senior IT executives, the higher the likelihood that the firm will achieve superior IT capability.
Table 1
Descriptive statistics.
Panel A
Year t − 1 to t
Pr(IT Capabilityit|NonCIOit - 1)
1997–1998
1998–1999
1999–2000
2000–2001
2001–2002
2002–2003
2003–2004
Average
200/790
144/744
138/727
108/693
119/693
142/716
105/686
Pr(IT Capabilityit|CIOit - 1)
25.32%
19.35%
18.98%
15.58%
17.17%
19.83%
15.31%
18.79%
106/170
137/216
149/233
168/267
161/267
135/244
157/274
62.35%
63.43%
63.95%
62.92%
60.30%
55.33%
57.30%
60.80%
Panel B
IT capable firmst
Year t − 1 to t
1997–1998
1998–1999
1999–2000
2000–2001
2001–2002
2002–2003
2003–2004
Titlet − 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
CIOit - 1
NonCIOit - 1
Non-IT capable firmst
Tobin's q
Tobin's q
Tobin's q
Tobin's q
(Mean)t
(StD.)t
(Mean)t
(StD.)t
1.591
1.789
1.947
1.778
1.521
1 568
1.421
1.384
1.299
1.044
1.410
0.245
1.241
1.360
1.193
2.231
2.906
2.479
1.275
1.160
1.164
1.164
1.027
0.834
1.312
0.998
1.017
1.197
1.100
1.966
2.205
1.905
1.703
1.411
1.514
1.289
0.965
1.118
1.179
1.322
1.197
1.308
1.510
5.460
5.116
4.388
2.917
1.697
2.167
1.216
0.813
0.951
0.811
1.181
0.884
1.021
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
31
Table 1 — Panel B provides descriptive statistics for the three variables of interest: power of senior IT
executives, IT capability, and Tobin's q. Results are reported for two consecutive years at a time. For
example, the first set shows the mean and standard deviation of Tobin's q for the group of firms that had a
CIO in 1997 and had a superior IT capability in 1998 were 1.591 and 1.193 respectively. The average and
standard deviation of the Tobin's q for the group of firms that had a Non-CIO in 1997 and had no superior IT
capability (non-IT capability) in 1998 were 1.966 and 5.460 respectively.
Comparative results for the remaining years in the sample do not seem to indicate that the group of
firms with CIO in t − 1 and superior IT capability in year t are doing better than their competitors. This
casual observation is further supported by the untabulated results of ANOVA analysis. This lack of support
for the second hypothesis is expected and validates the aforementioned position that a chain effect must
be considered in order to observe the impact of the structural power of senior IT executives on their
firm's competitive position. Recall that the impact of IT executives on performance is captured in
ϑqit
ϑqit ϑˆyit
=
. While the table provides a similar interaction effect this is the simple product of
ϑCIOit−1
ϑ̂yit ϑCIOit−1
ϑqit
ϑqit
the two isolated effects
and
. In other words, descriptive statistics suffer from the same
ϑyit
ϑCIOit−1
weakness that was observed in the review of prior literature. Furthermore, these results do not control for
confounding factors known to affect Tobin's q, such as R&D intensity (R&D), advertising (AD), and capital
expenditures (CAP).
Table 2
Two-stage regression analysis (IT capable vs. Non-IT capable firms).
Unrestricted model
Restricted model
Coefficient (Std. error)
Coefficient (Std. error)
A. First-stage dynamic logit regression with Wooldridge-type correlated RE specification
yit − 1
0.309 (0.111)***
CIOit − 1
0.142 (0.052)***
SIZEit − 1
0.113 (0.052)**
ROAit − 1
0.044 (0.032)
MVit − 1
0.053 (0.044)
yi0
0.382 (0.132)***
CIOi;–1
0.098 (0.037)***
SIZEi;–1
0.072 (0.021)***
ROAi;–1
0.023 (0.015)*
MV i;–1
0.026 (0.015)*
Industry dummies
Included
Year dummies
Included
ση
0.457 (0.147)***
Ln L
− 1025.1
Wald1
0.117
Wald2
0.065**
Wald3
0.000***
#Obs.
6720
0.386 (0.126)***
0.115 (0.035)***
0.081 (0.027)***
0.025(0.015)*
0.031 (0.016)*
Included
Included
0.462 (0.141)***
− 1007.3
0.122
0.069**
0.000***
6720
B. Second-stage linear regression
qit − 1
R&Dit
ADit
CAPit
ŷit
Adjusted R2
#obs.
0.210
0.152
0.117
0.130
0.297
0.85
6720
0.208
0.152
0.116
0.129
0.294
0.85
6720
(0.077)**
(0.059)**
(0.049)**
(0.065)**
(0.102)***
0.312 (0.095)***
0.191 (0.073)***
0.127 (0.050)**
(0.078)**
(0.058)**
(0.049)**
(0.063)**
(0.103)***
Note. Wald1 records the p-value of the Wald test for the joint exclusion of ROAit − 1 and MVit − 1. Wald2 records the p-value of the Wald
test for the joint exclusion of year effects. Wald3 records the p-value of the Wald test for the joint exclusion of industry effects. The
term xi;−1 denotes the firm average of lagged x variables, such as the firm average of lagged values of the explanatory variables. The
standard errors of the OLS parameter estimates in the second stage are corrected for a generated regressor using an extended Murphy
and Topel (2002) estimator. ***, **, and * denote significance at the 1, 5, or 10% level, respectively.
32
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
4.2. Econometric results — IT executives, superior IT capability, and Tobin's q
Empirical results reported in Table 2 consist of two parts. Panel A shows the results of RE estimation as
a fully specified model. Panel B presents the results of the second stage linear regression. The RE
estimations of superior IT capability Eq. (1) are presented in the restricted and the unrestricted models.
Both models allow the IT capability to be dependent on lagged IT capability (yit − 1), the initial values of
IT capability status (yi0), firm-specific time-varying exogenous variables (CIOit − 1, Sizeit− 1, ROAit − 1, and
MVit − 1) and unobserved permanent firm heterogeneity (ση). The unrestricted model shows that one of
the time-varying exogenous covariates (Sizeit − 1) appears to be statistically significant. Based on the
insignificant results from the Wald test for joint exclusion of ROAit − 1 and MVit − 1, they were excluded
from the restricted version of the model. 6
Consistent with the first hypothesis (H1), the results of restricted RE estimates indicate that power of
senior IT executives is positively and statistically significantly associated with the firm's ability to achieve
IT capability superiority in the following year. The estimated parameter for CIO (δ1) is 0.191 it is
statistically significant (p-value < 1%). This indicates that the probability that a firm with a CIO will achieve
superior IT capability next year is approximately 19% higher than that of firm with a Non-CIO. In addition
to this, the parameter related to the path dependence of the IT capability (ρ) is 0.312 and highly significant
(p-value < 1%). This means that the probability that an IT capable firm will maintain its IT capability
superiority next year is 31.2% higher than that of a non-IT capable firm. 7 Lastly, there is some evidence
of time effects and very strong evidence of industry effects (Wald test in the restricted model; are
p-value = 0.069 and p-value = 0.000 respectively). 8
The results of the second stage linear regression strongly support the second hypothesis (H2). The
coefficient for the predicted IT capability status (γ) is 0.297 and it is statistically significant (p-value < 1%).
The other control variables are moderately significant and the explanatory power of the model is
relatively high (adjusted R 2 is 85%). As a robustness check we estimated Eq. (2) as a separate model and
it produced insignificant results (results have not been tabulated). Contrasting the robust evidence from
the second stage linear regression with the lack of evidence in descriptive statistics as well as from
the estimation of Eq. (2) as a separate model further validated our position. The impact of a senior IT
executive's contribution on the firm's competitive position is channeled through his/her success in
helping the company develop superior IT capability.
4.3. Econometric results – IT executives, enduring IT capability, and Tobin's q
The main premise of RBV and IT business value research is that only IT resources and capabilities that
are valuable and durably heterogeneous are rent yielding (Barney, 1991; Bhatt and Grover, 2005;
Ravichandran and Lertwongsatien, 2005). Durable heterogeneity of superior IT capability, in the context
of RBV, reflects a firm's ability to resist competitors' attempts to imitate or improve on its IT capability
(Barney, 1991). Several recent studies have shown that enduring superior IT capability is more value
adding than a non-enduring IT capability (Stratopoulos and Lim, 2010; Chen et al., forthcoming, and Lim
et al., forthcoming). As an additional validation for the hypotheses we consider the link between the
power of IT executives and a firm's ability to develop an enduring IT capability. Achieving and sustaining
such IT capability superiority will be possible for firms that can leverage feedback cycles of organizational
6
If the Wald test is significant, we can conclude that the parameters associated with the variables are not zero, so that the
variables should be included in the model.
7
According to Lim et al. (forthcoming) a positive and statistically significant estimate of ρ would identify the presence of
persistence in the firm's ability to achieve superior IT capability. That is, a firm that had superior IT capability in t − 1 is more likely
to achieve IT superiority in t. Therefore, achieving superior IT capability is a path dependent process. This has important strategic
implications, since path dependence means that competitors would find it difficult to acquire or develop similar capability.
According to RBV, if IT capability is path dependent then it is rent yielding.
8
Furthermore, as indicated by the highly significant estimates of ση (0.462), we note that unobserved permanent heterogeneity
accounts for a significant portion of the overall variance. Please note that a high level of unobserved permanent heterogeneity does
not imply a problem with model fit.
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
33
learning in a way that enables them to build stronger or reconfigured IT capabilities (Bhatt and Grover,
2005). Senior IT executives interacting with the firm's top management team are more likely to have the
business and IT knowledge needed to nurture such organizational learning. Therefore we expect that the
power of IT executives will be associated with a firm's ability to achieve enduring superior IT capability,
and firms with enduring IT capability will outperform firms with non-enduring IT capability.
Adopting the approach used in prior studies (Stratopoulos and Lim, 2010; Chen et al., forthcoming;
and Lim et al., forthcoming) we classify firms into three groups: Firms with enduring IT capability, nonenduring IT capability, and non-IT capable. This classification is based on the cross-sectional evolution of a
firm's recognition in IW500 over four-year rolling windows (i.e., 1997–00, 1998–01, 1999–02, 2000–03,
and 2001–04). For each window we classify a firm as a having an enduring IT capability if it has been
recognized in IW500 all years within a four-year rolling window. A firm is classified as having a nonenduring IT capability if it has appeared less than four times in a four-year rolling window. Non-IT capable
firms are those that have not been recognized in any of the 4 years.
The two-stage regression analysis of firms with enduring IT capability versus firms with nonenduring IT capability is reported in Table 3 and clearly validates the position of this study. Results
show that the probability that a firm with a high ranking senior IT executive (CIOit − 1) will achieve an
enduring IT capability superiority is 17% higher than that of a firm with a Non-CIOit − 1. Additionally, the
contribution of IT executives on Tobin's q through superior and enduring IT capability is statistically
Table 3
Two-stage regression analysis (firms with enduring vs. non-enduring IT capability).
Unrestricted model
Restricted model
Coefficient (Std. error)
Coefficient (Std. error)
A. First-stage dynamic logit regression with Wooldridge-type correlated RE specification
yit − 1
0.382 (0.121)***
CIOit − 1
0.166 (0.057)***
SIZEit − 1
0.125 (0.055)**
ROAit − 1
0.071 (0.049)
MVit − 1
0.082 (0.050)
yi0
0.414 (0.155)***
CIOi;–1
0.103 (0.039)***
SIZEi;–1
0.077 (0.029)***
ROAi;–1
0.038 (0.016)**
MV i;–1
0.029 (0.012)**
Industry dummies
Included
Year dummies
Included
ση
0.379 (0.117)***
Ln L
− 998.1
Wald1
0.123
Wald2
0.055**
Wald3
0.000***
#Obs.
474
0.415 (0.153)***
0.106 (0.040)***
0.079 (0.30)***
0.041 (0.017)**
0.035 (0.015)**
Included
Included
0.383 (0.119)***
− 1001.3
0.129
0.063**
0.000***
474
B. Second-stage linear regression
qit − 1
R&Dit
ADit
CAPit
^
yit
Adjusted R2
#obs.
0.228
0.197
0.104
0.110
0.355
0.82
474
0.226
0.194
0.103
0.111
0.353
0.82
474
(0.076)***
(0.079)**
(0.040)**
(0.046)**
(0.104)***
0.390 (0.119)***
0.171 (0.056)***
0.132 (0.052)**
(0.075)***
(0.080)**
(0.040)**
(0.048)**
(0.105)***
Note. Wald1 records the p-value of the Wald test for the joint exclusion of ROAit − 1 and MVit − 1. Wald2 records the p-value of the Wald
test for the joint exclusion of year effects. Wald3 records the p-value of the Wald test for the joint exclusion of industry effects. The
term xi;−1 denotes the firm average of lagged x variables, such as the firm average of lagged values of the explanatory variables. The
standard errors of the OLS parameter estimates in the second stage are corrected for a generated regressor using an extended Murphy
and Topel (2002) estimator. ***, **, and * denote significance at the 1, 5, or 10% level, respectively.
34
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
significant. That is, the coefficient for the predicted superior and enduring IT capability status (γ) is 0.355
(p-value < 1%).
Perhaps an even stronger validation to the importance of using a chain effect approach for measuring
the impact of IT executives on firm performance comes from the comparison of firms with non-enduring
IT capability and non-IT capable firms on Table 4. Recall that according to prior evidence (Stratopoulos
and Lim, 2010; Lim et al., forthcoming) a non-enduring IT capability superiority is not a source of
competitive advantage. Therefore, a senior IT executive who delivers a non-enduring IT capability is not
making a value adding contribution to his/her company. Consistent with this, the results show that while
the power of senior IT executives has a moderate effect on a firm's ability to achieve non-enduring IT
capability, the effect of non-enduring IT capability on firm performance is not statistically significant. The
probability that a firm with a CIOit − 1 will achieve a non-enduring IT capability superiority is 9% higher
than that of firm with a Non-CIOit − 1. The coefficient for the predicted non-enduring IT capability status
(γ) is 0.068 and it is not statistically significant.
4.4. Additional tests
As robustness check we estimate an endogenous self-selection model with fixed effects by using
an extension to Heckman's (1979) correction to control for the self-selection bias. The predicted values from
Table 4
Two-stage regression analysis (firms with non-enduring vs. non-it capability).
Unrestricted model
Restricted model
Coefficient (Std. error)
Coefficient (Std. error)
A. First-stage dynamic logit regression with Wooldridge-type correlated RE specification
yit − 1
0.226 (0.114)**
CIOit − 1
0.087 (0.044)**
SIZEit − 1
0.101 (0.054)*
ROAit − 1
0.028 (0.017)
MVit − 1
0.037 (0.023)
yi0
0.314 (0.134)**
CIOi;–1
0.051 (0.024)**
SIZEi;–1
0.048 (0.024)**
ROAi;–1
0.015 (0.008)*
MV i;–1
0.020 (0.012)*
Industry Dummies
Included
Year Dummies
Included
ση
0.623 (0.229)***
Ln L
− 1184.7
Wald1
0.106
Wald2
0.058**
Wald3
0.000***
#Obs.
1640
0.322 (0.138)**
0.063 (0.027)**
0.054 (0.026)**
0.016 (0.008)*
0.022 (0.013)*
Included
Included
0.625 (0.228)***
− 1179.2
0.113
0.061**
0.000***
1640
B. Second-stage linear regression
qit − 1
R&Dit
ADit
CAPit
ŷit
Adjusted R2
#obs.
0.141
0.092
0.155
0.208
0.068
0.62
1640
0.139
0.088
0.152
0.215
0.077
0.61
1640
(0.072)*
(0.042)**
(0.092)
(0.107)*
(0.047)
0.231 (0.117)**
0.090 (0.046)**
0.113 (0.062)*
(0.074)*
(0.046)**
(0.095)
(0.109)*
(0.041)
Note. Wald1 records the p-value of the Wald test for the joint exclusion of ROAit − 1 and MVit − 1. Wald2 records the p-value of the Wald
test for the joint exclusion of year effects. Wald3 records the p-value of the Wald test for the joint exclusion of industry effects. The
term xi; −1 denotes the firm average of lagged x variables, such as the firm average of lagged values of the explanatory variables. The
standard errors of the OLS parameter estimates in the second stage are corrected for a generated regressor using an extended Murphy
and Topel (2002) estimator. ***, **, and * denote significance at the 1, 5, or 10% level, respectively.
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
35
Table 5
Results for Endogenous Self Selections model with fixed effects using Heckman's (1979) extension.
B. Second-stage linear regression
Panel A: IT capable vs. non-IT capable firms
qit − 1
R&Dit
ADit
CAPit
λ̂it
Adjusted R2
#obs.
Panel B: firms with enduring vs. non-enduring
qit − 1
R&Dit
ADit
CAPit
λ̂it
Adjusted R2
#obs.
Unrestricted model
Restricted model
Coefficient (Std. error)
Coefficient (Std. error)
0.201
0.147
0.108
0.114
0.263
0.88
6720
0.213
0.152
0.117
0.130
0.297
0.90
6720
(0.107)**
(0.065)**
(0.059)**
(0.060)**
(0.137)**
0.215
0.181
0.098
0.103
0.318
0.88
474
(0.102)**
(0.090)**
(0.44)**
(0.051)**
(0.144)**
0.131
0.083
0.149
0.202
0.047
0.66
1640
(0.072)*
(0.042)**
(0.091)
(0.120)*
(0.029)
(0.101)**
(0.066)**
(0.049)**
(0.058)**
(0.132)**
IT capability
0.213 (0.103)**
0.176 (0.088)**
0.095 (0.043)**
0.098 (0.049)**
0.311(0.132)**
0.86
474
Panel C: firms with non-enduring vs. non-IT capability
qit − 1
0.122 (0.073)*
R&Dit
0.071 (0.042)*
ADit
0.144 (0.096)
CAPit
0.209 (0.124)*
λ̂it
0.056 (0.035)
Adjusted R2
0.64
#obs.
1640
Note: the standard errors of the OLS parameter estimates in the second stage are corrected for a generated regressor using an
extended Murphy and Topel (2002) estimator. ***, **, and * denote significance at the 1, 5, or 10% level, respectively.
this first-stage regression ( ŷit ) are used to compute the inverse of Mill's ratio (IMR) as follows:
λ ŷit = f ŷit = F ŷit , if yit = 1 and −λ ŷit = −f −ŷit = F −ŷit , if yit = 0, where f ŷit and F ŷit represent
the probability density function and cumulative density function, respectively. Including the IMR in the
second-stage regression provides a control for potential selectivity bias in our examination of the effects
of firm's choices regarding IT executives and IT capability in Eq. (2′).
qit = μqit−1 + β1 R&Dit + β2 ADit + β3 CAPit + η λ̂it + αi + ψt + εit
ð2′Þ
where λ̂it = λ ŷit is the IMR. The sign of the estimate of η in Eq. (2′) is determined by the sign of the
correlation between the error terms in Eq. (1) and qit = μqit − 1 + β1R & Dit + β2ADit + β3CAPit + αi + ψt + εit.
Results of the endogenous self-selection model with fixed effects using Heckman's (1979) extension are
reported in Table 5. As expected, the estimated coefficients of λ̂it are positive. These results suggest a
positive correlation between the processes governing firm's choice regarding power of IT executives, firm's
ability to develop superior IT capability, and the resulting improvement in firm performance. For example,
Table 5 Panel A reports results for the second stage regression analysis based on IT capable vs. non-IT
capable firms. The coefficient estimates of λ̂it in the unrestricted and restricted models are respectively
0.263 and 0.297 with t-statistics of 1.99 and 2.17 in absolute values. The estimated coefficients are
statistically significant at the 5% level. 9
This study uses Tobin's Q as a proxy for firm performance because it is likely to capture such
intangibles as good management skills and superior IT capability. As a robustness check we run the
analysis using ROA as the performance measure in the second stage. The results reported in Table 6 are
9
For a more general discussion on endogeneity and self-selection and how they relate to this study see Appendix C.
36
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
Table 6
Two stage regression analysis with ROA as dependent variable in second stage.
B. Second-stage linear regression with ROA as the dependent variable
Panel A: IT capable vs. non-IT capable firms
ROAit − 1
R&Dit
ADit
CAPit
ŷit
Adjusted R2
#obs.
Note: first stage results are same as in Table
Unrestricted model
Restricted model
Coefficient (Std. error)
Coefficient (Std. error)
0.165 (0.086)*
0.089 (0.038)**
0.075 (0.042)*
0.107 (0.059)*
0.211 (0.095)**
0.74
6720
2 — Panel A.
0.182
0.102
0.093
0.114
0.228
0.77
6720
(0.094)*
(0.042)**
(0.047)**
(0.063)*
(0.090)**
Panel B: firms with enduring vs. non-enduring IT capability
ROAit − 1
0.186 (0.109)*
R&Dit
0.173(0.098)*
ADit
0.086 (0.041)**
CAPit
0.092 (0.048)*
ŷit
0.269 (0.118)**
Adjusted R2
0.71
#obs.
474
Note: first stage results are same as in Table 3 — Panel A.
0.193
0.181
0.095
0.107
0.270
0.75
474
(0.097)**
(0.096)*
(0.043)**
(0.058)*
(0.113)**
Panel C: firms with non-enduring vs. non-IT capability
ROAit − 1
0.087 (0.052)*
R&Dit
0.039 (0.024)
ADit
0.116 (0.069)*
CAPit
0.185 (0.114)
ŷit
0.039 (0.024)
2
Adjusted R
0.46
#obs.
1640
Note: first stage results are same as in Table 2 — Panel A.
0.083
0.057
0.124
0.192
0.051
0.49
1640
(0.042)**
(0.034)*
(0.066)*
(0.115)*
(0.031)
Note: the standard errors of the OLS parameter estimates in the second stage (Table 6 Panels A to C) are corrected for a generated
regressor using an extended Murphy and Topel (2002) estimator. ***, **, and * denote significance at the 1, 5, or 10% level, respectively.
slightly weaker but consistent with the position of this study. The relative weakness is due to the fact
that ROA is much noisier measurement compared to Tobin's q. Results based on contrasting of IT
capable and non-IT capable firms (Table 6 Panel A) show that the contribution of IT executives on ROA
through superior IT capability is statistically significant. That is, the coefficient for the predicted IT
capability status (γ) is 0.228 (p-value < 5%). Similarly, results based on the contrast of firms with
superior and enduring IT capability versus firms with superior but non-enduring IT capability (Table 6
Panel B) show that the coefficient for the predicted superior and enduring IT capability status is 0.277
(p-value < 5%). Finally, results based on the contrast of firms with non-enduring IT capability versus
firms non-IT capability (Table 6 Panel C) show that the effect of non-enduring IT capability on ROA is
not statistically significant.
5. Discussion and conclusion
Evaluating the business value of IT remains one of the most interesting questions for researchers and
professionals. This interest is motivated by the realization that ability to extract value from investment in
IT resources and capabilities, while feasible, is far from certain. In this study we propose coupling the role
of IT executives with IT capability in order to evaluate their effect on firm performance. More specifically
we theorize and find empirical evidence to support the following positions: first, there is a positive
relationship between the hierarchical power of senior IT executives and the likelihood that the firm will
develop superior IT capability. Second, the contribution of IT capability to a firm's competitive advantage
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
37
is much stronger in firms with powerful senior IT executives as they are the driving force that may ensure
the continuous renewal of IT capability.
This study makes several contributions to IT business value literature. First, it suggests the need to
extend the prior literature on IT capability and firm performance with a search for factors that have a
multiplicative effect on the payoffs from IT capabilities. Second, existing literature is leveraged to theorize
and empirically validate the proposition that the power of senior IT executives works multiplicatively
with IT capability. Third, a relatively new area of research is illuminated, such as the value adding
contribution of incumbent senior IT executives. Fourth, the study demonstrates that senior IT executives
who leverage their power to develop a superior and enduring IT capability are making a much stronger
contribution to their company than IT executives who develop a superior but non-enduring IT capability.
Finally, identifying antecedents of IT capability, i.e., factors that lead a firm to achieve and sustain superior
IT capability, as well as recognizing the role that these factors play on the rent yielding capacity of IT
capability helps to illuminate the proverbial black box of IT business value research.
Naturally, like similar studies, there are limitations that must be acknowledged. First, since
InformationWeek has been a well-respected and widely used source of secondary information on IT capability
(Bharadwaj, 2000; Santhanam and Hartono, 2003), it was assumed that firms listed in IW500 are a proxy for
firms with superior IT capability. However, we cannot confirm that the IW500 firms are independently
evaluated each year. Second, Finkelstein (1992) suggests that official title, number of titles, and compensation
are a proxy of hierarchical power. While we use official title and number of titles, compensation was excluded
because only 8% of IT managers' compensation data is publicly available. Nevertheless, since the empirical
results are consistent with the anecdotal evidence offered by the IT leaders such as PNC, Intel, Harrah's
Entertainment and many others, these limitations do not seem to compromise the main message of our study.
Future research could attempt to resolve some of these limitations or extend our research by considering
other characteristics of senior IT executives such as experience, education, connection to other business and IT
executives and how they affect their ability to add value to their firm through IT capability. Finally, as
mentioned in the introduction this study has important practical implications for senior business executives
who want to compete through IT enabled strategies. We suggested that bestowing more power on senior IT
executives increases the likelihood that the firm will develop and sustain superior IT capability, the sine qua
non for an IT enabled strategy. Related to this observation, future research could evaluate the extent to which
firms reciprocate by rewarding the contribution of senior IT executives to their firm's ability to achieve and
sustain superior organizational IT capability.
Acknowledgments
The authors would like to thank Andreas Nicolaou (Editor) and the three anonymous referees for their
careful reviews and constructive suggestions. We also thank the anonymous reviewers and participants at
the 2011 International Conference on Enterprise Systems, Accounting and Logistics. The authors
acknowledge the financial support provided by the Social Sciences Humanities Research Council of
Canada (SSHRC) and the University of Waterloo.
Appendix A. Variable descriptions
Variables
Definition
yit
qit
DV-stage 1
DV-stage 2
CIOit
SIZEit
ROAit
MVit
R & Dit
CAPit
ADit
IV
CV
CV
CV
CV
CV
CV
1 for firms with superior IT capability in t, otherwise 0.
Tobin's q: a ratio of market value [(fiscal year-end market value of equity) + (liquidating value of the
firms' outstanding preferred stock) + (current liabilities) − (current assets) + (book value of
inventories) + (long-term debt)] to book value of total assets.
1 if a senior IT executive has the formal title of CIO or CIO with additional official titles, otherwise 0.
Natural logarithm of total assets.
Return on assets in year t − 1.
Market-to-book-value.
Research and development expenditure/net sales.
Capital expenditure/net sales.
Advertising expense/net sales.
38
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
Appendix B. Estimation methodology
In Eq. (2), ŷit is the predicted value of the first-stage regression. Instead of using a full-information
maximum likelihood, which would be extremely intricate in our case, we adopt a two-step procedure in
which the first-stage regression is estimated by using the random-effect (RE) approach proposed by
Wooldridge (2005). The second stage regression is then estimated by ordinary least squares (OLS).
Although the resulting parameter estimates in the second-stage regression are consistent, their estimated
standard errors would tend to be under-estimated leading to spuriously high values of t-statistics for the
significance of each individual estimated parameter in Eq. (2). This is because ŷit in Eq. (2), being estimated
from the first-stage regression is a generated regressor according to Pagan (1986). To correct for this
generated regressor bias, we extend the correction formula for the covariance matrix estimates of the
estimated parameters in the second stage originally suggested by Murphy and Topel (2002) to our twostage model.
Let the first-stage regression's parameters be denoted as θ1 = (ρ, δ′, η′, φ′)′, where η is a vector of
coefficients on the firm dummy variables and φ is a vector of coefficients on the year dummy variables. Let
the second-stage regression's parameters be denoted as θ2 = (μ, β′, γ, α′, ψ′)′, where α is a vector of
coefficients on the firm
dummy
variables
and ψ is a vector of coefficients on the year dummy variables.
jpffiffiffiffiffiffi
k
ffi
based on the sample log-likelihood function form the first-stage
Let V1 = Asy Cov NT θ̂1 −θ1
jpffiffiffiffiffiffiffi
k
based on the sample log-likelihood function form the
regression, ln L1. Let V2 = Asy Cov NT θ̂2 −θ2
second-stage regression, ln L2. Further define the following two matrices
C=E
1 ∂lnL2 Þ ∂lnL2 Þ
1 ∂lnL2 Þ ∂lnL1 Þ
and R = E
NT
NT
∂θ2
∂θ1′
∂θ2
∂θ1′
Then under standard regularity conditions for the validity of both log-likelihood functions, the secondstep maximum likelihood estimator of θ2 is consistent and asymptotically normally distributed with
Murphy and Topel (2002) corrected asymptotic covariance matrix
MT
V2
=
1 V + V2 CV1 C ′ −RV1 C ′ −CV1 R′ V2
NT 2
Its consistent estimate is given by
MT
V̂ 2 =
1 h
V̂ 2 + V̂ 2 Ĉ V̂ 1Ĉ ′ −R̂ V̂ 1Ĉ ′ −Ĉ V̂ 1R̂′ ÞV̂ 2 NT
with matrices with hat being matrices of respective consistent estimates.
Appendix C. Endogeneity and self selection
Since the dataset for this study is based on IW500, a relevant question could be raised on how
representative the sample is. We address this question by using an extended (dynamic) version of twostage endogenous self-selection Heckman procedure to control for the effect of this sample selection bias
on the parameter estimates in our second-stage regression. Sample selection bias and endogeneity bias
refer to two slightly distinct concepts. In general, sample selection bias refers to problems where the
dependent variable is observed only for a restricted, nonrandom sample. In the context of this study, we
observe superior IT capability (yit = 1) for firms self-selecting to participate in the IW500 survey.
Conversely, we observe no-IT capability (yit = 0) for firms self-selecting not to participate in the IW500
survey. Endogeneity refers to the fact that an independent variable included in the model is potentially a
choice variable for the firm, correlated with unobservables relegated to the error term. The dependent
variable, however, is observed for all observations in the data. Here the firm's choice regarding the power of
IT executives may be endogenous if the decision is correlated with unobservables related to IT capability
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
39
status. Failure to control for this correlation will yield an estimated IT executive power effect on IT
capability superiority that is biased.
The problem with IT executive power and IT capability superiority, and a host of other problems, can be
treated either as a self-selection problem or as an endogeneity problem. The appropriate model depends on
how we believe the power of IT executives affects firm's ability to achieve superior IT capability.
(1) If power of IT executives power has merely an intercept effect on firm's IT capability, then the
appropriate model includes power of senior IT executives as a right-hand-side variable and pools the
entire sample of superior and inferior IT capable firms. Because the entire sample is used, there are
no sample-selection issues.
(2) If power of IT executives has not only an intercept effect but also a slope effect, then a sample
selection model is called for. To proceed, split the sample into two groups based on IT capability and
estimate the performance regression for each subsample. If IT executive power is the only
potentially endogenous variable in the model, the two separate equations may be estimated via OLS,
accounting for the fact that each sample is a nonrandom sample of all firms. This is accomplished via
an extended Heckman's selection correction model using either ML estimation or two-step
estimation. In a two step estimation, in the first stage a dynamic probit model is used to predict the
probability of developing superior IT capability and in the second stage, the inverse Mills' ratio is
included as a regressor. This is the approach used in Section 4.3.
(3) If the power of IT executives is endogenous and firms self-select to develop superior IT capability,
then we should instrument for IT executive power using a two-step method. This is the main
approach used in this study.
References
Adams R, Hermalin B, Weisbach M. The role of board of directors in corporate governance: a conceptual framework and survey. J Econ
Lit 2010;48(1):51–107. cbe.
Applegate LM, Elam JJ. New information systems leaders: a changing role in a changing world. MIS Q 1992;16(4):469–90.
Armstrong CP, Sambamurthy V. Information technology assimilation in firms: the influence of senior leadership and IT
infrastructures. Inf Syst Res 1999;10(4):304–27.
Barney J. Firm resources and sustained competitive advantage. J Manage 1991;17(1):99–120.
Bassellier G, Benbasat I, Reich BH. The influence of business managers' IT competence on championing IT. Inf Syst Res 2003;14(4):
317–36.
Bharadwaj A. A resource-based perspective on information technology capability and firm performance: an empirical investigation.
MIS Q 2000;24(1):169–96.
Bharadwaj AS, Bharadwaj SG, Konsynski BR. Information technology effects on firm performance as measured by Tobin's q. Manage
Sci 1999;45(7):1008–24.
Bhatt GD, Grover V. Types of information technology capabilities and their role in competitive advantage: an empirical study. J
Manage Inf Syst 2005;22(2):253–77.
Brynjolfsson E, Hitt L. Paradox Lost? Firm-Level Evidence on the Returns to Information Systems Spending. Manage Sci 1996;42(4):
541–58.
Brynjolfsson E, Hitt LM. Beyond the productivity paradox. Commun ACM 1998;41(8):49–55.
Brynjolfsson E, Hitt LM. Beyond computation: information technology, organizational transformation and business performance. J
Econ Perspect 2000;14(4):23–48.
Brynjolfsson E, Hitt LM. Computing productivity: firm-level evidence. Rev Econ Stat 2003;85(4):793–808.
Chari MDR, Devaraj S, David P. Research note the impact of information technology investments and diversification strategies on firm
performance. Manage Sci 2008;54(1):224–34.
Chatterjee D, Richardson VJ, Zmud RW. Examining the shareholder wealth effects of announcements of newly created CIO positions.
MIS Q 2001;25(1):43–70.
Chen DQ, Mocker M, Preston DS, Teubner A. Information systems strategy: reconceptualization, measurement, and implications. MIS
Q 2010;34(2):233–59.
Chen C, Lim JH, Stratopoulos TC. it capability and a firm's ability to recover from losses: evidence from the economic downturn of the
early 2000s. J Inf Syst 2011;25(2).
Cheng Q. What determines residual income? Account Rev 2005;80(1):85–112.
Chung KH, Pruitt SW. A simple approximation of Tobin's q. Financ Manage 1994;23(3):70–4.
Curley M. The IT transformation at intel. MIS Q Exec 2006;5(4):155–68.
Dehning B, Stratopoulos T. DuPont analysis of an IT-enabled competitive advantage. Int J Account Inf Syst 2002;3(3):165–76. Oct.
Dehning B, Stratopoulos TC. Determinants of a sustainable competitive advantage due to an IT-enabled strategy. J Strateg Inf Syst
2003;12(1):7–28.
Finkelstein S. Power in top management teams: dimensions, measurement, and validation. Acad Manage J 1992;35(3):505–38.
Galliers R, Leidner DE. Strategic information management: challenges and strategies in managing information systems. ButterworthHeinemann; 2003.
40
J.-H. Lim et al. / International Journal of Accounting Information Systems 13 (2012) 21–40
Grover V, Jeong S-R, Kettinger WJ, Lee CC. The chief information officer: a study of managerial roles. J Manage Inf Syst 1993;10(2):107–30.
Hambrick DC, Mason PA. Upper Echelons: the organization as a reflection of its top managers. Acad Manage Rev 1984;9(2):193–206.
Hamilton BH, Nickerson JA. Correcting for endogeneity in strategic management research. Strategic Organization 2003;1(1):51–78.
Heckman JJ. Sample selection bias as a specification error. Econometrica 1979;47(1):153–61.
Hitt LM, Brynjolfsson E. Productivity, business profitability, and consumer surplus: three different measures of information
technology value. MIS Q 1996;20(2):121–42.
Karimi J, Gupta YP, Somers TM. The congruence between a firm's competitive strategy and information technology leader's rank and
role. J Manage Inf Syst 1996;13(1):63–88.
Khallaf A, Skantz TR. The effects of information technology expertise on the market value of a firm. J Inf Syst 2007;21(1):83–105.
Khallaf A, Skantz T. Does long term performance improve following the appointment of a CIO? Int J Account Inf Syst 2011;12(1):57–78.
Lev B. Some economic determinants of time-series properties of earnings. J Account Econ 1983;5(1):31–48.
Lim JH, Stratopoulos TC, Wirjanto T. Path dependence of dynamic information technology capability: an empirical investigation.
Journal of Management Information Systems (forthcoming).
Marlin S. Double duty. Information Week [Internet]; 2004 [cited 2010 Jul 9];(November 8). Available from: http://www.informationweek.
com/shared/printableArticle.jhtml;jsessionid=3W51RGNKNA0SDQE1GHRSKH4ATMY32JVN?articleID=52200063.
Masli A, Richardson VJ, Sanchez JM, Smith RE. Returns to IT excellence: evidence from financial performance around information
technology excellence awards. International Journal of Accounting Information Systems 2011;12:189–205.
Mata F, Fuerst W, Barney J. Information technology and sustained competitive advantage: a resource-based analysis. MIS Q
1995;14(4):487–505.
Milgrom P, Roberts J. The economics of modern manufacturing: technology, strategy, and organization. Am Econ Rev 1990;80(3):
511–28.
Miller DJ. Technological diversity, related diversification, and firm performance. Strateg Manag J 2006;27(7):601–19.
Mintzberg H. Power in and around organizations. Prentice-Hall; 1983.
Morrow JL, Sirmon DG, Hitt MA, Holcomb TR. Creating value in the face of declining performance: firm strategies and organizational
recovery. Strateg Manage J 2007;28(3):271–83.
Mueller DC. The Dynamics of company profits: an international comparison. Cambridge University Press; 1990.
Muhanna WA, Stoel MD. How do investors value IT? an empirical investigation of the value relevance of IT capability and IT spending
across industries. J Inf Syst 2010;24(1):43–66.
Murphy KM, Topel RH. Estimation and inference in two-step econometric models. J Bus Econ Stat 2002;20(1):88–97.
Newbert SL. Empirical research on the resource-based view of the firm: an assessment and suggestions for future research. Strateg
Manage J 2007;28(2):121–46.
Pagan A. Two stage and related estimators and their applications. Rev Econ Stud 1986;53(4):517–38.
Powell TC, Dent-Micallef A. Information technology as competitive advantage: the role of human, business, and technology resources.
Strateg Manag J 1997;18(5):375–405.
Raghunathan B, Raghunathan TS. Relationship of the rank of information systems executive to the organizational role and planning
dimensions of information systems. J Manage Inf Syst 1989;6(1):111–25.
Ravichandran T, Lertwongsatien C. Effect of information systems resources and capabilities on firm performance: a resource-based
perspective. J Manage Inf Syst 2005;21(4):237–76.
Ravichandran T, Liu Y, Han S, Hasan I. Diversification and firm performance: exploring the moderating effects of information
technology spending. J Manage Inf Syst 2009;25(4):205–40.
Roberts PW, Dowling GR. Corporate reputation and sustained superior financial performance. Strateg Manag J 2002;23(12):1077–93.
Rosencrance L. Intel names co-CIOs. Computerworld [Internet]; 2002 [cited 2010 Jul 9];(February 21). Available from: http://www.
computerworld.com/s/article/68511/Intel_names_co_CIOs.
Sambamurthy V, Zmud RW. Foundation FER. IT management competency assessment: a tool for creating business value through IT.
Financial Executives Research Foundation; 1994.
Santhanam R, Hartono E. Issues in linking information technology capability to firm performance. MIS Q 2003;27(1):125–53.
Scott JT, Pascoe G. Purposive diversification of R&D in manufacturing. J Ind Econ 1987;36(2):193–205.
Shaver JM. Accounting for endogeneity when assessing strategy performance: does entry mode choice affect FDI survival?
Management Science 1998;44(4):571–85.
Stratopoulos TC, Lim JH. IT innovation persistence: an oxymoron? Commun ACM 2010;53(5):142–6.
Tanriverdi H. Performance effects of information technology synergies in multibusiness firms. Manage Inf Syst Q 2006;30(1):57–77.
Wang L, Alam P. Information technology capability: firm valuation, earnings uncertainty, and forecast accuracy. J Inf Syst 2007;21(2):
27–48.
Westerman G, Curley M. Building IT-enabled innovation capabilities at intel. MIS Q Exec 2008;7(1):33–48.
Weston R. Where to find innovators. Information Week; 1998. September 14.
Wooldridge JM. Simple solutions to the initial conditions problem in dynamic, nonlinear panel data models with unobserved
heterogeneity. J Appl Econ 2005;20(1):39–54.
Zhu K. The complementarity of information technology infrastructure and e-commerce capability: a resource-based assessment of
their business value. J Manage Inf Syst 2004;21(1):167–202.