Outsourcing Decisions in Global Supply Chains:
An Exploratory Multi-Country Survey
, Ph.D., CSCP
Assistant Professor of Supply Chain Management
Department of Supply Chain Management, N361 North Business Complex
The Eli Broad College of Business, The Eli Broad School of Management
Michigan State University, East Lansing, MI 48824-1122
Phone: 517-432-6437, Fax: 517-432-1112
This article has been published in the International Journal of Production Research.
Schoenherr, Tobias (2010). “Outsourcing decisions in global supply chains: an exploratory
multi-country survey.” International Journal of Production Research, 48: 2, 343 — 378.
Outsourcing Decisions in Global Supply Chains:
An Exploratory Multi-Country Survey
Outsourcing has become a necessity for most companies in today’s competitive
environment, which is also evidenced by a growing interest by academics worldwide. However,
to date very few multi-country studies exist that compare and contrast outsourcing decisions and
practices by organizations in different countries. The present research aims to contribute to this
area by reporting and analyzing results from a survey conducted in 15 countries. Using a total
dataset of 806 responses, interesting empirical exploratory insights are reported on the evolving
approaches and concepts of outsourcing decision-making in different countries. More
specifically, this study provides characteristics of the most important suppliers to respondents
across the 15 nations, and explores rationales for outsourcing, highlighting country-specific
similarities and differences. In addition, four hypotheses are developed and tested, linking
contract specificity, purchase risk, supplier responsiveness, and procedural rigor to purchase
performance. Arguments are primarily based on transaction cost economics. And finally,
differences on these main variables across the 15 countries are explored.
Keywords: Outsourcing; Global supply chain; Purchase performance; Supplier
responsiveness; Contract specificity; Purchase risk; Procedural rigor; Survey research
Outsourcing has become inevitable for most companies in today’s competitive
marketplace, and can be a key enabler for global supply chain success. However, much is still to
be learned by both researchers and decision makers. Following the call for papers for this special
issue, it is the goal of the present manuscript to provide interesting empirical results arising from
research on the evolving approaches and concepts of outsourcing decisions. The empirical bases
and behavior in outsourcing strategies and relationships will be explored with data collected in a
large-scale multi-country survey.
The Institute for Supply Management (ISM) defines outsourcing as “a version of the
make-or-buy decision in which an organization elects to purchase an item that previously was
made or a service that was performed in-house” (Monczka et al. 2005, p. 10). Outsourcing has
many facets, and terms such as ‘cosourcing’ and ‘offshore outsourcing’ have been introduced to
differentiate aspects of this evolving concept (Jahns et al. 2006, cf. also Sanders et al. 2007). For
the purpose of this study we define outsourcing as procuring items from an outside supplier. Data
collected with the outsourcing module of the Global Manufacturing Research Group (GMRG)
survey were used to provide insight into approaches and concepts of outsourcing decisions
across 15 nations. At the time of this study, 806 useable responses were received.
The primary objective and contribution of the present work is to provide an overview of
current sourcing and outsourcing decisions and practices in 15 countries. More specifically, we
first provide a characterization of the most important supplier of our respondents across the
different countries; their size and the rationale behind them being classified as the most
important supplier to the buying organization are discussed. Second, we summarize rationales for
the decision to outsource to this particular supplier; we attempt to identify similarities and
differences in clusters of countries. Third, we develop and test four hypotheses relating
behavioral sourcing variables to purchase performance. These variables consist of contract
specificity, purchase risk, supplier responsiveness, and procedural rigor. The latter objective
provides the main context for this manuscript. And fourth, we conduct supplemental tests that
explore whether the 15 countries in our sample differ significantly on these five main variables,
as well as a potential different impact of the four behavioral sourcing variables on performance.
The paper proceeds as follows. Section 2 provides a literature review discussing the
major theoretical angles outsourcing has been looked at, and highlighting some of the most
recent relevant works. The third section provides our research framework and develops our four
main hypotheses. Section 4 describes the methodology used; data collection, data and construct
measure analysis, as well as respondent characteristics are discussed. Section 5 provides results
and their discussion. This section is divided into insights related to the characteristics of the most
important supplier, reasons for outsourcing, hypothesis tests, country-specific differences on the
main constructs of the study, and overall insights. Section 6 concludes and offers avenues for
2. Literature review
The literature on outsourcing has been proliferating, especially in recent years, as it
became a necessity for most companies to remain competitive and focus on their core
competencies. Outsourcing has been described as a key feature of today’s global economy
(Gunasekaran and Kobu 2007, cf. also Yusuf et al. 2006). Outsourcing, oftentimes also referred
to as the ‘make-or-buy’ decision (Ellram et al. 2008), has been looked at especially with an
economic and strategy lens. This section is intended to provide an overview of the different
theoretical and disciplinary angles outsourcing has been analyzed with. It also highlights the
strong tradition of the International Journal of Production Research in publishing outsourcing
Transaction cost economics (TCE) offers one of the earliest theoretical foundations for
the analysis of outsourcing decisions (Balakrishnan et al. 2008). The key attributes of TCE are
asset specificity, uncertainty and frequency, which are also relevant determinants in the
procurement decisions of an industrial organization. For a recent review of TCE as it relates to
outsourcing, including the operationalization of TCE, different styles of outsourcing,
qualifications to those, and the main take-aways of TCE for supply chain management, please
see Williamson (2008). As an illustrative recent study, Ellram et al. (2008) provide an
examination of the offshore outsourcing decision of professional services from a TCE
Another major theoretical anchor that has been used to explain outsourcing decisions is
the resource-based view (RBV) of the firm. The RBV assesses firm resources based on the four
attributes of valuable, rare, inimitable and non-substitutable (Barney 1991, Wernerfelt 1984).
While the fields of strategy and general management have applied the theory to a large extent, it
also has potential for explaining operations management concepts (Amundson 1998, St. John,
Cannon and Pouder 2001). More specifically, the RBV is an important theory for the study of
outsourcing decisions, since valuable, rare, inimitable and non-substitutable resources can result
in a competitive advantage, suggesting not to outsource the associated tasks (cf. McIvor 2009).
As an illustrative recent study, Holcomb and Hitt (2007) utilize the RBV, in combination with
TCE, to develop a model of strategic outsourcing.
Related to the RBV is the core competency concept (Prahalad and Hamel 1990), which
states that activities, products or services should not be outsourced if they represent a core
competence of the firm, i.e. if they are of strategic importance. On the other hand, non-core
activities should be outsourced so that the organization can focus on what they do best; done
right, outsourcing can result in “game-changing levels of value” (Doig et al. 2001, p. 25). The
core competency concept has also been used in combination with TCE to inform the outsourcing
decision (Arnold 2000).
Outsourcing can also be looked at with the lens of agency theory. As such, outsourcing is
not recommended when agents are risk averse, the length of the relationship is long, or the
uncertainty in cost or quality is high (Balakrishnan et al. 2008, citing Lee et al. 2002). Other
frameworks that have been used in the analysis of outsourcing decisions include the knowledge-
based view (KBV) of the firm, which is related to the KBV and provides an incentive for vertical
integration, and social exchange theory, which suggests outsourcing when partnership advantage
is high and relative dependence is low (Balakrishnan et al. 2008).
There are both benefits and drawbacks of outsourcing. On the one hand, outsourcing has
been described as a key enabler for a responsive supply chain (Gunasekaran et al. 2008), offering
access to cutting-edge technology and the use of high-powered performance contracts (Novak
and Stern 2008). Outsourcing can also increase flexibility, facilitate market penetration, and
accelerate reengineering efforts (Lau and Zhang 2006). Outsourcing partners can be the source
for best-in-class skills and fresh ideas, as well as objective creativity (Harland et al. 2005). On
the other hand, outsourcing can also have significant drawbacks, such as loss of control (cf.,
Gunasekaran et al. 2007, Sharif et al. 2007), and the danger of failure can be high (Doig et al.
2001, Dabhilkar and Bengtsson 2008). Offshoring risks include market volatility risk, the risk of
incomplete specifications, the inability to measure performance, and general risks associated
with doing business internationally (Ellram et al. 2008). Further risks are associated with the
development of new management competencies, capabilities and decision-making processes
(Harland et al. 2005). The outsourcing decision should therefore not be taken lightly, and a
careful approach is necessary (Schoenherr et al. 2008). In addition, the need to better account for
and manage outsourcing partnerships is crucial (Amaral et al. 2004). Numerous studies have
explored effective strategies for doing so (e.g., Bardhan et al. 2007, Ruiz-Torres et al. 2008,
Youngdahl and Ramaswamy 2008).
The International Journal of Production Research (IJPR) has had a strong tradition of
publishing insightful works related to outsourcing. For example, Dekkers (2000) provided
decision models for outsourcing and core competencies in manufacturing, Coman and Ronen
(2000) formulated a production outsourcing problem as a linear programming model for the
theory of constraints, and Wu et al. (2005) discussed an outsourcing decision model for
sustaining long-term enterprise performance. More recently, Serrato et al. (2007) developed a
Markov decision model to evaluate outsourcing in reverse logistics, and Ruiz-Torres et al.
(2008) dealt with the problem of finding outsourcing strategies considering the trade-off between
outsourcing cost and average tardiness. Mishra et al. (2008) offered a mixed integer
programming model for integrated planning and scheduling across the outsourcing supply chain
and demonstrated how such models can be used to make strategic decisions. Chan et al. (2009)
discussed the relevance of outsourcing and leagile strategies in performance optimization of an
integrated process planning and scheduling model, and Dhawan et al. (in press) considered the
minimization of total cost with regular and emergency outsourcing sources.
3. Research framework
The primary objective of this research is to provide an overview of current sourcing and
outsourcing decisions and practices in 15 countries. More specifically, we first provide a
characterization of the most important supplier of our respondents across the different countries;
their size and the rationale for them being the most important supplier are discussed. Second, we
summarize the reasons for outsourcing to this supplier; we attempt to identify similarities and
differences in clusters of countries. Third, we develop and test four hypotheses relating
behavioral sourcing variables to purchase performance. These variables are contract specificity,
purchase risk, supplier responsiveness, and procedural rigor. The latter objective, which
provides the main context for this paper, is graphically depicted in Figure 1. And fourth, we
conduct supplemental tests that explore whether the 15 countries in our sample differ
significantly on these five main variables. The remainder of this section develops and motivates
the four hypotheses.
Insert Figure 1 about here
3.1 Contract specificity
Outsourcing arrangements are usually governed by contracts, which outline the
obligations, rights and responsibilities of the parties involved. Careful attention should be paid to
the conditions and clauses in the contract, due to their ability to impact performance, quality and
efficiency. Contracts in call center outsourcing, for example, have recently been analyzed by Ren
and Zhou (2008) and Aksin et al. (2008). A contract should be very specific and account for
various eventualities. Within the context of our study, contract specificity refers to how detailed,
comprehensive and thorough a contract in a sourcing relationship is defined. Specifically, we
consider the statement of various contract clauses in the agreement with the focal supplier. Such
clauses can, for example, deal with financial repayment, exclusivity, damage for poor technical
performance, and damage for late delivery. These clauses can bring structure to the relationship,
reduce uncertainty, and provide objective measures and procedures for performance evaluation.
Viewing contracting from a TCE perspective (Williamson 2008), a more specific contract
can reduce uncertainty inherent in the outsourcing relationship. With higher contract specificity
better control is maintained, and the possibility of opportunistic behavior or abuse is reduced
(Amaral et al. 2004). In these settings the buyer’s perceived performance of the sourcing
relationship should therefore be better, compared to relationships governed by lower contract
specificity. Having established clear performance goals, along with penalty and termination
clauses, should reduce the risk of non-performance. This is in line with the argument provided in
Ellram et al. (2008), who propose that greater difficulty in verifying contractual performance is
associated with a lower probability to outsource offshore. Based on this discussion we suggest
the following hypothesis.
H1: Contract specificity is related to the buyer’s purchase performance, with higher
specificity being associated with better performance.
3.2 Purchase risk
Risk has been a central variable considered in outsourcing decision making. Risks
associated with outsourcing include the loss of control over the task, function or item outsourced,
a potential degradation of critical capabilities, increased dependency, and financial vulnerability
(Sanders et al. 2007). Additional risks emerge out of market volatility, incomplete specifications,
and the inability to measure performance (Ellram et al. 2008). Risk is usually heightened with a
more involved outsourcing arrangement, as more control is transferred to the supplier.
Considering an agency theory perspective, outsourcing is therefore not recommended when
agents are risk averse (Balakrishnan et al. 2008, citing Lee et al. 2002). In addition, a heightened
risk is generally associated with greater uncertainty present in the relationship. Uncertainty,
which can be defined as the source of disturbances (Williamson 2008) or the degree of volatility
and unpredictability in the supply market (Ellram et al. 2008), has been suggested to take away
some of the advantages of outsourcing (Balakrishnan et al. 2008). In uncertain situations it seems
more prudent to conduct the activity in-house. This was also captured in a proposition by Ellram
et al. (2008), who showed that a more volatile supply market reduces the likelihood of the
professional services category being offshore outsourced. Greater risk-taking has also been
associated with poorer performance (Avlonitis and Salavou 2007).
We suggest that a greater risk and uncertainty associated with the sourcing relationship
should lead to a more involved rapport between the parties. This should be demonstrated by the
buying firm’s less efficient dealing with the supplier in these instances, compared to other
vendors. With an increase in purchase risk the buyer’s purchase performance should therefore
be lower. Our second hypothesis is formulated accordingly.
H2: Purchase risk is related to the buyer’s purchase performance, with higher risk being
associated with lower performance.
3.3 Supplier responsiveness
Supplier responsiveness can be defined as the supplier’s willingness to meet the needs of
the buying firm (Carr and Smeltzer 2000), and has been a valuable concept measuring the
performance of the buyer-supplier relationship (Handfield and Bechtel 2002). Supplier
responsiveness can manifest itself by, for example, (1) the supplier’s products working, enduring
and fitting to the buyer’s specifications, (2) the supplier providing production forecasts and
schedules, or (3) the existence of joint efforts to stabilize production plans. Increased supplier
responsiveness can lead to beneficial outcomes, such as cycle time reduction (Handfield et al.
1998) and the buying firm’s ability to better meet customer expectations (Goodman et al. 1995).
It is therefore not surprising that this concept has been the focus of numerous studies, which
looked at inhibitors of responsiveness (Holweg 2005) and ways to increase it (Handfield and
Bechtel 2000), as well as the impact of supply chain complexity (Handfield et al. 1998).
Relating to the outsourcing decision, we suggest that the supplier’s sourcing
responsiveness is positively related to the buying firm’s purchase performance. With a more
responsive supplier the buyer should need to devote less time and resources to the management
of the relationship, and thus perceive it as being more efficient. At the same time, frequent and
timely communication can reduce uncertainty between buyer and supplier. From a TCE
perspective, this will further add to a better perception of the relationship. We therefore suggest
H3: Supplier responsiveness is related to the buyer’s purchase performance, with higher
responsiveness being associated with better performance.
3.4 Procedural rigor
Procedural rigor or procedural control refers to the extent to which guidelines, policies or
transaction norms are available for the buyer to deal with the sourcing relationship to the
supplier. This rigor can, for example, be assessed by looking at (1) whether the product
specifications were carefully specified before the singing of the contract, (2) whether it is
generally easy to determine the performance of the supplier, and (3) whether there is a standard
approach when solving a problem with the supplier. Procedural control was described as one of
the four distinct constructs underlying the activities of industrial buyers (Bunn 1994). A study by
Hunter et al. (2006) suggested procedural control influencing the buyer’s search for information
and their proactive focusing. Especially in an outsourcing context, procedural rigor has been
shown to be of particular importance (Adeleye et al. 2004). Decision rules can positively
influence behavior (Ramaswami 1996) and make the dealings with the supplier more efficient
(Johnston and Lewin 1996). Employing the TCE perspective, procedural rigor can also reduce
uncertainty and opportunistic behavior that may be inherent in the buyer-supplier relationship.
This is in line with arguments in Ellram et al. (2008), who proposed that the more uncertain a
firm is about its requirements, the less likely it will offshore outsource their professional services
category. We therefore formulate our fourth hypothesis as follows:
H4: Procedural rigor is related to the buyer’s purchase performance, with higher rigor
being associated with better performance.
4.1 Data collection
Data were collected with a large-scale worldwide survey as part of the Global
Manufacturing Research Group (GMRG). The GMRG is a multi-national community of
academic researchers dedicated to the study and improvement of manufacturing practices world-
wide via coordinated surveys (cf., Wacker and Sheu 2006). The same questionnaire, translated
into the appropriate language, is administered by members of the group in their respective
countries. Once collected, data are combined into a single dataset, providing the basis for a
powerful analysis of current worldwide manufacturing practices and trends. Rigorous translating
and back-translating rounds are employed to ensure the equivalency of the questionnaire versions
in different languages. Since the conception of the group in 1985, the survey has already been
conducted three times on a global basis. Numerous publications have resulted out of this effort,
with the International Journal of Production Research being a primary outlet of this work (e.g.,
Pagell and Sheu 2001, Pagell et al. 2007, Sheu and Wacker 2001, Wacker and Sheu 2006,
Wacker and Sprague 1995). Clay Whybark, one of the founders of the group, provides a
comprehensive overview and summary of the GMRG, questionnaire development and design,
and methodology (Whybark 1997). Additional detail is provided in Whybark and Vastag (1993).
For the present study a subset of the GMRG data, collected in its fourth round, was used.
Primarily answers to questions in two of the questionnaire’s five modules, outsourcing and
demographics, were employed for the present manuscript. Besides general company-level data,
the outsourcing module asked respondents to focus on their most important supplier, and then
answer the subsequent questions based on this supplier. Questions included whether the supplier
was international, the reason for classifying it as most important, the percentage of total plant
materials this supplier is responsible for, and the buying organization’s relative size compared to
the supplier. Information was also gathered about the predominant reason to outsource to this
particular supplier. In addition, data were collected for several multi-item constructs, which we
labeled as contract specificity, purchase risk, supplier responsiveness and procedural rigor. And
lastly, the purchase performance of the respondent’s firm was assessed by a question asking
whether resources, such as labor, material and capital equipment, are more or less efficiently
used with the focal supplier, relative to other suppliers.
4.2 Data analysis and construct measures
Primary tests to detect differences and to assess the hypotheses were conducted via
contingency table analyses. This approach to test hypotheses and derive
insights was successfully applied, for example, in Schoenherr (2008), Hartley et al. (2004) and
Min and Galle (2003).
The independent variables in the hypotheses were measured using multi-item constructs.
As such, to assess the concept of contract specificity, respondents were asked to indicate the
likelihood of various contract clauses to be legally existing and binding in contract agreements
with the focal supplier. These included clauses related to financial repayment, exclusivity,
damage for poor technical performance, and damage for late delivery. Respondents were asked
to indicate the likelihood of their existence on a scale ranging from ‘does not exist/none’
(value=0) to ‘very likely’ (value=5). The composite measure for purchase risk was built by
questions to which the respondent had to indicate their level of technological, behavioral and
market risk associated with the focal supplier. Per the definitions provided, technology risk is the
risk associated with the newness of the technology for the supplier, behavioral risk is the risk
associated with contract difficulties that may emerge with this supplier, and market risk is the
risk associated with the market not developing as expected. Respondents had to indicate a risk
level ranging from ‘no risk’ (value=1) to ‘very risky’ (value=5).
To gather data for the concept of supplier responsiveness individuals were asked to
indicate how well the focal supplier engages in a variety of activities, compared to other
suppliers. The activities consisted of (1) the provision of production forecasts, plans, schedules
and supply requirements, (2) how well the supplier’s products work, endure and fit the firm’s
specifications, and (3) whether joint efforts are conducted to stabilize production schedules. A
scale ranging from ‘worst’ (value=1) to ‘best’ (value=5) was used. For the final construct,
procedural rigor, respondents to the survey were asked to evaluate a series of statements related
to the contract specifications with the particular supplier. As such, individuals had to assess (1)
whether the product specifications were carefully specified before contract singing, (2) whether
it is easy to determine the performance of the supplier, and (3) whether there is a standard
approach when solving a problem with the supplier. A four-point scale was used ranging from
‘completely agree’ (value=1) to ‘completely disagree’ (value=4). Reliability of theseconstruct
measures was assessed via Cronbach’s alpha (Cronbach 1951).
4.3 Respondent characteristics
For the outsourcing module, which provided data for this manuscript, a total of 806
responses from 15 countries were received. A breakdown of the data by country is presented in
Table 1. The average number of responses per country was 54, with however some countries
delivering significantly more data than others. For example, both Fiji and Korea had responses of
above 100, while Austria, Sweden and Albania were at the low end with fewer than 20
responses. The interpretation of the results for those countries with a lower record count thus
needs to be conducted cautiously. Nevertheless, most countries had at least 50 responses, and
thus provide a good basis for statistical evaluation on the country-level.
The average firm size in our sample was 538 employees, with however some wide
disparities. While the average number of employees per company in most countries was below
500 individuals, companies in Austria, and especially Korea, had a significantly larger workforce
than their counterparts. However, subsequent analysis revealed that these widely different results
were a result of outliers. Removing the eight companies with the largest number of employees
for Korea lead to an average firm size of 483 individuals for the country, which is comparable to
others. Similarly for Austria, removing the records with the three largest employee numbers
reduced the average company size to 362.
A similar picture is revealed when looking at the firm size distribution of the aggregate
data presented in Table 2. Most firms were either small or medium-sized companies, with only
7.8% of the respondents working in firms with more than 1,000 employees.
Insert Table 1 about here
Insert Table 2 about here
Before proceeding a discussion of the industries represented seems warranted. Since
different countries use different standard industrial classification (SIC) codes to denote the
respective industries, GMRG designated their own set of SICs. Responses were coded by the
survey administrator based on the primary products offered by the company (respondents had to
list their top four product lines). Most frequently represented was the electronic and other
electrical equipment and component industry (US SIC 36, 17.85%), followed by fabricated metal
products, except machinery and transportation equipment (US SIC 34, 13.27%). The third-most
frequently represented industry produced industrial and commercial machinery as well as
computer equipment (US SIC 35, 12.80%). Food and kindred products (US SIC 20, 9.48%) and
miscellaneous manufacturing industries (US SIC 39, 9.00%) followed. All remaining SIC codes
were represented with a percentage of 5% or less.
5.1 Characteristics of the most important supplier
In the outsourcing module respondents were asked to focus on their most important
supplier, and to use this supplier as a reference when answering subsequent questions. In a first
question respondents had to indicate whether this supplier is international or domestic.
Considering the entire sample, domestic and international suppliers were about equally
distributed, with 52.8% of the respondents indicating their most important supplier to be
international. However, examining the answers across the individual countries clearly showed
some nations as relying more heavily on international supply, at least when using the fact of their
most important supplier being international or domestic as a proxy (Table 3). A nonparametric
contingency table analysis was used to statistically test this difference, which resulted in a
significant relationship between the two variables across the entire sample (Pearson Χ
(14, N =
782) = 143.88, p = 0.00, Cramér’s V = 0.43). To investigate potential differences for individual
countries, binominal tests were conducted to assess the equality of the proportions. Respondents
in eight of the 15 countries exhibited a significantly higher or lower proportion of an
international supplier being their most important sourcing partner. More specifically, our data
suggest that respondents in Austria, China, Fiji, Ghana, Macedonia and Taiwan are more
internationally oriented, as indicated by them having a significantly higher proportion of
international suppliers as their most important one. In contrast, Korea and Poland account for a
significantly lower proportion of international suppliers. The remaining countries had about an
equal share of domestic and international suppliers.
Additional information was provided in terms of the percentage on the total plant
materials this most important supplier possesses. On average, the supplier accounted for more
than 40% of the plant’s material (ranging from 22% to 67%), illustrating their importance to the
proper operation of the facility (Table 3). The most important suppliers for respondents in Ghana
and Albania accounted for more than half of the plant’s total material, illustrating the large
dependence of these firms on the supplier. In contrast, Sweden, Germany and Italy had the
lowest percentage of plant sales allocated to their most important supplier, proposing a greater
diversification of the supply base in these more industrialized countries. The standard deviations
across the countries were about equal in magnitude, providing a consistent picture and
suggesting equal deviations or discrepancies.
Insert Table 3 about here
For most of the respondents across the countries, the focal supplier was much smaller or
smaller in size. Exceptions form China, Fiji and Taiwan, where most often the focal supplier was
about the same size or larger. Relative percentages are shown in Table 4, with the cell having the
highest frequency per country highlighted. A nonparametric contingency table analyses
confirmed the differences between the countries statistically (Pearson Χ
(56, N = 768) = 302.37,
p = 0.00, Cramér’s V = 0.31).
Insert Table 4 about here
An additional question asked for further explanation why this supplier was classified as
most important. Except for China, Ghana and Taiwan, the majority of respondents stated this
particular supplier representing the largest sales volume for the company as the reason for the
classification. The supplier having the latest new product technology was the primary reason for
most respondents in Ghana and Taiwan, whereas it was the fact that it was required by the parent
company for respondents in China. The remaining reasons of providing the latest manufacturing
technology (especially mentioned by respondents in Albania, Ghana and Macedonia) and having
the longest contract with the supplier (especially mentioned by respondents in Albania,
Macedonia and Poland) were mentioned with lower frequencies. The relative frequencies are
presented in Table 5, with the cell having the largest proportion for each country highlighted.
Analysis via a two-way contingency table confirmed the significance of the countries differing
on this variable (Pearson Χ
(70, N = 770) = 336.07, p = 0.00, Cramér’s V = 0.30).
Insert Table 5 about here
5.2 Reasons for outsourcing
Interesting to explore are also the rationales behind the decision to outsource or to not
produce the product and/or service internally. Respondents were provided with a set of reasons
judged to represent an exhaustive group of alternatives, and were asked to indicate the primary
motivation why the firm does not produce these products in-house. Table 6 breaks down the
results by reason and country, with the cell exhibiting the highest percentage per country
highlighted. A nonparametric contingency table analysis indicated that the differences between
the countries are statistically significant (Pearson Χ
(126, N = 732) = 518.68, p = 0.00, Cramér’s
V = 0.28). The primary reason to outsource for most of our respondents was the lack of resources
(material) at the needed location (21.2%) – the majority of respondents in seven of our 15
countries stated this as being the principal argument. The unavailability of resources, which can
be material and/or labor, is a compelling reason to outsource. Rather than the buying form
investing in these capabilities on- or off-site, it may be more efficient for the outside vendor to
fulfill these tasks. The second-most frequent reason to outsource was the lack of time to acquire
the resources (16.5%). An outside supplier can again provide efficiencies in obtaining these
resources, especially if they are not the buyer’s strength or competency. For the majority of
respondents in three of our countries this was the pivotal reason. No access to natural resources
represents the third-most frequent argument to outsource (13.3%), and was the primary reason
for the majority of respondents in two countries. Similar as above, an outside vendor specializing
in these resources may possess better capabilities to obtain the needed input. Acquiring the
resources from a supplier may also be more efficient than the buying firm investing in these
capabilities and establishing a facility at the location of the natural resources. Further mentioned
explanations were the lack of specific skilled labor (8.3%), a long-lasting contract with the
supplier (7.2%), a lower price (6.4%) and the lack of a patent for the needed technology (6.4%).
Insert Table 6 about here
5.3 Impact of behavioral sourcing variables on purchase performance
The independent variables in our hypotheses, linking behavioral sourcing variables to
purchase performance, were operationalized using multi-item construct measures. Three or four
survey questions, assessed on Likert-type scales, represented measurements for each construct.
The actual items, their means and standard deviations, as well as the construct measure reliability
are presented in Table 7. As can be seen, the Cronbach alpha reliabilities of the construct
measures were satisfactory. The means and standard deviations of the aggregate constructs for
each country are provided in Table 8. We will refer to this table as we proceed with the
description of the results and the discussion.
Insert Table 7 about here
Insert Table 8 about here
As described above, chi-squared (X
) contingency tables were used to test the hypotheses.
The dependent variable was a five-level performance scale, on which respondents indicated
whether resources, such as labor, material and capital equipment, were more or less efficiently
used with the focal supplier, relative to other suppliers. The four constructs contract specificity,
purchase risk, supplier responsiveness and procedural rigor served as independent variables. For
parsimony, and to facilitate data interpretation, each independent variable was classified into a
high-level and a low-level of aggregate values on the measures (cf., Hartley, Lane and Hong
The results for H1, presented in Table 9, provide support for the suggested relationship
between contract specificity and purchase performance. While very few respondents judged
themselves of falling in the lower two performance groups (indicating less efficiency practiced in
dealing with the focal supplier, compared to other suppliers), a clearer picture is provided for the
three higher performance groups. For example, when looking at the entries in the rows labeled
“% within LCS” and “% within HCS”, for the two contract specificity levels, one can see that
respondents who had higher contract specificity (HCS) in their outsourcing agreement reported
better purchase performance. In contrast, individuals with low contract specificity (LCS) were
more heavily represented in lower-level performance groups (most notably in the group
characterizing the performance with the focal supplier as having the same level of efficiency
when compared to other supplier relationships).
Insert Table 9 about here
No significant differences were detected between purchase risk and purchase
performance, failing to support H2. The assessment of risk in an outsourcing relationship
apparently does not influence performance. In fact, as illustrated in Table 10, the distribution of
performance groups among the two risk levels is about the same. This can be seen when
comparing the rows “% within LR” and “% within HR” – values are almost identical across the
five performance groups. However, when comparing the “% within Perf.” across the low risk
(LR) and high risk (HR) groups, higher risk tends to be associated with less efficient
performance (55.2 (HR) vs. 44.8 (LR), and 65.4 (HR) vs. 34.6 (LR), for the last two performance
groups), and lower risk tends to be associated with more efficient purchasing (45.8 (HR) vs. 54.2
(LR), for the “2-10% more efficient” performance group). While this visual comparison seems
compelling, the differences are not statistically significant (p>0.10).
Insert Table 10 about here
Table 11 provides results supporting our third hypothesis, linking supplier responsiveness
with purchase performance. More specifically, the data suggest that higher supplier
responsiveness is associated with better performance. This is for example visible when
comparing the rows labeled “% within LRe” and “% within HRe” across the five performance
groups. A larger proportion of the two performance groups characterized by more efficiency
comes also from the high responsiveness (HRe) group (55.3 (HRe) vs. 44.7 (LRe), and 66.8
(HRe) vs. 33.2 (LRe), for the last two performance groups).
Insert Table 11 about here
Our fourth hypothesis was also supported, confirming the suggested relationship between
procedural rigor and purchase performance (Table 12). This can especially be seen when
comparing the two responsiveness levels across the second and fifth performance groups; higher
procedural rigor (HRi) is associated with better performance. For example, most respondents
(77.3%) in the second performance group (10-2% less efficiently) characterized their outsourcing
relationship by low rigor, whereas most respondents (58.0%) in the fifth performance group (10-
20% more efficiently) characterized their relationship by high rigor.
Insert Table 12 about here
In a follow-up analysis we were interested in exploring whether the levels of our five
main variables differ depending on whether the sourcing agreement is with a domestic or
international supplier. As discussed above, about half of our respondents indicated their most
important supplier to be international. Independent sample t tests were conducted to evaluate this
aspect. Three of the five tests were statistically significant. Respondents characterized the
relationship with international suppliers as being higher in contract specificity (t(656)=5.37,
p=0.00), higher in supplier responsiveness (t(769)=3.44, p=0.00), and higher in procedural rigor
(t(757)=-2.98, p=0.00). No significant differences were detected across the purchase risk and the
purchase performance variables.
5.4 Country specific differences on the main construct variables
To evaluate differences among the 15 countries on the median change in the five main
variables included in the hypotheses, Kruskal-Wallis tests were conducted. This section
summarizes these efforts and provides additional insight. Once a positive result for the Kruskal-
Wallis test has been received, follow-up pairwise comparisons were conducted using the Mann-
Whitney U test (Green and Saalkind 2003). With 15 countries represented in our sample and five
main variables, this represents a total of 420 pairwise comparisons for the four main independent
variables (105 possible combinations/comparisons for each construct), and an additional 105
comparisons for the dependent variable considered in our hypotheses.
The test for the construct contract specificity led to a statistically significant result, which
suggested differing levels of this variable across countries, X
(14, N=660)=175.96, p=0.00.
Table 8 provides the means and standard deviations for the construct values across the various
countries, while Figure 2 illustrates the result in a boxplot. Highest values on contract specificity
belong to Sweden and Taiwan, followed by Fiji and Albania. Poland, Italy and Macedonia
exhibit the lowest values on this variable. Almost two-thirds of the 105 comparisons indicated a
statistically significant difference (Table 13). More specifically, 3.81% were significant at the
0.01 level, an additional 12.38% at the 0.05 level, and a further 47.62% at the 0.1 level.
Appendix A provides details about which comparisons showed statistically significant
Insert Figure 2 about here
Insert Table 13 about here
Assessing potential differences between the 15 countries on the purchase risk variable
resulted in a positive outcome with the Kruskal-Wallis test, X
(14, N=774)=57.55, p=0.00.
Nevertheless, the boxplot, presented in Figure 3, suggests that these differences, although
statistically significant, are rather subtle. Lower levels are experienced by Albania, Austria, Fiji,
Germany, Ghana, Italy and Sweden, whereas Australia, China and Korea exhibit higher values.
A total 37.14% of the follow-up Mann-Whitney U tests were statistically significant, with 12%
being significant at the 0.01 level, an additional 11.43% at the 0.05 level, and an additional
13.33% at the 0.1 level (Table 13). Table 8 above summarizes the means and standard deviations
of this variable across the countries, and Appendix A offers details about which Mann-Whitney
U test indicated a significant relationship.
Insert Figure 3 about here
The Kruskal-Wallis test, evaluating whether the 15 countries differ significantly on their
levels of supplier responsiveness, was significant, X
(14, N=774)= 56.85, p=0.00. However, the
boxplot in Figure 4 indicates again that these differences are rather small. Lowest values were
received by Fiji and Sweden, with highest values for Australia and Taiwan. About two-fifths of
the 105 follow-up Mann-Whitney U tests were statistically significant (6.67% at the 0.01 level,
15.24% at the 0.05 level, and 17.14% at the 0.1 level) (Table 13). Means and standard deviations
are given in Table 8, and details about the Mann-Whitney U tests can be found in Appendix A.
Insert Figure 4 about here
A Kruskal-Wallis test was conducted also for our last independent construct, procedural
rigor. The result suggested that countries differ significantly on their procedural rigor employed
when dealing with their most important supplier, X
(14, N=761)=123.33, p=0.00. While most
countries exhibited very low values on this variable, indicating high procedural rigor,
Macedonia was characterized by particularly high rigor, followed by Albania, Austria, Hungary,
Italy and Mexico. Lower procedural rigor existed in Korea and Poland. Table 8 provides means
and standard deviations of this variable across the 15 countries. Over half (59.05%) of the
follow-up Mann-Whitney U tests were statistically significant, with 6.67% at the 0.01 level, a
further 14.29% at the 0.05 level, and an additional 38.10% at the 0.1 level. Which one of the 105
comparisons were statistically significant is summarized in Appendix A.
Insert Figure 5 about here
Overall, looking at all 420 Mann-Whitney U tests for the four independent variables
together, about half were statistically significant at least at the 0.1 level (209 tests or 48.76%).
More specifically, 122 or 29.05% of the tests were significant at the 0.01 level, an additional 56
or 13.33% percent were significant at the 0.05 level, and a further 31 or 7.38% were significant
at the 0.1 level. Countries differed most significantly on the contract specificity construct,
followed by procedural rigor. The tests for the remaining two constructs, supplier
responsiveness and purchase risk, turned out significant in about 40% of the cases (Table 13).
The last set of tests concerned country-specific differences on the purchase performance
variable. The Kruskal-Wallis test also returned statistically significant results here, indicating
that countries differ in regards to their perceived performance in sourcing, X
p=0.00. Means and standard deviations are given in Table 8, which are also graphically depicted
in Figure 6. A wide dispersion of answers existed in all countries. Ghana and Albania saw their
efficiency in the most positive light, whereas especially Germany had a rather pessimistic
perception of their purchase performance. Interesting in Germany is also the particularly wide
dispersion of the values from highest to lowest. When Mann-Whitney U tests were conducted
one-third or 35 of these 105 additional tests were statistically significant (16.19% at the 0.01
level, 4.76% at the 0.05 level, and 12.38% at the 0.1 level). The second table in Appendix A
provides further detail.
Insert Figure 6 about here
6.1 Characteristics of the most important supplier
Our results showed that, while the most important supplier of our respondents were
equally likely to be domestic and international when considering our sample as a whole,
significant differences in international orientation or dependence existed among the individual
countries. Nations with a significantly larger proportion of international suppliers included
Austria, China, Fiji, Ghana, Macedonia and Taiwan. This internationally-focused group is
contrasted by a domestically-focused group consisting of Macedonia and Poland. A third group is
comprised of countries in which respondents were equally likely to have an international or a
domestic supplier as their most important sourcing partner. This balanced group includes
Albania, Australia, Germany, Hungary, Italy, Mexico and Sweden.
Classical trade theories inform this choice between domestic and international
outsourcing (Chen 2009). While an international supplier may be able to provide comparable
products for a lesser price, due to for example lower wages, additional costs associated with
transacting across national boundaries and cultures need to be considered. Classical trade theory
thus suggests the choice for an international supplier when the benefits of sourcing
internationally outweigh its costs (Melvin 1985, Ruffin 1990).
The internationally-focused countries thus derive added benefit by offshore outsourcing
their requirements. This is understandable for smaller nations, such as Austria and Macedonia,
and lesser developed countries, such as Fiji and Ghana. The remaining two countries in the
internationally-focused group, China and Taiwan, may source internationally since the product is
specialized or more effectively procured offshore. The domestically-focused group, Korea and
Poland, has primarily domestic suppliers as their most important sourcing partner. For these
countries, the benefits of sourcing internationally do not outweigh the costs associated with this
approach. A large group of countries has an equal share of international and domestic suppliers
as their most significant vendor, suggesting an equal reliance and importance of both national
and international suppliers. It also suggests that these countries have the capability to source
domestically, which is complemented by the opportunity to do so internationally as well.
Managers in the various countries can use these findings as a benchmark to identify whether they
source more or less internationally than the average firm. Potential disparities in overall company
performance could be attributed to this difference.
The significance of the focal supplier was demonstrated by it accounting for over 40% of
the total plant materials acquired. For Ghana and Albania this percentage was above 50%,
whereas it was below 30% for Sweden, Germany, Italy and Taiwan. This rough exploratory
insight suggests that emerging and developing countries tend to rely more heavily on one
supplier, whereas for more developed nations the dependence is less, as indicated by the
supplier’s lower share on total plant materials provided. The latter is also indicative of a more
diversified and stratified supply base.
The primary reason for respondents to classify the focal supplier as the most important
was the fact that it represented the largest sales volume. This observation is consistent across
almost all countries in our sample. Exceptions form Taiwan and Ghana, which both seek the
latest new product technology from their focal supplier, and thus classify them as most
important. This result suggests that desired capabilities do not exist internally, and that it
therefore has to be procured from outside. China formed another exception. Most respondents in
this country indicated the supplier being required by the parent company as making them the
Given above discussion it was surprising to see that, for most respondents, the focal
supplier is smaller than their firm. This is true across most countries, with however most
respondents in China, Fiji and Taiwan indicating their focal supplier to be the same in size or
larger. The relative size can have an impact on the outsourcing strategies of firms, since a larger
supplier has the potential to exert pressure on the buying firm, whereas the power balance is
shifted to the buyer in the relationship with a smaller supplier. However, smaller suppliers may
not be as flexible in changing output volume or satisfying increasing demands.
6.2 Reasons for outsourcing
Companies in our sample outsourced for a variety of reasons, with the lack of resources
at the needed location being the primary rationale. This was mentioned by respondents in seven
countries as their primary motivation. Only for Australia was a lower price the determining
factor for outsourcing. This is surprising, since price and cost advantages are often mentioned as
the primary reason to outsource in the literature (e.g., Ross et al. 2005). However, as was noted
by Arya et al. (2008), the decision to outsource can be far more complex. Not possessing the
patent for the needed technology was the primary reason to outsource for Macedonian firms.
Outsourcing was therefore a means to obtain the technology. This suggests that these firms want
to take advantage of the latest technology, for which patents still exist, and to thus obtain a
competitive advantage. It is noteworthy that Macedonia was the dominant country providing this
reason – 31.4% of its respondents noted the lack of patents as the primary motivation to
outsource, whereas the percentage in other countries was only 12% or less. This shows that
outsourcing strategy in Macedonian firms is heavily influenced by the utilization of patented
The lack of specific capital equipment was the predominant reason for firms in Ghana to
outsource. Related capabilities are apparently not yet well developed, which is probably also true
due to the often significant investment needed for capital equipment. For a developing country,
such as Ghana, this may be especially challenging. The primary reason for Albania, and
especially Korea, was the lack of natural resources. Investing in a facility enabling this access
was apparently not feasible, economical or aligned with company strategy. Firms in these
countries therefore rather rely on the expertise of an outside supplier. Austria and Hungary noted
not having enough time to acquire the needed resources as their primary outsourcing motivation.
Although they may have possessed the capability to do so, these countries decided to rather let a
third party obtain the resources. This enables the buying firm to focus on what they can do best,
on their core competencies. Lack of skilled labor was the primary motivation for respondents in
China and Taiwan to outsource. Similarly as above, although the capability to obtain or train
skilled labor may have been present, companies decided for an outside provider, allowing the
firm to focus on their core competencies.
6.3 Impact of behavioral sourcing variables on purchase performance
Hypothesis 1, suggesting higher contract specificity being associated with better purchase
performance, was supported. Respondents with high contract specificity inherent in their
outsourcing agreements judged their performance to be better, on average, when compared to
respondents with low contract specificity. This observation confirms the importance of having
clear and unambiguous contract language addressing contingencies in place. It also shows that
the more detailed and thorough the contract is developed, the more efficient the management of
the buyer-supplier relationship will be, as perceived by the buying firm. More involved up-front
work in specifying the contract details can therefore pay off in the end, since less time will be
needed to manage the partnership after contract implementation.
The link between purchase risk and performance, stipulated in Hypothesis 2, was not
supported. While a visual examination of the results seemed to suggest this relationship, it was
not statistically significant. One can explain the nonexistence of the relationship as follows. In
situations associated with high risk and uncertainty, buyers may not want to rely on suppliers,
but rather conduct the activity in-house, to gain increased control and leverage. This is consistent
with suggestions by Balakrishnan et al. (2008), who noted that risk and uncertainty take away
some of the advantages of outsourcing, and by Ellram et al. (2008), who showed that a more
volatile supply market reduces the likelihood of offshore outsourcing. Since respondents were
asked to identify an outsourcing relationship, by its very nature, it may have already been
characterized by a low risk level; otherwise the activity would have been performed in-house. In
any case, the level of risk inherent with the particular outsourcing agreement apparently does not
influence the perceived performance of the relationship, and a higher risk does not negatively
impact the buyer’s efficiency in dealing with the supplier.
Hypothesis 3, linking supplier responsiveness and performance, was supported. This
seems logical, since when a supplier is more responsive, it leaves the buyer with less work and
effort in potentially following up to prior inquiries. Buying companies should therefore entice the
supplier to be more responsive, by for example joint projects, closer integration, or supplier
development initiatives. Respondents that characterized their supplier as responsive reported
better performance. The result highlights the benefits of supplier responsiveness, which was
operationalized by good communication, information sharing and collaboration. Investing in the
buyer-supplier relationship and developing the supplier is therefore a wise investment, since it
will most likely make the supplier more responsive. At the same time, increased communication
has the potential to reduce risk and uncertainty, and to increase the efficiency with which the
relationship is managed.
We also found support for Hypothesis 4, relating procedural rigor to performance. The
presence of procedures brings structure to the relationship and decreases risk and uncertainty.
Both parties can be more confident in relationship governance, which also increases efficiency.
Lack of procedural rigor was associated with lower performance. Buyers are therefore
encouraged to develop procedures, guidelines and behavioral norms, which we have shown to
make the dealings with suppliers more efficient.
A follow-up analysis explored whether international suppliers had lower or higher values
on our five main variables, compared to domestic suppliers. International outsourcing
relationships were characterized by higher contract specificity, higher supplier responsiveness,
and higher procedural rigor, when compared to domestic outsourcing. Higher contract specificity
and higher procedural rigor make sense in international arrangements, since transactions take
place across borders, and oftentimes cultures and viewpoints. It is therefore crucial to have a
common understanding about the relationship and its procedures, as well as the rules by which
the partnership is governed. Interesting was the finding that international outsourcing
arrangements are characterized by better supplier responsiveness. This is surprising, since the
likely geographical and cultural proximity of domestic suppliers would speak for an opposite
relationship. However, contract specificity and procedural rigor may in turn facilitate supplier
responsiveness, in that they provide structure for the relationship, including reporting and
communication requirements. No differences were detected on the purchase risk and the
purchase performance variables across the two groups. This is also an intriguing finding, since it
suggests that both domestic and international outsourcing arrangements can be equally
successful, and that risk is not reduced by staying within one’s home country.
6.4 Country specific differences on the main construct variables
Values on the four independent variables and the dependent variable from above were
also compared across countries to detect significant differences. The construct contract
specificity exhibited high values especially in Sweden and Taiwan, followed by Fiji and Mexico.
Particularly low values were recorded from respondents in Poland and Italy. Mexico, Poland and
Sweden were the countries most different from the others, as indicated by their values on
contract specificity being significantly different to all other countries (Appendix A). Contract
specificity was said to reduce uncertainty in the outsourcing arrangement. We therefore
compared our results and classification of the countries to scores of the uncertainty avoidance
index developed by Hofstede (1991). However, no relationship or pattern could be detected. We
would have expected countries classified as high in uncertainty avoidance to employ more
specific contracts. Based on the Hofstede values, however, the wish for uncertainty avoidance
does not explain higher levels in contract specificity in our data. This is an interesting finding,
and future research is encouraged to explore this relationship further.
Risk inherent in the outsourcing arrangement is relatively level across the countries, with
however China and Korea receiving the highest levels. The risk was lowest in Germany and
Italy, as indicated by the purchase risk construct. Overall, these are intriguing findings, since
they suggest that the purchase risk in Ghana is relatively comparable to the risk in Germany. It
shows that, on average, developing and emerging countries are very well up to par with
industrialized nations when it comes to risk management. This is true at least as perceived by
purchasing professionals in the respective countries. The countries that differed most from others
were Austria, with a higher risk value, and Taiwan, with a lower risk value.
Supplier responsiveness and procedural rigor were also fairly level across the countries.
For supplier responsiveness, Taiwan and Australia were at the high end, and Sweden and Fiji at
the low end. Highest values for procedural rigor were received by Poland and Korea, and lowest
by Austria and Hungary. Similar as above, these are interesting and not quite expected findings.
We would have thought the differences to be more pronounced. Countries that exhibited the
most differences to other nations were Sweden, Australia and Austria for supplier
responsiveness, and Mexico and Taiwan for procedural rigor.
The differences in countries were more pronounced when examining the purchase
performance variable. Interestingly, Ghana and Albania judged their performance with the focal
supplier as best among our sample, while respondents in Germany recorded a relatively low
performance score. Our data suggest that industrialized countries seem to exhibit lower means on
performance than developing and emerging nations. Considering however how the question was
worded, the results indicate greater discrepancies in performance (efficient use of resources to
manage the relationship) among suppliers of Ghanaian and Albanian respondents, whereas the
assessment of the relationship between the German respondents and their most important
supplier is comparable with other suppliers. When looking at what country appeared most often
in significant relationships on purchase performance, Taiwan is at the top of the list with being
significantly different to almost all other countries in terms of performance. Germany is still
significantly different to ten other countries, followed by Australia which is different to 7 other
7.1 Insights for academia
The test of hypothesis 1 supported the TCE perspective, in that higher contract specificity
can decrease uncertainty and the risk of opportunistic behavior. Purchase performance was better
in these instances. This is consistent with recent findings by Ellram et al. (2008), who suggested
lower levels of contract specificity to be associated with a lower probability to outsource
Support is also provided for the agency theory perspective, according to which
outsourcing is not recommended when agents are risk averse (Balakrishnan et al. 2008, citing
Lee et al. 2002). As a corollary, outsourcing is feasible when agents are not risk averse, or when
there is little risk to begin with. The latter was the situation among our respondents (H2).
The TCE perceptive was also supported by Hypothesis 3, which linked supplier
responsiveness to performance. Uncertainty and risk can be taken out of the transaction between
buyer and supplier, strengthening the perception of the relationship by both parties, and
increasing trust. A similar effect of reducing uncertainty and risk has procedural rigor, which was
suggested to influence performance in Hypothesis 4. This further strengthens arguments inherent
in the TCE – by introducing procedures, rules and guidelines, transactions are made more secure,
enabling their more efficient handling.
Interesting insight was received when comparing the five main variables across countries.
Contrary to our expectation we found that higher levels of uncertainty do not necessarily lead to
higher contract specificity. This was at least the case when using Hofstede’s (1991) uncertainty
avoidance index as a proxy for the level of uncertainty present in a country. No relationships or
patterns were detected. Future academic research is therefore encouraged to examine this issue
closer, and bring clarity to this discussion. Surprising was also the finding that countries in our
sample differed only slightly on their perceived purchase risk. We would have expected this risk
to be lower in developed and industrialized countries. However, this expectation was not
confirmed, which may however be due to our operationalization of the construct. The low
variances of supplier responsiveness and procedural rigor across the countries were also
unanticipated – we would have expected a clearer picture differentiating countries into clusters.
Results for the purchase performance variable indicate the focal supplier being about up-to-par
with other suppliers in industrialized countries, especially Germany, at least as measured by the
buyer’s perception of the efficiency of the sourcing relationship. More discrepancies exist in
developing and emerging countries, such as Ghana and Albania.
7.2 Insights for practitioners
Results from the hypothesis tests can also provide useful insight for practitioners. As
such, higher contract specificity was associated with better purchase performance. Detailed and
clear contract language should therefore be used when formalizing the outsourcing agreement,
incorporating clauses to address a variety of contingencies. Of particular consideration should be
clauses related to financial repayment, exclusivity, damage for poor technical performance, and
damage for late delivery. This can lead to lower levels of uncertainty, reduce the risk of
opportunistic behavior, and increase the control and confidence in the arrangement. When
contract specificity cannot be assured, the better option may be to perform the activities in-house.
In selecting potential areas to outsource, care should be taken to primarily focus on
aspects being associated with a low level of risk. With higher risk involved, activities should be
conducted in-house, to increase control and leverage. As shown by our data, the link between
purchase risk and performance was not statistically significant, which however can be explained
by high-risk activities not being chosen for outsourcing in the first place. The relationships
considered by our respondents were associated with lower levels of risk, which in turn did not
have an impact on performance.
Not to be neglected is supplier responsiveness, which had a positive impact on
performance. As such, open and frequent communication, information sharing and collaboration
should be encouraged between buyers and suppliers. These actions will also decrease uncertainty
and risk inherent in the relationship, and have to potential to make the dealings between buyer
and supplier more efficient.
A similar effect was recorded for procedural rigor, which positively impact performance.
Procedures and rules provide structure, and increase each party’s confidence in and reliance on
the relationship. Whenever possible, outsourcing agreements should be accompanied with a set
of specific rules and guidelines of how issues are handled and what processes are in place. This
will also enable to more easily determine and assess performance, since objective standards are
A follow-up analysis compared the five main variables across outsourcing relationships
that involved an international vs. a domestic supplier. Higher contract specificity, higher supplier
responsiveness, and higher procedural rigor were the characteristics of international outsourcing
arrangements, whereas the two groups did not differ on purchase risk and purchase performance.
Practitioners are therefore encouraged to ensure clear and precise contract language, as well as
unambiguous procedural rigor, especially when sourcing internationally. Since the partnerships
spans borders, and also often cultures, languages and viewpoints, it is important to have the
appropriate governance mechanisms spelled out. This in turn can explain why supplier
responsiveness is also higher among international outsourcing relationships. Due to the structures
put in place via contract specificity and procedural rigor, clear requirements and lines of
communication exist, leading the buyer to perceive the supplier as more responsive.
The comparison of construct values across countries reveals some valuable insight for
practitioners. Buyers in Sweden, Taiwan, Fiji and Mexico appreciate the inclusion of various
contract clauses, as demonstrated by high values on the contract specificity construct. On the
other hand, this is relatively unimportant for purchasers in Poland, Italy and Austria. This insight
can be used by suppliers wanting to become the outsourcing partner of companies in those
countries. As such, when soliciting the buyers business in Sweden and Taiwan, suppliers should
emphasize their willingness to include and adherence to a variety of contract clauses. This will
most likely make the supplier more acceptable, since such high value is placed on contract
specificity by buyers in these countries. On the other hand again, this issue should be relatively
unimportant for suppliers soliciting business in Poland, Italy and Austria. Resources should
therefore be better spent on relationship aspects other than contract specificity.
Purchase risk in the different countries is comparable, with however China and Korea
exhibiting higher risk, and Germany and Italy possessing lower risk. All other countries fall with
their values in between. For practitioners this suggests that outsourcing arrangements in any of
these countries are relatively comparable when it comes to purchase risk. The same is true for
supplier responsiveness and procedural rigor, which received relatively level values across the
countries. Slight differences exist for supplier responsiveness, with Taiwan and Australia at the
high end, and Sweden and Fiji at the low end of supplier responsiveness. Highest values on
procedural rigor were received in Poland and Korea, while lowest values were recorded for
Austria and Hungary. Suppliers can use this as a benchmark for requirements in these countries.
This exploratory study provides interesting overall insights and benchmarks into sourcing
and outsourcing decisions, practices and behavior. Findings are based on primarily small and
medium sized companies in 15 countries in various parts of the world, ranging from developed to
developing nations. Our study makes several contributions. First, using data collected from 806
worldwide company representatives, we investigated sourcing and outsourcing decisions in 15
countries, providing insights for academic researchers and practitioners. Most international
studies to date have taken a narrower look and focused on just a few key countries. This study,
however, investigated a broad set of countries, and provided one of the few published
comparisons that consider five or more countries. Second, we provided a characterization of the
most important supplier of our respondents across the different countries, their size and the
rationale for them being the most important supplier. Third, we summarized the reasons for
outsourcing to this supplier and attempted to identify similarities and differences in clusters of
countries. Fourth, building on the TCE, we developed and tested four hypotheses relating
behavioral sourcing variables to purchase performance. These variables consisted of contract
specificity, purchase risk, supplier responsiveness, and procedural rigor. And fifth, we
conducted supplemental tests that explore whether the 15 countries in our sample differ
significantly on these five main variables. Overall insights were provided.
Several limitations need to be mentioned. First, although this study provided insight into
outsourcing decision making by respondents in 15 countries, most of the nations were located in
Europe or Asia. No countries from North and South America, as well as the Middle East, were
included, since data were not available. Second, some countries are relatively small in terms of
industrial output, such as Albania and Fiji, and most respondents came from primarily small
companies (over 60% of the firms had less than 200 employees), possibly skewing the results.
Future studies should extend the current research to countries not yet studied, distinguish
countries based on their industrial output, and seek responses from larger firms.
And third, the data used were not specifically collected for this paper, but represent
secondary data collected as part of the GMRG. The fact that an outsourcing module was included
in this round of the survey provided an exciting and fruitful basis to contribute to this special
issue. However, we had to work with the variables and measurement items collected, and the
operationalization of the constructs could be much improved. It served the purpose of an
exploratory study well, but future research should follow-up with these initial findings,
employing a more rigorous approach, including a focused development of the measurement
instrument and primary data collection.
7.5 Future research
Future research in this area is exciting and multifarious. Three areas for further
exploration are suggested. First, our study showed that there are significant differences in how
our sample of 15 countries approach sourcing and outsourcing decisions. Key findings were
highlighted, and differences and similarities were discussed. Future research should take a more
focused look on a smaller set of countries and conduct a more detailed and centered comparison.
The present study provides a starting point for this exploration, highlighting which countries,
based on this sample, differ significantly, and which comparisons would therefore especially be
noteworthy. Second, we did not investigate why some countries differ more than others on our
variables considered. Various frameworks exist that can provide guidance to bring light into this
aspect, for example the dimensions of power distance, individualism, masculinity, uncertainty
avoidance, and long-term orientation, as introduced by Hofstede (1991). Future research in this
area should especially be stimulating. And third, it will be very insightful to compare and
contrast countries on a broader set of sourcing and outsourcing variables. This study considered
the constructs collected with the outsourcing module of the GMRG project questionnaire.
Additional concepts seem worthy of further exploration, for example the extent to which certain
sourcing and outsourcing aspects are practiced, such as single- vs. multi-sourcing, countries
especially sourced from, as well as risks involved in the outsourcing decision. Sourcing and
outsourcing in global supply chains remains an exciting topic for investigation. It is hoped that
the present study provides motivation and a starting point for further exploration.
Adeleye, B.C., Annansingh, F. and Nunes, M.B., Risk management practices in IS outsourcing:
an investigation into commercial banks in Nigeria. Int. J. Inform. Manage., 2004, 24, 167-180.
Aksin,O.Z., de Véricourt, F. and Karaesmen, F., Call center outsourcing contract analysis and
choice. Manage. Sci., 2008, 54, 354-368.
Aksin, O.Z. and Masini, A., Effective strategies for internal outsourcing and offshoring of
business services: an empirical investigation. J. Oper. Manage., 2008, 26, 239-256.
Amaral, J., Billington, C.A. and Tsay, A.A., Outsourcing production without losing control.
Supply Chain Manage. Rev., 2004, 8, 44-52.
Amundson, S.D., Relationships between theory-driven empirical research in operations
management and other disciplines. J.Oper. Manage., 1998, 16, 341-359.
Arnold, U., New dimensions of outsourcing: a combination of transaction cost economics and
the core competencies concept. Europ. J. Purch. Suppl. Manage., 2000, 6, 23-29.
Arya, A., Mittendorf, B. and Sappington, D.E.M., The make-or-buy decision in the presence of a
rival: strategic outsourcing to a common supplier. Manage. Sci., 2008, 54, 1747-1758.
Aubert, B.A., Beaurivage, G., Croteau, A.-M. and Rivard, S., Firm strategic profile and IT
outsourcing. Information Systems Frontiers, 2008, 10,129-143.
Avlonitis, G.J. and Salavou, H.E., Entrepreneurial orientation of SMEs, product innovativeness,
and performance. J. Bus. Res., 2007, 60, 566-575.
Balakrishnan, K., Mohan, U. and Seshadri, S., Outsourcing of front-end business processes:
quality, information, and customer contact. J. Oper. Manage., 2008, 26, 288-302.
Bardhan, I., Mithas, S. and Lin, S., Performance impacts of strategy, information technology
applications, and business process outsourcing in U.S. manufacturing plants. Prod. Oper.
Manage., 2007, 16, 747-762.
Barney, J.B., Firm resources and sustained competitive advantage. J. Manage., 1991, 17, 99-120.
Bunn, M.D., Key aspects of organizational buying: conceptualization and measurement. J. Acad.
Marketing Sci., 1994, 22, 160-169.
Carr, A.S. and Smeltzer, L.R., An empirical study of the relationships among purchasing skills
and strategic purchasing, financial performance, and supplier responsiveness. J. Supply Chain
Manage., 2000, 36, 40-54.
Chan, F.T.S., Kumar, V. and Tiwari, M.K., The relevance of outsourcing and leagile strategies in
performance optimization of an integrated process planning and scheduling model. Int. J. Prod.
Res., 2009, 47, 119-142.
Chen, S.-H.S., A transaction cost rationale for private branding and its implications for the
choice of domestic vs offshore outsourcing. J. Int. Bus. Stud., 2009, 40, 156-175.
Conan, A. and Ronen, B., Production outsourcing: a linear programming model for the Theory-
Of-Constraints. Int. J. Prod. Res., 2000, 38, 1631-1639.
Cronbach, L.J., Coefficient alpha and the internal structure of tests. Psychometrika, 1951, 16,
Dabhilkar, M. and Bengtsson, L., Invest or divest? On the relative improvement potential in
outsourcing manufacturing. Production Planning & Control, 2008, 19, 212-228.
Dekkers, R., Decision models for outsourcing and core competencies in manufacturing. Int. J.
Prod. Res., 2000, 38, 4085-4096.
Dhawan, A., Srinivasan, S., Rajib, P. and Bidanda, B., Minimising total cost with regular and
emergency outsourcing sources: a neuro-dynamic programming approach. Int. J. Prod. Res., in
Doig, S.J., Ritter, R.C., Speckhals, K. and Woolson, D., Has outsourcing gone too far? McKinsey
Q., 2001, 25-37.
Ellram, L.M., Tate, W.L. and Billington, C., Offshore outsourcing of professional services: a
transaction cost economic perspective. J. Oper. Manage., 2008, 26, 148-163.
Goodman, P., Fichman, M., Lerch, F. and Snyder, P., Customer–firm relationships, involvement,
and customer satisfaction. Acad. Manage. J., 1995, 38, 1310–1324.
Green, S.B. and Salkind, N.J., Using SPSS for Windows and Macintosh, 2003 (Prentice Hall:
Upper Saddle River, NJ).
Gunasekaran, A. and Kobu, B., Performance measures and metrics in logistics and supply chain
management: a review of recent literature (1995-2004) for research and applications. Int. J.
Prod. Res., 2007, 45, 2819-2840.
Gunasekaran, A., Ngai, E.W.T. and McGaughey, R.E., Information technology and systems
justification: a review for research and applications. Eur. J. Operat. Res., 2007, 173, 957-983.
Gunasekaran, A., Lai, K.-h. and Cheng, T.C.E., Responsive supply chain: a competitive strategy
in a networked economy. OMEGA, 2008, 36, 549-564.
Handfield, R.B. and Bechtel, C., The role of trust and relationship structure in improving supply
chain responsiveness. Ind. Marketing Manage., 2002, 31, 367-382.
Handfield, R., Krause, D., Scannell, T. and Monczka, R., An empirical investigation of supplier
development: reactive and strategic processes. J.Oper. Manage., 1998, 17, 39–58.
Harland, C., Knight, L., Lamming, R. and Walker, H., Outsourcing: assessing the risks and
benefits of organizations, sectors and nations. Int. J. Oper. Prod. Manage., 2005, 25, 831-850.
Hartley, J., Lane, M.D. and Hong, Y., An exploration of the adoption of e-auctions in supply
management. IEEE Trans. Eng. Manage., 2004, 51, 153-161.
Hofstede, G.H. Cultures and Organizations: Software of the Mind. 1991 (McGraw-Hill: London,
Holcomb, T.R. and Hitt, M.A., Toward a model of strategic outsourcing. J.Oper. Manage., 2007,
Holweg, M., An investigation into supplier responsiveness: empirical evidence from the
automotive industry. Int. J. Log.Manage., 2005, 16,1 96-119.
Hunter, G.K., Bunn, M.D. and Perreault Jr., W.D., Interrelationships among key aspects of the
organizational procurement process. Int. J. Res. in Marketing, 2006, 23, 155-170.
Jahns, C., Hartmann, E. and Bals, L., Offshoring: dimensions and diffusion of a new business
concept. J. Purch.. Suppl. Manage., 2006, 12, 218-231.
Johnston,W.J. and Lewin, J.E., Organizational buying behavior: toward an integrative
framework. J. Bus. Res., 1996, 35, 1-15.
Lau, K. and Zhang, H.J., Drivers and obstacles of outsourcing practices in China. Int. J. Phys.
Distr. Log. Manage., 2006, 36, 776-792.
Lee, J.-N., Huynh, Q.M., Kwok, C.W.R., Pi, S.-M., The current and future directions of IS
outsourcing. In: Hirschheim, R., Heinzl, A., Dibbern, J. (Eds.), Information Systems
Outsourcing: Enduring Themes, Emergent Patterns and Future Directions, 2002, pp. 195–220
McIvor, R., How the transaction cost and resource-based theories of the firm inform outsourcing
evaluation. J. Oper. Manage., 2009, 27, 45-63.
Melvin, J., Domestic taste differences, transportation costs, and international trade. J. Int. Econ.,
1985, 18, 65-82.
Min, H. and Galle, W.P., E-purchasing: profiles of adopters and nonadopters. Ind. Marketing
Manage., 2003, 32, 227-233.
Mishra, N., Choudhary, A.K. and Tiwari, M.K., Modeling the planning and scheduling across
the outsourcing supply chain: a Chaos-based fast Tabu-SA approach. Int. J. Prod. Res., 2008, 46,
Monczka, R.M., Markham, W.J., Carter, J.R., Blascovich, J.D. and Slaight, T.H., Outsourcing
strategically for sustainable competitive advantage. CAPS Research Report, 2005.
Novak, S. and Stern, S., How does outsourcing affect performance dynamics? Evidence from the
automobile industry. Manage. Sci., 2008, 54, 1963-1979.
Pagell, M., Krumwiede, D.W. and Sheu, C., Efficacy of environmental and supplier relationship
investments – moderating effects of external environment. Int. J. Prod. Res., 2007, 45, 2005-
Pagell, M. and Sheu, C., Buyer behaviours and the performance of the supply chain: an
international exploration. Int. J. Prod. Res., 2001, 39, 2783-2801.
Prahalad, K. and Hamel, G., The core competence of the corporation. Harv. Bus. Rev., 1990, 68,
Ramaswami, S.N., Marketing controls and dysfunctional employee behavior: a test of traditional
and contingency postulates. J. Marketing, 1996, 60, 105-120.
Ren, Z.J. and Zhou, Y.-P., Call center outsourcing: coordinating staffing level and service
quality. Manage. Sci., 2008, 54, 369-383.
Ross, W.T., Jr., Dalsace, F. and Anderson, E., Should you set up your own sates force or should
you outsource it? Pitfalls in the standard analysis. Bus. Hor., 2005, 48, 23-36.
Ruffin, R., The Ricardian factor endowment theory of international trade. Int. Econ. J., 1990, 4,
Ruiz-Torres, A.J., López, F.J., Ho, J.C. and Wojciechowski, P.J., Minimizing the average
tardiness: the case of outsource machines. Int. J. Prod. Res., 2008, 46, 3615-3640.
Sanders, N.R., Locke, A., Moore, C.B. and Autry, C.W., A multidimensional framework for
understanding outsourcing arrangements. J. Suppl. Chain Manage., 2007, 43, 3-15.
Schoenherr, T., Diffusion of online reverse auctions for B2B procurement: an exploratory study.
Int. J. Oper. Prod. Manage., 2008, 28, 259-278.
Schoenherr, T., Tummala, V.M.R. and Harrison, T.P., Assessing supply chain risks with the
analytic hierarchy process: providing decision support for the offshoring decision by a US
manufacturing company. J. Purch. Suppl. Manage., 2008, 14, 100-111.
Serrato, M.A., Ryan, S.M. and Gayt n, J., A Markov decision model to evaluate outsourcing in
reverse logistics. Int. J. Prod. Res., 2007, 45, 4289-4315.
Sharif, A.M., Irani, Z. and Lloyd, D., Information technology and performance management for
build-to-order supply chains. Int. J. Prod. Oper. Manage., 2007, 27, 1235-1253.
Sheu, C. and Wacker, J.G., Effectiveness of planning and control systems: an empirical study of
US and Japanese firms. Int. J. Prod. Res., 2001, 39, 887-905.
St. John, C.H., Cannon, A.R., Pouder, R.W., Change drivers in the new millennium: implications
for manufacturing strategy research. J. Oper. Manage., 2001, 19, 143-160.
Wernerfelt, B., A resource-based view of the firm. Strat. Manage. J., 1984, 5, 171-180.
Wacker, J.G. and Sheu, C., Effectiveness of manufacturing planning and control systems on
manufacturing competitiveness: evidence from global manufacturing data. Int. J. Prod. Res.,
2006, 44, 1015-1036.
Wacker. J.G. and Sprague, L.G., The impact of institutional factors on forecast accuracy:
manufacturing executives’ perspective. Int. J. Prod. Res., 1995, 33, 2945-2958.
Whybark, D.C., GMRG survey research in operations management. Int. J. Oper. Prod. Manage.,
1997, 17, 686-696.
Whybark, D.C. and Vastag, G., Global Manufacturing Practices: A Worldwide Survey of
Practices in Production Planning and Control. 1993 (Elsevier: Amsterdam, The Netherlands).
Williamson, O.E., Outsourcing: transaction cost economics and supply chain management. J.
Suppl. Chain Manage., 2008, 44, 5-16.
Wu, F., Li, H.Z., Chu, L.K. and Sculli, D., An outsourcing decision model for sustaining long-
term performance. Int. J. Prod. Res., 2005, 43, 2513-2535.
Youngdahl, W. and Ramaswamy, K., Offshoring knowledge and service work: a conceptual
model and research agenda. J. Oper. Manage., 2008, 26, 212-221.
Yusuf, Y., Gunasekaran, A. and Wu, C., Implementation of enterprise resource planning in
China. Technovation, 2006, 26, 1324-1336.
Table 1. Responses and total employees per country.
Responses per Country Total Employees
n % of N Mean S.D.
15 1.86 33.67 38.42
Australia 30 3.72 436.53 912.92
Austria 17 2.11 1098.87 1886.24
China 57 7.07 490.46 587.98
Fiji 110 13.65 303.37 445.97
Germany 55 6.82 285.85 1679.28
Ghana 63 7.82 147.66 234.81
Hungary 53 6.58 567.74 1178.21
Italy 54 6.70 155.89 280.45
Korea 114 14.14 1803.46 5076.90
Macedonia 39 4.84 185.33 649.00
Mexico 79 9.80 248.28 442.88
Poland 57 7.07 347.74 809.10
Sweden 13 1.61 479.08 739.52
Taiwan 50 6.20 264.12 276.88
Total 806 100.00 538.23 2161.09
Table 2. Aggregate total employee data.
Total Employees Count Percentage
≤ 25 139 18.19
26-50 98 12.83
51-100 131 17.15
101-200 120 15.71
201-400 125 16.36
401-1000 96 12.57
> 1000 55 7.20
Total 764 100
Table 3. The most important supplier.
Most important supplier: international or
Importance of this supplier in terms of
percentage on total plant materials
Mean Standard Deviation
Albania 46.7 53.3 n.s. 0.54 0.26
Australia 63.3 36.7 n.s. 0.42 0.23
Austria 93.3 6.7 *** 0.31 0.20
China 82.5 17.5 *** 0.38 0.21
Fiji 66.4 33.6 *** 0.42 0.16
Germany 38.9 61.1 n.s. 0.26 0.18
Ghana 63.9 36.1 ** 0.67 0.23
Hungary 42.3 57.7 n.s. --- ---
Italy 45.1 54.9 n.s. 0.29 0.20
Korea 25.7 74.3 *** 0.31 0.26
Macedonia 80.0 20.0 *** 0.52 0.25
Mexico 44.2 55.8 n.s. 0.50 0.25
Poland 15.8 84.2 *** 0.46 0.29
Sweden 69.2 30.8 n.s. 0.22 0.18
Taiwan 82.0 18.0 *** 0.29 0.28
Total 52.8 47.2 0.41 0.26
*** p < .01; ** p < .05; * p < .10.
Table 4. Relative size of the supplier (in percent).
Smaller About the
Albania 20.0 40.0 6.7 20.0 13.3
Australia 10.0 46.7 20.0 16.7 6.7
Austria 33.3 13.3 33.3 6.7 13.3
China 8.8 19.3
36.8 26.3 8.8
Fiji 2.8 1.8 38.5 39.4 17.4
55.6 18.5 14.8 9.3 1.9
48.1 35.2 3.7 5.6 7.4
30.8 7.7 21.2 17.3
32.7 32.7 10.2 18.4 6.1
Korea 25.7 27.6 5.7 10.5
48.5 3.0 12.1 6.1
Mexico 20.5 32.1 21.8 19.2 6.4
30.9 29.1 21.8 9.1 9.1
30.8 23.1 15.4 15.4 15.4
Taiwan 4.1 16.3 22.4 53.1 4.1
Total 23.3 25.1 18.6 20.6 12.4
Table 5. Primary reason for classifying this supplier as most important (in percent).
Albania 33.3 6.7 6.7 20.0 20.0 13.3
66.7 16.7 0.0 3.3 6.7 6.7
76.9 7.7 15.4 0.0 0.0 0.0
China 17.5 24.6
28.1 3.5 8.8 17.5
50.9 8.2 3.6 9.1 27.3 0.9
66.7 5.6 3.7 7.4 5.6 11.1
36.8 3.5 31.6 3.5 7.0
66.7 2.0 7.8 0.0 7.8 15.7
66.0 14.0 0.0 4.0 2.0 14.0
47.6 13.6 18.4 3.9 1.9 14.6
Macedonia 37.1 5.7 0.0 20.0 22.9 14.3
53.2 10.4 15.6 3.9 0.0 16.9
Poland 25.0 14.3 7.1 10.7 17.9 25.0
Sweden 33.3 25.0 0.0 8.3 8.3 25.0
44.0 6.0 16.0 6.0 4.0
Total 45.1 15.5 9.0 9.0 9.6 11.9
Table 6. Primary reasons for outsourcing (in percent).
No resources at
Lack of specific
Required by parent
No patent for
Not enough time to
Lack of specific
No access to
Albania 28.6 7.1 --- 7.1 --- 21.4 --- 7.1 28.6 ---
Australia 10.0 3.3 6.7 6.7 30.0 --- 13.3 6.7 10.0 13.3
Austria 15.4 --- 7.7 --- --- 7.7
30.8 15.4 --- 23.1
21.1 10.5 1.8 15.8 5.3 17.5 --- --- 12.3
27.3 15.5 0.9 3.6 0.9 10.0 13.6 9.1 18.2 0.9
35.2 7.4 --- 1.9 7.4 14.8 3.7 9.3 20.4
Ghana 8.8 7.0 7.0 --- 3.5 3.5 29.8
31.6 3.5 5.3
Hungary 9.8 --- --- 7.3 --- ---
41.5 7.3 14.6 19.5
32.6 --- 7.0 7.0 11.6 7.0 4.7 7.0 23.3
Korea 10.6 4.3 6.4 7.4 4.3 2.1 14.9 2.1 41.5 6.4
Macedonia 22.9 2.9 ---
31.4 5.7 8.6 17.1 --- 2.9 8.6
25.0 6.9 2.8 8.3 5.6 5.6 12.5 --- 9.7 23.6
Poland 44.9 2.0 6.1 2.0 20.4 16.3 4.1 --- 2.0 2.0
30.8 --- 15.4 7.7 --- 7.7 15.4 --- 7.7 15.4
22.0 2.0 12.0 4.0 20.0 20.0 --- 10.0 4.0
Total 21.2 8.3 4.2 6.4 6.4 7.2 16.5 5.7 13.3 10.7
Table 7. Construct measurement items.
Construct Variable Measurement Item / Survey Question Mean S.D.
α = 0.81
SPEC1 Financial repayment if contract is terminated prior to its ending
SPEC2 Exclusivity clause (that is, are you an exclusive supplier by
SPEC3 Damage for poor technical performance 3.01 1.58
SPEC4 Damage for late delivery 2.84 1.55
α = 0.73
RISK1 Technology risk associated with this supplier 2.10 0.92
RISK2 What is your behavioral risk associated with this supplier? 2.27 0.93
RISK3 What is the market risk associated with this supplier's products? 2.42 0.94
α = 0.68
SUPPL1 Provides its production forecasts, plans, schedules and supply
SUPPL2 How well does this supplier products work, endure and fit your
SUPPL3 Does your firm and this supplier develop joint efforts to stabilize
of production schedules?
α = 0.62
PROC1 The product specifications were carefully specified before
PROC2 It is easy to determine the performance of this supplier. 1.72 0.75
PROC3 There is a standard approach when solving a problem with this
Table 8. Construct values by country.
Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D.
Albania 2.63 1.82 2.16 0.83 3.62 0.59 1.17 0.39 4.00 1.04
Australia 2.39 1.29 2.40 0.78 3.78 0.77 1.63 0.49 3.41 1.12
Austria 2.11 1.28 2.44 0.71 3.74 0.63 1.13 0.35 3.62 1.12
China 2.66 1.09 2.65 0.69 3.75 0.63 1.66 0.48 3.93 1.02
Fiji 3.26 0.99 2.08 0.34 3.35 0.43 1.47 0.50 3.87 0.79
Germany 2.28 1.05 2.02 0.65 3.38 0.74 1.65 0.48 2.98 0.90
Ghana 2.73 0.93 2.09 0.63 3.57 0.69 1.44 0.50 4.12 1.20
Hungary 2.26 1.11 2.36 0.73 3.36 0.68 1.14 0.35 3.65 1.00
Italy 1.85 1.37 2.02 0.69 3.36 0.78 1.31 0.47 3.64 0.82
Korea 2.77 1.26 2.58 0.88 3.55 0.85 1.78 0.41 3.87 1.03
Macedonia 2.16 1.26 2.36 0.67 3.76 0.79 1.16 0.37 3.85 0.91
Mexico 3.12 1.04 2.17 0.83 3.60 0.70 1.30 0.46 3.88 1.06
Poland 1.04 1.02 2.28 0.69 3.36 0.82 1.84 0.37 3.64 0.95
Sweden 3.84 1.20 2.28 0.76 3.26 0.61 1.60 0.52 3.33 1.30
Taiwan 3.84 0.96 2.29 0.84 4.00 0.79 1.44 0.50 3.84 1.20
Total 2.57 1.33 2.27 0.73 3.54 0.73 1.49 0.50 3.75 1.03
Table 9. The impact of contract specificity on buyer’s purchase performance.
15 10 111 92 72 300
% within LCS
5.0 3.3 37.0 30.7 24.0 100.0
% within Perf.
55.6 37.0 55.5 48.4 41.1 48.5
% of Total
2.4 1.6 17.9 14.9 11.6 48.5
12 17 89 98 103 319
% within HCS
3.8 5.3 27.9 30.7 32.3 100.0
% within Perf.
44.4 63.0 44.5 51.6 58.9 51.5
% of Total
1.9 2.7 14.4 15.8 16.6 51.5
27 27 200 190 175 619
% of Total
4.4 4.4 32.3 30.7 28.3 100.0
(4, N = 619) = 9.68, p < 0.05, Cramér’s V = 0.13
Table 10. The impact of purchase risk on buyer’s purchase performance.
13 9 123 122 100 367
% within LR
3.5 2.5 33.5 33.2 27.2 100.0
% within Perf.
44.8 34.6 50.8 54.2 49.5 50.7
% of Total
1.8 1.2 17.0 16.9 13.8 50.7
16 17 119 103 102 357
% within HR
4.5 4.8 33.3 28.9 28.6 100.0
% within Perf.
55.2 65.4 49.2 45.8 50.5 49.3
% of Total
2.2 2.3 16.4 14.2 14.1 49.3
29 26 242 225 202 724
% of Total
4.0 3.6 33.4 31.1 27.9 100.0
(4, N = 724) = 4.33, p > 0.10, Cramér’s V = 0.08
Table 11. The impact of supplier responsiveness on buyer’s purchase performance.
19 20 154 101 67 361
% within LRe
5.3 5.5 42.7 28.0 18.6 100.0
% within Perf.
65.5 74.1 63.9 44.7 33.2 49.8
% of Total
2.6 2.8 21.2 13.9 9.2 49.8
10 7 87 125 135 364
% within HRe
2.7 1.9 23.9 34.3 37.1 100.0
% within Perf.
34.5 25.9 36.1 55.3 66.8 50.2
% of Total
1.4 1.0 12.0 17.2 18.6 50.2
29 27 241 226 202 725
% of Total
4.0 3.7 33.2 31.2 27.9 100.0
(4, N = 725) = 53.11, p < 0.01, Cramér’s V = 0.27
Table 12. The impact of procedural rigor on buyer’s purchase performance.
12 17 98 90 63 280
% within LRi
4.3 6.1 35.0 32.1 22.5 100.0
% within Perf.
52.2 77.3 50.0 54.2 42.0 50.3
% of Total
2.2 3.1 17.6 16.2 11.3 50.3
11 5 98 76 87 277
% within HRi
4.0 1.8 35.4 27.4 31.4 100.0
% within Perf.
47.8 22.7 50.0 45.8 58.0 49.7
% of Total
2.0 0.9 17.6 13.6 15.6 49.7
23 22 19 166 150 557
% of Total
4.1 3.9 35.2% 29.8 26.9 100.0
(4, N = 557) = 11.59, p < 0.05, Cramér’s V = 0.14
Table 13. Summary of Mann-Whitney U tests.
the 0.01 level
the 0.05 level
the 0.1 level
Contract specificity 67 63.81% 4 (3.81%) 13 (12.38%) 50 (47.62%)
Purchase risk 39 37.14% 13 (12.38%) 12 (11.43%) 14 (13.33%)
Supplier responsiveness 41 39.05% 7 (6.67%) 16 (15.24%) 18 (17.14%)
Procedural rigor 62 59.05% 7 (6.67%) 15 (14.29%) 40 (38.10%)
Figure 1. Research framework.
Figure 2. Boxplot for contract specificity.
Figure 3. Boxplot for purchase risk.
Figure 4. Boxplot for supplier responsiveness.
Figure 5. Boxplot for procedural rigor.
Figure 6. Boxplot for purchase performance.
* *** ***
** ** **
** *** *
** *** **
* *** **
** *** * *
*** *** *** ***
*** *** *** *
** *** * **
*** *** * *
*** *** ***
*** ** *** *** ** *** ***
** *** ** **
* *** ** **
*** ** ** **
*** *** *** ** *** ** ***
** ** *** *** *** ***
*** *** *** *** ***
*** *** *** *** *** *** *** *** *** *** ***
*** * * **
** *** **
*** *** *** *** *** *** *** *** ***
** *** *** * *** *** *** *** *** *** ** ***
** *** * ** ** **
** ** * *** **
* *** *** ** *** *** *** *** *** *** *** ***
* ** *
*** ** *** *** *** ** *** ***
** *** *** ** ***
** *** ** ** *** *** *** *** ***
** *** * ** *** ** * *
** *** * ** ***
*** ** ** *** *** *** * *
*** p < .01; ** p < .05; * p < .10.
Comparison of contract
specificity, purchase risk,
supplier responsiveness and
procedural rigor across countries
*** *** *
*** * ***
* ** *
*** * *** *** ** *** *** *** *** *** *** *** ***
*** p < .01; ** p < .05; * p < .10.
Comparison of purchase