See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265688246

The Role of Operations Executives in Strategy Making

Article in Journal of Operations Management · November 2014

DOI: 10.1016/j.jom.2014.09.006

CITATIONS READS

28 529

3 authors:

Lieven Lode E. Demeester Arnoud De Meyer

Singapore Management University Singapore Management University
17 PUBLICATIONS 389 CITATIONS 136 PUBLICATIONS 7,341 CITATIONS

SEE PROFILE SEE PROFILE

Jovan Grahovac
Purdue University
10 PUBLICATIONS 336 CITATIONS

SEE PROFILE

All content following this page was uploaded by Lieven Lode E. Demeester on 18 October 2022.

The user has requested enhancement of the downloaded file.

 Journal of Operations Management 32 (2014) 403–413

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

The role of operations executives in strategy making

Lieven Demeester a,∗ , Arnoud De Meyer b , Jovan Grahovac c,1
a
Lee Kong Chian School of Business, Singapore Management University, 50 Stamford Road, 178899 Singapore, Singapore
b
Singapore Management University, 81 Victoria Street, 188065 Singapore, Singapore
c
College of Business, University of Illinois at Urbana-Champaign, 1206 South 6th Street Champaign, IL 61820, United States

a r t i c l e i n f o a b s t r a c t

Article history: Creating competitive advantage based on operations capabilities is likely to require much analysis and
Available online 16 September 2014 communication within the operations function. At the same time, much communication and joint strate-
gizing with the top and other functional executives is likely to be needed as well. Hence, given that
Keywords: operations executives have limited time and also have to perform many other routine tasks, they need
Operations strategy to manage two tradeoffs. The first one is between the time spent on strategy making and the time spent
Strategy process
on everything else. The other is within strategy making, between the time spent on “functional delib-
Information processing
eration” within the operations function and “top-level communication” with other executives. Using a
Contingency theory
survey of 134 operations executives, we find that an increase in the time the operations executive spends
on strategy making is positively associated with performance in complex and hostile environments and
when the relative strength of the operations function within the firm is low. Within the operations exec-
utive’s strategy making, an increased emphasis on top-level communication is positively associated with
performance in environments that are complex, stable (less uncertain), or hostile.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction can best adjust their involvement in strategy making as the context
changes is the question we address in this paper.
It is well established that operations executives (OEs) contribute We define the OE as the person in charge of the operations func-
to competitive advantage when they play an active role in strat- tion and responsible for the resources and processes used in the
egy making (Brown et al., 2007; Papke-Shields and Malhotra, 2001; production and delivery of a firm’s goods and services. The OE’s job
Swamidass and Newell, 1987; Wheelwright and Hayes, 1985). The entails a number of monitoring, coordination, and other tasks (e.g.
types of decisions made in this role are also well known from Fayol, 1949) beside what we refer to as strategy making, that is, the
descriptions of the process of operations strategy (Skinner, 1969; involvement in the analysis and formulation of business level and
Fine and Hax, 1985; Hill, 1989; Menda and Dilts, 1997). What is not operations strategies as well as in the analysis and planning of how
well understood, though, is how and to what extent various inter- to best implement them. Importantly, like everyone else, OEs have
nal and external contingencies affect the optimal involvement of limited time, attention, and ability to process and communicate
the OE in strategy making. For example, Clancy and Kieff (2004) information at their disposal (Simon, 1947), and these limitations,
found that increased communication with other functional execu- in turn, imply that OEs have to prioritize and balance their numer-
tives was necessary to formulate successful strategies in an industry ous tasks. Moreover, these priorities and balances probably need
facing a threat of commoditization. A different need was suggested periodic reevaluation and adjustment.
by reports on the retail industry’s response to the emergence of In this paper, we focus on two broad tradeoffs, or balances, in
the Internet. Here, a number of OEs fell short in developing newly the OE’s job. The first is between strategy making and all other
needed logistics capabilities (e.g. Wall Street Journal, 1999), imply- tasks. The second occurs within strategy making, between what we
ing that these OEs may have failed to devote sufficient time to the call functional deliberation and top-level communication. We define
analysis, planning, and execution within their function. How OEs functional deliberation (FD) as the analysis and communication
with subordinates in the operations function and external parties.
In contrast to the functional focus of FD, top-level communication
(TLC) is inherently cross-functional and consists of various engage-
∗ Corresponding author. Tel.: +65 6828 0729.
ments with the CEO and senior executives of other functions such as
E-mail addresses: [email protected] (L. Demeester),
[email protected] (A. De Meyer), [email protected] (J. Grahovac). marketing, research and development, finance, etc. In essence, the
1
Tel.: +1 217 265 0266. second tradeoff corresponds to a tradeoff between strengthening

http://dx.doi.org/10.1016/j.jom.2014.09.006
0272-6963/© 2014 Elsevier B.V. All rights reserved.
404 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413

operations resources and tightening the strategic linkages between advantage (Teece et al., 1997). Specifically, our analysis raises
operations and the rest of the firm. the question whether this process could benefit from mech-
To understand these tradeoffs and their impact on competitive anisms to adjust itself and become more decentralized and
advantage, we synthesize insights from operations strategy (e.g. functionally focused in some contexts and more collaborative and
Skinner, 1969, 1974, 1985; Hayes and Wheelwright, 1979, 1984; cross-functional in others. Perhaps it is not surprising that this
Ferdows and De Meyer, 1990; Miller and Roth, 1994), competi- exploration of the strategy making process should come from an
tive strategy including the influential resource-based view (RBV) of investigation of the role of executives in operations, as the invest-
the firm (e.g. Barney, 1986, 1989; Dierickx and Cool, 1989; Porter, ments in equipment, people, and technology are typically so big and
1980, 1996), and more recent studies that seek to integrate these important that the fields of competitive and operations strategy are
two literatures (Paiva et al., 2008; Schroeder et al., 2002). We then natural candidates for co-evolution and cross-pollination.
hypothesize about how these tradeoffs are likely to be affected The rest of the paper is structured as follows: Section 2 reviews
by the complexity, uncertainty, and hostility of the firm’s envi- the relevant literatures and Section 3 synthesizes these literatures
ronment, three external variables which are most commonly used to develop eight hypotheses regarding the impact of external and
in the literature on strategic planning (e.g., Dess and Beard, 1984; internal context on the balance between different components of
Tushman and Nadler, 1978; Miller and Friesen, 1983). In addition, the OE’s role. Section 4 describes the data and the measures, and
we hypothesize about the impact of the relative strength of the Section 5 presents the results. Section 6 concludes the paper with
operations function within the firm as an internal context variable a discussion of the results and their implications.
(Wheelwright and Hayes, 1985).
We test our hypotheses using responses from multiple surveys
with respondents from a range of different industries. We find sup- 2. The role of operations executives in strategy making:
port for 6 out of the 8 hypotheses. Specifically, an increase in the insights from the literature
time the OE spends on strategy making is positively associated
with performance in complex and hostile environments and when The notion that businesses face tradeoffs in what they can do
the relative strength of the operations function within the firm is and hence need to choose how to compete has been a corner-
perceived as low. We also find that within the OE’s strategy making, stone of competitive and operations strategy since the early days
an increased emphasis on TLC is positively associated with perfor- (e.g. Skinner, 1969, 1974; Porter, 1980; Hayes and Wheelwright,
mance in environments that are complex, stable (less uncertain), 1979, 1984). At the level of operations, typical tradeoffs involve
or hostile. cost, quality, variety, responsiveness, etc. and result in the need
These results resonate with the evidence from the litera- to select specific priorities or a “manufacturing task” that should
ture, business press, and our contacts with executives. Regarding subsequently guide facilities, capacity, technology, quality manage-
the above examples, commoditization of an industry is likely to ment, and other important choices (Hayes and Wheelwright, 1979,
increase the hostility and complexity of the environment, resulting 1984; Skinner, 1969, 1974, 1985). The conception of operations
in the need for an increase in strategy making and especially TLC. On strategy as a pursuit of coherent and mutually reinforcing oper-
the other hand, the rise of the Internet caused a great deal of uncer- ational choices in support of the overall firm strategy has entered
tainty for the retail industry, suggesting the need for more FD. More the mainstream of managerial thinking and practice (e.g., Fine and
recently, in China, many shopping mall operators started to under- Hax, 1985; Hill, 1989; Miller and Roth, 1994). Importantly, this con-
perform as their industry moved from the munificent and relatively ception also implies a top-down approach in which competitive
simple (in terms of the recipe for success) environment of rapid priorities for the operations function stem from the overarching
growth to the one characterized by overcapacity and new compe- business level strategy (Porter, 1980; Hayes and Wheelwright,
tition from online retailers such as Alibaba. OE’s who were once 1984).
busy managing the construction of new and routine operations of This classical view of strategy as positioning on tradeoffs –
the existing shopping malls had to rethink the more complex model implicitly assumed to be static – was supplemented over time with
of mixed online and physical retail in a more saturated market. more dynamic views. Barney (1986, 1989) proposed that firms pur-
An OE of a major international developer told us: “While we were suing an attractive position may compete away their profits as they
building shopping malls, Alibaba and other online retailers were try to acquire necessary resources in the so-called “strategic factor
developing a new business model. We need to spend much more markets.” Firms can hence earn rents only if they are lucky, act
time on redesigning and running the malls differently in order to faster or based on superior information, or if they pursue opportu-
create more of a special experience. We also need to interact more nities for which they already have some resources that other firms
with our colleagues and partners to understand the nature of on- do not have. A closely related view proposes that only resources and
line sales, and how we can align the way we run the malls with it.” capabilities that are assembled over time and cannot be bought in
In other words, complexity and hostility seem linked to a need for strategic factor markets can be a source of sustained rents (Dierickx
more strategy making by the OE and especially more TLC. and Cool, 1989; Teece et al., 1997). The two views are referred to as
While contingency theories for the process of operations strat- the resource-picking and capability-building perspectives within
egy have been proposed before (e.g. Wheelwright and Hayes, 1985; the resource-based view (RBV) in strategy (Makadok, 2001).
Mills et al., 1995), we provide a different approach by distinguishing Mirroring the capability-building perspective, Ferdows and De
between FD and TLC, and we test our hypotheses empirically. Our Meyer (1990) have documented the cumulative nature of some
results also help interpret insights from the literature on strategic operations capabilities. Motivated by the success of Japanese man-
planning in the context of the process of operations strategy. ufacturers which seemed to defy operational tradeoffs in the last
Finally, our analysis could also be relevant to other functional decades of the 20th century, some researchers have proposed that
executives because they tend to face similar tradeoffs. Essentially, the general excellence-based approach to operations strategy may
we unpack strategy making and the role of a specific func- be an alternative or even superior to the one based on positioning
tional executive in it to derive and test specific contingency-based (Hayes and Wheelwright, 1984; Nakane and Hall, 1991; Womack
hypotheses. By doing so, we further explore strategy making as et al., 1990; Corbett and Van Wassenhove, 1993). These so-called
a dynamic capability (e.g. Eisenhardt and Martin, 2000), that is, “world-class” operations capabilities have also been found to con-
a process that can be purposefully designed to help a firm select tribute to firm performance in empirical studies (e.g., Rosenzweig
and accumulate resources that create and sustain competitive et al., 2003; Rosenzweig and Easton, 2010).
 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413 405

In response to the tension between the classical and For practice, answers to such questions can provide guidance on
capabilities-based perspective, Porter (1996) has proposed that how operations executives and top management teams can adjust
the ever-changing technologies and best practices effectively shift the frequency and structure of various planning processes when
various tradeoffs that firms face. Hence maintaining general opera- internal or external context changes. For example, one of the ways
tional effectiveness is a necessary condition for staying in the game we have seen an operations executive at a large ingredient producer
but not sufficient for sustained profits, which still require firms to in Asia raise TLC is through the implementation of a sales-and-
anticipate and occupy unique profitable positions in this changing operations-planning routine in which sales, operations and new
world. product development executives met bi-monthly to discuss oper-
These arguments and the RBV in general imply a more nuanced ational and strategic issues.
role of the OE in the strategy process than the classical top-down So far, only a few studies have examined how internal and exter-
view. The first task of the OE is the accumulation of operations nal factors influence the role of OEs in the process of strategy.
resources and capabilities, whether those related to general best Wheelwright and Hayes (1985) have posited that the manufactur-
practices needed to keep up with the moving efficiency frontiers ing function passes through four stages of development in which
(Ferdows and De Meyer, 1990; Hayes and Pisano, 1996), or those the role of the OE within the firm steadily increases, with increasing
in support of the firm’s unique positioning (Porter, 1996). We posit emphasis on top-level communication. Other studied have focused
that this task requires a substantial amount of functional delibera- on external factors. Swamidass and Newell (1987) and Ho (1996)
tion (FD). Accumulating resources and capabilities requires a great empirically examined the effect of environmental uncertainty on
deal of careful deliberation within the operations function as well as the role of manufacturing managers in the process of strategy. Both
information exchange with external parties (Schroeder et al., 2002). studies found a diminished role in uncertain environments but also
Moreover, some capabilities such as those in total quality manage- generally positive association between a larger role of the OE and
ment (TQM) make it easier to acquire others such as those related firm performance. However, these studies did not consider a con-
to responsiveness and flexibility (Ferdows and De Meyer, 1990), tingency framework in which environmental conditions and the
implying the need for careful sequencing of learning efforts. Finally, role of manufacturing managers interacted to explain performance.
the development of higher order combinative capabilities benefits Finally, Mills et al. (1995) proposed a contingency approach to the
from both exploration and exploitation (Kristal et al., 2010), further process of operations strategy and stressed the need to link it with
increasing the difficulty of the OE’s first task. the contingency theories for the general process of strategy. Our
The second task of the OE is to make sure that operations study is different in two ways. First, we do not treat the firm’s
resources and those in other functions work well together and strategy-making mode as a context variable. Rather, we treat one
we posit that top-level communication (TLC) is the main way in part of it – the role of the OE – as a managerial variable that can be
which the OE can contribute to such coherence across functions. influenced. Second, the parsimonious characterization of context
For example, if marketing executives are not aware of advances in and strategy process outcomes allows us to test several hypotheses
the flexibility of operations, they will likely fail to increase the vari- regarding the interaction between the context, the strategy-making
ety of product offerings believing that the production costs would process, and firm performance.
be prohibitive (de Groote, 1994). Similarly, if the firm’s internal
resources are to inform the selection of future strategies (Barney, 3. Information processing, resources, and the job of the
1986, 1989), then the OE needs to communicate to the top man- operations executive
agement team what resources the operations function has and to
help evaluate the viability of different strategic choices through In this section, we synthesize the supporting literature to
the lens of these resources. For example, if a company has a high- develop a model that links the OE’s strategy-making activities, their
performing offshore plant in Vietnam, the growing local economy strategically important outcomes, and firm performance (Fig. 1).
may offer an opportunity to enter the Vietnamese market by build- We then use this model to derive hypotheses regarding the impact
ing on the network relationships, knowledge of the market for of context on the optimal allocation of the OE’s time spent on these
local talent, and the proximity to customers provided by the exist- activities.
ing plant management. An appropriate amount of TLC will likely To streamline the analysis, we abstract additional layers of
allow the company to become aware of this opportunity and take complexity that arise in large corporations and we consider a func-
advantage of it. Importantly, empirical studies have found that, tionally organized firm with a single business unit. The OE is in
while challenging (Menda and Dilts, 1997), cross-functional com- charge of the operations function and has direct access to the chief
munication contributes to competitive advantage and performance executive officer (CEO) and the heads of other functions such as
(Paiva et al., 2008; Papke-Shields and Malhotra, 2001; Swink et al., marketing, R&D, finance, etc. The OE also has direct lines of commu-
2007). nication with the subordinates in the operations function as well as
The need to perform two distinct and demanding tasks poses with external parties such as suppliers, vendors, service providers,
important challenges to the OE from the standpoint of bounded etc.
rationality (Simon, 1947). To address questions of this type,
research in organization theory and strategic decision making often
conceptualizes organizations as hierarchies of boundedly ratio-
nal information processors in which the structure and intensity of
information flows need to respond optimally to the environment
(e.g. Galbraith, 1973, 1977). The implication for our analysis is that
the limited ability to communicate and process information forces
the OE to trade the time spent on FD against that spent on TLC.
Hence, we arrive at the key questions in this paper: What is the right
balance between the two tasks and how might this balance change
depending on the context that the firm faces? In addition, how
are these two strategy-related tasks in aggregate balanced against
the basic coordination, control, and other management tasks (e.g.,
Fayol, 1949) that an OE has to perform? Fig. 1. The job of the operations executive.
406 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413

We focus on two bounded-rationality-induced tradeoffs faced In essence, problems with many moving parts and strong inter-
by the OE, shown in Fig. 1 using dashed lines. The first tradeoff is actions among them require harnessing the expertise of multiple
between the time spent on strategy making and the time spent on experts and their extensive communication. Consequences for the
what we label “execution”, which includes organizing, command- OE’s strategy making activities (Fig. 1) are rather straightforward.
ing, coordinating, and controlling (Fayol, 1949). While it stands to The accumulation of functional resources and capabilities is likely
reason that both strategy making and execution should contribute to be more challenging as the OE needs to solve difficult problems
to firm performance, the tradeoff between the two raises the ques- in collaboration with functional experts, workers, and external
tions of their optimal balance and how this balance is affected by parties. As a result, we expect the optimal time spent on FD to
context. Studies by Wheelwright and Hayes (1985), Swamidass and increase. At the same time, the OE is likely to spend significantly
Newell (1987), and Ho (1996) have implied that context has an more time communicating with the CEO and other functional heads
impact but, as mentioned above, no specific hypotheses have been on the formulation of appropriate strategies as this problem is
tested. now more difficult as well. In complex environments, customers
The other tradeoff is between TLC and FD as two distinct are more segmented, competitors more closely watched, and the
strategy-making activities. TLC consists of the OE’s communica- nature of competition oligopolistic (Dess and Beard, 1984; Miller
tion with the CEO and/or other functional heads. As argued in and Friesen, 1983), all requiring careful positioning and high coher-
Section 2, this activity increases the cross-functional coherence ence across functions. In addition, TLC is likely to be required on
of firm resources residing in different functions (Brown et al., an ongoing basis to ensure that dense and unanticipated inter-
2007; de Groote, 1994; Menda and Dilts, 1997; Papke-Shields actions are resolved and cross-functional coherence maintained.
and Malhotra, 2001) and helps the firm select better strategies. Finally, the accumulation of functional resources and capabilities
FD consists of the remaining strategy-making activities including that in a simple environment may entail only FD is likely to require
individual analysis and deliberation, communication with subordi- some TLC to prevent imposing unanticipated negative externalities
nates, and communication with external parties such as vendors, on other functions (Siggelkow and Rivkin, 2005). As a result, with
suppliers, consultants, and distributors. FD thus represents the the requirements for TLC exceeding those for FD, we formulate the
strategy making activities within the function performed indepen- following two hypotheses:
dently from the rest of the firm. As described in Section 2, this
activity increases the stock of resources and capabilities residing Hypothesis 1a. Higher environmental complexity is associated
in the operations function (Ferdows and De Meyer, 1990; Kristal with a more positive relationship between the time the OE spends
et al., 2010; Schroeder et al., 2002). on strategy making and firm performance.
The distinct strategically important outcomes of TLC and FD that Hypothesis 1b. Higher environmental complexity is associated
ultimately contribute to firm performance are shown in Fig. 1. FD with a more positive relationship between the share of top-level
helps accumulate operations resources and capabilities that help communication in the OE’s strategy making and firm performance.
the firm produce and deliver goods or services. Similarly, FD in
other functions can be expected to help accumulate the respective Numerous studies including those by Goll and Rasheed (1997),
resources and capabilities that facilitate marketing, new product Miller and Cardinal (1994), Miller and Friesen (1983), and Priem
development, recruitment, etc. TLC, on the other hand, steers the et al. (1995) have suggested that more strategic planning is
accumulation of functional resources and capabilities toward cross- associated with higher performance in dynamic environments.
functional coherence and synergies. TLC also helps the firm select Researchers have also suggested that uncertain environments
strategies that leverage these resources and capabilities residing call for loose coupling or incrementalism (Fredrickson, 1984;
in different functions because the respective functional heads are Fredrickson and Mitchell, 1984; Mintzberg, 1973), decentralization
most familiar with them. A central question in this study is how (Siggelkow and Rivkin (2005), and simple forms of control with
the optimal mix of TLC and FD changes as the firm faces different less communication between the constituent parts (Eisenhardt
contingencies. and Bhatia, 2002; Eisenhardt and Martin, 2000). Importantly,
environmental dynamism, characterized by unpredictable mar-
3.1. Impact of external context: complexity, uncertainty, and kets, rapid technological development, and competitive ambiguity
hostility (Davis et al., 2009; Hough and White, 2003), makes it relatively
more likely that a significant portion of strategic decision mak-
Environmental complexity, uncertainty, and hostility have been ing occurs during implementation, given the risks of premature
widely used in the literature (e.g. Brews and Hunt, 1999; Dess et al., commitment to any particular course of action.
1997; Goll and Rasheed, 1997; Miller and Friesen, 1983) to describe The implication for TLC is that whereas formulating strate-
external context. Dess and Beard (1984) demonstrated that it is gies is likely to entail some additional effort in considering more
along these dimensions that context tends to vary the most. Envi- risks and scenarios, the considered strategies are likely to be
ronmental complexity refers to the number of factors and their simpler and emphasize flexibility and postponement of commit-
interactions that affect firms. Environmental uncertainty relates to ments. Moreover, we can expect significantly less TLC in assuring
the unpredictability of these factors and is high in dynamic, fast- cross-functional coherence because of the likely implementation
changing environments. Environmental hostility is a measure of of simple rules and an emphasis on flexibility across functions. The
the difficulty of survival and growth in an environment. consequence is that we expect little or no increase in the need for
Complex environments require more information processing TLC in dynamic environments.
in general (Tushman and Nadler, 1978) and in the strategy pro- In contrast, dynamism is likely to increase the need for FD. Basic
cess in particular (Miller and Friesen, 1983). Dess et al. (1997), operations capabilities are more at risk of losing their value and new
Hart (1992), and Mintzberg and Waters (1985) have argued that ones may need to be built quickly in uncertain, dynamic environ-
complex environments require a balance between the initiative of ments (Loch et al., 2006; Klingebiel and De Meyer, 2013). Moreover,
the organizational members and central guidance. This proposition the likely focus on flexibility tends to entail non-trivial learning
was echoed by Nickerson and Zenger (2004), Siggelkow and Rivkin and resource accumulation processes that require a great deal of
(2005), and Loch et al. (2006) who found extensive information expertise, time, and effort (Ferdows and De Meyer, 1990). Hence
sharing and participative or consensus-based decision making to uncertain environments can be expected to call for decentraliza-
be optimal organizational responses to complexity. tion of the strategy process in that the OE’s contribution to strategy
 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413 407

making is likely to take place mainly within the operations function operations function within the firm. We draw on applications of
and away from the center. Hence the following pair of hypotheses: the RBV to operations strategy (e.g. Paiva et al., 2008; Rosenzweig
et al., 2003; Schroeder et al., 2002) to define this contingency as the
Hypothesis 2a. Higher environmental uncertainty is associated
contribution of operations resources and capabilities to the com-
with a more positive relationship between the time the OE spends
petitive advantage of the firm relative to that of the other functional
on strategy making and firm performance.
resources. The four-stage development path of the manufactur-
Hypothesis 2b. Higher environmental uncertainty is associated ing function proposed by Wheelwright and Hayes (1985) predicts
with a less positive relationship between the share of top-level that the primary focus on operational improvements is gradually
communication in the OE’s strategy making and firm performance. replaced by a focus on the alignment with the rest of the firm and
the pursuit of competitive advantage as the relative strength of
Interestingly, there are fewer studies and less agreement in the manufacturing function within the firm increases. Somewhat
the literature on strategy making in hostile environments than puzzlingly, this important managerial question has received little
with respect to other contextual variables. The inconsistent results further attention in the literature.
could be due to different research designs. Defining hostility as “the When the operations function is relatively underdeveloped, fix-
degree of threat to the firm posed by the multifacetedness, vigor, ing this problem is likely to be high on the OE’s agenda. Dearth of
and intensity of the competition and the downswings and upswings operations resources is likely to result in high production costs and
of the firm’s principal industry,” Miller and Friesen (1983) found possibly quality problems, directly hurting the firm’s performance.
that relative to samples of poor performers, samples of success- The implication is that gathering and careful analysis of information
ful firms showed a more positive correlation between increases about the resources that need to be developed are relatively more
in environmental hostility and increases in analysis. Slevin and important than the OE’s other activities in this situation. Similarly,
Covin’s (1997) results were similar but Goll and Rasheed (1997) when operations resources and capabilities are highly developed
found that organizational rationality was more strongly associated and have an important and supportive role in the value chain, the OE
with performance in highly munificent environments character- is likely to be facing technological limits to and diminishing returns
ized by high sales growth. from further resource and capability accumulation, implying that
Like Miller and Friesen (1983), we expect that firms in hos- the OE’s time and effort are likely to be more useful elsewhere. In
tile environments typically have to carefully assess their strengths, summary, we expect the OE to spend less time on FD as the relative
resources, and opportunities in order to formulate good strategic strength of the operations function increases.
responses to the difficulties that they face. Hence we expect a sig- Resources and capabilities that are housed in the operations
nificant increase in the need for TLC. In addition, turnarounds in function are likely to create strategic opportunities for the firm
hostile environments are likely to entail realignment and search for (Barney, 1989; Dierickx and Cool, 1989). Hence one can expect the
new synergies, with the implication that additional TLC is likely to OE to assume a more prominent role in the formulation of strategies
be needed in the search for cross-functional coherence (e.g. Clancy and engage in more TLC as the relative strength of the operations
and Kieff, 2004). In contrast, munificent or non-hostile environ- function increases (Wheelwright and Hayes, 1985). However, if
ments are typically those in which firms have to manage rapid operations resources and capabilities are highly developed and
growth, with the implication that speedy and efficient execution serve as an anchor for firm strategies, cross-functional coherence
of the existing strategy is likely the best course of action. with resources in operations might be achieved through the TLC
The impact of environmental hostility on the need for FD is less of executives in other functions with a CEO who is already aware
straightforward. On the one hand, hostility can be a sign of a mature of operations resources, and less through TLC with the OE. In sum-
or even declining industry with low or negative sales growth (Dess mary, the effect of the relative strength of the operations function
and Beard, 1984) in which the development of operations capabili- on TLC is ambiguous, as the OE will have more influence on strategy
ties through FD has reached diminishing returns. On the other hand, but perhaps with less need for communication. Combined with the
the environment can be hostile immediately after the emergence negative effect on FD, these arguments lead to the following pair of
of a dominant design in a growing industry, and the competition hypotheses:
to upgrade production techniques and reduce costs can become
fierce (Jovanovic and MacDonald, 1994; Klepper, 1996). Similarly Hypothesis 4a. Higher relative strength of operations resources
ambiguous predictions can be derived by considering high-growth, and capabilities within the firm is associated with a less positive
munificent environments. Such environments can be seen as call- relationship between the time the OE spends on strategy making
ing for rapid accumulation of operations resources and capabilities and firm performance.
in order to facilitate growth. Alternatively, it can also be that most Hypothesis 4b. Higher relative strength of operations resources
technological and production problems have been solved and the and capabilities within the firm is associated with a more positive
main priority is simple replication of well-established recipes in relationship between the share of top-level communication in the
additional locations or production facilities. Given the positive OE’s strategy making and firm performance.
effect on the need for TLC and the ambiguous effect on the need
for FD, we formulate the following pair of hypotheses: 4. Sample and measures
Hypothesis 3a. Higher environmental hostility is associated with
a more positive relationship between the time the OE spends on We tested the hypotheses using an approach similar to that
strategy making and firm performance. of Miller and Friesen (1983): We investigated whether interac-
tions between changes in firm context and changes in the OE’s
Hypothesis 3b. Higher environmental hostility is associated with strategy-making activities are associated with firm performance.
a more positive relationship between the share of top-level com- This measurement approach is not based on longitudinal data but
munication in the OE’s strategy making and firm performance. relies on retrospection by managers, which has been shown to be
a valid technique for management research (Miller et al., 1997).
3.2. Impact of internal context: relative strength of operations By adapting existing instruments to measure external context
and by constructing new ones for internal context and the OE’s
An important internal contingency that is likely to affect the role in strategy making, we developed a questionnaire and pilot-
OE’s allocation of time and effort is the relative strength of the tested it with five OE’s from different industries. To obtain a sample
408 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413

with large context diversity we generated a web-based version compared to four years earlier. Four items were used to measure the
and an identical hard-copy version of the questionnaire, and we change in the uncertainty of the environment: uncertainty about
used two different databases. In an INSEAD database of past par- customer preferences, competitors, regulators, and suppliers. Five
ticipants in executive programs, we identified approximately 700 items were used to measure the change in complexity: change in
OEs (as apparent from job titles) with valid e-mail addresses and the numbers of competitors, technological designs, customer seg-
collected 125 web-based responses, obtaining a response rate of ments, suppliers, and regulatory bodies. The final set of four items
17%. We also mailed the survey to an additional 900 past partici- was designed to measure the change in hostility of the company’s
pants for whom we did not have a valid e-mail address and 1300 OEs environment: change in survival chances, competitive rivalry, cus-
drawn from an external South-East Asian marketing database. Each tomer loyalty, and profit margins.
of these mailings generated an additional 35 responses. Despite the To extract and score factors previously identified in the liter-
low response rate of the mailings we included the responses in the ature as meaningful and significant, we applied principal factor
sample as no bias between sub-samples was found, non-response analysis on the 13 items. After a standard power-of-three promax
biases are known to rarely impact correlations or regressions (Dey, rotation, which allows for correlation between factors, this analysis
1997), and the removal of the responses from the sample did not generated factor loadings that were generally consistent with the
substantially change the hypothesis test results (reported in an previously validated constructs of (1) complexity, (2) uncertainty
annex available upon request). (or dynamism) and (3) hostility.
To increase the likelihood that strategy-making involved mul- Table 1 shows that the first factor, which we will refer to as
tiple top-level and functional managers, we restricted our sample complexity, loads strongly on the number of competitors, num-
to firms with 50 or more employees. We thus retained one large ber of technological designs and number of customer segments.
data set of 167 observations to extract and score the external con- It also loads strongly on competitor rivalry and negatively on cus-
text factors of complexity, uncertainty and hostility and a smaller tomer loyalty. The complexity factor seems to correspond to the
data set of 134 observations for the testing of hypotheses. The data notion of an increasingly complex, increasingly competitive busi-
set reduction was a result of missing data for relevant variables ness environment. The factor with the second highest eigenvalue
(21 observations) and from eliminating respondents who stated to loads strongly on all measures of increased environmental uncer-
have no responsibilities in operations (9 observations) and respon- tainty and is simply referred to as uncertainty. The factor loadings
dents from business units that had been in business for less than for the third factor provide for a straightforward interpretation
four years (3 observations), with four years being the period over as environmental hostility. It loads negatively on increased sur-
which changes were to be observed. Of those 134 OEs, 51% reported vival chances of a typical company, increased customer loyalty and
directly to the head of a business unit, 25% reported to the head increased profit margins. It also loads positively on increased rivalry
of a region and the remaining 24% reported to the head of opera- and negatively on increased numbers of suppliers and regulators,
tions. The reported number of employees in the business units of consistent with unfavorable market situations. The eigenvalue of
respondents ranged from 50 to 45,000 with a median of 575. The this third factor is below the usual cut-off value of 1 but above
sample contained respondents from companies in a range of indus- 0.7, which has been suggested as a cut-off value by other authors
tries (process industry: 29%, assembly/production: 26%, finance and (Joliffe, 1972). Because of its straightforward interpretation as a
insurance: 10%, transportation and warehousing: 5%, and other ser- commonly used factor in comparable studies, we retain the fac-
vices: 30%) operating in a range of geographies (worldwide: 53%, tor for our main analysis but we also report on tests without
Asia-Pacific: 21%, Europe and Africa: 17%, Americas: 4.5%, Middle it.
East: 4.5%). To test hypotheses regarding the impact of a change in the
relative strength of operations resources and capabilities, the
4.1. Context variables questionnaire also contained three items aimed at measuring
this construct, each one employing a seven-point Likert scale.
To measure the firm’s external context along the dimensions Respondents were asked to indicate their level of agreement (or
of complexity, uncertainty, and hostility, we adapted instruments disagreement) with statements of increased relative contribution
from previous studies in the strategy-making literature (Miller and of the operations function to (1) competitive advantage, (2) firm
Friesen, 1983; Slevin and Covin, 1997). These instruments rely strategic assets, and (3) firm value, compared to four years ear-
on managerial perceptions, a measurement method that has been lier. The Cronbach alpha coefficient for the three items was 0.87,
found to be accurate with respect to objective measures of exter- above the usual cut-off value of 0.7 for construct reliability. The
nal context (Bourgeois, 1985). We posed 13 questions, each using scale is constructed by averaging the three items and is referred to
a seven-point Likert scale to assess the change in external context as strength of operations.

Table 1
Factor loadings for external context variables (N = 167).

Variable (increase in [. . .] during last four years) Complexity factor 1 (e.v. = 1.90) Uncertainty factor 2 (e.v. = 1.06) Hostility factor 3 (e.v. = 0.79)

Number of competitors 0.57a 0.12 −0.04

Number of technological designs 0.57 −0.01 −0.10
Number of customer segments 0.50 0.00 −0.05
Number of regulators 0.36 −0.08 −0.25
Number of suppliers 0.16 0.02 −0.27
Competitor rivalry 0.53 0.05 0.29
Customer loyalty −0.33 0.21 −0.34
Survival chances of typical company 0.15 0.01 −0.52
Profit margins −0.11 −0.16 −0.42
Uncertainty about customer preferences 0.07 0.41 −0.01
Uncertainty about competitors 0.10 0.68 0.06
Uncertainty about regulators 0.03 0.55 −0.04
Uncertainty about suppliers −0.06 0.63 −0.01
a
Variables with loadings > 0.30 are in bold face.
 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413 409

Table 2
Measuring increases in the share of top-level communication.

Questionnaire items Top-level communication (range: −3 to 3) (sum of 3 components below)

Percentage of top-level communication, now and 4 years earlier 1, if percentage of functional deliberation decreased and percentage of top-level
Percentage of functional deliberation, now and 4 years earlier communication increased
−1, if the opposite is indicated
0, otherwise
“For strategy making, I now spend relatively more time researching options by −1, if agreement is indicated
myself or with my staff, while four years ago the focus was more on 0, if neutrality is indicated
discussions with the upper management team” 1, if disagreement is indicated
(7 point Likert scale: strongly disagree – neutral – strongly agree)
“Compared to four years ago, I now spend proportionally less time on strategic −1, if agreement is indicated
discussions with my colleagues and upper management, and more on 0, if neutrality is indicated
strategic decision making within my own team” 1, if disagreement is indicated
(7 point Likert scale: strongly disagree – neutral – strongly agree)

4.2. Role of operations executives in strategy making ways of measuring an increased, steady, or decreased share of TLC
in strategy-making activities (see Table 2).
To obtain a measure for the change in time spent on strategy The three measurements allowed the construction of Guttman
making, we provided the following definition of strategy making: scales with coefficients of reproducibility larger than the suggested
“The process by which a company makes decisions about long term cut-off value of 0.9 (Anderson et al., 1983, p. 259). For example,
commitments (investments, large or long-term contracts, public if respondents’ answers for the three measurements are classified
commitments, etc.)”. Next, we asked the respondents to estimate as agreement or disagreement with an increased share of FD (by
the percentage of time they spent on strategy making, now and four treating neutral answers as disagreement), only 5% of “agreements”
years earlier. To avoid measuring the change in the respondent’s had to switch to “disagreements” or vice versa to create a perfectly
position instead of the position’s role, we asked the respondents reproducible Guttman scale (in which all observations are consis-
who did not hold the same position four years earlier to compare tent with a ranking among the three measurements by which an
their situation with that of their predecessor at that time. This ques- “agreement” on a higher ranked measure implies “agreement” in
tion is comparable to what Miller and Friesen (1983) asked CEOs all lower-ranked measures but not the other way around). If the
when inquiring about change in the strategy making approach of same procedure is applied to construct a Guttman scale indicat-
“top managers”, “senior management”, and “managers”. In other ing an increased share of TLC only 9% of answers needed to be
words, the respondents were asked to recall the strategy-making switched (the differences in the number of switches required is
approaches of other members in their organization in the past. t- simply related to the different treatment of the neutral answers
tests for means also revealed that these respondents (n = 81) did not for the two scales). The coefficients of reproducibility of 0.95 and
differ significantly from others (n = 53) with respect to independent 0.91 for these two Guttman scales indicate that our measures are
variables (four context variables and two strategy making vari- reliably measuring a one-dimensional construct and we performed
ables), except for perceiving a slightly higher increase in hostility, tests of our hypotheses that use these two scales. To leverage the
which can be expected to correlate with shorter tenures. more fine-grained information available in the measurements for
To obtain a scale for the increase (or decrease) in the percentage our main tests, we constructed a combined scale by allowing each of
of time spent on strategy making, we calculated the natural log- the three measurements in Table 2 to take on one of three instead of
arithm of the ratio of the current percentage over the percentage two values, −1, 0, or 1, and by adding these values for each respon-
from four years earlier, a scale we refer to as strategy making. With dent as done in a regular Guttman scale. This scale, which we refer
the use of the ratio of the percentages instead of their difference, an to as top-level communication is thus limited to discrete values ran-
OE who changed from 5% four years earlier to 25% now registers a ging from −3 to +3, with −3 indicating a clear increase in the share
higher score than an OE who changed from 20% to 40%. We take the of FD and +3 indicating a clear increase in the share of TLC.
logarithm of the ratio to obtain a symmetrical scale that does not
assign excess weight to very high or very low ratios. This scale has 4.3. Performance
a mean value of 0.62 and a standard deviation of 0.66. Hence the
average increase in the amount of time spent on strategy making To measure performance, we obtained subjective assessments,
by the respondents over the four years is 86%. a method that is commonly used in comparable studies (e.g. Brews
To obtain estimates of changes in the share of top-level com- and Hunt, 1999; Dess et al., 1997). We asked respondents to com-
munication (TLC) in the OE’s strategy-making activities, we asked pare their business unit’s performance during the last two fiscal
respondents what percentage of their time spent on strategy years with the performance of their industry along three dimen-
making was devoted to TLC and what percentage to functional sions: sales growth, return on sales and return on assets. All three
deliberation (FD), now and four years earlier. The definition for TLC dimensions were measured on a seven-point Likert scale in which
was: “exchanging information with the person you report to and the fourth point corresponded to the industry average. We aver-
with peers or colleagues in other functions”. The definition pro- aged out the three measures to generate an aggregate performance
vided for FD was: “personal research, analysis and decision-making measure. As this measure relates to the respondent’s industry, there
as well as information exchange with staff, subordinates and exter- was no need to obtain industry controls for performance effects.
nal sources”. To assure sufficient reliability for this new construct, However, as performance may be related to organizational size,
we also included two additional questions to measure increases data on the number of employees and sales-range (four ranges:
in the ratio of FD to TLC directly. These last two measures were $0–$10 M, $10–$100 M, $100–500 M, above $500 M) were also col-
obtained directly from items in which the respondents were asked lected. The sales range variable correlated most with performance
to state their level of agreement (or disagreement) with statements so that variable was used as a control for size.
aimed at reflecting an increase in the ratio of FD to TLC (the bottom With performance assessed in the two most current years, and
two left cells in Table 2 contain the exact wording of the two state- changes in context and role of the OE assessed in a four-year
ments). In total, these measurements provided three independent retrospection, the measurements were specified to allow for a
410 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413

Table 3
Descriptive statistics and correlationsd .

Variablesa Mean Standard Complexity Uncertainty Hostility Strength of Strategy Top-level com- Sales Performance
Deviation operations making munication range

Complexityb 0.01 0.84 1

Uncertaintyb 0.02 0.83 0.33 1
Hostilityb 0.00 0.71 −0.11 −0.04 1
Strength of operationsb 1.28 1.16 0.16 0.20 0.00 1
Strategy makingb 0.61 0.66 0.09 −0.01 −0.05 −0.01 1
Top-level communicationb −0.10 1.70 −0.26 −0.12 −0.00 −0.24 −0.11 1
Sales range 3.03 0.92 −0.08 −0.05 −0.07 −0.13 0.05 0.03 1
Performancec 0.76 1.02 −0.02 −0.12 −0.18 0.04 0.14 0.01 0.15 1
a
Responses from OEs in business units 4 years old or older.
b
Change in the last 4 years.
c
During last two fiscal years.
d
Correlations that are significant at p ≤ 0.05 are bold faced.

time-lagged performance effect of adjustments to the OE’s role in role in strategy making (strategy making and top-level commu-
strategy making. nication). The interaction terms include the product of each of the
four context variables with each of the two variables for the OE’s
5. Analysis and results role in strategy making. To control for company size, the sales range
was included in all regressions.
5.1. Correlations To check normality conditions, we performed an information
matrix test for the levels of heteroskedasticity, skewness, and kur-
Table 3 contains the descriptive statistics and correlations of the tosis and had to reject the normality conditions required for OLS
different variables. We comment on the significant correlations, at the p ≤ 0.1 levels for two of our three regression models (Mod-
indicated in bold face in the table. els 1 and 2 in Table 4). Hence we used the Huber/White method
The promax rotation of factors resulted in a correlation between (Huber, 1967; White, 1980) instead of OLS to obtain robust standard
environmental complexity and uncertainty, which is consistent errors for the regression coefficients. To generate coefficients that
with the view that these two conditions frequently occur together. are easy to interpret and to avoid hard-to-interpret coefficients for
There are also positive correlations between the strength of opera- non-interaction terms when interaction terms are significant, all
tions and uncertainty, and between the strength of operations and variables were standardized.
complexity. One potential explanation is that the capabilities of an Table 4 contains the results for three models. The first one does
operations function become more critical in complex and uncertain not contain interaction terms and is the base model. The second
environments, a notion that may be of interest for further research. contains the interactions between context and strategy making,
There is no significant correlation between top-level commu- which are highly significant as a set and thus included in a third
nication and strategy making, supporting the notion that these model. The third model tests for interactions between context and
are independent constructs. Top-level communication correlates the share of TLC in strategy making. The table shows the coefficients
negatively with complexity (p ≤ 0.01), indicating that OEs increase and their t-statistics for each of the independent variables. Variance
the share of FD in more complex environments. This result indi- inflation factors did not reveal any multi-collinearity concerns.
cates that OEs tend to look for solutions to complexity within the The changes in R-squared values indicate that the interaction
operations function. We will see later that, even though common, between changes in context and changes in the role of the OE in
this reflexive response to complexity is associated with low perfor- strategy making can explain significant amounts of variation in
mance. Top-level communication also correlates negatively with perceived performance. R-squared increases from 0.076 in Model
the strength of operations. There could be a tendency for OEs to 1 to 0.252 in Model 3, with all interaction terms included. An F-test
spend more time on FD as the operations function is perceived to on the R-squared values of Models 1 and 2 indicates that the strat-
grow in strength and importance, thereby spending more time on egy making interaction terms are jointly significant at the p ≤ 0.01
what is considered of value to the firm. An alternative hypothesis level. Using the R-squared values of Models 2 and 3, a similar F-test
may be that an operations function growing in value may simply finds the top-level communication interaction terms significant at
draw more attention from the OE, because of the number of issues the p ≤ 0.05 level. The explanatory power of the interaction effects
that have to be resolved. Regardless of the causal explanation, our is comparable to those reported in other studies that gauged the
findings will indicate that this co-movement of context and process interaction between the strategy process, environment and perfor-
is not significantly associated with high performance. mance (Goll and Rasheed, 1997; Priem et al., 1995).
Finally, performance is negatively correlated with hostility. This In Model 3, six of the eight hypotheses are supported at a level of
result is not surprising given that executives at poorly performing p ≤ 0.05 or higher and the coefficients of the interaction terms have
companies may attribute part of their problems to the hostility of the correct sign for the two remaining hypotheses. Fig. 2 summa-
the environment even when the performance is defined relative to rizes the hypotheses and the level of empirical support in Model 3
the competitors in the same industry. (see Table 4).

5.2. Regression analysis 5.3. Robustness tests

To test the hypotheses regarding the role of OEs in strategy mak- To check the robustness of our findings, we tested Model 3 with-
ing, we use moderated regression analysis in which we compare out the hostility factor, with alternative scales for the time spent on
regression models with and without the interaction terms of inter- strategy making (1 alternative) and the share of TLC (2 alternatives),
est. Performance is the dependent variable and the independent with the outliers removed, and without the responses from low
variables include four context variables (complexity, uncertainty, response-rate populations. We report these tests in an annex which
hostility, strength of operations) and two variables for the OE’s is available upon request. As could be expected, significance levels
 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413 411

Table 4
Moderated regressions.

Dependent variable: performance Model 1 Model 2 Model 3

linear regressions (with robust standard errors)
independent variables

ˇ t-stat ˇ t-stat ˇ t-stat

Constant −0.43 −1.56 −0.42 −1.50 −0.43 −1.44

Sales range 0.15 1.64 0.14 1.57 0.15 1.65
Complexity 0.01 0.13 −0.02 −0.19 0.01 0.14
Uncertainty −0.07 −0.75 −0.09 −1.06 −0.07 −0.84
Hostility −0.19* −2.27 −0.20* −2.48 −0.16* −2.14
Strength of operations 0.07 0.63 0.01 0.10 0.01 0.08
Strategy making 0.09 0.90 0.03 0.32 0.02 0.18
OE top-level communication 0.01 0.13 −0.03 −0.41 −0.04 −0.47

Strategy making × complexity 0.17* 2.22 0.22** 2.66

Strategy making × uncertainty 0.06 0.71 0.02 0.33
Strategy making × hostility 0.12 1.53 0.17* 2.10
Strategy making × strength of operations −0.27*** −4.18 −0.26*** −3.96

Top-level communication × complexity 0.25*** 3.41

Top-level communication × uncertainty −0.16* −2.46
Top-level communication × hostility 0.11* 1.98
Top-level communication × strength of 0.02 0.25
operations

Observations 134 134 134

R-squared 0.076 0.184 0.252
F-value 2.08 ‘ 3.30*** 4.49***
*
p ≤ 0.05.
**
p ≤ 0.01.
***
p ≤ 0.001.

Fig. 2. Hypotheses with level of empirical support.

slightly change, but in all of the six robustness checks hypothe- OEs have a tendency to reduce TLC in complex environments. This
ses H1a, H1b, H2b and H4a maintain support at p ≤ 0.1 level or is perhaps not surprising because a larger number of technological
above. Both H3a and H3b maintain the level of support in 5 out designs and customer segments may lead to unfocused factories
the 6 tests. We conclude that the empirical support for six of the (Skinner, 1974), and OEs may then be inclined to increase their
eight hypotheses appears robust. FD to strategize about problems stemming from such complexity.
However, we find that in complex environments, it is an increase in
6. Discussion TLC that leads to better performance (see Table 4). Hence, it appears
that an effective way to avoid the trap of unfocused factories in
Our results provide empirical support for a contingency view complex environments is to infuse the processes for firm-level and
of the OE’s role in strategy making that can ultimately help in operations-function strategizing with sufficient TLC by the OE, an
achieving and sustaining competitive advantage. The results also approach that was not the norm among the firms in our sample.
appear to be the first to empirically validate the important tradeoff The second important insight is related to the negative impact of
between top-level communication (TLC) and functional delib- the relative strength of the operations function on the relationship
eration (FD) in strategy making. In addition, the findings have between the OE’s strategy making and firm performance. This find-
important implications for some widely accepted concepts in oper- ing indicates that it is useful to increase the time spent on strategy
ations strategy and can help us interpret contingency theories and making when the operations function is struggling to build capa-
findings from the strategic planning literature in the particular con- bilities and, vice versa, to focus on coordination, control and other
text of the OE’s job. non-strategic tasks, when the operations function is contributing
The first set of implications relates to the relationship between strongly. Even if intuitive to some top executives, this finding comes
complexity and the share of TLC in the OE’s strategy-making activ- as a useful reminder that when a function is not contributing to
ities. Inspecting pair-wise correlations only (Table 3), we find that competitive advantage, more often than not, the solution lies in
412 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413

revisiting the functional strategy and not in more emphasis on Acknowledgments

execution.
This result also alerts executives not to presume that a pro- We are grateful to Murat Kristal and several anonymous referees
gression along Wheelwright and Hayes’ (1985)four stages implies who have provided feedback on earlier versions of this paper.
a gradual increase in the OE’s time spent on strategy making.
While the OE’s influence over business unit strategy is likely References
to increase, we find that in high-performing companies the OE
tends to spend less time on strategy making and more on other Anderson, A.B., Basilevsky, A., Hum, D.P.J., 1983. Measurement: theory and tech-
tasks as the operations function becomes a source of competitive niques. In: Rossi, P.H., Wright, J.D., Anderson, A.B. (Eds.), Handbook of Survey
Research. Academic Press, London, pp. 244–251.
advantage.
Barney, J., 1986. Strategic factor markets: expectations, luck, and business strategy.
Our findings regarding complexity and uncertainty add insight Manag. Sci. 32 (10), 1231–1241.
to the literature on strategic planning. These two dimensions are Barney, J.B., 1989. Asset stocks and sustained competitive advantage: a comment.
Manag. Sci. 35, 1511–1513.
often treated similarly as they both require an increased amount of
Bourgeois, L.J., 1985. Strategic goals, perceived uncertainty, and economic perfor-
planning (Dess et al., 1997; Hart, 1992; Miller and Friesen, 1983). mance in volatile environments. Acad. Manag. J. 28 (3), 548–573.
We indeed find a correlation between perceived environmental Brews, P.J., Hunt, M.R., 1999. Learning to plan and planning to learn: resolving the
complexity and uncertainty and we find a positive relationship planning school/learning school debate. Strateg. Manag. J. 20 (10), 889–913.
Brown, S., Squire, B., Blackmon, K., 2007. The contribution of manufacturing strategy
between strategy making and firm performance under both con- involvement and alignment to world-class manufacturing performance. Int. J.
ditions (significant for complexity, correct sign for uncertainty). Oper. Prod. Manag. 27 (3), 282–302.
However, we also theorized that the importance of TLC would Clancy, K.J., Kieff, J.R., 2004. Listen and learn. Mark. Manag. 13 (4), 16–18.
Corbett, C.J., Van Wassenhove, L.N., 1993. Trade-offs? What trade-offs? Compe-
increase in complex (H1b) and decrease in uncertain environ- tence and competitiveness in manufacturing strategy. Calif. Manag. Rev. 4 (35),
ments (H2b), marking a clear difference between the two kinds of 107–122.
external conditions. Complexity confronts the firm with difficult de Groote, X., 1994. Flexibility and marketing/manufacturing coordination. Int. J.
Prod. Econ. 36 (2), 153–167.
problems that have many moving parts and dense interactions, Davis, J.P., Eisenhardt, K.M., Bingham, C.B., 2009. Optimal structure, market
which can only be uncovered and mapped through collaboration dynamism, and the strategy of simple rules. Adm. Sci. Q. 54 (3), 413–452.
between different functional experts. Therefore picking a good Dess, G.G., Beard, D.W., 1984. Dimensions of organizational task environments. Adm.
Sci. Q. 29 (1), 52–73.
strategy requires intense communication across the top manage- Dess, G.G., Lumpkin, G.T., Covin, J.G., 1997. Entrepreneurial strategy making and firm
ment team. Moreover, as functional heads go back to their functions performance: tests of contingency and configurational models. Strateg. Manag.
to build resources and capabilities, they need to keep communicat- J. 18 (9), 677–695.
Dey, E.L., 1997. Working with low survey response rates: the efficacy of weighting
ing in order to mitigate the risk of unanticipated interactions. In
adjustments. Res. High. Educ. 38 (2.).
contrast, uncertainty is likely to call for the pursuit of flexibility Dierickx, I., Cool, K., 1989. Asset stock accumulation and sustainability of competitive
of resources and implementation of simple rules and strategies at advantage. Manag. Sci. 35 (12), 1504–1511.
the firm level. Hence the focus shifts toward maintaining general Eisenhardt, K.M., Bhatia, M.M., 2002. Organizational complexity and computation.
In: Baum, J.A.C. (Ed.), The Blackwell Companion to Organizations. Blackwell,
functional excellence through FD. The support we found for these Oxford, pp. 442–466.
hypotheses draws attention to the difference between complex and Eisenhardt, K.M., Martin, J.A., 2000. Dynamic capabilities: what are they? Strateg.
uncertain environments. Manag. J. 21, 1105–1121.
Fayol, H., (C. Storrs, Trans.) 1949. General and Industrial Management. Pitman, Lon-
Our result that more involvement in strategy making by the OE don.
is associated with high performance in hostile environments seems Ferdows, K., De Meyer, A., 1990. Lasting improvements in manufacturing perfor-
broadly consistent with the studies by Miller and Friesen (1983) and mance. J. Oper. Manag. 9 (2), 168–184.
Fine, C.H., Hax, A.C., 1985. Manufacturing strategy: a methodology and an illustra-
Slevin and Covin (1997). The finding that the same is valid for larger tion. Interfaces 15 (6), 28–46.
shares of TLC within this strategy-making activity adds another Fredrickson, J.W., 1984. The comprehensiveness of strategic decision processes:
layer of detail to the optimal response of top management teams extension, observations, future directions. Acad. Manag. J. 27 (3), 445–466.
Fredrickson, J.W., Mitchell, T.R., 1984. Strategic decision processes: comprehen-
to environmental hostility. It appears that challenges in these envi-
siveness and performance in an industry with an unstable environment. Acad.
ronments require strategic solutions and attention to TLC, while Manag. J. 27 (2), 399–423.
the opportunities in non-hostile or munificent environments seem Galbraith, J.R., 1973. Designing Complex Organizations. Addison-Wesley, Reading,
MA.
best leveraged through an emphasis on the development of func-
Galbraith, J.R., 1977. Organizational Design. Addison-Wesley, Reading, MA.
tional resources and capabilities as well as attention to other tasks Goll, I., Rasheed, A.M.A., 1997. Rational decision-making and firm performance: the
such as coordination and control. moderating role of environment. Strateg. Manag. J. 18 (7), 583–591.
To the extent that other functional executives face similar trade- Hart, S.L., 1992. An integrative framework for strategy-making processes. Acad.
Manag. Rev. 17 (2), 327–351.
offs, our analysis seems generalizable to their roles in strategy Hayes, R.H., Pisano, G.P., 1996. Manufacturing strategy: at the intersection of two
making as well and may stimulate similar research in other func- paradigm shifts. Prod. Oper. Manag. 5 (1), 25–41.
tional counterparts to operations strategy. If generalizable, the Hayes, R.H., Wheelwright, S.C., 1979. Link manufacturing process and product life
cycles. Harv. Bus. Rev. 57 (1), 133–140.
analysis may be seen as describing aspects of the strategy mak- Hayes, R.H., Wheelwright, S.C., 1984. Restoring Our Competitive Edge: Competing
ing process by the entire top management team rather than the Through Manufacturing. Wiley, New York.
OE exclusively. Eisenhardt and Martin (2000) have argued that Hill, T.J., 1989. Manufacturing Strategy: Text and Cases. Richard D. Irwin, Homewood,
IL.
dynamic capabilities, which help firms achieve and sustain com- Ho, C.-F., 1996. A contingency theoretical model of manufacturing strategy. Int. J.
petitive advantage (Teece et al., 1997), are sets of specific and Oper. Prod. Manag. 16 (5), 74–98.
identifiable processes. Our results resonate with this view as they as Hough, J.R., White, M.A., 2003. Environmental dynamism and strategic decision-
making rationality: an examination at the decision level. Strateg. Manag. J. 24
they raise the question whether the strategy making process by the
(5), 481–489.
top management team may include more decentralized work with Huber, P.J.,1967. The behavior of maximum likelihood estimates under nonstandard
functional focus in uncertain environments and more collaborative conditions. In: Proceedings of the Fifth Berkeley Symposium on Mathematical
Statistics and Probability, vol. 1. University of California Press, Berkeley, CA, pp.
and cross-functional efforts in those that are complex and hostile.
221–223.
An additional somewhat intriguing question is whether to involve Joliffe, I.T., 1972. Discarding variables in principal components analysis: I. Artificial
the executives from relatively weaker functions more intensely in data. Appl. Stat. 21 (1), 160–173.
strategy making. We leave these questions for further research, not- Jovanovic, B., MacDonald, G.M., 1994. The life cycle of a competitive industry. J. Polit.
Econ. 102 (2), 322–347.
ing that researchers seem to have only begun to unpack this and Klepper, S., 1996. Entry, exit, growth, and innovation over the product life cycle. Am.
other strategically important processes. Econ. Rev. 86 (3), 562–583.
 L. Demeester et al. / Journal of Operations Management 32 (2014) 403–413 413

Klingebiel, R., De Meyer, A., 2013. Becoming aware of the unknown: decision making Priem, R.L., Rasheed, A.M.A., Kotulic, A.G., 1995. Rationality in strategic decision
during the implementation of a strategic initiative. Organ. Sci. 24 (1), 133–153. processes, environmental dynamism and firm performance. J. Manag. 21 (5),
Kristal, M.M., Huang, X., Roth, A.V., 2010. The effect of an ambidextrous supply chain 913–929.
strategy on combinative competitive capabilities and business performance. J. Rosenzweig, E.D., Easton, G.S., 2010. Tradeoffs in manufacturing? A meta-analysis
Oper. Manag. 28 (5), 415–429. and critique of the literature. Prod. Oper. Manag. 19 (2), 127–141.
Loch, C.H., De Meyer, A., Pich, M.T., 2006. Managing the Unknown. John Wiley & Rosenzweig, E.D., Roth, A.V., Dean Jr., J.W., 2003. The influence of an integration
Sons, Hoboken, NJ. strategy on competitive capabilities and business performance: an exploratory
Makadok, R., 2001. Toward a synthesis of the resource-based and dynamic-capability study of consumer products manufacturers. J. Oper. Manag. 21 (4), 437–456.
views of rent creation. Strateg. Manag. J. 22 (5), 387–401. Schroeder, R.G., Bates, K.A., Junttila, M.A., 2002. A resource-based view of manu-
Menda, R., Dilts, D., 1997. The manufacturing strategy formulation process: linking facturing strategy and the relationship to manufacturing performance. Strateg.
multifunctional viewpoints. J. Oper. Manag. 15 (4), 223–241. Manag. J. 23, 105–117.
Miller, C.C., Cardinal, L.B., 1994. Strategic planning and firm performance: a synthesis Simon, H.A., 1947. Administrative Behavior. Macmillan, New York, NY.
of more than two decades of research. Acad. Manag. J. 37 (6), 1649–1665. Siggelkow, N., Rivkin, J.W., 2005. Speed and search: designing organizations for
Miller, C.C., Cardinal, L.B., Glick, W.H., 1997. Retrospective reports in organizational turbulence and complexity. Organ. Sci. 16 (2), 101–122.
research: a reexamination of recent evidence. Acad. Manag. J. 40 (1), 189– Slevin, D.P., Covin, J.G., 1997. Strategy formation patterns, performance, and the
204. significance of context. J. Manag. 23 (2), 189–209.
Miller, D., Friesen, P.H., 1983. Strategy-making and environment: the third link. Skinner, W., 1969. Manufacturing – missing link in corporate strategy. Harv. Bus.
Strateg. Manag. J. 4 (3), 221–235. Rev. 47 (3), 136–145.
Miller, J.G., Roth, A.V., 1994. A taxonomy of manufacturing strategies. Manag. Sci. 40 Skinner, W., 1974. The focused factory. Harv. Bus. Rev. 52 (3), 113–121.
(3), 285–304. Skinner, W., 1985. Manufacturing, The Formidable Competitive Weapon. John Wiley
Mills, J., Platts, K., Gregory, M., 1995. A framework for the design of manufacturing & Sons, New York.
strategy processes. Int. J. Oper. Prod. Manag. 15 (4), 17–49. Swamidass, P.M., Newell, W.T., 1987. Manufacturing strategy, environmental uncer-
Mintzberg, H., 1973. Strategy-making in three modes. Calif. Manag. Rev. 16 (2), tainty and performance: a path analytic model. Manag. Sci. 33 (4), 509–524.
44–53. Swink, M., Narasimhan, R., Wang, C., 2007. Managing beyond the factory walls:
Mintzberg, H., Waters, J.A., 1985. Of strategies, deliberate and emergent. Strateg. effects of four types of strategic integration on manufacturing plant perfor-
Manag. J. 6 (3), 257–272. mance. J. Oper. Manag. 25 (1), 148–164.
Nakane, J., Hall, R.W., 1991. Holonic manufacturing: flexibility – the competitive Teece, D.J., Pisano, G., Shuen, A., 1997. Dynamic capabilities and strategic manage-
battle in the 1990s. Prod. Plan. Control 2 (1), 2–13. ment. Strateg. Manag. J. 18, 509–533.
Nickerson, J.N., Zenger, T.R., 2004. A knowledge-based theory of the firm—the Tushman, M.L., Nadler, D.A., 1978. Information processing as an integrating concept
problem-solving perspective. Organ. Sci. 15 (6), 617–632. in organizational design. Acad. Manag. Rev. 3 (3), 613–624.
Paiva, E.L., Roth, A.V., Fensterseifer, J.E., 2008. Organizational knowledge and the Amazon vice president retires amid firm’s expansion problems., 1999. Wall Street J.
manufacturing strategy process: a resource-based view analysis. J. Oper. Manag. 234 (September (46)), B6 (Eastern Edition).
26 (1), 115–132. Wheelwright, S.C., Hayes, R.H., 1985. Competing through manufacturing. Harv. Bus.
Papke-Shields, K., Malhotra, M., 2001. Assessing the impact of the manufactur- Rev. 63 (1), 99–109.
ing/operations executive’s role on business performance through strategic White, H., 1980. A heteroskedasticity-consistent covariance matrix estimator and a
alignment. J. Oper. Manag. 19 (1), 5–22. direct test for heteroskedasticity. Econometrica 48, 817–830.
Porter, M.E., 1980. Competitive Strategy. Free Press, New York. Womack, J.P., Jones, D.T., Roos, D., 1990. The Machine that Changed the World.
Porter, M.E., 1996. What is strategy? Harv. Bus. Rev. 74 (6), 61–78. Macmillan, New York.

View publication stats