Trade-offs or cumulative enhancement?
A meta-analysis on the relationships between competitive capabilities
and their effect on business performance
Master Thesis, Technology Management
2
Preface
With pleasure and pride I hand you the result of my thesis project for the master program Technology Management. This meta-analytic research is the product of five months of thinking, writing, discussing, coding, discussing and rewriting. In this preface I would like to specially thank the following people:
First, I thank my first supervisor, Boyana Petkova, for her guidance, knowledge and dedication. Secondly, I thank Nick Ziengs, who, though not involved on paper, was a major help in the realization of this thesis. Thirdly, I thank my second supervisor, Dr. Ruël, for the examination of this thesis (much of it in her own free time).
Last, but certainly not least, I would like to thank my family, girlfriend and friends for their endless support and believe. Without you I would not have been able to stand where I stand today. Period.
3
Abstract
Competitive capabilities are defined as the resources that firms can exploit for competitive advantage. Literature is unclear on how competitive capabilities relate to each other. On one hand, authors state that trade-offs exist between competitive capabilities. On the other hand, authors state that the competitive capabilities built upon each other. There is consensus in literature that competitive capabilities have a positive effect on business performance. This meta-analytic research investigates the relationships between the competitive capabilities and shows that trade-offs are rare. The results provide evidence for the existence of cumulative enhancement of competitive capabilities. This research also shows that the effect of competitive capabilities on business performance is generally positive though not in every situation, which challenges the general belief.
4
1. Introduction
Competitive capabilities are defined as the resources that firms can exploit for competitive advantage (Schoenherr et al., 2012). In the light of manufacturing they are defined as components of production competence (Vickery et al., 1993), and are referred to as manufacturing capabilities. In this paper we only use the term ‘competitive capabilities’ and look at them from a manufacturers perspective. Although competitive capabilities have been studied to a great extent in various fields of business literature (Schoenherr et al., 2012, Sarmiento, 2011, Rosenzweig and Easton, 2010, Rosenzweig and Roth, 2004, Safizadeh et al., 2000, White, 1996, Skinner, 1969), there remains considerable debate as how these competitive capabilities relate to each other (Roth, 1996, Ferdows and De Meyer, 1990, Skinner, 1969). Moreover, it is unclear how the competitive capabilities relate to business performance (Schoenherr et al., 2012, Rosenzweig and Easton, 2010, Rosenzweig and Roth, 2004, Ebert and Tanner, 1996, White, 1996). In this paper we aim to increase the understanding regarding these two issues through an extensive quantitative meta-analysis on existing studies.
Literature has been unclear how the competitive capabilities relate to each other. On one side, Skinner (1969) introduced the trade-off model, stating that enhancement of one capability results in a decrease of other capabilities. On the other side, authors theorize that the manufacturing capabilities cumulatively build upon each other and thus enhance each other (Roth, 1996, Ferdows and De Meyer, 1990). There is empirical support for both contradicting perspectives (Flynn and Flynn, 2004, Rosenzweig and Roth, 2004, Pagell et al., 2000, Safizadeh et al., 2000), which indicates different results when looking at different studies. Yet, it is unclear how large this heterogeneity is and what specifically causes it.
Similarly, literature has been unclear how the competitive capabilities affect business performance. Although generally speaking literature agrees that as the competitive capabilities rise, business performance will increase (Schoenherr et al., 2012, Rosenzweig and Easton, 2010, Rosenzweig and Roth, 2004, Ebert and Tanner, 1996, White, 1996), resources spent on the ‘wrong’ competitive capabilities may very well decrease business performance. In order to achieve the highest possible business performance, some authors state that firms should invest according to available resources and based on the firm’s priorities on competitive capabilities (Roth, 1996, Ferdows and De Meyer, 1990). Yet, others argue that it depends on external factors such as consumer preferences which competitive capability is the most important and has the largest effect on business performance (Schoenherr et al., 2012, Porter, 1985). It is unclear what the relationship between the competitive capabilities and business performance is, although a lot heterogeneity in the results is to be expected, given that investment on the wrong competitive capabilities may very well decrease business performance.
5
to quality in order to be successful. In other words, there are reasons that a trade-off between competitive capabilities could exist, simultaneously there are reasons why a trade-off between competitive capabilities should not exist.
By performing a meta-analysis on existing studies that report empirical relationships between the various competitive capabilities and between competitive capabilities and business performance, we try to provide insights in the above stated inconsistencies. By quantitatively combining a large body of existing empirical studies, we are able to shed light onto these issues which are difficult to address by each individual study. The contingency approach used, in order to possibly explain the heterogeneity in the results, makes us to introduce moderating variables to analyze existing studies.
6
2. Theoretical framework
In this section, we will define and describe the competitive capabilities and business performance. We will also lay out theoretical insights that discuss the relationships between the various competitive capabilities, and these competitive capabilities and business performance. As we will explain, combined, these theories suggest heterogeneity in the combined results of multiple studies. Therefore, we will introduce and discuss possible moderating variables that can help explain this heterogeneity. The research questions are stated in this section as well.
2.1 Competitive capabilities
Literature hands us a large quantity of different competitive capabilities, not all are influenced primarily by the manufacturing department. White (1996) starts with thirty-one different capabilities (Vickery et al., 1993) and reduces them to quality, cost, flexibility, delivery dependability, and speed, based on responsibility of manufacturing of at least fifty percent for performance of the capability. The most commonly used competitive capabilities however are quality, delivery, flexibility, and cost (Schoenherr et al., 2012, Rosenzweig and Easton, 2010, Slack et al., 2010, White, 1996).
Quality is about “doing things right”, or, to be consistent in conformance to customers’ expectations (Slack et al., 2010). Delivery means “doing things in time for
customers to receive their goods or services exactly when they are needed, or at least when they were promised” (Slack et al., 2010). Flexibility means “being able to change the operation in some way” (Slack et al., 2010). Change in operation can have four types of
requirements: product/service flexibility, mix flexibility, volume flexibility, and delivery flexibility. Cost is defined as the cost of producing goods and services, or the way a firm efficiently uses its financial inputs (Slack et al., 2010).
The four classical competitive capabilities are complemented in literature by two other competitive capabilities: innovation and sustainability (Weerawardena and Mavondo, 2011, Dangelico and Devashish, 2010). Innovation is defined as the ability to offer new products fast to the market (technical innovation) (Daft and Becker, 1978). Firms compete on (environmental) sustainability since an increase in demand of ‘green products’ is recognized (Dangelico and Devashish, 2010). Given that the ability to produce these green products is an alternative way of having a competitive advantage, being sustainable is recognized as yet another competitive capability. It is defined as using resources to meet the needs of the present without compromising the ability of future generations to meet their own needs (Linton et al., 2007).
7
Table 1: The competitive capabilities used and their basic definitions Competitive capability Basic definition
Quality Be consistent in conformance to customers’ expectations (Slack et al., 2010)
Delivery Doing things in time for customers to receive their goods or services exactly when they are needed, or at least when they were promised (Slack et al., 2010) Flexibility Being able to change the operation in some way
(Slack et al., 2010)
Cost The cost of producing goods and services, or the way a firm efficiently uses its financial inputs (Slack et al., 2010)
Innovation The ability to offer new products fast to the market (Daft and Becker, 1978)
Sustainability Using resources to meet the needs of the present without compromising the ability of future generations to meet their own needs (Linton et al., 2007)
2.2 Business performance
Venkatraman and Ramanujam (1986) state “The narrowest conception of business
performance centers on the use of simple outcome-based financial indicators that are assumed to reflect the fulfillment of the economic goals of the firm”. These indicators are
typically sales growth, profitability (e.g. ROI) and earnings per share (Venkatraman and Ramanujam, 1986). Other generally used measures are return on assets (Zhou et al., 2008), the firm’s profit level (Rosenzweig and Roth, 2004), and market share (Kaynak and Hartley, 2008). In the context of competitive capabilities, business performance is mostly defined financially as well. White (1996) for instance uses Return On Investment (ROI) as a business performance measure. This is in line with the work of Venkatraman and Ramanujam (1986) when stating that financial performance is most used as a business performance measurement in strategy research.
2.3 Contradicting theories
Skinner (1969) was the first to suggest that organizations were able to be competitive with the use of competitive capabilities. He stated however that such capabilities were affected by trade-offs; organizations could not pursue multiple competitive capabilities at once. This resulted in the concept of a so called ‘focused factory’ that strategically focused on only one capability at a time (Skinner, 1974). The focused factory theory was further stressed by Hayes and Wheelwright (1984) when stating that it could be potentially dangerous for the survival of a company to pursue high performance on multiple competitive capabilities. Based on this new theory several authors have performed cross-sectional survey studies focused on a sole competitive capability and its relationship with overall performance of the firm (i.e.: Swamidass and Newell, 1987).
8 first, followed by dependability, speed and eventually cost efficiency. Ferdows and De Meyer (1990) mention Skinners trade-off model in their writings and state: “According to this
theory, unless there is slack in the system, improvement of one of the generic capabilities is possible only at the expense of the others”. This is followed by the notion that firms operating
according to industry standards will not have slack in their system. Resources will have to be made available in order to improve on a capability without decreasing performance on other capabilities. This process takes time and therefore they conclude that firms have to increase performance on multiple capabilities in a specific order so to come to the best overall performance possible. This is line with the work of Schmenner and Swink (1998) when they provide the theory of performance frontiers. Such a frontier is defined as “the maximum
performance that can be achieved by a manufacturing unit given a set of operating choices”.
By providing these theories, Schmenner and Swink (1998) mention that it depends on whether or not there is slack in the system a firm will experience trade-offs.
Porter (1985) provides another view on the matter with the introduction of his generic competitive strategies. In order to be a high performing firm, one should choose a competitive strategy depending on environmental forces. These competitive strategies can lead to trade-offs within the competitive capabilities and to improved performance on multiple capabilities simultaneously.
Yet another theory combines the theories of trade-offs and cumulative capabilities (as is the sand-cone model), naming it “a hybrid model of competitive capabilities” (Hallgren et al., 2011). The hybrid model shows that quality and delivery can be cumulatively enhanced where after cost and flexibility are enhanced separately since these capabilities do not relate to each other significantly.
Given the contradicting theories outlined above it is unclear whether or not there are trade-offs present in the relationships between the competitive capabilities and whether increased performance of a competitive capability results in higher overall business performance. Hence, we aim to answer the following research questions:
1. a) Are there trade-offs present in the relationships between the competitive capabilities or do they enhance each other?
b) Do the competitive capabilities positively affect business performance?
Since theory suggest different results, it is expected that these relationships are heterogeneous when numerous studies on the matter are taken into account, which is the case when performing a meta-analysis. We will explore the heterogeneity and test it with use of moderating variables.
2.4 Moderating variables
The contingency approach leads us to use a number of moderating variables that can help explain the expected heterogeneous results. Rosenzweig and Easton (2010) used a similar approach and looked at the influence of research related variables such as the unit of analysis. In our research we look at environmental variables instead to explain the heterogeneity.
9
capabilities follow different patterns in differentially industrialized environments. This is the first reason to use the region of the firms researched in the competitive capabilities studies as a moderating variable. Rosenzweig and Easton (2010) too used geographical region as a possible source of influence and base their reasons on the notion that different regions use different initiatives to increase performance of manufacturing firms. Asian-pacific manufacturers for instance started to use JIT and TQM quite consistently from 1980 on. In line with this we use geographical region as a moderating variable.
Secondly, we use time as a moderating variable. The rise of literature on the trade-offs in competitive capabilities which started with Skinner (1969) provides the reason to do so. The theory of trade-offs from 1969 showed that there was evidence for a pure trade-off model existing in the performance of competitive capabilities. Studies such as the one of White (1996) provide evidence for this trade-off model not to exist. Moreover, the same can be said here as with the moderating variable of geographical region; since manufacturers started using different initiatives throughout the years, it can be that as a result relationships between the capabilities changed. Hence it is possible that the year the research was conducted has an influence on the relationships of investigation.
Thirdly, we use type of business as a moderating variable. An early investigation of the sample used in this research showed that there are studies that focus solely on manufacturing firms and others that focus on manufacturing and service firms. Another possible source of heterogeneity lies within the differences between these two types of firms. Daft (2007) shows that in the service industry rapid response time is generally necessary but also that quality is difficult to measure. Though there are far more differences between the types, this is reason enough to be a possible source of heterogeneity. Since the importance of the various competitive capabilities are different it is possible that this results in different correlations between the competitive capabilities.
Given the introduction of the moderating variables of time, geographical region, and type of business, we first need to know the amount of heterogeneity in the results. Secondly we have to find whether this heterogeneity can indeed be explained by these moderating variables. Hence, we state the following research questions:
2. a) How heterogeneous are the results for the relationships between the competitive capabilities and between the competitive capabilities and business performance?
10
3. Methodology
For answering our research questions we perform a meta-analysis of correlations. This is a procedure to systematically analyze existing studies and thereby refining and extending theory (Nair, 2006). It enables us to describe the distribution of correlations between two given variables. To do so we use existing studies and extract the usable data from these studies.
The studies are not identical when it comes to research methods and various errors might be present. It is important to determine these errors and adjust the outcomes accordingly by using study information such as sample size and reliability estimates (Nair, 2006). Each of the studies used in this paper provides one or multiple correlation(s) to estimate an association between constructs. Due to sampling errors and other artifacts the found correlation is only an estimate of this association. Meta analytical research provides us with relations that are closer to the true associations (Hunter and Schmidt, 2004). Hunter & Schmidt (2004) provide eleven artifacts that affect the true correlation between constructs of which the artifact ‘sampling error’ is one of the most commonly observed. Given that a meta-analysis uses a large body of existing studies the sampling errors effectively cancel each other out because of the randomness of the errors in the individual studies (Nair, 2006).
The remainder of this section will first describe the formation of the sample, and the sample itself. Thereafter we will describe, in two separate stages, the methodological procedures needed to come to answering our research questions.
3.1 Sample
The sample consists of data extracted from studies researching or providing (among other things) one or more relationship(s) between competitive capabilities. We searched the abstracts of thirteen key Operations Management journals (see table 2). We used the keywords stated in table 3 and table 4 where ‘OR’ is used for the terms within each table and ‘AND’ is used for combining the terms of both tables. There was no limitation on when these studies were performed. The electronic search resulted in a total of 1450 studies which were further reviewed thereafter. Scanning of these papers on correlations between constructs used in our research resulted in 109 studies. Finally these 109 studies were carefully examined on constructs and underlying items. This resulted in a total of 50 papers of which we extracted the usable data. This amount of studies is more than previous meta-analytic investigation (Mackelprang and Nair, 2010, Rosenzweig and Easton, 2010). The data extracted consists of the effect sizes, concepts of the reported competitive capabilities, construct reliability measures (either Cronbach’s Alphas or composite reliability measures), year of research, type of business (manufacturing or service), and country of studied firms. Based on the reported concepts of the competitive capabilities we classified the effect sizes in the competitive capabilities of this research. The concepts that provide the basis for this classification can be found in tables 5 and 6 for the competitive capabilities and business performance respectively. These steps were done separately by two researchers to inter-rater reliability of 96%. On the remaining 4% of the effect sizes a discussion lead to a consensus regarding the classification of the corresponding competitive capabilities.
11
items) and ‘process flexibility’ (four items) of which we calculated the average. A similar alteration is made for relationships between competitive capabilities and business performance when studies report two measures of business performance that are both within the definition of business performance as is outlined in the theoretical section of this paper. This is for instance the case for the study of Kristal et al. (2010) where correlations are given between competitive capabilities and both ‘profit level’ and ‘market share’. The same yields for studies reporting separately on ‘speed’, which, in our research is part of the competitive capability ‘delivery’. This is for instance the case in the study of Avella et al. (2011).
For the studies reporting no construct reliability, because of the construct depending on mere two items for instance, we provided a reliability measure that is calculated as the average of the construct reliabilities given in all other studies concerning the same construct.
Although we liked to research ‘sustainability’ as a competitive capability as well, the found studies provided too little results to come to a conclusion regarding sustainability in relation with other capabilities and in relation with business performance.
Since there are two studies that make use of the same sample database with the same sample size (Menor et al., 2007, Rosenzweig and Roth, 2004), and since the weight of a relationship between constructs depends on this sample size (Mackelprang and Nair, 2010), we divided the sample size for the results of these papers regarding the relationships between identical constructs in both papers by two.
Furthermore, subgroups are created using the moderating variables of year of research, region of the researched firms, and type of business (manufacturing and service). The latter moderating variable is scrapped due to the fact that there are very few studies reporting results based on data of both manufacturing and service firms, and there are none that report results based on data from solely service providing firms. Subgroups based on region of the researched firms are created with the regions being USA, Europe, Asia, and the rest (table 6). Subgroups based on year of research are created with the timeframes being 1990-1999, 2000-2009, and 2010-2012. This definition of subgroups leads to reasonably sized groups which do not differ too much in terms of size.
Table 2: Operations Management journals searched and number of studies found per journal
Journal Number of studies
Journal of Operations Management 16
International Journal of Operations & Production Management 9 International Journal of Production Economics 7 International Journal of Production Research 5
Decision Sciences 3
Production and Operations Management 3
Strategic Management Journal 3
Total Quality Management 2
International Journal of Physical Distribution and Logistics Management 1 Manufacturing and Service Operations Management 1 International Journal of Logistics Management 0 International Journal of Purchasing and Materials Management 0
Journal of Supply Chain Management 0
Management Science 0
12
Table 3: Search terms Table 4: Search terms
13
Table 5: The constructs of this research, the underlying concepts and the studies using these concepts
Construct Concepts Studies
Quality Product performance Avella et al. (2011), Narasimhan et al. (2010), Kristal et al. (2010), da Silveira and Sousa (2010), Squire et al. (2006), Boyer and Lewis (2002), Prajogo et al. (2008), Kaynak and Hartley (2008), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Koufteros and Marcoulides (2006), Vachon and Klassen (2008), Ariss and Zhang (2002), Filippini (1998), Ward et al. (1995)
Product durability Avella et al. (2011), Narasimhan et al. (2010), Arauz et al. (2009), Kristal et al. (2010), da Silveira and Sousa (2010), Oltra and Flor (2010), Squire et al. (2006), Prajogo et al. (2008), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Oltra et al. (2005), Koufteros and Marcoulides (2006), Vachon and Klassen (2008)
Product reliability Avella et al. (2011), Li et al. (2011), Narasimhan et al. (2010), Arauz et al. (2009), Kristal et al. (2010), Wong et al. (2011), da Silveira and Sousa (2010), Boyer and Lewis (2002), Prajogo et al. (2008), Größler and Grübner (2006), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Koufteros and Marcoulides (2006), Ward et al. (1995)
Conformance to design specifications
Avella et al. (2011), Narasimhan et al. (2010), Kristal et al. (2010), da Silveira and Sousa (2010), Rosenzweig and Roth (2004), Oltra and Flor (2010), Squire et al. (2006), Boyer and Lewis (2002), Prajogo et al. (2008), Größler and Grübner (2006), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Vachon and Klassen (2008)
Percentage returned defective
Arauz et al. (2009), Wong et al. (2011), Oltra and Flor (2010), Squire et al. (2006), Kaynak and Hartley (2008), Dean Jr. and Snell (1996), Oltra et al. (2005), Kaynak (2003), Mapes (1997), Ward et al. (1995), McKone et al.(2001)
Product meets market needs Zhou et al. (2008), Wong et al. (2011), Rosenzweig et al. (2003), Koufteros and Marcoulides (2006)
Dependability Speed of delivery Avella et al. (2011), Malhotra and Mackelprang (2012), Narasimhan et al. (2010), Wong et al. (2011), Kristal et al. (2010), da Silveira and Sousa (2010), Oltra and Flor (2010), Squire et al. (2006), Boyer and Lewis (2002), Größler and Grübner (2006), Dean Jr. and Snell (1996), Oltra et al. (2005), Danese and Kalchschmidt (2011), Ariss and Zhang (2002), Narasimhan and Jayaram (1998), Ward and Duray (2000), Ward et al. (1995)
Average lead time Avella et al. (2011), Wong et al. (2011), Kristal et al. (2010), da Silveira and Sousa (2010), Squire et al. (2006), Größler and Grübner (2006), Danese and Kalchschmidt (2011), Vachon and Klassen (2008), Mapes (1997), McKone et al.(2001)
Reliability of delivery Avella et al. (2011), Malhotra and Mackelprang (2012), Narasimhan et al. (2010), Wong et al. (2011), da Silveira and Sousa (2010), Rosenzweig and Roth (2004), Oltra and Flor (2010), Squire et al. (2006), Boyer and Lewis (2002), Größler and Grübner (2006), Rosenzweig et al. (2003), Oltra et al. (2005), Danese and Kalchschmidt (2011), Vachon and Klassen (2008), Ariss and Zhang (2002), Narasimhan and Jayaram (1998), Ward and Duray (2000), Ward et al. (1995)
Percentage delivered on time
Wong et al. (2011), Squire et al. (2006), Dean Jr. and Snell (1996), Filippini (1998), Mapes (1997), McKone et al.(2001)
Flexibility Volume flexibility Avella et al. (2011), Narasimhan et al. (2010), Wong et al. (2011), Kristal et al. (2010), da Silveira and Sousa (2010), Rosenzweig and Roth (2004), Oltra and Flor (2010), Menor et al.(2007), Squire et al. (2006), Boyer and Lewis (2002), Größler and Grübner (2006), Avittathur and swamidass (2007), Rosenzweig et al. (2003), Boyer et al. (1997), Vachon and Klassen (2008), Liu et al. (2009), Ariss and Zhang (2002), Narasimhan and Jayaram (1998)
Mix flexibility Avella et al. (2011), Patel et al. (2012), Malhotra and Mackelprang (2012), Narasimhan et al. (2010), Camisón and Villar (2010), Wong et al. (2011), Kristal et al. (2010), da Silveira and Sousa (2010), Menor et al.(2007), Squire et al. (2006), Boyer and Lewis (2002), Größler and Grübner (2006), Avittathur and swamidass (2007), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Boyer et al. (1997), Vachon and Klassen (2008), Liu et al. (2009), Narasimhan and Jayaram (1998)
Process flexibility Avella et al. (2011), Patel et al. (2012), Camisón and Villar (2010), Kristal et al. (2010), Oltra and Flor (2010), Menor et al.(2007), Avittathur and swamidass (2007), Rosenzweig et al. (2003), Boyer et al. (1997), Liu et al. (2009), Ward and Duray (2000), Ward et al. (1995)
Customizability Narasimhan et al. (2010), Camisón and Villar (2010), Wong et al. (2011), Oltra and Flor (2010), Menor et al.(2007), Boyer and Lewis (2002), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Boyer et al. (1997), Ariss and Zhang (2002)
Cost Production cost Hollos et al. (2012), Wong et al. (2011), Kristal et al. (2010), Rosenzweig and Roth (2004), Grawe et al. (2009), Boyer and Lewis (2002), Oltra et al. (2005), Vachon and Klassen (2008), Ward and Duray (2000)
Manufacturing cost Avella et al. (2011), Narasimhan et al. (2010), Kristal et al. (2010), Squire et al. (2006), Größler and Grübner (2006), Rosenzweig et al. (2003), Danese and Kalchschmidt (2011), McKone et al.(2001), Narasimhan and Jayaram (1998)
Product cost Hollos et al. (2012), Narasimhan et al. (2010), Wong et al. (2011), Kristal et al. (2010), Rosenzweig and Roth (2004), Oltra and Flor (2010), Dean Jr. and Snell (1996), Rosenzweig et al. (2003), Danese and Kalchschmidt (2011), Vachon and Klassen (2008), Ariss and Zhang (2002), Ward et al. (1995)
Labor cost Avella et al. (2011), Hollos et al. (2012), Oltra and Flor (2010), Boyer and Lewis (2002), Größler and Grübner (2006), Dean Jr. and Snell (1996), Vachon and Klassen (2008)
Overhead cost Narasimhan et al. (2010), Wong et al. (2011), Oltra and Flor (2010), Größler and Grübner (2006), Ward et al. (1995)
Inventory cost Avella et al. (2011), Wong et al. (2011), Boyer and Lewis (2002), Größler and Grübner (2006), Ward and Duray (2000), Ward et al. (1995)
Innovation Number of newly introduced products
Hsu and Sabherwal (2012), Malhotra and Mackelprang (2012), Chen et al. (2010), Narasimhan et al. (2010), Martínez-Costa and Martínez-Lorente (2008), Menor et al.(2007), Prajogo et al. (2008), Koufteros et al. (2001), Sadikoglu and Zehir (2010), Koufteros and Marcoulides (2006), Mapes (1997)
Lead time to introduce new products
Hsu et al. (2011), Chen et al. (2010), Narasimhan et al. (2010), Camisón and Villar (2010), Arauz et al. (2009), Prajogo et al. (2008)
First to market Hsu et al. (2011), Prajogo et al. (2008), Sadikoglu and Zehir (2010)
Offer innovative products Hsu et al. (2011), McDermott and Prajogo (2012), Camisón and Villar (2010), Arauz et al. (2009), Menor et al.(2007), Grawe et al. (2009), Prajogo et al. (2008), Koufteros et al. (2001), Koufteros and Marcoulides (2006)
14
Table 6: Business performance, the underlying concepts and the studies using these concepts
3.2 Meta-analytic procedures
Similar to other meta-analyses performed in the field of Operations Management (Mackelprang and Nair, 2010, Nair, 2006), we use the meta-analytic procedures of Hunter and Schmidt (2004). The procedure is divided into two stages. The first stage examines the relationships between the competitive capabilities and between the competitive capabilities and business performance so to answer research questions 1a and 1b. The second stage investigates the heterogeneity of combined study results and aims to explain this heterogeneity by use of the moderating variables of year of research and region of the researched firms so to answer research questions 2a and 2b.
3.2.1 Stage-I
Stage-I considers the relationships between the competitive capabilities, and between the competitive capabilities and overall business performance. This stage consists of 10 steps which are outlined in table 7 (steps 10 and 12 are only used for stage-II calculations). We can detect whether the population correlations significantly differ from zero by evaluating the ratio of the average population correlations divided by the correlations’ standard deviations (Hunter and Schmidt, 2004). This ratio is also known as RATIO1 (Mackelprang and Nair, 2010, Nair, 2006), and the result of this calculation should be equal to or greater than two to safely state that the population correlations at hand are not lesser than or equal to zero. This will show that the mean correlation of the population at hand is approximately 2.0 standard deviations above nil. We calculate RATIO1 for all relationships between the competitive capabilities and for the relationships between the competitive capabilities and business performance. By doing so we can evaluate whether there are possibly trade-offs present in the relationships between the competitive capabilities (RATIO1 results are lower than two) and whether the competitive capabilities have a positive effect on business performance (RATIO1 results are equal to or higher than two). The same can be said the other way round (there are no trade-offs present when RATIO1 scores are equal to or higher than two for the relationships between the competitive capabilities and the competitive capabilities do not always have a positive effect on business performance when RATIO1 scores are lower than two).
Construct Concepts Studies
Business performance
Return On Investment (ROI)
Chen et al. (2010), Kristal et al. (2010), Kaynak (2003), Kaynak (2003), Boyer et al. (1997), Narasimhan and Jayaram (1998)
Return On Assets (ROA) Li et al. (2011), Zhou et al. (2008), Oltra and Flor (2010), Rosenzweig et al. (2003)
Return On Sales (ROS) Li et al. (2011), Kristal et al. (2010), Boyer et al. (1997), Merschmann and Thonemann (2011), Narasimhan and Jayaram (1998)
Profit level Li et al. (2011), Hsu et al. (2011), Hsu and Sabherwal (2012), Patel et al. (2012), McDermott and Prajogo (2012), Kristal et al. (2010), Martínez-Costa and Martínez-Lorente (2008), Rosenzweig and Roth (2004), Oltra and Flor (2010), Grawe et al. (2009), Kaynak (2003), Avittathur and swamidass (2007), Dean Jr. and Snell (1996), Kaynak (2003)
Market share Hsu et al. (2011), McDermott and Prajogo (2012), Camisón and Villar (2010), Kristal et al. (2010), Martínez-Costa and Martínez-Lorente (2008), Menor et al.(2007), Grawe et al. (2009), Kaynak (2003), Kaynak (2003)
15
Table 7: Methodological steps, their input variables, formulas and purposes
Process steps Input variables Formula Purpose
Step 1. Attenuation factor
1a. Reliability of the first competitive capability (or business performance) ( )
1b. Reliability of the second competitive capability (or business performance) ( )
The attenuation factor is
used to correct the correlation for measurement error, create the error variance across studies and to weight the studies. Step 2. Correct
study correlations
2a. Attenuation factor ( )
2b. Study correlations ( ) The corrected correlations are used in calculating RATIO1, which is used to identify significant population correlations. Step 3. Individual
study weights
3a. Study sample size ( )
3b. Attenuation factor ( ) The study weight is used to find the average corrected correlations, average error variances and variance of the corrected correlations. Step 4. Corrected
study sampling error
4a. Weighted sample mean correlations ( ̅)
4b. Study sample size ( ) 4c. Attenuation factor ( )
̅ Each study’s corrected sampling error variance is used to calculate the weighted mean sampling error variance across studies.
Step 5. Weighted mean sampling error variance
5a. Study weight ( ) 5b. Study error variances ( )
̅ The weighted mean error variances is used to estimate the population standard deviation.
Step 6. Weighted mean corrected correlations
6a. Study weight ( )
6b. Corrected study correlations ( ) ⃑ The corrected correlations is weighted mean used to find both the variance of the corrected correlations as well as RATIO1
Step 7. Variance of the corrected correlations
7a. Study weight ( )
7b. Corrected study correlations ( ) 7c. Weighted mean corrected correlations ( ⃑)
[ ⃑] The variance of the
corrected correlations is used to estimate the population SD
Step 8. Estimate the population SD
8a. Variance of the corrected ( )
correlations
8b. mean error variances ( ̅)
[ ̅ ] The estimate of the
population standard deviation is used to calculate RATIO1
Step 9. Calculate RATIO1
9a. Average corrected correlations ( ⃑)
9b. Estimated population standard deviation ( )
⃑ RATIO1 values greater than 2 imply that a positive correlation exists between the variables considered Step 10. Calculate
RATIO2
10a. Weighted mean sampling error variances ( ̅)
10b. Variance of the corrected correlations ( )
̅ RATIO2 values greater than
or equal to 0.75 imply that there is only one population correlation and that the relationship is not subject to moderating factors
Step 11. Credibility interval
11a. Estimated population standard deviation ( )
11b. Average corrected correlations ( ⃑)
11c. Z-value of desired credibility level ( )
⃑
The credibility interval returns the endpoints whereby the percentage selected of the values in the correlation distribution are contained
Step 12. Heterogeneity
12a. Study weight ( )
12b. Corrected study correlations ( ) 12c. Weighted mean corrected correlations ( ⃑)
12d. degrees of freedom ( )
[ ⃑] [ ⃑]
16
3.2.2 Stage-II
Stage-II considers the evaluation of the heterogeneity of the results. We first need to assess the extent of this heterogeneity to thereafter possibly explain it with use of the moderating variables of year of research and region of the researched firms. We use RATIO2 ( ̅ ; where
̅ is the corrected estimate of the sampling error variability, and is the corrected estimate of
the study correlation variability) to detect if moderation effects have an influence on the correlation results of stage-I (Mackelprang and Nair, 2010, Nair, 2006). When the score is greater than or equal to 0.75, it is reasonably safe to state that the calculated correlation represents the true population correlation. If it is less than 0.75 it can be concluded that there are moderating variables that affect these correlations (Nair, 2006). This is based on Orlitzky et al. (2003)when stating that “if 75 per cent or more of the observed variance of correlations
across studies is due to artifacts, then probably all of it is artifactual variance (on the grounds that the remaining 25 per cent is likely due to artifacts not corrected for). Thus, in cases where 75 per cent or more of the variance is explained by artifacts, including sampling error variance, moderators are unlikely to have caused a real variation in observed correlations”. Thus, RATIO2 scores show whether or not the results are heterogeneous. As
with RATIO1 calculations, we calculate RATIO2 for all relationships between the competitive capabilities and for the relationships between the competitive capabilities and business performance.
We also need to assess the amount of heterogeneity when the results are in fact heterogeneous. For this we use the measure (Huedo-Medina et al., 2006). It results in a percentage which indicates the amount of heterogeneity of the used sample. generally is a complement to Cochran’s Q statistic, on which it is build, and is given to overcome the issues of power of the Q statistic when sample sizes are low (Huedo-Medina et al., 2006). Since most of our researched relations, especially when moderating variables are introduced, have a small number of studies, we solely provide the measure to provide conclusions on the amount of heterogeneity. Since a benchmark of is missing for our and related research we use the benchmark of Huedo-Medina et al. (2006), which states that an percentage of 75%, 50% and 25% means a high, medium and low heterogeneity respectively (we use low heterogeneity , medium heterogeneity , and high heterogeneity ). When the calculations of result in less than nil per cent the variations in the results are a product of chance and is scored 0% (Huedo-Medina et al., 2006). By introducing the found moderating variables and dividing the sample for each relation in groups we aim to explain the heterogeneity in the results.
17
18
4. Results
This section provides the results of the meta-analytic procedures as described in the methodology section and answers the research questions stated in the theoretical framework. First the results of stage-I are provided followed by the results of stage-II.
4.1 Stage-I results
The stage-I results for the relationships between the competitive capabilities are presented in table 8 where the results for the ratios that reach the before mentioned thresholds are underlined and put in bold. For for-instance the relationship ‘quality – dependability’ the number of studies found to report on this relation is 23. The overall combined sample size is 5527. It has a weighted mean corrected correlation of 0.485 with an estimated standard deviation of 0.236. The credibility interval ( ⃑ , with 1.96) for this relation is from 0.008 to 1.000 and the sample has a minimum and maximum of 0.017 and 1.000 respectively. The calculation of RATIO1 results in 2.061, which is above the threshold of 2, hence it is underlined and put in bold.
Table 8: Stage-I results for the relationships between the competitive capabilities
As can be seen from the RATIO1 scores in table 8, for six of the ten relations we can be reasonably sure that the correlation is positive, hence there is no trade-off found between the constructs of these relations. For the relationships where the corresponding RATIO1 scores do not surpass the threshold of 2, the weighted mean corrected correlations are not significantly higher than nil. This answers research question 1a (are there trade-offs present in the relationships between the competitive capabilities?) in the following way: only when the competitive capability ‘cost’ is involved can there be trade-offs in the performance with respect to the other competitive capabilities. Though the mean weighted corrected correlations for the relationships involving cost are all above nil, the credibility- and min/max intervals both show that there are circumstances in which a trade-off is possible and there are circumstances in which no trade-off will be found.
19
A similar table of results is constructed for the relationships between the competitive capabilities and business performance (table 9).
Table 9: Stage-I results for the relationships between the competitive capabilities and business performance
As can be seen from the RATIO1 scores of table 9, for only the competitive capabilities ‘flexibility’ and ‘innovation’ we can be reasonably sure that they have a positive effect on business performance. The weighted mean corrected correlations between quality and business performance, dependability and business performance, cost and business performance and between all capabilities combined and business performance are not significantly higher than nil. This answers research question 1b (Do the competitive capabilities positively affect business performance?) in the following way: flexibility and innovation have a positive effect on business performance no matter the situation. Quality, dependability and cost do not have a positive effect on business performance in all possible situations.
4.2 Stage-II results
Stage-II results considering the heterogeneity of the results of relations between the competitive capabilities and between the competitive capabilities and business performance are given in table 10. The results of the RATIO2 scores that reach the threshold of 0.75 are underlined and put in bold. The table furthermore shows the scores of each relation.
As can be seen from the RATIO2 scores of table 10, only for the relationship ‘flexibility – innovation’ we can state that the results are homogeneous. For all other relationships the results are heterogeneous. The same is reflected in the scores. We see a score of 0% for the relationship ‘flexibility – innovation’. The results of the relationship ‘flexibility – business performance’ have a medium score of heterogeneity (57%). Results of all other relationships are highly heterogeneous, all scoring above 75% on the measure. This answers research question 2a (How heterogeneous are the relationships between the competitive capabilities and between the competitive capabilities and business performance?) in the following way: results for all relationships are highly heterogeneous except for the
20 relationship between flexibility and business performance, which shows a medium amount of heterogeneity, and for the relationship between flexibility and innovation, which shows no signs of heterogeneity.
Table 10: RATIO2 and I2 results for all relationships
21
4.2.1 Region of the researched firms
We first consider the moderating variable of region of the researched firms. Sometimes dividing the sample into sub-groups based on geographical region makes the RATIO1 scores reach the threshold of two, while in contrast, the scores on RATIO1 for the relationship in general did not (see figure 1). This is especially noticeable for the relation between flexibility and cost (see table 12). The amount of heterogeneity, based on the scores of the measure, lowers for this relation when it is divided into subgroups (see figure 1). In general, considering geographical region the RATIO1 scores are above the threshold of two in 32 out of 49 results (63%) versus 7 out of 15 (47%) results considering no moderating variables. RATIO2 scores are now above the threshold of 0.75 in 12 out of 49 results (24%) versus none of the results when not considering moderating variables. The amount of heterogeneity, based on the measure, is nil in 9 out of 49 results (18%), low in 8 out of 49 results (16%), medium in 9 out of 49 results (18%), and high in 23 out of 49 results (47%) (together these percentages reach 99% due to rounded off percentages). Considering no moderating variables the results show medium heterogeneity in 1 out of 15 times (7%) and high heterogeneity in 14 out of 15 times (93%). This answers research question 2b (Can the heterogeneity be explained by the moderating variables of time, geographical region, and type of business?) for region of the researched firms in the following way: geographical region explains the heterogeneity in the results to a certain amount. Heterogeneity decreases with use of this moderating variable, however it has not disappeared completely. The Kurskal-Wallis test shows moderate significance on the relation between flexibility and cost (H(3) = 6.429, P = 0.092) with a mean weighted corrected correlation of 0.454 for the group USA, 0.457 for the group Europe, 0.624 for the group Asia and 0.250 for the rest. Other relations do not report significant results on the this test. Therefore the correlations between the subgroups for all relations, except for the relationship between flexibility and cost, do not differ significantly from each other.
Figure 1: Percentages of results reaching thresholds (RATIO1 and RATIO2) and percentages of results per class of heterogeneity (based on the result of the I2 measure) for the general sample and for the subgroups based on region of the researched firms
47% 0% 0% 0% 7% 93% 65% 24% 18% 16% 18% 47% 0% 20% 40% 60% 80% 100%
RATIO1 RATIO2 No heterogeneity Low heterogeneity Medium heterogeneity
High heterogeneity
22
4.2.2 Year of research
Secondly, we consider the moderating variable of year of research. In general, RATIO1 scores are above the threshold of two in 24 out of 42 results (57%) versus 7 out of 15 results (47%) when not considering moderating variables (see figure 2). RATIO2 scores are now above the threshold of 0.75 in 7 out of 42 results (17%) versus none of the results when not considering moderating variables (see figure 2). The amount of heterogeneity, based on the measure, is nil in 5 out of 42 results (12%), low in 3 out of 42 results (7%), medium in 7 out of 42 results (17%), and high in 27 out of 42 results (64%). Considering no moderating variables the results show medium heterogeneity in 1 out of 15 times (7%) and high heterogeneity in 14 out of 15 times (93%) (see figure 2). This answers research question 2b (Can the heterogeneity be explained by the moderating variables of year of research, region of the researched firms, and type of business?) for year of research in the following way: year of research explains the heterogeneity in the results to a certain amount. It does so to a lesser amount than the moderating variable of region of the researched firms. Moreover, the Kruskal-Wallis test for variance shows significant results for the relationship between quality and innovation (H(2) = 6.133, P = 0.047) with a mean weighted corrected correlation of 0.145 for the group 1990-1999, 0.470 for the group 2000-2009 and 0.318 for the group 2010-2012, and for the relationship between all competitive capabilities and business performance (H(2) = 9.286, P = 0.01) with a mean weighted corrected correlation of 0.07 for the group 1990-1999, 0.262 for the group 2000-2009 and 0.335 for the group 2010-2012. For these relationships the weighted mean corrected correlations of the subgroups, based on year of research, differ significantly from each other. For all other relationships the correlations do not differ significantly from each other.
Figure 2: Percentages of results reaching thresholds (RATIO1 and RATIO2) and percentages of results per class of heterogeneity (based on the result of the I2 measure) for the general sample and for the subgroups based on year of research
47% 0% 0% 0% 7% 93% 57% 17% 12% 7% 17% 64% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
RATIO1 RATIO2 No heterogeneity Low heterogeneity Medium heterogeneity
High heterogeneity
23
Table 11: Stage-II results for all relationships and subgroups
24
Table 12: Stage-II results for all relationships and subgroups, continued
25
Table 13: Stage-II results for all relationships and subgroups, continued
26
5. Discussion
Examination of past studies on the relationships between competitive capabilities resulted in considerable contradiction. On one side authors argue that trade-offs in these relationships are present. On the other side they argue that competitive capabilities built upon each other. These studies show consensus on the effect of competitive capabilities on business performance: competitive capabilities have a positive effect on business performance. Our study investigates the relationships between the competitive capabilities and shows that the results are heterogeneous.
The same yields for the effect of these capabilities on business performance. This challenges the status quo. By introducing the moderating variables of year of research and region of the researched firms we are able to explain this heterogeneity to a certain extent.
Most of the capabilities (quality, delivery, flexibility, cost, and innovation) are positively correlated to each other. Whenever cost is part of a relationship however, the results show that trade-offs are possible, though rare.
The effects of the competitive capabilities on business performance differ for each competitive capability. Only flexibility and innovation have a positive effect on business performance in all situations. Other competitive capabilities show mean correlations which are positive, though the lower bounds of the credibility intervals drop below zero, which means that quality, dependability and cost do not have a positive effect on business performance in all situations. However, these negative effects are rare and small in the worst cases. In general, regarding the relationship between all competitive capabilities and business performance, we cannot be sure that this relationship is positively correlated. Hence, competitive capabilities do not have a positive effect on business performance in any given situation.
The results for both the relationships between the competitive capabilities and between the competitive capabilities and business performance are highly heterogeneous. The relationship between flexibility and innovation is an exception. Flexibility and innovation are positively correlated and do not vary across studies.
The year a study was conducted as well as the region of the studied firms explain this heterogeneity to a certain amount. Subgroups based on these moderating variables show a lesser amount of heterogeneity in the relationships results than the results of the complete sample.
Theoretical and practical implications
Our study has several implications for theory and business practice.
First, the pure trade-off model introduced by Skinner (1969). Our results show that trade-offs between competitive capabilities are only possible when one of the considered competitive capabilities is cost. This is in line with the ‘sand-cone model’ introduced by Ferdows and De Meyer (1990) which states that increasing in cost efficiency is only possible when all other capabilities are in place. Our results suit the sand-cone model and reject the trade-off model.
27
uncertain outcome on business performance. Resources can be spend on the ‘wrong’ competitive capabilities which results in a (small) negative effect on business performance.
Third, we have shown that innovation is regarded as a competitive capability in numerous studies. Also, we have recognized that firms can create competitive advantage by means of being environmentally sustainable.
Fourth, the year a study has been conducted as well as the region of the researched firms has an effect on the heterogeneity in the results of the relationships researched in this paper. Though the correlations of the subgroups based on these moderating variables do not differ significantly for most relationships, we did show that we can be more certain about the relationships when considering these moderating variables.
Fifth, we showed the different concepts that are being used in literature in defining the competitive capabilities. Different authors use different combinations of these concepts which causes heterogeneous results.
Limitations and directions for future research
Our study has several limitations and offers several directions for future research.
First, our results are limited by the sample size. Though the number of studies is enough when considering the general relationships between the competitive capabilities and between the competitive capabilities and business performance, they lack in quantity sometimes when considering the subgroups based on the moderating variables. In some of the subgroup relations the number of studies used for calculating the results is two. We base our conclusions on the standard deviation estimates which are particularly small in some of these relations. We cannot exclude chance in having an effect on these results, which decreases the reliability.
Second, we primarily use studies that calculate their results based on questionnaires. Therefore the results in our study are not based on objective performance measures. This is a common shortcoming in survey-based research (Schoenherr et al., 2012). Future research can objectively investigate the relationships researched in this paper by relying on observations instead of surveys.
Third, our sample did not entail studies on service providing firms. Though we recognized that the differences between types of business can have an effect on the results we were unable to investigate it. Future research which focuses on differences in relationships between competitive capabilities and between competitive capabilities and business performance based on subgroups created by industry types is desired. This will increase our understanding of the relationships and will provide best practices for managers in specific situations.
Fourth, future research is also wanted in the fields of cumulative capabilities theory. We showed that there is evidence supporting the sand-cone model of Ferdows and De Meyer (1990), but we also showed that there are more competitive capabilities to be taken into account than are outlined within the model. And, to be conclusive on this theory, longitudinal research is desired to show performance on multiple capabilities within firms over the years.
28 theory, though numerous studies report on it. Sustainability as a competitive capability is considered far less than any other competitive capability.
Sixth, future research should provide standardized definitions of the constructs used in this paper to enhance Operations Management research. Today, different authors use different definitions of concepts which makes it hard to put research into context and to compare results.
Conclusion
29
References
Schoenherr, T., Power, D., Narasimhan, R., Samson, D., 2012. Competitive Capabilities among Manufacturing Plants in Developing, Emerging, and Industrialized Countries: A Comparative Analysis. Decision Sciences. 43, 37-72
Vickery, S.K., Droge, C., Markland, R.E., 1993. Production Competence and Business Strategy: Do They Affect Business Performance?. Decision Sciences. 24, 435-456
Sarmiento, R., 2011. A note on ‘trade-off and compatibility between performance: definitions and empirical evidence’. Int J Prod Res. 49, 4175
Rosenzweig, E.D., Easton, G.S., 2010. Tradeoffs in Manufacturing? A Meta-Analysis and Critique of the Literature. Production and Operations Management. 19, 127-141
Rosenzweig, E.D., Roth, A.V., 2004. Towards a Theory of Competitive Progression: Evidence from High-Tech Manufacturing. Production & Operations Management. 13, 354-368
Safizadeh, M.,H.., Ritzman, L.P., Mallick, D., 2000. Revisiting Alternative Theoretical Paradigms in Manufacturing Strategy. Production and Operations Management. 9, 111-126 Skinner, W., 1969. Manufacturing--missing link in corporate strategy. Harv. Bus. Rev. 47, 136-145
White, G.P., 1996. A meta-analysis model of manufacturing capabilities. J. Oper. Manage. 14, 315-331
Ferdows, K., De Meyer, A., 1990. Lasting improvements in manufacturing performance: In search of a new theory. J. Oper. Manage. 9, 168-184
Roth, A V, 1996. Competitive progression theory: Explanation and evidence in
Manufacturing Strategy in a Global Context, C. Voss ed. London Business School Press, London, UK
Ebert, R.J., Tanner, G., 1996. Coaligning market requirements, quality capabilities, and human resource management: How will.. Journal of Quality Management. 1, 243
Flynn, B.B., Flynn, E.J., 2004. An exploratory study of the nature of cumulative capabilities. J. Oper. Manage. 22, 439-457
Pagell, M., Melnyk, S., Handfield, R., 2000. Do Trade-offs Exist in Operations Strategy? Insights from the Stamping Die Industry. Bus. Horiz. 43, 69
Porter, M E, 1985. Competitive Advantage. The Free Press, New York
30 Weerawardena, J., Mavondo, F.T., 2011. Capabilities, innovation and competitive advantage. Industrial Marketing Management. 40, 1220-1223
Dangelico, R.M., Devashish, P., 2010. Mainstreaming Green Product Innovation: Why and How Companies Integrate Environmental Sustainability. 95, 471-486
Daft, R L, Becker, S.W., 1978. The innovative organization : innovation adoption in school organizations. Elsevier, New York
Linton, J.D., Klassen, R., Jayaraman, V., 2007. Sustainable supply chains: An introduction. J. Oper. Manage. 25, 1075-1082
Venkatraman, N., Ramanujam, V., 1986. Measurement of Business Performance in Strategy Research: A Comparison of Approaches. The Academy of Management Review. 11, 801-814 Zhou, K.Z., Li, J.J., Zhou, N., Su, C., 2008. Market orientation, job satisfaction, product quality, and firm performance: evidence from China. Strategic Manage. J. 29, 985-1000 Rosenzweig, E.D., Roth, A.V., 2004. Towards a Theory of Competitive Progression: Evidence from High-Tech Manufacturing. Production & Operations Management. 13, 354-368
Kaynak, H., Hartley, J.L., 2008. A replication and extension of quality management into the supply chain. J. Oper. Manage. 26, 468-489
Skinner, W., 1974. The focused factory. Harv. Bus. Rev. 52, 113-121
Hayes, R H, Wheelwright, S.C., 1984. Restoring our competitive edge : competing through manufacturing. John Wiley, New York
Swamidass, P.M., Newell, W.T., 1987. Manufacturing Strategy, Environmental Uncertainty and Performance: A Path Analytic Model. Management Science. 33, pp. 509-524
Roth, A.V., Miller, J.G., 1992. Success factors in manufacturing. Bus. Horiz. 35, 73-81 Cleveland, G., Schroeder, R.G., Anderson, J.C., 1989. A Theory of Production Competence*. Decision Sciences. 20, 655-668
Ferdows, K., De Meyer, A., 1990. Lasting Improvements in Manufacturing Performance: In Search of a New Theory. J. Oper. Manage. 9, 168-184
Schmenner, R.W., Swink, M.L., 1998. On theory in operations management. J. Oper. Manage. 17, 97-113
Hallgren, M., Olhager, J., Schroeder, R.G., 2011. A hybrid model of competitive capabilities. International Journal of Operations & Production Management. 31, 511-526
31
Nair, A., 2006. Meta-analysis of the relationship between quality management practices and firm performance—implications for quality management theory development. J. Oper. Manage. 24, 948-975
Hunter, J E, Schmidt, F.L., 2004. Methods of meta-analysis : correcting error and bias in research findings. Sage, Thousand Oaks, Calif.
Mackelprang, A.W., Nair, A., 2010. Relationship between just-in-time manufacturing practices and performance: A meta-analytic investigation. J. Oper. Manage. 28, 283-302 Kristal, M.M., Huang, X., Roth, A.V., 2010. The effect of an ambidextrous supply chain strategy on combinative competitive capabilities and business performance. J. Oper. Manage. 28, 415-429
Avella, L., Vazquez-Bustelo, D., Fernandez, E., 2011. Cumulative manufacturing capabilities: an extended model and new empirical evidence. Int J Prod Res. 49, 707-729
Menor, L.J., Kristal, M.M., Rosenzweig, E.D., 2007. Examining the Influence of Operational Intellectual Capital on Capabilities and Performance. Manufacturing & Service Operations Management. 9, 559-578
Narasimhan, R., Swink, M., Viswanathan, S., 2010. On Decisions for Integration Implementation: An Examination of Complementarities Between Product-Process Technology Integration and Supply Chain Integration. Decision Sciences. 41, 355-372 da Silveira, Giovani J. C., Sousa, R.S., 2010. Paradigms of choice in manufacturing strategy. International Journal of Operations & Production Management. 30, 1219-1245
Squire, B., Brown, S., Readman, J., Bessant, J., 2006. The Impact of Mass Customisation on Manufacturing Trade-offs. Production & Operations Management. 15, 10-21
Boyer, K.K., Lewis, M.W., 2002. Competitive Priorities: Investigating The Need For Trade-Offs In Operations Strategy. Production & Operations Management. 11, 9-20
Prajogo, D.I., McDermott, P., Goh, M., 2008. Impact of value chain activities on quality and innovation. International Journal of Operations & Production Management. 28, 615-635 Dean Jr., J.W., Snell, S.A., 1996. The Strategic use of Integrated Manufacturing: an Empirical Examination. Strategic Manage. J. 17, 459-480
Rosenzweig, E.D., Roth, A.V., Dean Jr., J.W., 2003. The influence of an integration strategy on competitive capabilities and business performance: An exploratory study of consumer products manufacturers. J. Oper. Manage. 21, 437
Koufteros, X., Marcoulides, G.A., 2006. Product development practices and performance: A structural equation modeling-based multi-group analysis. Int J Prod Econ. 103, 286-307 Vachon, S., Klassen, R.D., 2008. Environmental management and manufacturing
32 Ariss, S.S., Zhang, Q., 2002. The impact of flexible process capability on the product–process matrix: an empirical examination. Int J Prod Econ. 76, 135-145
Filippini, R., 1998. Trade-off and compatibility between performance: Definitions and empirical evidence. Int J Prod Res. 36, 3379-3406
Ward, P.T., Duray, R., Keong Leong, G., Sum, C., 1995. Business environment, operations strategy, and performance: An empirical study of Singapore manufacturers. J. Oper. Manage. 13, 99-115
Arauz, R., Matsuo, H., Suzuki, H., 2009. Measuring changes in quality management: An empirical analysis of Japanese manufacturing companies. Total Quality Management & Business Excellence. 20, 1337-1374
Oltra, M.J., Flor, M.L., 2010. The moderating effect of business strategy on the relationship between operations strategy and firms' results. International Journal of Operations &
Production Management. 30, 612-638
Oltra, M.J., Maroto, C., Segura, B., 2005. Operations strategy configurations in project process firms. International Journal of Operations & Production Management. 25, 429-448 Li, Y., Wang, L., Liu, Y., 2011. Organisational learning, product quality and performance: the moderating effect of social ties in Chinese cross-border outsourcing. Int J Prod Res. 49, 159-182
Wong, C.Y., Boon-itt, S., Wong, C.W.Y., 2011. The contingency effects of environmental uncertainty on the relationship between supply chain integration and operational performance. J. Oper. Manage. 29, 604-615
Größler, A., Grübner, A., 2006. An empirical model of the relationships between
manufacturing capabilities. International Journal of Operations & Production Management. 26, 458-485
Kaynak, H., 2003. The relationship between total quality management practices and their effects on firm performance. J. Oper. Manage. 21, 405
Mapes, J., 1997. Performance trade-offs in manufacturing plants. International journal of operations production management. 17, 1020
McKone, K.E., Schroeder, R.G., Cua, K.O., 2001. The impact of total productive maintenance practices on manufacturing performance. J. Oper. Manage. 19, 39-58
Malhotra, M.K., Mackelprang, A.W., 2012. Are internal manufacturing and external supply chain flexibilities complementary capabilities?. J. Oper. Manage. 30, 180-200
Danese, P., Kalchschmidt, M., 2011. The role of the forecasting process in improving forecast accuracy and operational performance. Int J Prod Econ. 131, 204-214
33
Ward, P.T., Duray, R., 2000. Manufacturing strategy in context: environment, competitive strategy and manufacturing strategy. J. Oper. Manage. 18, 123-138
Avittathur, B., Swamidass, P., 2007. Matching plant flexibility and supplier flexibility: Lessons from small suppliers of U.S. manufacturing plants in India. J. Oper. Manage. 25, 717-735
Boyer, K.K., Keong Leong, G., Ward, P.T., Krajewski, L.J., 1997. Unlocking the potential of advanced manufacturing technologies. J. Oper. Manage. 15, 331-347
Liu, Y., Li, Y., Wei, Z., 2009. How organizational flexibility affects new product
development in an uncertain environment: Evidence from China. Int J Prod Econ. 120, 18-29 Patel, P.C., Terjesen, S., Li, D., 2012. Enhancing effects of manufacturing flexibility through operational absorptive capacity and operational ambidexterity. J. Oper. Manage. 30, 201-220 Camisón, C., Villar, A., 2010. An examination of the relationship between manufacturing flexibility and firm performance. International Journal of Operations & Production Management. 30, 853-878
Hollos, D., Blome, C., Foerstl, K., 2012. Does sustainable supplier co-operation affect performance? Examining implications for the triple bottom line. Int J Prod Res. 50, 2968-2986
Grawe, S.J., Chen, H., Daugherty, P.J., 2009. The relationship between strategic orientation, service innovation, and performance. International Journal of Physical Distribution & Logistics Management. 39, 282-300
Hsu, I., Sabherwal, R., 2012. Relationship between Intellectual Capital and Knowledge Management: An Empirical Investigation. Decision Sciences. 43, 489-524
Chen, M., Lin, H., Michel, J.G., 2010. Navigating in a hypercompetitive environment: the roles of action aggressiveness and TMT integration. Strategic Manage. J. 31, 1410-1430 Martínez-Costa, M., Martínez-Lorente, A.R., 2008. Does quality management foster or hinder innovation? An empirical study of Spanish companies. Total Quality Management &
Business Excellence. 19, 209-221
Koufteros, X., Vonderembse, M., Doll, W., 2001. Concurrent engineering and its consequences. J. Oper. Manage. 19, 97-115
