St. Petersburg University
Graduate School of Management
Master in Management Program
Factors of Influence on the Stability of Strategic Alliances
Master thesis by the 2nd year student
Anastasiia Reusova
Concentration — Master in Management
Research advisor:
Nikolay A. Zenkevich, Associate Professor
St. Petersburg
2016
1
ЗАЯВЛЕНИЕ О САМОСТОЯТЕЛЬНОМ ХАРАКТЕРЕ ВЫПОЛНЕНИЯ
ВЫПУСКНОЙ КВАЛИФИКАЦИОННОЙ РАБОТЫ
Я, __Реусова Анастасия Игоревна__, студент второго курса магистратуры
направления «Менеджмент», заявляю, что в моей магистерской диссертации на тему
«Факторы влияния на устойчивость стратегических альянсов», представленной в службу
обеспечения программ магистратуры для последующей передачи в государственную
аттестационную комиссию для публичной защиты, не содержится элементов плагиата.
Все прямые заимствования из печатных и электронных источников, а также из
защищенных ранее выпускных квалификационных работ, кандидатских и докторских
диссертаций имеют соответствующие ссылки.
Мне известно содержание п. 9.7.1 Правил обучения по основным образовательным
программам высшего и среднего профессионального образования в СПбГУ о том, что «ВКР
выполняется индивидуально каждым студентом под руководством назначенного ему
научного руководителя», и п. 51 Устава федерального государственного бюджетного
образовательного
учреждения
высшего
образования
«Санкт-Петербургский
государственный университет» о том, что «студент подлежит отчислению из СанктПетербургского
университета
за
представление
курсовой
или
выпускной
квалификационной работы, выполненной другим лицом (лицами)».
_______________________________________________(Подпись студента)
_________________26.05.2016______________________ (Дата)
STATEMENT ABOUT THE INDEPENDENT CHARACTER OF
THE MASTER THESIS
I, __Anastasiia Reusova__, (second) year master student, program «Management», state
that my master thesis on the topic «Factors of Influence on the Stability of Strategic Alliances»,
which is presented to the Master Office to be submitted to the Official Defense Committee for the
public defense, does not contain any elements of plagiarism.
All direct borrowings from printed and electronic sources, as well as from master theses,
PhD and doctorate theses which were defended earlier, have appropriate references.
I am aware that according to paragraph 9.7.1. of Guidelines for instruction in major
curriculum programs of higher and secondary professional education at St.Petersburg University
«A master thesis must be completed by each of the degree candidates individually under the
supervision of his or her advisor», and according to paragraph 51 of Charter of the Federal State
Institution of Higher Education Saint-Petersburg State University «a student can be expelled from
St.Petersburg University for submitting of the course or graduation qualification work developed
by other person (persons)».
________________________________________________(Student’s signature)
__________________26.05.2016_____________________ (Date)
2
Table of Contents
АННОТАЦИЯ .............................................................................................................................. 4
ABSTRACT ...................................................................................................................................5
PREFACE AND ACKNOWLEDGEMENTS ............................................................................6
LIST OF ABBREVIATIONS AND ACRONYMS .....................................................................7
INTRODUCTION .........................................................................................................................8
CHAPTER 1. STABILITY IN STRATEGIC ALLIANCES ..................................................13
1.1 Strategic alliances as a form of cooperation ............................................................................13
1.2 Explaining strategic alliance success .......................................................................................17
1.3 Strategic alliance stability in academic literature ....................................................................20
1.4 Strategic alliance stability conceptualization ..........................................................................28
1.5 Factors of strategic alliance stability: theoretical perspectives ...............................................32
1.6 Chapter 1 concluding remarks .................................................................................................37
CHAPTER 2. STRATEGIC ALLIANCE STABILITY FACTORS HYPOTHESES
DEVELOPMENT AND METHODOLOGY ............................................................................39
2.1 Strategic alliance stability factors and hypotheses ..................................................................39
2.2 Data collection: resources and restrictions ..............................................................................46
2.3 Chapter 2 concluding remarks .................................................................................................53
CHAPTER 3. EMPIRICAL STUDY ON STRATEGIC ALLIANCE STABILITY
FACTORS ....................................................................................................................................55
3.1 Data analysis and measurements ............................................................................................. 55
3.2 Measurement model development and assessment .................................................................58
3.3 Structural model specification and assessment .......................................................................68
3.4 Modeling results and analysis..................................................................................................72
3.5 Chapter 3 concluding remarks .................................................................................................78
CONCLUSION AND IMPLICATIONS ...................................................................................80
LIMITATIONS AND FURTHER RESEARCH ......................................................................84
LIST OF REFERENCES............................................................................................................86
APPENDICES.............................................................................................................................. 99
Appendix 1: Survey Cover Letter .............................................................................................. 99
Appendix 2: Survey Questionnaire Questions .........................................................................100
Appendix 3: Cronbach’s Alpha Test Results ...........................................................................105
Appendix 4: Initial Confirmatory Factor Analysis Output ......................................................107
Appendix 5: Modification Indices ........................................................................................... 109
Appendix 6: Final Confirmatory Factor Analysis Output .......................................................113
Appendix 7: Structural Model Output .....................................................................................115
3
АННОТАЦИЯ
Автор
Реусова Анастасия Игоревна
Название
магистерской
диссертации
Факультет
Факторы влияния на устойчивость стратегических альянсов
Направление
Менеджмент
Высшая школа менеджмента
подготовки
Год
2016
Научный
Зенкевич Николай Анатольевич, кандидат физ.-мат. наук, доцент
руководитель
Описание
Целью
исследования
цели, задач и
устойчивостью стратегических альянсов и межорганизационными
основных
факторами.
результатов
академической
Задачи
является
выявление
исследования:
литературы
(1)
дать
взаимосвязей
на
основании
определение
между
анализа
устойчивости
стратегических альянсов; (2) разработать концептуальную модель
факторов устойчивости стратегических альянсов; (3) на основании
эмпирического
исследования
сделать
выводы
о
связях
между
устойчивостью стратегических альянсов и ее факторами.
Результаты показали, что долгосрочная ориентация партнеров значимо
влияет на внешнюю устойчивость, тогда как взаимодополняемость
ресурсов и доверие, а также внешняя устойчивость влияют на
внутреннюю устойчивость стратегических альянсов. Более того,
подтвердилось
предположение
о
положительном
влиянии
взаимодополняемости ресурсов на доверие, и доверия – на долгосрочную
ориентацию партнеров.
Ключевые
Устойчивость стратегических альянсов, внутренняя устойчивость,
слова
внешняя
устойчивость,
взаимодополняемость
доверие,
ресурсов,
долгосрочная
теория
игр,
теория
ориентация,
ресурсного
преимущества.
4
ABSTRACT
Master Student's Name
Anastasiia Reusova
Master Thesis Title
Factors of Influence on the Stability of Strategic Alliances
Faculty
Graduate School of Management
Main field of study
Management
Year
2016
Academic Advisor's Name Nikolay A. Zenkevich, Associate Professor
Description of the goal, The goal of the study is to identify relationships between strategic
tasks and main results
alliance stability and inter-organizational strategic alliance stability
factors. Research objectives are: (1) based on academic literature
analysis on strategic alliances, define the term of strategic alliance
stability; (2) develop a conceptual model of strategic alliance
stability factors; (3) based on empirical research, make conclusions
about relationships between strategic alliances stability and its
factors.
Results show that partners’ long-term orientation significantly and
positively
influences
external
stability,
while
resource
complementarity and trust influence strategic alliance internal
stability. Moreover, the assumptions on the positive association
between resource complementarity and trust, and between trust and
long-term orientation have been supported.
Keywords
Strategic alliance stability, external stability, internal stability,
trust, long-term orientation, resource complementarity, game
theory, resource-advantage theory
5
PREFACE AND ACKNOWLEDGEMENTS
Writing this Master thesis was truly an exciting academic journey, a great experience I
have gained during my Master studies at GSOM. It was a challenging and exciting process, and I
would like to briefly thank people who have helped me in accomplishing this journey by investing
their time and encouraging my interest to the study.
By grasping this opportunity, I kindly express my gratitude to Nikolay A. Zenkevich, my
research advisor, Anton Grigoryev, without whose support I would have never completed the
empirical part of a research, as well as Siegfried Leidig, Andrew Ewing, Victor Wan, Saha
Banibrata, Joe Kittel and Anil Nell who have spent a significant amount of time providing me with
a managerial perspective on the issue of strategic alliance stability.
Anastasiia Reusova,
Saint Petersburg,
June, 2016
6
LIST OF ABBREVIATIONS AND ACRONYMS
AVE
Average variance extracted
CR
Construct reliability
ES
External stability
IJV
International joint venture
IS
Internal stability
JV
Joint venture
LTO
Long-term orientation
MI
Modification index
MM
Measurement model
R-A
Resource-advantage
RBV
Resource-based view
SA
Strategic alliance
SAS
Strategic alliance stability
SEM
Structural equation modeling
SM
Structural model
SR
Standardized residuals
T
Trust
TCE
Transaction-cost economy
WOS
Wholly-owned subsidiary
7
INTRODUCTION
Research background. The last 50 years have shown an immense growth in emergence of
strategic alliances. Hence, the attention of researchers and practitioners to this issue has been
increasing (Christoffersen, 2013). The growth of the number of strategic alliances was especially
noticeable starting from the 80s, which corresponded to the growing relational and institutional
complexity of managing these forms of collaboration. Strategic alliances are widely recognized to
be a helpful form on inter-organizational relationships that aids firms in standing against the
competition in a complex business environment (Akkaya, 2007) and in creating customer value
(Iyer, 2002; Umukoroa, Sulaimonb, Kuyeb, 2009). However, some scholars estimate the failure
rate of strategic alliances to mount to 60-65% due to unmet objectives, failed expectations or other
reasons (Geringer and Hebert, 1991; Umukoroa, Sulaimonb, Kuyeb, 2009; Gibbs, Humphries,
2016). At the same time, growing competition, raising research and development costs, shortening
product life cycles lead to a new surge in emergence of strategic alliances (Gibbs, Humphries,
2016).
Academic studies on strategic alliances have been carried out for at least 50 years with the
past 30 years being the most intense (Umukoroa, Sulaimonb and Kuyeb, 2009). The earliest studies
on strategic alliances include (Friedmann and Kalmanoff, 1961; Franko, 1971). With the passage
of time, along with strategic alliances growing prevalence, the studies on strategic alliances were
progressing from the most broad to increasingly specific, addressing specific types of alliances
(e.g., joint ventures, international joint ventures, non-equity alliances, international strategic
alliances) and specific issues in alliances (e.g., motivation for collaboration, alliance performance,
alliance stability).
The interest to strategic alliances is not only academic, because (Vyas, Shelburn, Rogers,
1995) it is crucial for a partner entering an alliance to have a thorough understanding of an alliance
along with its requirements, objectives, expectations and expected benefits.
Strategic alliance stability is in the focus of this particular study. Stability of long-term
cooperative decisions, and strategic alliance stability in particular, is recognized to be a
fundamental problem that is studied in academic literature for the last 30 years. The problem of
strategic alliance stability is widely recognized not only by scholars, but also by practitioners
(Zenkevich, Koroleva, Mamedova, 2014a, b). The drawback of most of the researches on the topic
is in viewing strategic alliance stability as a static (Jiang, Li and Gao, 2008) and one-dimensional
concept (Zenkevich, Koroleva, Mamedova, 2014a), while relationships between partners in an
8
alliance are certainly dynamic, and managing this dynamics is challenging (Douma et. al. 2000;
Buffenoir, Bourdon, 2013).
Problem statement. Many issues related to strategic alliance stability remain arguable. On
the one hand, strategic alliance stability is well studied in game theory as a part of cooperative
decisions stability (Zenkevich, Koroleva, Mamedova, 2014a,b). On the other hand, there has been
a significant number of attempts to study strategic alliance stability determinants, e.g., trust,
partners’ goal congruence, governance mechanisms (Jiang, Li and Gao, 2008; Deitz et al, 2010;
Christoffersen, Plenborg and Robson, 2014; Isidor et al, 2015; Qing, Zhang, 2015). However,
many of those studies show contradictory results due to different reasons, in particular, because
the concept of strategic alliance stability lacks precision (Jiang, Li and Gao, 2008). Moreover,
there is a scarcity of papers to provide managers with a comprehensive tool for strategic alliance
stability management.
This research contributes to the field by attempting to identify relationships between
strategic alliances stability and its factors, connecting the two fields of studies: game theory and
resource-advantage theory. Viewing the problem from a game theory perspective, the research
adopts a strategic alliance stability definition and conceptualization provided by Zenkevich,
Koroleva, Mamedova (2014a). On another hand, the paper contributes to a series of studies
dedicated to identify strategic alliance stability factors using resource-advantage theory.
Research goal and objectives. This study is aimed at providing an integrated approach to
the concept of SAS stability and its factors. Therefore, the research goal is the following:
Research goal
Identification of relationships between strategic alliance stability and interorganizational strategic alliance stability factors.
In order to attain the research goal, several research objectives (RO) have to be addressed.
These ROs stem from the research goal and unfold the paper logic.
RO1
Based on academic literature analysis on strategic alliances, define the term of strategic
alliance stability.
RO2
Develop a conceptual model of strategic alliance stability factors.
RO3
Based on empirical research, make conclusions about relationships between strategic
alliances stability and its factors.
9
Research questions. In order to achieve goals and to complete objectives of the study, the
two research questions (RQ) were formulated.
RQ1
What are the relationships between strategic alliance stability inter-organizational
factors and different components of strategic alliance stability?
RQ2
What are potential indirect effects of strategic alliance stability factors on different
components of strategic alliance stability?
Research methodology. Research methodology is linked to the research process, which
starts from the theoretical overview of existing academic literature and building up a theoretical
base on the issue, and is followed by empirical analysis (Hussey et al. 1997). Therefore, a relevant
research methodology is critical for a successful research in order to adequately address research
goal and objectives, and provide the answer to research questions.
Research types can be classified by the purpose of the study, research process, research
logic, research outcome Hussey et al. (1997). The table below summarizes on these approaches.
Classification
Type of research
Purpose of the study
Exploratory, explanatory, descriptive,
Research design used in
this thesis
Explanatory/Exploratory
analytical or predictive
Process of the research
Quantitative or qualitative research
Quantitative research
Logic of the research
Deductive or inductive research
Deductive research
Outcome of the research
Applied or basic research
Applied research
Source: adapted from Hussey et al. (1997), Van Dijk (2014)
According to the purpose of the research, this paper can be classified as explanatory as the
paper is aimed at identifying the connections between SAS factors and SAS components based on
the analysis of existing literature on the matter. However, as the research is aimed at studying a
multi-component stability phenomena in connection to strategic alliance stability factors, which
has not been done previously, the research has an exploratory purpose as well.
As for the process of the research, it builds upon theoretical model development which
comes as a result from academic literature analysis, and then puts the theoretical model under
empirical test. Therefore, a research can be classified as deductive (Hussey et al., 1997).
Due to the reason that strategic alliance stability is considered by management
practitioners, but not thoroughly understood by them, it is expected that the results of a research
will have an applied nature.
10
Scope and limitations of the study. The data for an empirical part of the research was
collected through a web-based questionnaire. As the questionnaire was web-based, a link to it was
distributed to companies that might have potentially been involved into strategic alliances by
email.
Survey respondents were European companies’ employees that were involved in strategic
alliances. There was no particular focus on a type of a strategic alliance or on the industry an
alliance operates in. The database of contact details that was used to approach respondents had
been compiled of different sources, particularly from SDC Platinum and Amadeus (Bureau van
Dijk) database. The total number of respondent equaled 184, however, later, the sample was
decreased to 175 observations.
The analysis provided in the paper makes a contribution to theoretic literature on strategic
alliances as a form of long-term collaboration and draws practical conclusions for the use of
strategic alliance managers. Specifically, results reported in this research indicate that there are
different relationships between different strategic alliance stability components and strategic
alliance stability factors, which implies that there is a rationale for viewing strategic alliance
stability as a multidimensional construct.
Several limitations of the study have their place. Firstly, strategic alliances are studied in
general, and the differentiation among different types of alliances is not made. However, it might
be true that in different types of alliances the factors that determine their stability are not the same
(Jiang, Li and Gao, 2008). Secondly, no differentiation between industries an alliance operates in
was made. Third, strategic alliance stability was regarded as a two-componential construct, while
theory allows to view it as a more fragmented phenomena. Lastly, alliance size was not considered.
All the limitations stem from difficulties connected to data collection and unwillingness of
companies to disclose details of their cooperative agreements.
Outline of the paper. There are 6 parts in this paper, namely, introduction, chapters from
one to three, conclusions and implications, limitations and further research. Chapter 1 provides the
overview of academic literature on strategic alliances overall, strategic alliance success and
stability as well as strategic alliance stability factors. In Chapter 2, conceptual model of strategic
alliance stability factors was developed and research hypotheses have been articulated. Moreover,
Chapter 2 describes research methodology, including data collection methods and results as well
as empirical methods used for conceptual model testing. Chapter 3 is dedicated to empirical
assessment of the conceptual model, done with structural equation modeling, which includes
measurement model development and assessment, CFA, and finally, SEM test itself. In the end of
11
the Chapter 3, SEM results are analyzed. Theoretical contributions are articulated and practical
implications are drawn in the following part. Lastly, an overview of research limitations and
further research directions is done.
12
CHAPTER 1. STABILITY IN STRATEGIC ALLIANCES
1.1 Strategic alliances as a form of cooperation
There is an increasing number of studies in academic literature dedicated to issues
associated with inter-firm cooperation (Umukoroa, Sulaimonb and Kuyeb, 2009). Strategic
alliances (SAs) can be defined as an “interfirm cooperative arrangements aimed at achieving the
strategic objectives of the partners” (Das and Teng 1998). There cooperative arrangements may
be embodied in a form of manufacturer-supplier relationships, purchasing agreements, joint
ventures, technology transfer agreements, outsourcing, etc. (Morgan and Hunt, 1994; Varadarajan
and Cunningham, 1995; Lambe, Spekman and Hunt, 2002).
The first mentioning of strategic alliances in academic literature goes back to the year 1923,
and mentioned by Hoxie (1923) in relation to trade unions. Since then, the concept and the nature
of strategic alliances has evolved quite noticeably and has been developing rapidly for the last 30
years (Ferreira, Storopoli and Serra, 2014; Gomes, Barnes and Mahmood, 2014). Multiple theories
are used to explain reasons for strategic alliances establishment, however, regardless the
theoretical approach to SA establishment, the primary reason for this is the net positive value of
the cooperation in a form of an alliance and expected benefits obtained by each partner (Qing,
Zhang, 2015). Indeed, in case potential partners do not find it beneficial to enter into an alliance,
they are unlikely to form the relationship or else, in case they are already in a relationship, they
will not have motivation to maintain cooperation (Umukoroa, Sulaimonb and Kuyeb, 2009). The
rest of the sub-chapter represents a theoretical overview on strategic alliances formation.
The emergence Williamson’s of transaction cost economics (TCE) in 1970s has largely
influenced the theory of cooperative relationships. This approach claims that firms search for ways
of cost reduction related to their activities. TCE assumes that forming a strategic alliance helps
firms reduce their transaction costs by reducing the uncertainty in dealing with their business
partners (Williamson, 1979). Transaction costs theory attempts to explain motivation for SA
creation as one of the ways to avoid ineffective transactions and costs associated with them.
Ineffective transactions can occur due to several reasons: firstly, the transaction can put a company
in a dependent-from-other-companies position (Kogut, 1988); secondly, ineffective markets where
transactions are conducted can be a reason for high transaction costs; thirdly, the inefficiency of
transactions may stem from company’s own inefficiency in operations. However, TCE theory has
its own limitations, and the most significant one is related to the fact that TCE views all the
relationships from the point of possible cost reductions, disregarding possibilities for value
creation (Lu et al, 2012).
13
Before late 1970s, strategic alliances were viewed as an auxiliary form of collaboration
between firms that aided new markets’ entry (Zenkevich, Koroleva, Mamedova, 2014a). Due to
this reason, most of the studies before that time were focused on issues of international joint
ventures (IJVs). Early studies in inter-organizational (IO) theory connected emergence of strategic
alliances with specific industrial conditions (e.g., Berg and Friedman, 1978; Boyle, 1968), e.g.
“convergent expectations and patterns of behavior, practice, shared beliefs, and mindsets” that lead
to benefits connected with collaborative efforts in form of strategic alliances. These benefits might
include reduced the transaction costs associated with partnering, fostering of cooperation through
better infrastructure, institutional trust, and assurance that ties will be formed (Adobor, 2011).
Looking at reasons for forming strategic alliances in broader terms, the reasoning can be summed
up with the following: risk sharing, economies on scale, competitive threat opposition, setting new
technological standards, new market entry, access to resources and competencies (Zenkevich,
Koroleva, Mamedova, 2014a). However, this field of studies, emphasizing industrial factors that
lead to emergence of strategic alliances, does not examine the issue of strategic alliance success
(Yeung, Petrosyan, 2006).
Resource dependency approach compares companies and their performance on the basis of
resources they have access to (Pfeffer, Nowak, 1976; Pfeffer and Salancik, 1978; Steensma and
Lyles, 2000). This theory concentrates solely on resources that might be obtained by a company
from the outside. From this perspective, strategic alliances are created to gain access to resources
of a partner-company and/or to increase control over partner-companies (Zenkevich, Koroleva,
Mamedova, 2014a).
In 1990s, the resource-based view (RBV) on firms got developed (Whipple, Frankel, 2000),
and scholars in the field of strategy have shifted their focus from examination of the external
alliance environment to internal resources and capabilities, which constitute firms’ competitive
advantage (Barney, 1991; Rumelt, 1991). Contrary to resource dependency concept, RBV is
focused on valuable internal resources of the company as important for most of the companies to
gain competitive advantage in the market (Barringer, Harrison, 2000). Following this logic,
absence of required resources and competences within the firm is pushing it to build up competitive
advantages by combining tangible and intangible resources in collaboration with other market
participants (Barney, 1991, 1992; Sanchez, 2003). Dyer and Singh (1998) suggest that with
increased globalization and intensified competition, it gets harder for firms to build up new
competencies, maintain and develop competitive advantage themselves, which explains why it is
beneficial for firms to enter into collaborative relationships, such as strategic alliances. The RBV
14
on strategic alliances was further developed into dynamic capabilities concept in the end of the
XX century.
Nevertheless, RBV has some limitations that hinder its practical application. As outlined
by Deitz et.al. (2010), there are two main limitations of RBV in regards to strategic alliances:
firstly, RBV assumes “demand homogeneity” (Barney, 1991; Peteraf, 1993), secondly, it neglects
subjective judgment and perception of decision-making agents, namely managers (Spender, 1996),
that make decisions on which partners to choose for an alliance, how the resources should be
combined, etc. (Foss, Ishikawa, 2007). E.g., resource-advantage (R-A) theory accounts for these
deficiencies and extends RBV by combining it with “marketing's heterogeneous demand theory”
(Alderson, 1957). A “distinctive competence” for a R-A theory is an “ability of a firm(s) to
combine lower order resources” in way that is hardly imitable by competition (Yeung, Petrosyan,
2006; Lambe et al., 2002).
According to the market power concept, companies get involved into strategic alliances in
order to improve their competitive position in the market relatively to the market (Kogut, 1988).
Such an idea is also incorporated in the R-A theory (Hunt, Lambe and Wittmann, 2002). Gaining
comparative competitive advantage involves not only building up companies’ superior position in
the market, but also an attempt to hinder their competitors’ attempts to do the same. Empirical
studies have shown that companies use SAs as a market-entry facilitation tool as well as a marketstructure alteration tool (Hagedoorn, 1993).
With a growing number of strategic alliances in the world, the social capital approach to
strategic alliances was offered in the beginning of XXI century as many then-current companies
began to have a wide web of strategic alliances of different forms with many counterparts. Within
the social approach, there are two distinctive research directions that were formed: relationship
approach and network structure approach. Relationship approach treats strategic alliances from
the viewpoint of social systems interactions because, in real life, strategic alliance establishment
is based not only on economic benefits estimation, but also on relationship characteristics like
trust, reputation, and communications. In this sense, social approach deals with deficiencies of a
TCE theory (Adobor, 2011). According to relationship approach, strategic alliances are
established, developed and terminated as a result of repetitive patterns of social interactions within
a relationship between partners (Seabright, Levinthal, Fichman, 1992). The network structure
approach studies a network that is formed around the company and its strategic alliances by
different market elements (companies) connected to each other. The social network of a company
affects its activities and behavior, e.g., the network might help company identify opportunities for
strategic alliance creation (Gulati, 1998; Wilkinson, Young, 2002). Zaheer and Venkatraman
15
(1995) suggest that social capital theory in respect to cooperative relationships can be successfully
combined with TCE, therefore, emphasizing the importance of both cost reduction and value
creation in cooperative relationships, representing a more balanced approach compared to the use
of each of them individually (Wu and Choi, 2004).
Table 1.1 summarizes discussions in the field of strategic alliances between years 1933 and
2012 with the most influential articles in each topic.
Table 1.1. Strategic alliances research topics
Discussion topics prevalent in the field of strategic alliances within different 5year periods and most influential papers
Time period
1933-1997
1998-2002
2003-2007
Topic discussed
Most influential articles
Performance and Porter (1990); Harrigan (1985); Harrigan (1986); Hamel and
competitive
Prahalad (1989); Killing (1983); Geringer (1989); Reich and
strategy
Mankin (1986); Hennart (1988); Ring and Van de Ven
(1994); Pfeffer and Salancik (1978); Hamel (1991);
Contractor and Lorange (1988); Harrigan (1988)
International JVs Kogut (1988); Parkhe (1991); Hladik (1985; Parkhe
(1993a); Geringer (1991); Kogut (1989; Parkhe (1993b);
Buckley and Casson (1988); Porter (1986); Osborn and
Baughn (1990)
Governance and
Porter (1985); Axelrod (1984); Powell (1990); Williamson
transaction costs
(1991); Williamson (1975); Porter (1980); Williamson
(1985); Borys and Jemison (1989)
Transaction costs Yan and Gray (1994); Killing (1983); Parkhe (1991);
Hennart (1988); Ring and Van de Ven (1994); Inkpen and
Beamish (1997); Parkhe (1993a); Hamel and Prahalad
(1989); Borys and Jemison (1989); Doz (1996); Hamel
(1991); Harrigan (1985); Williamson (1985)
Learning,
Granovetter (1985); Burt (1992); Hagedoorn (1993); Dyer
networks
and and Singh (1998); Powell et al. (1996); Eisenhardt and
access resources
Schoonhoven (1996); Williamson (1991); Nelson and
Winter (1982); Pfeffer and Salancik (1978); Cohen and
Levinthal
(1990); Mowery et al. (1996); Barney (1991)
JVs: structure and Kogut (1988); Kogut (1989); Ring and Van de Ven (1992)
Reciprocity
Inter-firm
Gulati (1995a); Gulati (1998)
coordination
Learning and
Nelson and Winter (1982); Hagedoorn (1993); Powell et al.
collaboration
(1996); Lane and Lubatkin (1998); Mowery et al. (1996);
Anand and Khanna (2000); Kale et al. (2002); Kogut and
Zander (1992); Barney (1991); Cohen and Levinthal (1990);
Koza and Lewin (1998); Eisenhardt and Schoonhoven
(1996); Khanna et al. (1998); Hamel (1991)
16
Table 1.1. Continued
Governance and
transaction costs
Oxley (1997); Parkhe (1993a); Ring and Van de Ven (1994);
Doz (1998); Hennart (1988); Williamson (1975); Inkpen
and Beamish (1997); Williamson (1985); Williamson
(1991); Kogut (1988); Doz (1996)
Gulati (1995a); Gulati (1998); Gulati (1995b); Gulati and
and Singh (1998); Zaheer, Gulati and Nohria (2000)
2008-2012
Alliance
formation
coordination
Social networks
Knowledge
transfer
and
learning
Dyer and Singh (1998)
Grant and Baden-Fuller (2004); March (1991); Kogut and
Zander (1992); Teece et al. (1997); Lane and Lubatkn
(1998); Mowery et al. (1996); Kale et al. (2000); Hamel
(1991); Barney (1991); Nelson and Winter (1982); Khanna
et al. (1998); Cohen and Levinthal (1990)
Governance and
Parkhe (1993a); Williamson (1975); Doz (1996);
transaction costs
Williamson (1985); Kale et al. (2002); Kogut (1988);
Anand and Khanna (2000)
Social networks
Burt (1992); Baum et al. (2000); Ahuja (2000);
Granovetter (1985); Uzzi (1997)
Alliance
Gulati (1995a); Gulati and Singh (1998); Gulati (1998);
formation
and Gulati (1995b)
coordination
Source: (Ferreira, Storopoli and Serra, 2014)
1.2 Explaining strategic alliance success
Considering alliance success from the RBV perspective (Barney, 1991, 1992; Conner,
1991; Peteraf, 1993; Wernerfelt, 1984), it is based on a fact that alliance partners contribute
immobile and heterogeneous resources to the alliance that have to be combined together.
Complementary resources can be defined as resources provided by partners that “fill out or
complete their resource assortments” (Das and Teng, 2000; Jap, 1999; Varadarajan and
Cunningham, 1995). The unique combinations of resources that exist within different alliances
largely define whether or not an alliance is a success or a failure because by combining resources
of each individual firm, partners can create unique, idiosyncratic resources that are developed
through the lifetime of an alliance (Hunt and Morgan, 1995; Jap, 1999; Anderson and Weitz, 1992;
Jap, 1999; Lambe, Spekman, and Hunt, 2000).
Considering the competence-based theory (Hunt, Lambe and Wittmann, 2002), alliances
that succeed develop specific competences that lead them to success. Competence is defined as
“an ability to sustain the coordinated deployment of assets in a way that helps a firm achieve its
goals” (Sanchez et al., 1996), therefore, an alliance competence is connected with partners
resources as well as partners capability of using these resources in attaining their strategic goals.
Lambe, Spekman, and Hunt (2000), Lado, Boyd, and Wright (1992) argue that competencies,
developed within an alliance are defined by the deployment of a lower-order resources, namely
17
alliance experience, alliance manager development capability, and partner vigilance capability.
Alliance experience in this regards, is an important factor that needs to be considered by partners
as it develops through time and helps alliance partners extract the most from the relationship
Lambe, Spekman, and Hunt (2000). However, the alliance potential could not be fully realized
without proper management that fosters alliance development though the employment of the
complementary resources, so they generate desirable synergies. Lastly, partners should be able to
identify potential resource complementarities and spot potential synergies in order to exploit
complementary resources in the most beneficial and advantageous way (Hunt, 1997).
Moreover, the importance of internal and external relational factors (e.g., relationships with
suppliers, customers, employees) for the alliance success cannot be underestimated. The impact of
relational factors on strategic alliance success are regarded in the frame of relational factors view.
Therefore, the quality of relationships in the alliance social network define whether an alliance is
going to be successful or not. Dwyer, Schurr and Oh (1987) argue that alliance development and
evolution greatly depend on the on-going relationships an alliance is a part of. Therefore, such
organizational factors as trust (Achrol, 1991; Wilson, 1995), commitment (Moorman, Zaltman,
and Deshpande, 1992), cooperation (Anderson and Narus, 1990), are regarded as antecedents of
alliance success according to relational factors view.
Looking at the alliance success from the competitive advantage theory viewpoint, an
alliance can only be successful if it provides partners a feasible opportunity to gain an advantage
over the competition (Porter, 1985; Hunt, Morgan, 1995; Hunt, Lambe and Wittmann, 2002; Hunt,
Arnett, 2003), otherwise there will be no need for partners to enter into an alliance and to remain
in the alliance, therefore, it will be terminated. It is important to note that competitive advantage
theory provides a dynamic view on SA success as firms have to maintain and improve their
competitive position constantly in order not only to constantly perform better than competition
(Jiang, Li and Gao, 2008).
While all the above mentioned approaches mostly concentrate on the way companies
exploit their resources reaching a competitive position in the market, TCE mainly focuses on
transaction costs management. According to TCE, to be successful, alliances should establish and
foster conditions that will decrease the costs of transactions within an alliance considering
opportunistic tendencies that naturally exist in the context of alliances (Williamson, 1975, 1985;
Das, Rahman, 2010). According to Williamson (1985), opportunism is “the incomplete or distorted
disclosure of information, especially to calculated efforts to mislead, distort, disguise, obfuscate,
or otherwise confuse”. As strategic alliances involve two or more partners, whose individual
strategic objectives should be aligned in order to reached, it is not surprising that partners’ are
18
prone to behaving opportunistically in order to reach their own objectives over the objectives of
other parties (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013). Therefore, a successful
alliance is able to be managed through establishing appropriate governance structures (Hennart
1988; Pisano and Teece 1989; Williamson 1991; Das, Rahman, 2010). An example of governance
mechanisms in alliances might include alliance-specific investments, mutual hostages, equity
involvement, etc. (Das, Rahman, 2010; Parkhe, 1993; Brown, Dev, Lee, 2000). In the context of a
complex environment alliance operates in and high uncertainty, partners might lose trust in each
other and deviate from the agreed or implied behavior that is not explicitly stated in the contract
(López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013).
Lastly, one of the latest theories that can be used to explain alliance stability and success
is resource-advantage (R-A) theory, introduced by Hunt (2000) and Hunt and Morgan (1995, 1996,
1997). A distinct quality of the R-A approach is that it considers dynamic perspective of
competition and firm behavior (Hunt, Arnett, 2003), and that is an integrative approach that
incorporates the abovementioned perspectives (Hunt, 2000; Hunt, Arnett, 2003; Hunt, Lambe and
Wittmann, 2002). Moreover, apart from being process-based, it builds upon and used in multiple
disciplines: marketing (Hunt and Arnett, 2001; Hunt, Lambe and Wittmann, 2002), management
(Hunt, 1995, 2000; Hunt and Lambe, 2000), economics (Hunt, 2000, 2001), general business (Hunt
and Duhan, 2001), and ethics (Arnett and Hunt, 2002).
A few more words should be said about resource-advantage theory as it integrates multiple
theories in one and seems to be one of the most. Originally, the theory was developed as a theory
of competition, which is based on several traditions and theories, as mentioned above. Therefore,
R-A theory can be called an embedded theory of competition because it considers not only
economic reasoning behind firms’ behavior, but also social impacts (Hunt, Arnett, 2003).
According to R-A theory, firms seek for advantages in resources and capabilities in order to
outperform competitors and attain over-the-average financial results compared to other companies
to finally gain market superiority, therefore, enforcing a dynamic view of competition, which
reflects reality to a large extent (Hunt and Arnett, 2001). Following R-A theory logic, firms are
limited by the restricted access to relevant information, principal-agent problem that exists
between firms and managers and ethical expectations posed on a firm (Hunt, Lambe and
Wittmann, 2002). Therefore, firms in real world do not seek to maximize their profits, but rather
perform better than some referent, e.g., another company or itself in previous periods.
As R-A theory explains firms’ behavior in the economy, it is also applied to the field of
strategic alliances (Hunt, Lambe and Wittmann, 2002). In application to SA success factors, R-A
theory focuses on resources alliances need in order to achieve success and outperform the
19
“referent” (e.g., other possible alliance forms that are available for partners, other alliances in the
market, other firms, etc.) and is suitable for explaining alliance cooperation and success. According
to R-A approach, resources are defined as “tangible and intangible entities available to the firm
that enable it to produce efficiently and/or effectively a market offering that has value for some
market segment(s)” (Hunt, Lambe and Wittmann, 2002).
By explaining SA success in the light of R-A theory, Hunt, Lambe and Wittmann (2002)
highlight the influence of the RBV approach on the development of the R-A theory. Given that RA theory relies upon RBV in explaining alliance success, it stresses the role of firms’ resources in
attaining alliance success. R-A theory argues that complementary and idiosyncratic resources
constitute an important part of the alliance success explanation because it helps understand the
nature of a competition an alliance is involved in (Lambe, Spekman and Hunt, 2002). Next, R-A
theory views competences as a combination of lower-order resources, a combination of which
leads an alliance to success. Relationship network that an alliance exists in has also become an
important part of an integrative resource-advantage approach. Therefore, an alliance is not torn out
from the social context, but is rather regarded in close connection with personal and organizational
interrelationships that exist between individuals within and outside the organizations. Lastly,
partner resources in an alliance should match the way they create synergies and provide partners
with benefits that cannot be attained by firms individually (Hunt 2000).
To summarize this sub-chapter, R-A theory provides an integrative framework to SA
success as it builds upon multiple traditional theories and approaches to the success of strategic
alliances. One of the most important advantages of the R-A theory in application to SA success is
its dynamic view on alliance performance, therefore, judging from this perspective, factors of
alliance success that are studied within R-A theory are related to the dynamic nature of alliance
operations.
1.3 Strategic alliance stability in academic literature
Despite all the advantages that strategic alliances can bring to partner companies, alliance
involvement might incur issue for individual firms in an alliance (Kolenak, 2007). It is not
uncommon that these issues within an alliance lead to the lack of stability, deteriorated
performance and can cause alliance premature termination (Geringer and Hebert, 1991;
Umukoroa, Sulaimonb, Kuyeb, 2009). Managing an alliance in a way that promotes cooperation
between partners and decreases opportunistic behavior becomes highly relevant when firms enter
alliances. Hence, strategic alliance stability is an important concept in relation to strategic
alliances.
20
The first milestone study related to strategic alliance stability was published in 1971 by
Franko, in which the author discussed the issue of strategic alliance stability as an opposite of SA
instability. Therefore, the paper examines strategic alliance instability, particularly, the instability
of joint ventures (JVs) in an international context. A brief summary of results of this study can be
found in Table 1.4. Nevertheless, after the first publication by Franko (1971), instability issue did
not attract enough attention from researchers until late 80’s – early 90’s (Fu, Lin, Sun, 2013). Later
on, SAS has been studied in multiple papers, however, there are still a lot of issues with SAS
definition, therefore, measurement (Jiang, Li and Gao, 2008). Oftentimes, strategic alliance
stability is not clearly defined in academic papers (Jiang, Li and Gao, 2008), and often defined as
an opposite to instability (Sim, Ali, 2000). Based on the literature review, a list of SAS definitions
was created (see Table 1.2).
In fact, the focus of researchers on strategic alliance stability has been split between two
general concepts: strategic alliance stability and strategic alliance instability (Jiang, Li and Gao,
2008). It appears that strategic alliance instability rather than strategic alliance stability was the
first and dominant focus of numerous studies (e.g., Franko, 1971; Killing, 1982, 1983; GomesCasseres, 1987; Inkpen and Beamish, 1997; Yan and Zeng, 1999; Das and Teng, 2000; Gill and
Butler, 2003; Nakamura, 2005), as it has been mentioned above. As a result, it is quite often when
authors do not conceptually differentiate between SA stability and instability, and sometimes
switch between the two in one study (e.g., Yan, 1998; Yan and Zeng, 1999).
Speaking of instability, it is conceptualized in terms of an outcome as a “termination, death
or failure” of an alliance (e.g., Franko, 1971; Killing, 1983; Kogut, 1989). However, among all the
cases of alliance termination, there are strategic alliances that end up their existence by achieving
their strategic goals or by terminating the collaborative agreement naturally, according to the plan
(Inkpen and Beamish, 1997; Hong, Yu, Zhichao, 2011; Jiang, Li and Gao, 2008). In these cases,
strategic alliances should not be characterized as unstable (Jiang, Li, Gao, 2008; Hong, Yu,
Zhichao, 2011). Christoffersen (2013) puts forward an example of a situation when partners sell
out an alliance. Theoretically, this can happen if partners are dissatisfied with alliance
performance, however, it is more likely to reflect the opposite – a stable and well-performing
alliance is viewed as a more attractive acquisition target, therefore, it can be sold out more easily.
Christoffersen (2013) insists that many alliances are formed as temporal entities that are meant for
sale, so their sell-off should not be viewed as poor performance or instability (Bowman and Hurry
1993; Kogut 1991).
Moreover, in cases when studies report high failure rates, it is uncommon that sufficient
explanation is provided to why strategic alliances get terminated after multiple years of successful
21
and stable performance (Yan, 1998). Continuing this line of logic, some studies report that after
controlling for size and age of strategic alliances in a form of JVs, their failure rate is very close
to that of wholly-owned subsidiaries (WOS) (e.g., Hennart et al, 1998; Delios and Beamish, 2004).
Hence, the termination of a strategic alliance cannot always be viewed as an indicator of SA
instability, and does not shed light on SA dissolution in particular (Jiang, Li and Gao, 2008).
Deitz et.al. (2010) in their empirical research argue that SAS should be viewed as one of
the strategic alliance outcomes, stressing the dynamic aspect of it. Authors focus on SAS defining
it as “the frequency of changes in contract or relationship status” consistent with (Inkpen and
Beamish, 1997). Moreover, authors view commitment, or cooperative intent, as another crucial
outcome SAs, particularly, for JVs. Partners’ commitment, which implies that partners would
rather demonstrate cooperative than opportunistic behavior (Das and Teng, 1998), represents a
managerial trade-off for alliance partners as they should not only consider their own perspective
in an alliance, but also consider other partners’ perspective in order to make an alliance a success
and not dissolve prematurely (Deitz et al, 2010).
Overall, it can be argued that academic literature on strategic alliance stability in particular
is less abundant compared to the number of studies on the instability issue (e.g., Kogut, 1989;
Beamish and Inkpen, 1995; Sim and Ali, 2000; Bidault and Salgado, 2001; Ernst and Bamford,
2005).
Finally, as a result of absence of one common strategic alliance stability definition, there
is no clear vision on how strategic alliance stability should be measured.
22
Table 1.2. Definitions of strategic alliance stability/instability
Academic paper
(Zenkevich,
Definition
Koroleva, “Strategic alliance stability should be understood as a success
Mamedova, 2014a)
of alliance performance during the period of alliance operations
under conditions of constant motivation of each partner firm to
maximize the results of cooperation.”
(Jiang, Li, Gao, 2008)
“…we define alliance stability as the degree to which an
alliance can run and develop successfully based on an effective
collaborative relationship shared by all partners.”
(Huang, 2003)
“Stability, means in the process of movement, or interference,
(Hong, Yu, Zhichao, 2011)
whether or not the system can keep its former state. As for the
specific strategic alliance, it means that the strategic alliance,
as an organization can keep its stable state, it is a dynamic
stability, relative stability.”
(Inkpen, Beamish, 1997)
“…joint venture is considered unstable if the partners’ equity
(Das, Teng, 2000)
holding in the joint venture changed (including take-over by
(Sim, Ali, 2000)
one partner) since the formation or the venture is terminated.
Termination as a result of a project ending was not included.”
(Qing, Zhang, 2015)
“…instability of such an [a competitive] alliance means short
and fragile cooperation, and the failure of alliance”
Source: augmented from (Zenkevich, Koroleva, Mamedova, 2014a)
As it can be concluded from the Table 1.2 above, many scholars connect strategic alliance
stability with its longevity and survival, absence of structural changes and reorganizations
(Beamish, 1984, 1988), etc. (Sim and Ali, 2000; Franko, 1971; Killing, 1983; Blogget, 1992).
Sometimes strategic alliance stability is regarded as a strategic alliance performance measure
(Jiang, Li and Gao, 2008). Nevertheless, while SAS is oftentimes regarded as an indicator and
measure of SA performance (Geringer and Hebert 1991), it is less widely used compared to other
ones (e.g., financial results, partner satisfaction with alliance results) (Sim, Ali, 2000). Despite the
fact that many approaches to strategic alliances exist, they are not mutually exclusive in nature.
These different approaches rather complement each other, and their existence proves the multisidedness and complexity of strategic alliance stability nature (Varadarajan, Jayachandran, 1999;
Jiang, Li and Gao, 2008; Yeung, Petrosyan, 2006).
Apart from the issues related to SAS definition and conceptualization, researchers tend to
focus on a particular type of a strategic alliance, e.g., JV or international JV (IJV) (Jiang, Li and
23
Gao, 2008). Summarizing on the particular issue of IJV stability, Sim and Ali (2000) have created
a short list of most important studies on this issue (see Table 1.2).
To summarize the approaches to SAS definition, the following scheme can be put forward
(see Table 1.3). Overall, it can be concluded that there is a lack of studies that view strategic
alliance stability as a dynamic concept, from a process-oriented perspective. However, the need to
view strategic alliance stability as a dynamic concept is widely acknowledged in the current
academic literature.
Table 1.3. Instability and stability, process-based and outcome-based researches examples
Instability
Process-oriented
Outcome-oriented
Stability
Killing, 1982, 1983
Jiang, Li and Gao, 2008
Yan, Zheng, 1999
Zenkevich,
Kogut, 1989
2014a,b
Sim, Ali, 2000
Qing, Zhang, 2015
Das, Teng, 2000
N/A
Koroleva,
Mamedova,
Lu, Beamish, 2006
Qing, Zhang, 2015
Table 1.4 provides a brief description of major empirical studies on strategic alliance
stability, according to (Sim, Ali, 2000). As can be seen from Table 1.4, many studies dedicated to
IJV stability in particular study it as an opposite to instability, applying outcome-oriented
approach. Based on the summary presented in the Table 1.4, it can be concluded that during the
XX century, apart from the issue of stability being studied as an opposite to instability, the
following factors were considered to be causing instability: partners’ conflicts (Franko, 1971),
cultural distance (Franko, 1971; Park and Ungson, 1997; Barkema et al, 1996; Killing, 1983),
opportunistism (Park and Ungson, 1997), role of management (Glaister and Buckley, 1998; Lee
and Beamish, 1995), ownership structure (Blodgett, 1992), etc. However, it still seems
inappropriate to study SAS through alliance termination or instability as it provides an outcomeoriented view on SAS, while it is more appropriate to consider the dynamics behind SAS (Jiang,
Li and Gao, 2008; Zenkevich, Koroleva, Mamedova, 2014a,b; Qing, Zhang, 2015). Therefore,
factors of “stability” examined in many researches in the past might be questioned as they refer
not to stability, but rather to an instability issue.
24
Table 1.4. Empirical studies on IJV stability
Author
Measure of Stability
Sample Size
Franko, 1971
Changes in equity
level (50%; 95%
cutoff; selling out;
liquidation)
US MNEs (159)
From Harvard
MNE
Database
Killing, 1983
Liquidation and
reorganization
37 IJVs in
developed
countries
Beamish 1984,
1988
Equity changes, major
reorganization
66 IJVs in
developing
countries
Games-Casseres,
1987
Liquidation,
conversion to whollyowned ventures
Survival, duration (No.
of years in operation)
US MNEs (180)
from Harvard
MNE Database
895 strategic
alliances in 23
industries
Ratio of terminated
JVs to all JVs which
survived
149 IJVs
Harrigan, 1988
Kogut, 1988
Stability Related
Findings
Partners’ conflict
increases instability.
Cultural differences
have little impact.
Strategy of global
concentration and single
product contribute to
instability.
Instability rate of 31%
in all IJVs; 15% in
dominant-controlled
IJVs and
50% in shared
management IJVs.
Differences in national
and corporate culture
have some impact. Firm
size and linkages with
parents affect stability.
Instability rate of 5%.
Higher instability rates
for IJVs in developing
countries than in
developed countries.
Instability rate for IJVs
was 30.6% and for
WOS 15.7%.
Size asymmetries have
little effect on IJV
survival and duration.
JV experience: negative
impact on duration, but
positive impact on
survival.
Vertical linkages:
negative effect on
survival.
Horizontal linkages: no
effect on survival and
duration.
Average IJV lifespan
was 3.5 years.
Instability rate of 46.3%
25
Table 1.4. Continued
Kogut, 1989
Ratio of terminated
from US JVs to all JVs
which merger and
survived
Geringer and
Hebert, 1991
Changes in equity
division survival
Blodgett, 1992
Changes in equity
structure
Beamish and
Inkpen, 1995
Unplanned equity
changes and major
reorganization
Lee and Beamish,
1995
Unplanned equity
changes and major
reorganizations
92 IJVs from US
merger and
acquisition
database
Instability rate 43%,
increased to 55% after
one year and to 70%
after 2 years. Ties
(linkages) between
partners contribute to
stability
(Horizontal linkages
have positive effect.
Vertical linkages have
no effect). Industry
growth and changes in
industry concentration
have positive effect on
instability.
69 IJVs in US, 48 Successful IJVs were
IJVs in Canada
more stable and
survived longer.
Stability measure has
least correlation with
subjective measure of
IJV performance.
1025 IJVs of 69
Highly unequal
firms (Merger
ownership structure
and Acquisitions contribute to Instability
Database)
(Dominant partnership
destabilizes IJV).
Restrictive host
government polices
contribute to IJV
stability.
Toppan Moore
Instability increases if
case
foreign partner attaches
study, 5
a high value to the
longitudinal
acquisition of local
studies, and 40
knowledge.
US-Japanese
IJVs
31 Korean IJVs
Instability rate of 19%
in
(low). Higher
LDCs
stability rates for IJVs
formed with
local private partners
than with local
Government partners.
26
Table 1.4. Continued
Barkema et al,
1996
Survival in terms of
longevity (No. of
years)
225 foreign
ventures
of 13 Dutch
firms
(1996-1988)
Barkema and
Vermeulen, 1997
Survival in terms of
longevity (No. of
years)
828 foreign
ventures
of 25 Dutch
firms in 72
countries (19611994).
Park and Ungson,
1997
Dissolution in terms of
liquidation and sale to
a third party
186 U.S. IJVs in
electronics
Glaister and
Buckley, 1998
Survival and duration
51 UK JVs
Hennart, Kim and
Zeng, 1998
Terminations in terms
of liquidation and
selloff
355 Japanese
stakes
in US
Longevity of foreign
ventures is negatively
related to cultural
distance.
Cultural distance has a
greater negative impact
on longevity of JVs than
for WOS.
Different dimensions of
cultural distance have
differential impact on
JV survival. Uncertainty
avoidance and longterm orientation have
higher negative effect
on JV survival than
masculinity. Power
distance and
individualism have no
effect.
Cultural distance in
general has no effect on
dissolution. USJapanese IJVs last
longer than US-US
IJVs.
Opportunistic threats
and rivalry enhance
dissolution.
Nature of management
control did not vary
with duration of JV.
30% termination rate.
Factors selloff different
from those in
liquidation.
Source: (Sim, Ali, 2000)
Strategic alliance stability and economic performance. The discussion on SAS definition
leads to an important point on alliance economic performance in line with connection to strategic
alliance stability. Mainly, there are two opinions that exist: strategic alliance stability is an
economic performance determinant, or economic performance is a determinant of strategic
alliance stability.
For example, Fu, Lin, Sun (2013) view strategic alliance performance as an antecedent of
strategic alliance stability, and find a positive association between strategic alliance performance
and strategic alliance stability as a result of their empirical study. Authors claim that economic
27
performance is in a core of any strategic alliance, thus, any alliance that does not meet economic
performance requirements, should eventually be terminated. Moreover, the authors argue that
participants “inclination to withdraw” resources from the alliance should be reduced in the light
of performance that meets their expectations and requirements and complemented by participants
motivation to positively contribute to joint cooperative actions with their partners (Ungson, 2001;
Fu, Lin, Sun, 2013).
However, the most popular view on a relationship between strategic alliance stability and
performance is that stability actually determines economic performance (Dussauge and Garrette,
1995; Beamish and Inkpen, 1995, Jiang, Li, Gao, 2008). Following the example of Sim and Ali
(2000), Jiang, Li, Gao (2008), this study adopts the view that SAS is a critical factor to alliance
economic success (Sim, Ali, 2000; Jiang, Li, Gao, 2008) and strategic goals achievement once the
alliance is established (Bidault and Salgado, 2001).
1.4 Strategic alliance stability conceptualization
The discussion about approaches to strategic alliance stability nature and definitions
inevitably leads to a conclusion that it is a multi-sided phenomenon in cooperative relationships.
These different sides of strategic alliance stability differ as the motivation of partners to enter and
to belong to a particular strategic alliance differs as well (Zenkevich, Koroleva, Mamedova,
2014a). At the same time, stability of cooperative relationships is well studied in game theory (e.g.,
Moor, 1971; Zenkevich, Petrosyan, Yang, 2009). Based on previous studies of cooperative
relationships stability in game theory (Moor, 1971; Zenkevich, Petrosyan, Yang, 2009; Gill,
Butler, 2003; Wong, Tjosvold, Zhang, 2005; Kumar, 2011), Zenkevich, Koroleva, Mamedova
(2014a, b) introduce several components of strategic alliance stability.
According to Zenkevich, Koroleva, Mamedova (2014a), strategic alliance stability has two
levels. On the first level, there is external and internal, or cooperative, stability. On the second
level, internal, or cooperative stability of strategic alliances, is comprised of motivational, strategic
and dynamic stability (Zenkevich, Koroleva, Mamedova, 2014a, b). The overall stability scheme
is presented in the Figure 1.1.
28
Strategic Alliance
Stability
External Stability
Internal Stability
Dynamic Stability
Motivational
Stability
Strategic Stability
Figure 1.1. Strategic alliance stability structure
Source: (Zenkevich, Koroleva, Mamedova, 2014a)
The concept of external stability allows to assess the stability of alliance as of a separate
economic entity as it is conventionally done, with the help of economic indicators. It is assumed
that a strategic alliance is externally stable in case its economic results show a raising trend.
Economic results of the strategic alliance might include its net profit, revenue, market share, etc.
If the trend is long-term, partner companies perceive a strategic alliance as a successful one, so
they have a lasting motivation to maintain cooperation. It is important to consider the long-term
trend because in a short-term perspective strategic alliance might experience losses (e.g., due to
initial stages of alliance implementation, unfavorable external conditions, etc.), which will be
perceived as “natural” and will not deteriorate participants cooperative intent, at least, to a
significant extent in case the long-term trend is positive.
However, as a strategic alliance is an agreement between companies which are eager to
attain their own objectives within the alliance, this explains the need of introduction of internal (or
cooperative) stability concept, which is well studied in game theory.
Internal stability of companies is well describes in numerous papers in the field of strategic
management (e.g., Gill, Butler, 2003; Wong, Tjosvold, Zhang, 2005; Kumar, 2011), while game
theory has thoroughly studied different components of internal stability of cooperative
relationships, and has developed a holistic approach for its assessment (Zenkevich, Petrosyan,
Yang, 2009).
29
An important assumption for internal stability conceptualization is that partners in a
strategic alliance are rational, this is why they enter a strategic alliance expecting that the benefits
of their cooperation will exceed possible benefits of their actions in case they kept operating
individually (Zenkevich, Koroleva, Mamedova, 2014a, Qing, Zhang, 2015).
Motivation to cooperate is essential to strategic alliance stability. Zenkevich, Koroleva,
Mamedova (2014a) in their paper explain that motivational stability means that partners find it
beneficial to actively contribute to alliance operations, or actively commit to alliance activities
(Kumar, Scheer, and Steenkamp, 1995) because such behavior will increase the overall benefits
of the alliance, hence, individual benefits of each partner (Gulati, Khanna and Nohria 1994; Sarkar
et al, 2001). Such definition of a strategic alliance stability is close to the understanding of
commitment introduces by Das and Teng (1998) and described above.
The importance of motivational stability is explained by the fact that success of alliance
operations is defined not only by economic factors and their trends, but also by relationships
among alliance participants (Deitz et al., 2010; Hunt, Lambe and Wittmann, 2002). Motivation for
further cooperation is supported by such factors as trust (Anderson, Weitz, 1989; Huo, Ye, Zhao,
2015), attention to cross-cultural differences (Doz, Hamel, 1998; Yan, Luo, 2001) as well as
common goals and objectives (Anderson, Weitz, 1989; Ozorhon et.al., 2008) and participants’
commitment (Kumar, Scheer, and Steenkamp, 1995). One can say that alliance partners are
committed to the alliance in case he contributes resources and capabilities necessary for alliance
success (Jiang, Li and Gao, 2008). Partners’ commitment has a positive influence on partners’
relationship because it indicates are loyal and long-term oriented while increasing reciprocity and
cooperation levels. Given these conditions, partners can expect that they are able to receive the
expected benefits during the time an alliance functions. (Zaheer, Venkatraman, 1995). If partners
are committed to the relationship, they are less likely to deviate from cooperation. On a contrary,
when partners are not committed to the alliance, they are not likely to establish a close cooperation
with each other, which destabilizes the relationship.
Strategic stability is well studied in game theory (Zenkevich, 2009). Assuming that
partners entering a strategic alliance are rational, when partners make a decision to form a strategic
alliance, it means that they find such form of collaboration to be the most beneficial for them
compared to all other opportunities in the market, including other partnerships and an opportunity
to operate alone. However, when the strategic alliance enters the implementation phase,
circumstances or partners’ access to information might change, etc., so one of the partners might
reconsider staying within an alliance not beneficial anymore and want to exit the alliance. Strategic
30
stability of a strategic alliance assumes that none of the partners find it beneficial to decline from
the cooperative agreement among partners, while other partners pertain to it.
Dynamic stability is examined in game theory along with strategic stability as a part of
internal stability of cooperative relationships (Zenkevich, 2009). Dynamic stability of strategic
alliances refers to benefits sharing in an alliance, or the payoff structure. Payoff structure is an
important issue for alliance partners as they are motivated not only through economic benefits
generated by an alliance as an economic entity, but also by benefits that are allocated to them
personally (Umukoroa, Sulaimonb and Kuyeb, 2009).
It has been mentioned by Franko (1971) that an alliance is stable rather than unstable when
partners agree to agree on the initial profit sharing mechanism and satisfied with it. At the stage of
alliance formation, partners form an understanding of what kind of benefits and in what quantity
they find to be fair for them in comparison with all the threats and possible disadvantages, such as
opportunity costs, that they are likely to face due to alliance participation and all the inputs they
have to make for cooperation. The alliance is dynamically stable in case when at each moment of
time the sum of gained and expected benefits by a partner corresponds to the amount and type of
benefits the partner had been expecting to gain when signing the contract for cooperation.
Dynamic stability assumes that this principle is supported for each of the partners in a strategic
alliance.
In case when a partner realizes that he will not be able to get all the expected benefits he
had been expecting from the alliance, his motivation to continue alliance participation might
decrease or even disappear (Zenkevich, Koroleva, Mamedova, 2014a). Therefore, a set of
measures in terms of governance and communication have to be undertaken to manage partner
satisfaction from the cooperation as well as perceived fairness of benefits sharing.
However, given that a well-developed pay-off structure is necessary for alliance success
and stability (Khanna et al., 1998), it is not a sufficient condition for it (Agarwal, Croson,
Mahoney, 2010).
Summarizing the abovementioned definitions of different stability components, strategic
alliance stability can finally be defined as a success of alliance performance during the period of
alliance operations under conditions of constant motivation of each partner firm to maximize the
results of cooperation (Zenkevich, Koroleva, Mamedova, 2014a).
Strategic alliance stability components in the focus of the study. Certain connections can
be made between strategic alliance stability components represented in the Figure 1.1 and existing
papers. For example, papers like (Jiang, Li and Gao, 2008; Zenkevich, Koroleva, Mamedova,
31
2014a,b; Qing, Zhang, 2015) view strategic alliance stability as an outcome of strategic alliance
activities and use it as a performance measure, which correlated with the way external stability is
conceptualized. External stability can be observed in a presence of a growing trend in economic
result. Hence, stability can be assessed as a series of alliance economic results as a dynamic
concept
As it has been discussed in sub-chapter 1.3, another traditional approach to strategic
alliance stability definition and assessment is connected to the SA longevity, premature
termination, changes in organizational form, partner commitment, satisfaction, etc. (Kogut, 1989;
Deitz et al, 2010; Christoffersen, 2013). Hence, the effects of motivational, dynamic, strategic
stability are often considered, however, not structured and not articulated explicitly, being defined
by a general term of “strategic alliance stability” (López‐Navarro, Callarisa‐Fiol and Moliner‐
Tena, 2013) or “commitment” (Deitz et al, 2010). One common characteristic of these approaches
to SAS definition, is that they focus on cooperation internal characteristics, that an outsider might
have difficulties in assessing due to scarcity of information disclosed by strategic alliances (Jiang,
Li and Gao, 2008).
For the purpose of this research, strategic alliance stability was analyzed as a multidimensional construct, however, the distinction among strategic alliance stability components was
made on the most aggregate level: between external and internal stability. Given the fact that not
much has been done in merging game theory approach to strategic alliance stability
conceptualization, which is comprehensive and all-inclusive, and broader managerial studies that
examine strategic alliance stability factors, the benefits of such SAS conceptualization within the
study are clear. Fist, such an approach pertain concept integrity. Second, conceptualizing stability
this way, there is an ability to identify differences in relationships between strategic alliance
stability factors and different strategic alliance stability components on the most aggregate level
to gain a general understanding about these interconnections. The third benefit is the feasibility of
further empirical analysis given the number of constructs to be analyzed in one study.
1.5 Factors of strategic alliance stability: theoretical perspectives
Partner firms can increase cooperation by altering the factors that affect it (Umukoroa,
Sulaimonb, Kuyeb, 2009), therefore, affect strategic alliance stability (Deitz et.al., 2010).
Pure economic view on partner cooperation and alliance success is believed to be
incomplete and limiting (Mellat-Parast, Digman, 2007; Nielsen, 2007). It is argued by researchers
(Lambe, Spekman and Hunt, 2002) that there are other factors that affect the extent to which
cooperation is maintained. Speaking of inter-organizational factors (Zenkevich, Koroleva,
32
Mamedova, 2014a), not considering factors of the external environment, the groups of factors that
affect alliance success are relational (e.g., trust) and non-relational (e.g., resource
complementarity), strategic (e.g., goal congruence, resource complementarity) and relational,
economic and non-economic (Lambe, Spekman and Hunt, 2002; Day, 1995; Ganesan, 1994; Hunt,
1997; Jap 1999; Morgan and Hunt, 1994; Varadarajan and Cunningham, 1995; Deitz et al, 2010).
Generally, studies on SAS either do not explicitly explain the rationale behind choosing
particular factors in explaining SAS (Jiang, Li and Gao, 2008), study stability from a pure
mathematical perspective (Qing, Zhang, 2015), apply a combination of theories, such as TCE and
RBV (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013), or use R-A theory (Deitz et al,
2010) to define a set of SAS factors. Overall, in examining SAS empirically, it seems logical to
refer to studies and models emerging from an integrative theory, such as R-A, or a combination of
theories, e.g. TCE and RBV.
Deitz et.al. (2010) adopt the resource-advantage theory, and examine “strategic and
relational factors” that affect alliance stability, particularly the stability of JVs, and ongoing
commitment, which is conceptually similar to strategic and motivational stability in the adopted
approach to SAS (Zenkevich, Koroleva, Mamedova, 2014a). Figure 1.2 depicts the conceptual
model (Deitz et al, 2010) put for testing. In this study, authors do not only test the influence of
resource complementarity and trust on alliance stability and cooperative intent (which correspond
to strategic and motivational stability in the context of this study), but also moderating effects of
alliance age and prior alliance experience.
In this paper, conducting the SEM on a sample of 219 observations, Deitz et al (2010) have
found the significant influence of resource complementarity on both components of SAS. The
same result was produced for the association between trust and JV stability, however, the
association between trust and commitment appeared to be marginally significant.
33
Prior JV
experience
JV age
(-)
Trust
(+)
(+)
JV stability
(+)
(-)
Resource
complementarity
(+)
Commitment
Figure 1.2. Effects of trust and resource complementarity on export JVs stability and
partner commitment
Source: (Deitz et al, 2010)
A similar study, (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013) extends the
traditional resource-based view by adding a transaction-costs economy approach to explain
international JV commitment. Such a way of combining the two approaches brings the approach
close to the resource-advantage explanation because, as was mentioned previously, to partner
cooperation in alliances, and build conceptual frameworks around partners’ commitment to the
relationship, defined as “the willingness of partners to invest in the JV the resources necessary for
its success” (Gulati, Khanna and Nohria 1994; Sarkar et al. 2001), which, in turn, matches the
concept of motivational stability. The authors of the paper come up with the following conceptual
model (see Figure 1.3).
34
Trust
(+)
Long-term orientation
(+)
Commitment
Resource
complementarity
Figure 1.3. The impact of long-term orientation, resource complementarity and trust on
commitment in IJVs
Source: (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013)
As a result of their empirical analysis, done with the help of PLS, on a sample of 85 Spanish
JVs, authors find that all the structural relationships in the scheme represented by Figure 1.3, apart
from Resource complementarity Long-term orientation are positive and significant, while the
denoted relationship is positive, but statistically insignificant. Therefore, the effect of resource
complementarity of long-term orientation has been found to be indirect. However, as can be seen
from the Figure 1.3, authors do not test the direct relationships neither between resource
complementarity and commitment, nor between trust and commitment.
Given multiple combinations of factors that have a potential impact of SAS and on the
alliance success overall (see sub-chapter 1.4), for the purpose of this study, the decision was made
to focus on three determinants of SAS: trust, long-term orientation, resource complementarity,
which were combined together in (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013; Deitz
et al, 2010). Moreover, these factors of SAS correspond to R-A understanding of SA success,
therefore, stability.
Trust can be conceptualized as “the willingness of one party to be vulnerable to the actions
of another party based on the expectation that the other will perform a particular action important
to the trustor, irrespective of the ability to monitor or control that other party” (Mayer et al, 1995).
Trust is believed to be an important factor for cooperation success (Das and Teng, 1998; Madhok,
1995) as trust enhances cooperation itself as well as lowers costs for coordinating activities,
reduces uncertainties and facilitates information exchange among partners (Smith, Carroll and
Ashford, 1995; Dyer, 1996; Gill and Butler, 2003; Deitz et al, 2010). Additionally, it has been
35
shown by Ganesan (1994), that in case when partners trust each other, they expect any possible
losses to be mitigated in the future (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013).
Besides, partners find it more comfortable to build long-term cooperative relationships with other
partners whom they trust (Jiang, Li and Gao, 2008). Contrary to the presence of trust in partner
relationships, in its absence, SAS is likely to disappear (Nielsen, 2007). Existing researches have
defined the importance of mutual trust for maintaining stable relationships among partners (Jiang,
Li and Gao, 2008; Anderson and Weitz, 1989). However, while the empirical evidence of trust
influence on cooperation, there is lack of evidence of trust influence directly on economic results
of the firm (Nielsen, 2007). The lack of evidence might be connected to the fact that trust is an
intangible construct that is hard to measure (Nielsen, 2007). Nevertheless, an indirect positive
impact on economic results might be assumed (Jap, 1999).
Strategic alliances are meant to be temporary (Das, Rahman, 2010), however, some of the
alliances may last for a long time. Generally, as for the date of establishment, the term of an
alliance can be determined or not determined. At the same time, partners’ time-horizon orientation
in an alliance matter. In case when partners are short-term oriented, the possibility of them
behaving opportunistically increases and cooperation suffers (Das, Rahman, 2010). López‐
Navarro, Callarisa‐Fiol, Moliner‐Tena (2013) argue that partners’ long-term orientation, defined
in terms of “attitude or vision of partners regarding the future benefits that the relationship can
bring them” (Ryu, Park and Min 2007; Sheth and Parvatiyar, 1992) fosters cooperation and
enhances alliance stability. Therefore, when partners are long-term oriented, they perceive their
own economic results and the economic results of the partner as interconnected, and they have an
aspiration that jointly partners are going to benefit from the cooperation in the long-run (Ganesan,
1994; Kelley and Thibaut, 1978).
Partners’ long-term orientation towards benefits sharing cause partners’ to be concerned
about the relationships (Das and Teng, 2000; Ganesan, 1994; López‐Navarro, Callarisa‐Fiol and
Moliner‐Tena, 2013) and think of the “shadow of the future”, which means that negative
experience of one of the partners in a relationship can cause him to demonstrate negative relational
tendencies in response to this experience, and vice versa (Das, Rahman, 2010). It is also assumed
that short-term oriented partners are likely to maximize their short-term gains for every transaction
within an alliance, therefore, would not consider partner’s perspective. Lack of long-term
orientation, as findings provided by (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013)
suggest, results in lack of partners’ commitment and partner involvement in an alliance. Moreover,
it is believed that long-term orientation enhances incentives alignment (Das, Rahman, 2010).
36
Resource complementarity is defined as the level to “which firms in an alliance are able
to eliminate deficiencies in each other's portfolio of resources” (Lambe, Spekman and Hunt, 2002).
The need for partners’ complementary skills and resources represents a motivation for the
formation of joint venture arrangements (Geringer, 1991; Hamel, 1991; Sim, Ali, 2000). Geringer
(1991) in particular, found that need for partners’ complementary resources (such as market
knowledge, market access, local identity, and marketing channel) is the most important partner
selection criterion. Firms with complementary resources, given that they cooperate, can increase
the other’s ability to attain business goals as they can provide their joint relationship with resources
and capabilities, both tangible and intangible, that will create synergies with those of the other
partner (Lambe, Spekman and Hunt, 2002). R-A theory implies that complementary resources of
each of the partners are combined to create a unique set of resources and capabilities within an
alliance to establish a competitive advantage over competitors (Hunt 1997; Hunt and Morgan,
1995, 1996, 1997; Jap, 1999). Summarizing on (Lambe, Spekman, and Hunt, 2000), resource
complementarity can be called a necessary condition for alliance success, based on which
distinctive alliance competences are developed.
However, not only resource complementarity affects partners’ ongoing cooperation, but it
is also considered by partners at a stage of choosing their partnership among available strategic
options (Jiang, Li and Gao, 2008). Therefore, firms should pay a close attention to resource
complementarity with their partners during the partner selection process and keep the notice of the
degree of complementarity during alliance implementation stage.
1.6 Chapter 1 concluding remarks
One of the issues discussed in academic literature and among practitioners is strategic
alliance stability. Strategic alliance stability is an important concept as it is assumed to be an
important strategic alliance characteristic, or an outcome, being positively associated with strategic
alliance performance (Sim, Ali, 2000; Jiang, Li, Gao, 2008).
Chapter 1 represents a literature review on strategic alliances, strategic alliance success and
strategic alliance stability. In this chapter, several theoretical approaches to issue of strategic
formation have been described, namely: resource-based theory, transaction cost economy, interorganizational theory, market power concept, social capital approach, resource-advantage theory.
A notion has to be made that regardless the theory used to explain strategic alliance formation, the
primary objective of partners that enter a strategic alliance is the desire to gain economic benefits
(Qing, Zhang, 2015). Finally, a strategic alliances was defined as a long-term cooperative
agreement between partner companies that stay legally independent from each other after alliance
37
formation, share cooperation benefits and governance control over defined objectives and are
continuously involved into one or more strategically important areas (Zenkevich, Koroleva,
Mamedova, 2014a).
Following the logic of the Chapter 1, factors that affect strategic alliance success were
analyzed from different theoretical perspectives: resource-based view, competence-based
approach, relational factors view, competitive advantage theory, transaction cost economy,
resource-advantage theory. Resource-advantage theory was identified to be the most all-inclusive
and all-embracing theory that incorporates other theories and provides a realistic view on the
success of strategic alliances as a dynamic and relative concept.
Next, strategic alliance stability was analyzed in relation to SA success. 3 main approaches
to SAS definition were derived from the literature on SAS: outcome-based as an opposite to SA
instability, process-based as an opposite to SA instability, process-based (dynamic) as strategic
alliance stability itself (see Table 1.2). Overall, it can be claimed that more recent studies rely on
the process-based stability definition, defining it as a self-sufficient concept rather than through
instability.
As a next step, strategic alliance stability was conceptualized considering game theory
approach, which seems to be the most all-inclusive. From the viewpoint of game theory, strategic
alliance stability consists of 2 components on a general level: external and internal stability.
Internal stability, in turn, consists of 3 components: dynamic, strategic and motivational stability.
In order for an alliance to be stable, it is important that all the components are present. Hence,
strategic alliance stability is defined as a success of alliance performance during the period of
alliance operations under conditions of constant motivation of each partner firm to maximize the
results of cooperation (Zenkevich, Koroleva, Mamedova, 2014a).
Lastly, guided by R-A theory and recent empirical studies by Deitz et.al. (2010), López‐
Navarro, Callarisa‐Fiol and Moliner‐Tena (2013), strategic alliance stability factors were
introduced for further consideration, namely: long-term organization, trust, resource
complementarity.
38
CHAPTER 2. STRATEGIC ALLIANCE STABILITY FACTORS
HYPOTHESES DEVELOPMENT AND METHODOLOGY
2.1 Strategic alliance stability factors and hypotheses
Cooperation between firms should be understood from various perspectives. However,
oftentimes hard data available for analysis of an alliance does not always represent a full picture
of alliance’s state, and does not allow drawing conclusions on such concepts as “trust, forbearance,
reciprocity and opportunism” (Christoffersen, 2013), which are important for understanding
strategic alliance stability (Zenkevich, Koroleva, Mamedova, 2014a,b; Jiang, Li and Gao, 2008).
Establishing and managing a strategic alliance requires sufficient resources of participants,
including their cooperative effort, time and financial resources. Apart from a great cost, some of
investments made in an alliance are specific and non-recoverable (Parkhe, 1993; Brown, Dev, Lee,
2000; Lambe, Spekman and Hunt, 2002; Das, Rahman, 2010), which means they cannot be easily
employed outside the alliance. For this reason, it is crucial to understand stabilizing and
destabilizing forces in alliances, so it is possible to prevent alliance instability and manage stability
in a systematic way (Jiang, Li and Gao, 2008).
As Jiang, Li and Gao (2008) and Zenkevich, Koroleva, Mamedova (2014a) mention, one
of peculiarities of multiple studies (Kogut, 1988, 1989; Deitz et al, 2010; López‐Navarro,
Callarisa‐Fiol and Moliner‐Tena, 2013) is that an explanation of stability is made for a specific
alliance type, which limits the implementation of findings to a specific alliance type (e.g., joint
ventures, buyer-supplier agreements). However, this paper considers strategic alliances in a
broader sense (see sub-chapter 1.4 for the definition).
As for the factors of strategic alliance stability, they are chosen on a basis of R-A theory,
as it seems to be the most comprehensive and dynamic approach (see sub-chapter 1.2), which
corresponds to the principles of game theory that views SAS as a dynamic and multi-sided concept
(see chapter 1.4). In particular, papers (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013;
Deitz et al, 2010) were considered in respect to SAS factors as these are recent studies examining
a particular issue of SAS. The paper by Zenkevich, Koroleva, Mamedova (2014a) was considered
for SAS conceptualization. SAS factors considered in the paper: long-term orientation, trust,
resource complementarity. SAS components: internal and external stability.
Long-term orientation. Studies show that the longer the “shadow of the future”, the less
likely it is that partners are going to engage into opportunistic activities because the consequences
such behavior might have are to be considered by them (Axelrod, 1984; Heide and Miner, 1992;
Das, Rahman, 2010). In turn, long-term orientation increases the shadow of the future, making
39
partners dependent on each others’ behavior, and their cooperation more vigorous (Das, Rahman,
2010).
Moreover, in case partners are long-term oriented, they stay committed to the alliance even
in case of temporary inequalities between them as they believe that all the inequalities will even
out in the long-run (Das, Rahman, 2010), therefore, partners will expect to at least be able to gain
the amount of benefits indicated by the alliance contract. Long-term orientation of partners also
decreases the urge, or the pressure, of gaining quick results. The importance of the absence of
pressure for quick results is especially important for strategic alliances as it is rare when it is
possible for them to start generation positive economic outcome right after establishment (Das,
Rahman, 2010; Zenkevich, Koroleva, Mamedova, 2014a). If the alliance horizon is set to be long,
partners are going to be willing to commit to the relationship and make efforts to preserve it (Ring
and Van de Ven, 1994).
As follows from the definition of external SAS, an alliance has to demonstrate an
increasing long-term trend in its economic results to be externally stable (Zenkevich, Koroleva,
Mamedova, 2014a,b). In case partners are long-term oriented, they are likely to believe in the
alliance perspective (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013) and, contrary to the
short-term orientation, will not be likely to behave opportunistically, which would have a
detrimental effect on alliance economic results of an alliance (Das, Rahman, 2010). Overall, longterm motivation appears to be important for both internal and external stability of SAs (Zenkevich,
Koroleva, Mamedova, 2014a).
The following hypotheses are put forward:
H1: Long-term orientation is positively associated with external stability of a strategic alliance
H2: Long-term orientation is positively associated with internal stability of a strategic alliance
Trust. Trust in partner relationships decreases uncertainties, therefore, positively affects
conflict resolution and enhances cooperation (Granovetter, 1985; Madhok, 1995; Deitz et al,
2010). Stemming from the TCE approach, and incorporated into R-A theory, trust reduces
transaction costs by developing a desirable transaction climate (Granovetter, 1985; Madhok, 1995;
Huo, Ye, Zhao, 2015). Without mutual trust, partners would be likely to behave opportunistically
by taking advantage of doubtful situations, not explicitly defined by the contract (Williamson,
1985), which would affect the cooperation between partners, in particular (Das, Rahman, 2010),
perceived payoff equality and fairness along with partners’ willingness to stay within an alliance
and commit to it. It has also been claimed by scholars that trust has an impact on the degree to
which partners are long-term oriented as even during the hard times for an alliance, partners would
40
believe that short-term losses would be compensated by long-term gains (Ganesan, 1994; Lee and
Dawes, 2005; Ryu, Park and Min, 2007; Yu and Pysarchik, 2002; Zhao and Cavusgil, 2006;
López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013; Jiang, Li, Gao, 2008).
However, the association between trust and alliance success in terms of alliance economic
performance is not clearly articulated in the literature. It is argued by Nielsen (2007) that trust has
rather an indirect impact on economic results of an alliance, therefore, on a sequence of economic
results in time as well.
The following hypotheses are put forward:
H3: Trust is positively associated with internal stability of a strategic alliance
H4: Trust is positively associated with long-term orientation in a strategic alliance
Resource complementarity. Resource complementarity is believed to be a crucial, corner
stone element to reach and maintain SAS (Deitz et al, 2010). Deitz et al. (2010) underline that
partners with complementary resources are able to combine them in a unique way to attain a
competitive advantage in the market through extracting value from valuable, rare, durable and
inimitable resource combinations (Barney 1991, 1992). When the competitive level of
complementarity is achieved, the probability that partners are willing to change the alliance form
or to exit the alliance should decrease significantly (Deitz et al, 2010).
Partners with complementary resources are seen as mutually dependent (Geringer, 1988)
as partners’ resource contribution is beneficial for each party by definition. It has been shown in
the study of Beamish (1988) that multinational companies are eager to find local partners with
complementary resources while expanding their business abroad. On the other hand, Park and
Ungson (1997) have shown that low resource complementarity is reflected in increased
termination rates of alliances.
By recognizing that a partner supplies resources that complement firm’s own ones, a firm
also recognizes the original value of its partner for the alliance and the interdependence between
partners. Therefore, in this sense resource complementarity leads to increased partners’ trust and
decrease opportunistic tendencies in a relationship (Morgan and Hunt, 1994; Sarkar, Echambadi,
Cavusgil, and Aulakh, 2001). Furthermore, López‐Navarro, Callarisa‐Fiol & Moliner‐Tena (2013)
find that resource complementarity influences partner commitment through trust, not finding the
support for a direct relationship.
Scholars have proposed and empirically tested the hypothesis that resource
complementarity positively influences partner intentions to remain in the JV and cooperative
41
intent, respectively (Deitz et al, 2010; Jiang, Li and Gao, 2008), and Deitz et al (2010) found
support for each case.
Not only resource complementarity is connected to partners’ internal cooperation, but it
also has been studied as an antecedent of a desirable economic performance due to synergies
created among complementarity resources (Luo, 1999; Lambe, Spekman and Hunt, 2002; Nielsen,
2007).
The following hypotheses have been put forward:
H5:Resource complementarity is positively associated with external stability of a strategic
alliance
H6: Resource complementarity is positively associated with internal stability of a strategic
alliance
H7: Resource complementarity is positively associated with partners’ trust
External and internal stability. There is a rationale to assume that external stability, the
proxy of which is an upward trend in alliance results (Zenkevich, Koroleva, Mamedova, 2014a,
b), is positively associated with internal stability of a SA. As a primary reason of alliance formation
is connected to economic benefits generation and gaining an expected financial return (Umukoroa,
Sulaimonb and Kuyeb, 2009; Qing, Zhang, 2015), it is expected that economic results are
considered by partners during the alliance implementation phase. Moreover, alliance success in
the real world is evaluated by partners in comparison with some referent: either another company,
industry, or itself at a different point of time (Hunt, Lambe and Wittmann, 2002). Therefore,
partners continuously evaluate alliance performance and make their decisions on the future
cooperation based on results of the assessment, deciding how to behave within an alliance, whether
or not to stay in the alliance, maintain the same alliance form, etc. (Qing, Zhang, 2015).
Therefore, there is a reason to put the following hypothesis forward:
H8:External stability is positively associated with internal stability
The conceptual model of strategic alliance is depicted in the Figure 2.1. Each arrow in the
conceptual model represents a causal relationship and corresponds to a certain hypothesis. Overall,
there are 8 hypotheses on the relationships between SAS factors and SAS components, the
connection between SAS components, and the connections between SAS factors. Note that a sign
(+) in the parenthesis stands for a positive association between constructs. Hypotheses for the
model are summarized in a Table 2.1.
42
Table 2.1. Research hypotheses summary
H1: Long-term orientation is positively associated with external stability of a strategic alliance
H2: Long-term orientation is positively associated with internal stability of a strategic alliance
H3: Trust is positively associated with internal stability of a strategic alliance
H4: Trust is positively associated with long-term orientation in a strategic alliance
H5: Resource complementarity is positively associated with external stability of a strategic
alliance
H6: Resource complementarity is positively associated with internal stability of a strategic
alliance
H7: Resource complementarity is positively associated with partners’ trust
H8: External stability is positively associated with internal stability
External stability
Long-term
orientation
H8 (+)
H4 (+)
Trust
Internal stability
H7 (+)
Resource
complementarity
Figure 2.1 Conceptual model: strategic alliance stability factors
Source: Adapted from (Deitz et al, 2010; López‐Navarro, Callarisa‐Fiol and Moliner‐Tena,
2013)
Given the number of hypotheses and a complex set of interconnections that exist among
constructs, it makes sense to increase model complexity gradually to test it. Hence, a deeper
understanding of relationships, direct and indirect effects of SAS factors on SAS components
might be obtained.
43
Therefore, the first model to be tested in the following Chapter 3 incorporates only direct
relationships between SAS factors and SAS components (see Figure 2.2), which are presented by
hypotheses H1, H2, H3, H5, H6.
External stability
Long-term
orientation
Internal stability
Trust
Resource
complementarity
Figure 2.2. Hypotheses scheme (1) for empirical test
After the model in the Figure 2.2 is tested, a direct impact of SAS factors on SAS
components can be determined. This differentiation needs to be made in order to define different
types of direct and indirect effects. For example, the conceptual model in the Figure 2.1 implies
that there is a number of mediators that might be present in the model. Mediator represents a
(latent) variable intervening a direct relationship between dependent and independent (latent)
variables (Kim, Kaye and Wright, 2001). In Figure 2.1, potential mediators are: Trust (for the
relationship between Resource complementarity and Internal stability) and External stability (for
the relationship between Long-term orientation and Internal stability). For the mediation effect to
be proven, it is important that a direct relationship between dependent and independent construct
is significant (Kim, Kaye and Wright, 2001; Hair, 2010), therefore, direct relationships between
SAS factors and SAS components are tested first and other relationships are introduced one by
one.
44
External stability
Long-term
orientation
H8 (+)
Internal stability
Resource
complementarity
Trust
Figure 2.3. Hypotheses scheme (2) for empirical test
In the hypotheses scheme (Figure 2.3), a new hypotheses (H8) is added to the set of
relationships, which allows to examine whether or not External stability is positively associated
with Internal stability, therefore, also examining indirect effect between Long-term orientation
and Internal stability as well.
External stability
Long-term
orientation
H8 (+)
Trust
Internal stability
H7 (+)
Resource
complementarity
Figure 2.4. Hypotheses scheme (3) for empirical test
45
Figure 2.4 represents the next set of hypotheses to be tested empirically, it is the last
modification of the conceptual model before the final version in the Figure 2.1. Comparing the
model in a Figure 2.4 with a model in a Figure 2.3, an additional hypothesis H7 is introduced. By
testing the model in Figure 2.4, it will be possible to make conclusions on whether or not Trust
plays a mediator role for the relationship between Resource complementarity and Internal stability.
2.2 Data collection: resources and restrictions
In this research, primary data was collected from the web-based questionnaire sent out to
European firms. Generally, collecting data from the questionnaire is very popular way to study the
phenomena of SAS, which is done in multiple studies, e.g., (Nielsen, 2007; López‐Navarro,
Callarisa‐Fiol and Moliner‐Tena, 2013; Deitz et al, 2010; Ozorhon et al, 2008). On the example of
the same studies, it can be concluded that deductive approach is also a common approach applied
across empirical papers on the topic as it allows to derive hypotheses from the literature that are
empirically tested later on.
Following the deductive approach, hypotheses were formulated based on existing literature
(see sub-chapter 2.1). Next, the questionnaire was constructed (see Appendix 2) and the survey
conducted. Consequently, questions in the survey corresponded to the latent constructs introduced
in sub-chapter 2.1. The web-based questionnaire provided access to primary and most up-to-date
data. Respondents were asked to give their answers on the alliance that had been functioning at
the moment of filling out the survey. In the survey, 7-point Likert (1932) type of scale was used,
as it provides internal scale assessment and is believed to be a powerful tool for data analysis (Hair,
2010).
The questionnaire itself consisted of 3 sections (see Appendix 2). Section A contained
questions on the basic information about alliances and respondents: country of operations, size,
age, alliance type, industry, respondents’ level of involvement in the alliance. In section A
respondents were asked to choose alliance characteristics among the options introduced.
Information provided by respondents in Section A was important to understand sample
characteristics in more details. Section B of the questionnaire included questions on SAS. The
answers to the questions in Section B were later used as data for indicators when measuring latent
constructs (external stability, internal stability). Section C was dedicated to SAS determinants.
Similar to answers in Section B, answers to Section C were used as data for indicators of SAS
determinants (long-term orientation, trust, resource complementarity). As for the Sections B and
C together, respondents were asked to evaluate their degree of agreement (1-Completely Disagree,
2-Disagree, 3-Somewhat Disagree, 4-Neutral, 5-Somewhat Agree, 6-Agree, 7-Completely Agree)
46
with the statements provided in the questionnaire. All the questions in a questionnaire were marked
as obligatory for submission, therefore, no missing values were generated.
Contact details of respondents were extracted from two databases: SDC Platinum and
Amadeus (Bureau van Dijk). Originally, a thousand email addresses of strategic alliances were
extracted from SDC Platinum, however, as, generally, many strategic alliances are short-term, it
is well-explained that 60% of email addresses from the extracted database did not exist at the
moment of survey distribution. Only one response was generated from the original distribution
attempt.
At a second attempt, a new database for contact addresses was compiled using Amadeus
Bureau van Dijk. The most of responses, therefore, were obtained from sending the survey out to
email addresses from Amadeus database. Particularly, the search was conducted using subdatabased (Owners – Shareholders, Owners - Immediate, Ultimate, Domestic Ultimate Owner,
Owners – Subsidiary, Auditors, Bankers, Managers) and filters for each of them: Step 1 – (“Search
entire database”) and (Very large + Large + Medium +Small Companies) and (All countries); Step
2 – (Legal form) and (Legal status) and (Country) and (Email Address). The peculiarity of this
source of contact information is that Amedeus does not specifically differentiate among alliances
and non-alliances. The cover letter for the survey (see Appendix 1) addressed potential respondents
explicitly specifying the assumption that a respondent should be somehow involved in a strategic
alliance. All the requirements were repeated for respondents at a survey starting page. After
deleting repeated email addresses, approximately 500 thousand email addresses were extracted
from Amadeus for further distribution.
Out of 1167 potential respondents who have opened the link to the survey, 184 complete
responses were obtained, which constitutes 15.77% of the original number. Afterwards, 9 answers
were deleted from the original sample due to the fact that alliances were short-term (shorter than
1 year).
The total number of respondents who took part in the survey equals 184. However, some
of the observations represented the alliances that were too young to draw any conclusions on their
stability (less than 1 year of functioning). As strategic alliance stability is applied for long-term
alliances, only the alliances that were at least one year of existence at the moment of respondent
filling in the questionnaire. Consequently, the sample size was decreased to 175 observations.
Respondents were managers of strategic alliances, managers of partner companies and
employers of both alliances and partner companies that operate in Europe. Raw data was collected
in a form of a survey created at surveygizmo.com.
47
Most of respondents (48,0%) described themselves as managers of companies that
participated in strategic alliances, while 39,4% of respondents were strategic alliance managers.
The rest of respondents were either employed by a company involved in strategic alliances (7,4%),
or worked in a strategic alliance (5,1%). No restrictions were posed on SA involvement of
respondents due to scarcity of data. See Table 2.2 for reference. Overall, it can be argued that
respondents were in a position to answer alliance-related questions by providing relevant
information because approximately 90% of respondents represented either alliance management
team or the management team of partner companies they were involved in strategic alliances.
Table 2.2. Respondents’ involvement
Respondent’s alliance involvement
N
Share
Partner (participant) company management team
84
48,0%
Strategic alliance management team
69
39,4%
13
7,4%
9
5,1%
175
100,0%*
Employed by a company that participates in a strategic alliance (not
management team)
Employed by a strategic alliance (not management team)
Total
*mistakes are possible due to rounding
Geographically (Table 2.3), alliances that respondents chose to provide information about,
were located in Europe, most of the alliances operating in Germany (18,3%), Bulgaria (16,0%)
and Finland (13,7%), representing Western, Eastern and Nothern Europe respectively. Generally,
however, it can be argued that the sample is homogenous in terms of geography.
48
Table 2.3. Alliance geography
Country
N
Share
Germany
32
18,3%
Bulgaria
28
16,0%
Finland
24
13,7%
Denmark
14
8,0%
Russia
11
6,3%
Estonia
10
5,7%
Czech Republic
7
4,0%
Austria
5
2,9%
Belgium
5
2,9%
France
5
2,9%
Spain
5
2,9%
Switzerland
5
2,9%
Bosnia and Herzegovina
5
2,9%
Italy
4
2,3%
Cyprus
4
2,3%
United Kingdom
4
2,3%
Croatia
2
1,1%
Greece
1
0,6%
Norway
1
0,6%
Serbia
1
0,6%
Slovakia
1
0,6%
Turkey
1
0,6%
175
100,0%*
Total
*mistakes are possible due to rounding
Speaking of the industry (see Table 2.4) alliances in a sample belong to, most of them are
concentrated in the business services industry (19%), machinery industry comes second (9,2%),
followed by chemical and allied products industry (5,2%). Overall, the sample constitutes of
alliances that are distributed across over 18 industries.
49
Table 2.4. Alliance industry
Industry
N
Share
Business Services
33
18,9%
Machinery
16
9,1%
Chemicals and Allied Products
9
5,1%
Computers, Peripheral Equipment and Software
9
5,1%
Metal and Metal Products
9
5,1%
Electronic and Electrical Equipment
7
4,0%
Health Services
7
4,0%
Food and Kindred Products
6
3,4%
Transportation and Shipping (except air)
6
3,4%
Drugs
4
2,3%
Textile Mill Products
4
2,3%
Transportation Equipment
4
2,3%
Wholesale Trade-Durable Goods
4
2,3%
Computer Processing and Data Preparation and Processing
3
1,7%
Investment and Commodity Firms, Dealers, Exchanges
3
1,7%
Paper and Allied Products
3
1,7%
Transportation by Air
3
1,7%
Wholesale Trade-Nondurable Goods
2
1,1%
Aerospace and Aircraft
1
0,6%
Coal Mining
1
0,6%
Computer Integrated Systems Design
1
0,6%
Measuring, Medical, Photo Equipment; Clocks
1
0,6%
Petroleum Refining and Related Industries
1
0,6%
Prepackaged Software
1
0,6%
Security Systems
1
0,6%
Telecommunications
1
0,6%
Transportation by Air
1
0,6%
Other
34
19,4%
Total
175
100,0%*
*mistakes are possible due to rounding
50
As it was mentioned previously, alliances that existed for less than 1 year were excluded
from further consideration. The summary on the alliances by their age is introduced in the Table
2.5. Most of the alliances in the sample (48,0%) exist for longer than 5 years, the lowest share of
alliances are between 3 and 5 years of existence.
Table 2.5. Alliance age
Number of years
N
Share
1-3 years
60
34,3%
3-5 years
31
17,7%
more than 5 years
84
48,0%
Total
175
100,0%*
*mistakes are possible due to rounding
As for the size of alliances in the sample, the most part of them (54.3%) belong to the
“micro” category, according to Eurostat classification, and have between 1 and 9 permanent
employees. The second biggest category of alliances (22.9%) in the sample in terms of size is
“small” alliances with 10-49 employees. The third biggest category of alliances (12.5%) in a
sample are “large” alliances with 250 or more permanent employees. The rest of the sample
(10.2%) is represented by “medium” alliances. Table 2.6 presents the alliance size summary for
the sample.
Table 2.6. Alliance size
Alliance size
N
Share
1-9 (micro)
95
54.3%
10-49 (small)
40
22.9%
50-249 (medium)
18
10.2%
250 or more (large)
22
12.5%
Total
175
100.0%*
*mistakes are possible due to rounding
Lastly, respondents were asked to classify their alliance into three categories: joint venture,
minority equity alliance or non-equity alliance. Such classification is general enough (Das,
Rahman, 2010), which is suitable for the purpose of this study. Judging by the Table 2.7, it can be
argued that most alliances in the sample (46.9%) are non-equity alliances, followed by joint
ventures (28.6%) and minority equity alliances (24.6%).
51
Table 2.7. Alliance type
Alliance type
N
Share
Joint venture
50
28.6%
Minority equity alliance (a member holds equity in the partner, or partners
cross-hold equity in each other)
43
24.6%
Non-equity alliance (does not involve any equity or the transfer of ownership)
82
46.9%
Total
175
100,0%*
*mistakes are possible due to rounding
Given the nature of variables under examination, the set of hypotheses and the type of
relationships among variables (see Figure 2.1 above), in particular that some variables act as both,
dependent and independent variables, and given the explanatory nature of the research the most
appropriate method for data analysis would be structural equation modeling (SEM). SEM is a
widely used tool in managerial researches because it enable the researcher to evaluate causal
relationships between constructs that cannot be measured directly (latent constructs) (Hair, 2010;
Grover and Malhotra, 2003), often describing theoretical concepts (Malhotra, Birks, 2007, p. 605),
connected with a complex set of interrelationships (Hair, Ringle, Sarstedt, 2011).
The advantages of SEM compared to other multivariate techniques include (Suhr, 2006):
(1) flexibility in terms of fields of application (e.g., psychology, economy, humanities); (2)
allowance for the assessment of a causal relationships set, defined a priori, based on theory; (3)
ability to test relationships among latent constructs (unobserved variables) contrary to traditional
techniques that allow to work only with measured variables; (4) explicit specification and
assessment of the error term; (5) allowance for multi-sided goodness-of-fit assessment for the
model; (6) decreased possibility of issues related to multicollinearity due to the fact that multiple
measured variables are used in order to describe one latent construct, therefore, SEM accounts for
high correlations among measured variables within one latent construct; (7) simultaneous
assessment of multiple linear regressions that provides a fuller picture of the examined
relationships.
The variables represented by ovals in the conceptual model (Figure 2.1) represent latent
constructs and will be referred to as “latent constructs” or “constructs” later on. Considering the
sample size that is sufficient for running the covariance-based SEM (CB-SEM), this study follows
the CB-SEM methodology for the conceptual model assessment. For this purpose, IBM SPSS
Amos 19 software package was used. Therefore, Chapter 3 reproduces the logic of a two-step
SEM-methodology.
52
2.3 Chapter 2 concluding remarks
Chapter 2 is dedicated to hypotheses development based on existing academic literature.
Unlike the majority of studies that view SAS as a one-dimensional construct, this research breaks
it down to two parts: external and internal stability. The choice of determinants was based on recent
conceptual models from recent studies (Deitz et al, 2010; López‐Navarro, Callarisa‐Fiol and
Moliner‐Tena, 2013) and results of empirical tests.
Overall, 8 hypotheses were formulated that covered theoretical relationships between
strategic alliance stability components and strategic alliance stability factors, between strategic
alliance stability components themselves, and between strategic alliance stability factors. The final
conceptual model is introduced in the Figure 2.1. Strategic alliance stability components (external,
internal stability) and their determinants (resource complementarity, trust, long-term orientation)
are incorporated in the model.
Then, the data collection process and the sample were described in the Chapter 2. Data was
collected on the basis of a web-based questionnaire that was distributed through emails. The main
part of questions is a survey (Section B and Section C) corresponded to latent constructs indicators
used for the following SEM analysis (see Appendix 2 for the questionnaire, Chapter 3 for SEM
analysis). All the indicators were measured with a 7-point Likert scale.
Two databases were used to extract contact information of potential respondents: SDC
Platinum and Amedeus Bureau van Dijk. Totally, over 500 thousand email addresses of European
companies were used for final distribution via email. One of the peculiarities of the final emails
set was that there was no indication of whether or not the recipient company is involved in a
strategic alliance. Therefore, the condition of alliance involvement was indicated as a major
requirement in the cover letter for the survey (see Appendix 1). All the requirements were repeated
for respondents at a survey starting page. Overall, the link was opened by 1167 potential
respondents who have opened the link to the survey, 184 provided full answers (15.77%).
However, 9 of them were deleted from the original sample due to the fact that alliances were shortterm (shorter than 1 year). Overall, a sample of 175 observations was obtained.
As for the sample, alliance information and respondents’ status were analyzed.
Approximately 40% of respondents represented SAs management team, nearly 50% represented
partner companies’ management team, which totals approximately 90% of respondents who are
certainly capable of providing relevant responses. The sample used for further analysis is
homogenous in terms of geography as all the alliances operated on the territory of Europe,
however, alliances differ by industry, size, type and age to a certain extent. These differences were
53
not considered in the research, given the initial complexity of the model and the scarcity of data,
which did not allow to cut the sample size in order to reach overall homogeneity.
54
CHAPTER 3. EMPIRICAL STUDY ON STRATEGIC ALLIANCE
STABILITY FACTORS
3.1 Data analysis and measurements
For the purpose of data analysis, IBM SPSS Statistics 19 and IBM SPSS Statistics Amos
19 were utilized to conduct data analysis. Apart from the analysis of responses presented in the
sub-chapter 2.4 (alliance country, industry, respondent’s status, alliance age and size), an analysis
of standard deviation (SD), minimum and maximum scores were analyzed.
Following the conceptual model and the chosen tool for an empirical study, a set of
questions was formulated in order to measure latent construct (see sub-chapter 3.2 and Appendix
2 for more detailed information). Each construct (external stability, internal stability, long-term
orientation, trust, resource complementarity) was measured with several indicators, which were
subject to further analysis. All the indicators were measured by a 7-point Likert (1932) scale, which
is commonly used in managerial studies (Hair, 2010). During the survey, respondents were asked
to evaluate statements using the following scale: 1-Completely Disagree, 2-Disagree, 3-Somewhat
Disagree, 4-Neutral, 5-Somewhat Agree, 6-Agree, 7-Completely Agree.
The results summary of descriptive statistics are presented in the Tables 3.1-3.5.
Table 3.1. Descriptive statistics. External stability
Indicator code (question)
ES-1 (Generally, there is a constant improvement
in alliance's economic result)
ES-2 (Most often, alliance meets its economic
objectives (e.g., revenue, net profit, additional
benefits generated for partner companies, etc.)
ES-3 (Overall, revenue trend of the alliance can be
characterized as rising)
ES-4 (Overall, net profit trend of the alliance can
be characterized as rising)
Mean
SD
N
5.10 (somewhat agree) 1.303 175
5.17 (somewhat agree) 1.205 175
5.14 (somewhat agree) 1.247 175
4.92 (somewhat agree) 1.297 175
Regarding external stability, average scores for indicators are varying around score 5,
meaning “somewhat agree”. Therefore, on average, it can be implied that alliances respondents
chose to talk about somewhat externally stable alliances. Standard deviations for each indicator
are comparable among each other. Only the net profit trend was assessed by respondents a little
below 5-score.
55
Table 3.2. Descriptive statistics. Internal stability
Indicator code (question)
IS-1 (There is a well-established procedure of how
benefits from the strategic alliance are shared
among participants)
IS-2 (There is a mutual understanding on how the
benefits from the strategic alliance should be
shared among alliance participants)
IS-3 (Participants are absolutely satisfied with this
form of cooperation compared to other possibilities
(e.g., different forms of cooperation with other
companies, such as other alliances)
IS-4 (Participants will continue cooperation in this
alliance form until the termination date)
IS-5 (Participants are involved in solving alliance
issues)
Mean
SD
N
5.35 (somewhat agree) 1.381 175
5.36 (somewhat agree) 1.390 175
4.77 (somewhat agree) 1.396 175
5.48 (somewhat agree) 1.203 175
5.50 (agree) 1.077 175
As for internal stability (Table 3.2), mean scores show a more positive respondents’
perception of it. Generally, the mean scores can characterize an average alliance in the sample as
internally “somewhat stable”. However, the question that indicates participants’ satisfaction from
alliance participation was given a lower score on average compare to other indicators, with a higher
standard deviation.
Table 3.3. Descriptive statistics. Long-term orientation
Indicator code (question)
LTO-1 (Participants believe that over the long run
the alliance will be profitable)
LTO-2 (Maintaining a long-term relationship among
the participants is important for them)
LTO-3 (Participants focus on long-term goals in this
alliance)
LTO-4 (Participants believe that any concessions
they make to help out among them will even out in
the long run)
LTO-5 (Participants expect working together for a
long time)
LTO-6 (Participants are willing to make sacrifices to
help out among them from time to time)
Mean
SD
N
5.80 (agree) 1.130 175
5.73 (agree) 1.137 175
5.46 (somewhat agree) 1.153 175
5.21 (somewhat agree) 1.111 175
5.66 (agree) 1.107 175
5.05 (somewhat agree) 1.205 175
Descriptive statistics shows (Table 3.3) that generally respondents have said that they agree
or somewhat agree that alliance partners in the alliance they have chosen for their responses, were
56
either long-term oriented or somewhat long-term oriented. The scores for indicators range from
5.05 to 5.8. The lowest score is associated with the indicator LTO-6, the highest – with LTO-1.
Table 3.4. Descriptive statistics. Resource complementarity
Indicator code (question)
RC-1 (Together, participants have been adding a
substantial value to the alliance)
RC-2 (Alliance participants bring to the table
resources and competencies that complement those of
other participants)
RC-3 (Strategic fit between participants could not be
better)
RC-4 (All participants contribute different resources
that help achieve their mutual goal)
RC-5 (Participants have complementary strengths that
are useful to their relationship)
RC-6 (Each participant has separate abilities that,
when combined together, enable them to achieve
goals beyond their individual reach)
Mean
SD
N
5.69 (agree) 0.939 175
5.68 (agree) 1.062 175
4.57 (somewhat agree) 1.456 175
5.30 (somewhat agree) 1.181 175
5.62 (agree) 1.038 175
5.67 (agree) 1.069 175
In general, judging by respondents’ responses, an average alliance in a sample has
complementary resources (Table 3.4). Means for every indicator are in line with each other, and
most of them exceed 5.5 score. However, some of the indicators show a lower complementarity in
resources than others (RC-3, RC-4). Such score allocation might have occurred due to the fact that
these respective questions ask respondents for a more profound understanding of the role of
resource complementarity for their alliance.
Table 3.5. Descriptive statistics. Resource complementarity
Indicator code (question)
T-1 (Participants believe that another participant
(other participants) is (are) honest)
T-2 (Participants consider each others perspective)
T-3 (Participants are always faithful)
T-4 (Partners found it necessary to be cautious in
dealing among themselves)
T-5 (Participant(s) are honest and truthful among
themselves)
T-6 (Participants interact with each other fairly and
justly)
Mean
SD
5.25 (somewhat agree) 1.341
N
175
5.20 (somewhat agree) 1.246
5.27 (somewhat agree) 1.170
4.52 (somewhat agree) 1.497
175
175
175
4.93 (somewhat agree) 1.320
175
5.06 (somewhat agree) 1.340
175
57
Overall, judging by descriptive statistics (Table 3.5), on average, partners somewhat trust
each other. Average scores exceed the medium 4-point, most average scores exceed 5 points
(somewhat agree). Hence, survey participants perceived that alliance participants have trust in each
other and their relationship, but the assessment of some of the manifestation of trust in partner
relationships vary in the sample more (SD for T-4 1.497 ) and assume the lowest averages (4.52)
among trust indicators.
A more profound analysis of latent constructs (validity, reliability, correlations) is provided
in the following sub-chapter 3.2.
3.2 Measurement model development and assessment
SEM represents a set of statistical techniques that combines factor and regression analysis,
which allows evaluating causal relationships among different sets of dependent and independent
constructs, organized in linear structural equations, at the same time. The peculiarity of SEM is
that it allows one construct to be independent in one set of relationships, but dependent in another
set (Malhotra, Birks, 2007; Hair, 2010). In SEM, latent constructs are measured by measured
variables, also denoted as observed variables or indicators.
SEM has become increasingly popular and constitutes a new standard of managerial
researches. Not only it allows to assess a set of inter-dependent relationships where one construct
might be both dependent and independent, SEM also allows to deal with non-normal data (Hair,
2010). Other benefits of SEM compared to traditional multivariate techniques are analyzed in the
sub-chapter 2.3.
The SEM process is traditionally divided into 2 stages: measurement model (MM)
development and assessment, and structural model (SM) specification and assessment (Hair,
2010). Hence, this approach is known as a two-step SEM process. Confirmatory factor analysis is
used to assess the MM fit and validity to ensure that the measures are satisfactory to serve as a
basis for a SM. After MM is proved to be adequately represent theory with the data obtained for
the study, SM assesses the structural theory, or a set of dependencies between latent constructs.
Therefore, the main difference between the CFA and the SM assessment is that the CFA is used
to assess the relationships between indicators and latent constructs, while the SM examines
relations between latent constructs.
This research follows a two-step SEM approach. The first step in this approach requires to
develop and assess the MM, the second step requires to specify and assess the MM (Hair, 2010).
This sub-chapter deals with the first step of the two-step SEM.
58
MM corresponds to the conceptual model (Figure 2.1) in terms of latent constructs that
need to be measured by a set of measured variables. To recap, latent constructs are the following:
external stability, internal stability, trust, long-term orientation, resource complementarity. Each
of them has a set of indicators, or measured variables, used for latent construct assessment. These
indicators and their codes are presented in the Appendix 2. All the measured variables in the MM,
hence, in SM, are reflective meaning that latent construct cause measured variables to occur, which
is a typical way of representing latent constructs (Hair, 2010). In the survey, participants were
asked to evaluate statements about their alliance on a 7-point Likert scale (from “1” – “Completely
disagree” to “7” – “Completely agree”). All the questions that participants were asked matched
theoretical understanding of the concepts.
Before the final launch of a survey, a small-scale test survey was conducted in order to
assess the difficulty level and identify potential sources of survey improvement (e.g., web-survey
navigation, level of question comprehensiveness, question wordings). One out of five test survey
respondents indicated that all the questions were formulated comprehensively and could be
answered by those practitioners who are involved in working with strategic alliances. Later on,
during the course of a survey, 2 respondent voluntarily noted that they found the questionnaire
easy and comprehensive, which was later supported by “real” respondents feedback in some cases,
therefore, supporting the notion that face validity of constructs is present.
External stability views SA as a separate economic entity, so it is possible for an external
observer to draw conclusions on its stability. Following external stability definition (see subchapter 1.4), it is assumed that a strategic alliance is externally stable in case its economic results
show a raising trend (Zenkevich, Koroleva, Mamedova, 2014a, b). Economic results of the
strategic alliance might include its net profit, revenue, market share, etc. Therefore, survey
participants were asked to evaluate statements about strategic alliance economic results (on a scale
from “1” – “Completely disagree” to “7” – “Completely agree”) from the most general to more
exact terms.
As discussed in sub-chapter 1.4, internal stability of a strategic alliance is a multidimensional construct, and is comprised of motivational, strategic and dynamic stability.
Therefore, each of these elements should be reflected in internal SAS measurement scale. Interpartner relationships play a great role in strategic alliance stability (Deitz et al., 2010), and their
constant mutual involvement in alliance activities is an important element of its stability that
eventually has an effect on alliance performance. The extent to which partners are involved into
alliance activities stem from their motivation to enhance alliance economic results, therefore, to
maximize their own benefits (Wong, Tjosvold, Zhang, 2005; Deitz et al, 2010; Gulati, Khanna,
59
and Nohria 1994; Sarkar et al. 2001; López‐Navarro, Callarisa‐Fiol, Moliner‐Tena, 2013). The
next element of internal stability is the dynamic stability, which is observed in cases when partners’
expected and gained benefits correspond to the benefits expected at the moment of signing the
contract (Zenkevich, Petrosjan, 2006; Kumar, 2011). According to the optimal decision principle
(Zenkevich, Petrosyan, Yang, 2009), the fact that the contract was signed among partners and they
have agreed on cooperation indicates that partners have accepted the rules of benefits sharing and
that they have a clearly established procedure of how benefits should be split among them. Hence,
in case of dynamic stability, the procedure of benefits sharing is also known to participants. Lastly,
if an alliance is strategically stable, all the participants prefer to stay within a particular alliance
given all other options available, and are likely to continue cooperation further without leaving the
alliance prematurely (Zenkevich, 2009; Zenkevich, Koroleva, Mamedova, 2014a, b). Therefore,
participants were asked to evaluate statements about partners’ contribution to the alliance, benefits
sharing and their attitude to the current alliance.
As it was discussed previously in the sub-chapter 1.5, trust is an important characteristic
of partner relationships in strategic alliances. In their study on the third-party supplier
relationships, Huo, Ye, Zhao (2015) claim that trust is indicated by one party’s assessment of
another’s honesty, eagerness to consider the party’s perspective. Another indication of presence
of trust in a relationship would be an outside observation of partners’ relationships that were
characterized as honest and truthful, fair and just (López‐Navarro, Callarisa‐Fiol and Moliner‐
Tena, 2013). Overall, if partners stay faithful to each other (Deitz et al, 2010), this is an indicator
of trust in a relationship. At a contrary, the fact that partners found it necessary to deal cautiously
with each other would indicate the absence of trust (López‐Navarro, Callarisa‐Fiol and Moliner‐
Tena, 2013).
Partners with long-term orientation hope for their relationship with each other to bring
them economic benefits in the future (López‐Navarro, Callarisa‐Fiol and Moliner‐Tena, 2013;
Ganesan, 1994; Kelley and Thibaut, 1978). As follows, due to the value that the cooperation
generates, partners would be concerned about their existing relationship (López‐Navarro,
Callarisa‐Fiol and Moliner‐Tena, 2013; Das and Teng, 2000). Logically, contrary to short-term
oriented firms who would push the partner to generate quicker results (Das, Rahman, 2010) and
try to get immediate benefit from each transaction (Das and Teng, 2000; Ganesan, 1994), longterm oriented partners would put their long-term goals before the quick gain (Das and Teng, 2000).
Moreover, there is evidence that partners with long-term orientation will adjust their behavior in
order to focus on the achievement of the long-term goals, e.g., partners will assist each other in
resolving issues because they believe that another partner will do the same for them (Griffith,
60
Harvey and Lusch, 2006; Lee and Dawes, 2005; Lusch and Brown, 1996). In other words, longterm orientation promotes the alignment in partners’ goals and actions (López‐Navarro, Callarisa‐
Fiol and Moliner‐Tena, 2013).
In case partners acknowledge that their resources are complementary, they are likely to
assume that each of them adds substantial value to the alliance jointly as well as that their resources
and competencies complement each other. Moreover, partners are likely to agree that the strategic
fit among them is the best possible and, therefore, they could not have found a partner with a better
strategic fit (Deitz et al, 2010), as the combination of resources among them creates a competitive
advantage through synergies (Hunt, Lambe and Wittmann, 2002) and helps attain their joint
objectives (Lambe, Spekman and Hunt, 2002; Hunt, Lambe and Wittmann, 2002). Moreover,
given that resources are complementary, it means that they should be distinct (Hunt, Lambe and
Wittmann, 2002) to create synergies between partners and provide more benefits to partners than
they could have gained operating individually (Lambe, Spekman and Hunt, 2002).
Initial confirmatory factor analysis. Before the CFA was done, measured variables with
reversed scores were recoded to directly reflect the latent construct. The re-coding was done by
subtracting the observed score from the highest value on a scale plus one, which totals 8.
Measurement model validity assessment is a crucial step in SEM as it compares the
empirical data against theoretical measurement model developed earlier in this chapter. The
assessment of measurement model validity contains several elements, which are: overall model fit
assessment and construct validity. When construct validity is analyzed, the attention is paid to
convergent, discriminant, nomological and face validity (Hair, 2010).
Following Van Dijk (2014), it was decided to run a preliminary test of construct reliability
analyzing each of the constructs separately from other ones. In a statistical sense, reliability
referred to as a percent of observations that are inconsistent among each other due to individual
differences that are natural for different respondents, given that the survey has been conducted
among individuals. This means that even a reliable survey will have different responses to the
same questions due to the fact that respondents’ opinion on questions will differ among
respondents, not because of the fact that the questionnaire questions were unclear or
incomprehensive. Therefore, a test for reliability was conducted for all 5 latent constructs.
The preliminary reliability analysis was conducted using Cronbach’s alpha coefficient. It
indicates that all latent construct taken separately, disregarding possible correlations between them
and potential cross-loadings are able to capture the concept described. Cronbach's alpha was
calculated in IBM SPSS Statistics 19 software package, and analyzed afterwards. In general
61
practice, Cronbach's alpha cut-off is 0.7, but small negative deviations are admissible (Cooper and
Schindler, 2006; Malhotra and Birks, 2006). The result of Cronbach’s alphas test is summarized
in the Table 3.1 below. More detailed results of the test can be found in Appendix 3.
However, for SEM is it important to keep the number of indicators for each latent construct
over 3 to keep the model overidentified (Hair, 2010). Therefore, in case it was possible, and
indicators could be kept within a construct, it was done this way. The total number of latent
variables in the MM equals to 8, and each of the constructs is assessed by at least 3 factors to keep
the model overidentified in order to make valid assessment of each latent variable and of the MM
overall (Hair, 2010).
Table 3.6. Cronbach’s Alpha Summary
Number of indicators
Cronbach’s alpha
External stability
4
0.844
Internal stability
5
0.813
Trust
6
0.905
Long-term orientation
6
0.899
Resource complementarity
6
0.846
All indicators
27
0.927
Latent construct
The results in the Table 3.6 indicate that all the latent constructs have Cronbach’s alpha
results above 0.7 threshold. Moreover, the composite Cronbach’s alpha of the whole dataset is
well above the threshold of 0.7.
CFA is the next step of the SEM analysis to ensure convergent, discriminant and face
validity along with construct reliability (Gerbing and Anderson, 1988) as well as the overall model
fit. Each indicator loading was treated as an a priori indicator for the latent construct it measures,
and all the latent constructs were allowed to be correlated as there was no ground for an assumption
that latent constructs are not correlated. The output for the MM after the initial CFA can be found
in the Appendix 4.
Measurement model fit assessment is necessary to identify, how well the observed data fits
the theoretical measurement model developed at earlier stages. The overall fit of the MM was
assessed by several indices to have a better understanding of the goodness-of-fit for the
measurement model. The rule of thumb suggests relying on, at least, one absolute fit index and
one
incremental
fit
index
apart
from
traditional
𝜒 2 results. Table 3.7 compares the expected MM fit indices for the good fit with the obtained ones.
62
Table 3.7. Initial Measurement Model Assessment
Expected*
𝜒2
𝜒 2 normed
CFI
RMSEA
Obtained
Significant p-values
expected
<2.0 – good fit
2.0-5.0 – acceptable fit
> 0.95 great
> 0.90 moderate
> 0.80 sometimes
acceptable
< .05 good
0.05 - 0.10 moderate
> 0.10 bad
728.409 (p = 0.000)
df = 314
2.32
0.853
0.087
90 percent confidence
interval RMSEA =
(0.079; 0.095)
*Source: (Hair, 2010; Van Dijk, 2014)
Results presented in the Table 3.7 demonstrate some issues with the MM fit. First of all,
the comparative fit index (CFI) is lower than the required 0.9 threshold for an adequate fit.
Moreover, root mean square error of approximation (RMSEA) is also considered to be moderate,
but not good enough. The 𝜒 2 normed, which is calculated as 𝜒 2 divided by the degrees of freedom,
suggests an acceptable, yet not perfect MM fit. Overall, given the mixed evidence, it makes sense
to improve the measurement model as, generally, the SM will have a less good fit than the MM.
To find the areas of MM improvement, construct validity is assessed along with
standardized residuals and modification indices.
The analysis of construct validity begins with the analysis of convergent validity (factor
loadings of greater than 0.5, preferably higher than 0.7). Appendix 4 that includes initial CFA
results, contains data on all the factor loading produced after the initial CFA. Table 3.8 summarizes
information on factor loadings that are lower than the ideal 0.7 cut-off.
No indicators have shown factor loadings below 0.5 threshold, however, 8 indicators out
of 27 have shown factors loading below the ideal cut-off point of 0.7 (see Table 3.8), which is not
critical, however, can cause issues with construct reliability.
63
Table 3.8. Initial CFA. Factor loadings between 0.5 and 0.7
Construct
Indicator
Estimate
Internal stability
IS3
0.685***
IS4
0.679***
IS5
0.523***
External stability
ES2
0.569***
Resource complementarity
RC1
0.566***
RC3
0.601***
Trust
T3
0.672***
Long-term orientation
LTO6
0.662***
*** significantly different from zero at the 0.001 level (two-tailed)
As a next step in assessing construct validity, average variance extracted (AVE) and
construct reliability (CR) (see formula (1), (2) respectively, Table 3.9 for the result of calculations)
were calculated manually using standardized factor loadings and squared multiple correlations
from the output. The calculations followed the formulas below.
𝐴𝑉𝐸 =
∑𝑛
𝑖=1 𝐿𝑖
2
,
𝑛
(1)
where 𝐿𝑖 is standardized factor loading, n is the total number of items loaded on a construct,
i is the number of an individual item.
2
𝐶𝑅 =
(∑𝑛
𝑖=1 𝐿𝑖 )
2
,
(2)
𝑛
(∑𝑛
𝑖=1 𝐿𝑖 ) + (∑𝑖=1 𝑒𝑖 )
where where 𝐿𝑖 is standardized factor loading, 𝑒𝑖 is the error variance term for each item,
n is the total number of items loaded on a construct, i is the number of an individual item.
Table 3.9. Initial CFA: Construct validity and reliability
Construct
ES
IS
RC
T
LTO
AVE
0.59
0.48
0.51
0.62
0.61
CR
DV
ES IS
RC
T
0.85
1
0.15
0.05
0.82
1
0.35
0.86
1
0.91
0.90
0.16
0.33
0.21
1
LTO
0.20
0.19
0.16
0.35
1
64
For an adequate construct validity, the AVE coefficient should at least reach the threshold
of 0.5, which would mean that the indicators explained at least half of the variance of the latent
construct. See Table 3.9 for results of these calculations for the model. Only one construct, IS does
not reach the threshold of 0.5, however, it is very close to it. This issue might be caused by the
proportion of lower than 0.7 factor loadings for indicators in the factor IS (see Table 3.8).
However, it is important to keep all the indicators for IS construct in order to keep the face validity
of it.
For good construct reliability, CR needs to be over 0.7, however, CR between 0.6 and 0.7
is also acceptable, especially in case when no other issues with the construct are indicated. High
construct reliability indicates that the measured variables represent the same latent construct. All
the CR coefficients calculated for latent constructs in the model surpass the desired threshold,
therefore, indicating good construct reliability (see Table 3.9).
The next step in construct validity analysis is the assessment of discriminant validity.
Statistically speaking, discriminant validity represents the extent to which a construct is distinctive
from other constructs. In other words, if the discriminant validity is proven, the conclusion can be
made that within the MM, a construct captures the phenomena other constructs do not. The analysis
of discriminant validity usually starts with the examination of inter-construct covariances, or
correlations if standardized measures are considered. The rule of thumb is that AVE for each latent
construct should exceed the squared interconstruct correlations associated with it. The information
on correlations and squared correlations can be addressed in the Table 3.9.
After the assessment of discriminant validity, nomological validity of constructs was
evaluated. Nomological validity of latent constructs means that they correlate with each other in a
meaningful way. Refer to Table 3.10 for inter-construct correlations.
Table 3.10. Initial CFA: Inter-construct correlations
ES
ES
IS
RC
T
LTO
IS
RC
T
LTO
1
0.39***
1
0.23* 0.59***
1
0.41*** 0.57*** 0.46***
1
0.45*** 0.44*** 0.40*** 0.59***
1
* significantly different from zero at the 0.05 level (two-tailed)
*** significantly different from zero at the 0.001 level (two-tailed)
Most of the correlations between constructs are statistically significant at 0.001
significance level. Not only the expected correlations between constructs are significant, the
65
correlations also have an expected sign, supported by the theory. According to Field (2005), strong
correlations exceed 0.5 (between LTO and T, between T and IS, between RC and IS), moderate
correlations are between 0.3 and 0.5 (between ES and IS, T and ES, LTO and ES, LTO and IS, RC
and T, LTO and RC). Correlations of 0.10 and below are considered to be low. No correlations in
the Table 3.10 are lower than 0.10, however, correlation between RC and IS is 0.23, which is
between 0.10 and 0.30, so it can be considered as weak-moderate.
The last piece of construct validity analysis is related to their face validity. Face validity or
constructs is based on the fact that all the measured variables were adapted from previous studies
and adapted to suit the needs of a research. Moreover, as a test survey launch has proven,
respondents were able to comprehend the logic of a questionnaire overall and of each question in
particular.
Some issues with MM specifications have already been discovered. However, a more
thorough analysis might be required to improve the MM. There are two additional methods for
this, apart from the analysis of path estimates, or standardized factor loadings, that are widely used
in statistical analysis. These are standardized residuals (SR) analysis and modification indices (MI)
analysis.
Standardized residuals are “computed for every covariance and variance in the observed
covariance matrix” and their number corresponds to the number of unique elements in the
covariance matrix (Hair, 2010). Standardized residuals that have values between |2.5| and |4.0|
generally require special consideration, but rarely require that an indicator should be eliminated
from the measurement model. However, standardized residuals greater than |4.0| usually indicate
some measurement issues to be treated as standardized residuals in this case show significant
differences between the observed and estimated covariance terms, therefore, indicating poor fit.
Therefore, the matrix of standardized residuals was analyzed to find standardized residuals
that would indicate the presence of model fit problems. None of the residuals in the matrix
exceeded |4.0|, and only two standardized residual absolute value fell between 2.5 and 4.0:
between RC2 and IS5 (SR = 3.138), and RC2 and T4 (SR = -2.405). However, these values are
not critical and to not require special treatment given the absence of other important problems with
constructs.
As a next step, modification indices were considered. The most problematic MIs in the
model correspond to 4 measured variables: T2, T4, RC1, RC2. The summary of modification
indices are presented in the Appendix 5. Therefore, to avoid potential problems with further model
interpretation, these doubtful variables were deleted from the model.
66
The model fit has improved, but did not reach desirable model fit indicators (e.g., CFI of
at least 0.9 for an adequate model fit). Therefore, MI for residuals within the same factor were also
assessed for potential problems similarly to what was done for cross-loadings. Some of them were
treated by drawing covariances between problematic error terms within the same latent construct
one by one starting from the most problematic ones. The covariances were drawn for errors of the
following indicators: T3 and T5, LTO1 and LTO2, IS1 and IS3, RC5 and RC6.
Final confirmatory factor analysis. The final CFA serves the purpose to ensure that MM
respecifications have lead to the improvement of MM fit, validity and reliability of constructs.
Overall, the procedure of the final CFA is conducted following the logic applied to the initial CFA.
The respecified CFA can be found in Appendix 6.
The new validity assessment procedure has to be held in order to ensure the new MM
overall fit along with construct validity.
The final MM overall fit assessment follows the same procedure described in the procedure
in paragraph 3.2.2. SPSS Amos output is reported in the Appendix 6. MM comparisons are
introduced in the Table 3.7.
Table 3.11. Measurement Models Comparison: Initial and Final CFA
Expected*
𝜒2
p < 0.05
𝜒 2 normed <2.0 – good fit
2.0-5.0 –
acceptable fit
Initial CFA
Final CFA
728.409 (p = 0.000)
df = 314
2.32
394.737 (p = 0.000)
df = 216
1.82
CFI
> 0.95 great
> 0.90 moderate
> 0.80 sometimes
acceptable
0.853
0.919
RMSEA
< .05 good
0.05 - 0.10
moderate
> 0.10 bad
0.087
90 percent confidence
interval RMSEA =
(0.079; 0.095)
0.069
90 percent
confidence interval
RMSEA = (0.058;
0.079)
*Source: (Hair, 2010; Van Dijk, 2014)
After MM respecifications that followed the initial CFA, the measurement model suggests
good overall fit. CFI, RMSEA, 𝜒 2 and 𝜒 2 normed have improved significantly.
67
Construct validity starts with convergent validity test and the analysis of standardized
factor loadings. All the loading in the final CFA prove to be statistically significant at a 0.001
level, however, one indicator, IS5 has a factor loading of 0.491, which is 0.09 points lower than
the desired 0.5 threshold. Nevertheless, it is highly desired to keep the indicator within the model
in order to keep construct integrity and face validity.
After the respecification, AVE and CR have improved. AVE coefficient for IS has
improved a little from 0.48 to 0.49, falling just 0.1 below the AVE threshold. Discriminant validity
was also confirmed for all the constructs. Refer to Table 3.12 for the results of these calculations.
Table 3.12. Final CFA: Construct validity and reliability
Construct
AVE
ES
IS
RC
T
LTO
CR
DV
ES IS
RC
T
0.85
1
0.15
0.05
0.83
1
0.36
0.71
1
0.88
0.90
0.59
0.49
0.50
0.64
0.60
0.16
0.36
0.25
1
LTO
0.20
0.20
0.13
0.36
1
As a next step, nomological validity was checked by addressing the correlations matrix for
constructs (see Table 3.13). Correlations are significant and meaningful, therefore, the
nomological validity can be argued. A slight increase in correlations between IS and RC, IS and
T, RC and T, T and LTO can be spotted, however, correlations between RC and ES, ES and T,
LTO and ES, LTO and RC have decreased slightly (see Table 3.10 for comparison).
Table 3.13. Final CFA: Inter-construct correlations
ES
ES
IS
RC
T
LTO
IS
RC
T
LTO
1
0.39***
1
0.22* 0.60***
1
0.40*** 0.60*** 0.50***
1
0.44*** 0.44*** 0.36*** 0.60***
1
* significantly different from zero at the 0.05 level (two-tailed)
*** significantly different from zero at the 0.001 level (two-tailed)
3.3 Structural model specification and assessment
After conducting CFA, the SM can be put forward for the analysis. The constructs used in
the scheme have proven their overall validity and reliability during the CFA procedure, therefore,
can be considered as a base for further SM analysis.
68
In principle, SEM represents a combination of linear equations that are used to test causal
relationships between latent constructs (Hair et al., 2010). As a final result, SEM is used to identify
to which extent the theoretically developed model fits observed data in the sample. The main
difference between CFA and SEM is that in SEM the attention is shifted to relationships between
latent constructs rather than the relationships between indicators and latent constructs.
Figure 3.1 provides a graphical representation of a SM, and matches the conceptual model.
In the figure, only causal relationships between latent constructs are shown, measured variables
are omitted for convenience of the reader. The number of latent constructs equals 5, each of them
is measured by at least 3 indicators. The overall number of indicators equals 23, which means the
model is over-identified.
εes
External
stability
Long-term
orientation
εlto
γ1
γ5
Trust
Internal
stability
β1
εis
Resource
complementarity
εt
Figure 3.1. Structural Model of Strategic Alliance Stability Factors
In the Figure 3.1 path coefficients 𝛽𝑖 , where i = 1, 2, 3, represent path coefficients on paths
stemming from the exogenous variable Resource complementarity. Path coefficients 𝛾𝑗 , where j =
1, 2, 3, 4, 5, evaluate causal relationships stemming from endogenous constructs (Trust, Long-term
orientation, External stability). The same path diagram can be represented by a set of equations
below:
69
𝐸𝑆 = 𝛾3 𝐿𝑇𝑂 + 𝛽2 𝑅𝐶 + 𝜀1
𝐼𝑆 = 𝛾5 𝐸𝑆 + 𝛾4 𝐿𝑇𝑂 + 𝛾2 𝑇 + 𝛽3 𝑅𝐶 + 𝜀2
{
𝐿𝑇𝑂 = 𝛾1 𝑇 + 𝜀3
𝑇 = 𝛽1 𝑅𝐶 + 𝜀4
(3),
where ES – External stability, IS – Internal stability, LTO – Long-term orientation, T –
Trust, RC – Resource complementarity, 𝛾𝑗 , 𝑗 𝜖 [1; 5] – path coefficients, denoting the effect of
endogenous construct on another endogenous construct, 𝜀𝑛 , 𝑛 𝜖 [1; 4] – error terms for endogenous
constructs, 𝛽𝑖 , 𝑖 𝜖 [1; 3] - path coefficients, denoting the effect of exogenous construct on
endogenous construct.
Based on the hypotheses that have been formulated (see sub-chapter 2.2), the model
includes 4 endogenous and 1 exogenous constructs.
Factor loadings for the SM are treated as not fixed, as suggested by most scholars to be an
appropriate treatment of factor loadings in SEM (Hair, 2010). Small fluctuations (less than 0.05)
between the loadings value after CFA and after SEM are expected and treated as normal.
The baseline for assessing structural model fit is the measurement model (MM) fit obtained
as a result of the CFA. In general, SM fit cannot be better than the MM fit (Hair, 2010), this
baseline can be used to identify suspicious results. At this stage, it is proven that the MM is valid
based on the results of the CFA, however, the need for assessment of relationships between
constructs that are driven by theoretical assumptions (see sub-chapter 2.1) require closer attention.
The procedure of the SM assessment is similar to the series of steps performed during the
CFA. Therefore, structural model assessment starts with the estimation of the overall model fit.
Following the logic, explained in sub-chapter 2.1, it was decided to test a series of SM in order to
better differentiate between direct and indirect effects in the final model (see Figure 2.1, Figure
3.1). The summary of the overall SM fit test is presented in the Table 3.14 below.
70
Table 3.14. Measurement and structural models comparison
Model Fit Indices
Expected*
𝜒2
p < 0.05
<2.0 –
good fit
2.0-5.0 –
acceptable
fit
CFI
> 0.95
great
> 0.90
moderate
> 0.80
sometimes
acceptable
RMSEA < .05
good
0.05 0.10
moderate
> 0.10
bad
PNFI
N/A
𝜒2
normed
Final
CFA
1.82
420.234
(p=0.000)
df = 219
1.92
415.471
(p=0.000)
df=218
1.91
SM
(ESIS,
RC T)
457.317
(p=0.000)
df=219
2.10
0.919
0.911
0.912
0.894
0.912
0.069
90 percent
confidence
interval
RMSEA =
(0.058;
0.079)
N/A
0.073
90 percent
confidence
interval
RMSEA =
(0.062;
0.083)
0.720
0.072
90 percent
confidence
interval
RMSEA =
(0.062;
0.083)
0.719
0.079
90 percent
confidence
interval
RMSEA =
(0.069;
0.089)
0.707
0.072
90 percent
confidence
interval
RMSEA =
(0.062;
0.083)
0.721
394
df = 216
SM (direct)
SM
(ESIS)
SM (ESIS,
RCT,
TLTO)
417.108
(p = 0.000)
df = 219
1.90
*Source: (Hair, 2010; Van Dijk, 2014)
Table 3.14 compares 5 models: final MM tested with CFA and 4 SMs, tested in a SEM
format (see Figures 2.2-2.4 for interim conceptual models). The final CFA is presented as a
benchmark for further SM assessment as the results for an overall model fit cannot usually be
better than those for the SM (Hair, 2010). The 4 structural models were introduced for comparison
among them in order to make a deeper analysis of relationships between constructs, which requires
the model to adequately fit the empirical data.
As can be concluded from the Table 3.14, the two best models out of 4 is the model where
only the direct relationships between SAS factors (LTO, T, RC) and SAS components (ES, IS) are
considered, and the last model where the full set of 8 hypotheses is reflected. Given the testing
result, the last model is comparable to the first one, however, the PNFI (parsimonious normed-fit)
71
index, used for model comparison is slightly better than for the last model as well as other
coefficients (𝜒 2 , 𝜒 2 normed, RMSEA). PNFI is a goodness-of-fit index, ranging from 0 to 1,
therefore, the higher the value of it, the better the model fits the data (Mulaik et al, 1989).
Summarizing on the Table 3.14, it can be concluded that the model with a full set of
hypotheses is overall reflective of the data used for analysis. For the Amos output with factor
loadings, see Appendix 7.
Following the logic of CFA, construct validity should be assessed to ensure that the results
of the SM are reliable. AVE, CR were assessed for the final model as it is the model that
incorporates all the hypotheses meant to be tested (see Table 3.15 for results). DV is not assessed
as all the covariances among constructs in the model are substituted with causal relationships.
Overall, results are consistent with MM assessment with CFA and indicate adequate validity and
reliability.
Table 3.15. AVE, CR for SM
Construct
ES
IS
RC
T
LTO
AVE
0.59
0.49
0.60
0.64
0.60
CR
0.85
0.82
0.77
0.88
0.90
3.4 Modeling results and analysis
Modeling results show that most, but not all of the specified relationships are statistically
significant. However, only 2 relationships out of 8 have demonstrated statistical insignificance,
therefore, it can be claimed that, overall, theoretical model adequately fits the data. Table 3.16
summarizes information on hypotheses, structural relationships, their magnitude and statistical
significance.
72
Table 3.16. Modeling results. Path coefficients and their significance
Hypothesis
H1
H2
H3
H4
H5
H6
H7
H8
Structural relationship
Long-term orientation External stability
Long-term orientation Internal stability
Trust Internal stability
Trust Long-term orientation
Resource complementarity External stability
Resource complementarity Internal stability
Resource complementarity Trust
External stability Internal stability
Estimate
0.433***
0.025 (ns)
0.355**
0.609***
0.056 (ns)
0.377***
0.450***
0.171*
ns – not significant
*significantly different from zero at the 0,05 level (two-tailed)
**significantly different from zero at the 0,01 level (two-tailed)
***significantly different from zero at the 0,001 level (two-tailed)
All the significant effects of SAS determinants on both SAS components correspond to
theoretical assumptions. SEM has shown that SAS determinants have different effects on the
components of SAS. More specifically, Trust and Resource complementarity have a direct positive
effect on Internal SAS, the effect of Resource Complementarity on External stability is indirect
and minor (see Table 3.18), while Long-term orientation is the only significant and direct
determinant of External stability. These results partially correspond to findings revealed by
previous studies. Speaking of Trust and Resource complementarity effects on Internal stability,
results of an empirical test go in line with findings by Deitz et al (2010) that find a direct and
significant effect of Resource complementarity on the intent to stay within a joint venture as well
as partner commitment. It has also been proven by the same authors that Trust is positively
associated with commitment. However, authors find marginal support for the causal relationship
between Trust and commitment. Clearly, there is a difference in stability conceptualization chosen
in this paper and in the paper by Deitz et al (2010). As it has been discussed in sub-chapter 1.5,
the term “commitment” used by authors is similar to “motivational stability” term used in this
paper, while “intern to remain in a joint venture” better matches the concept of “strategic stability”
(see the definition in sub-chapter 1.4). Therefore, there is a rationale to assume that different
components within Internal stability (see Figure 1.1) are affected differently by Trust, however,
Internal stability overall is positively affected by it.
Contrary to the expected results predicted by theory, Resource complementarity did not
manifest a significant effect on External stability, which contradicts the assumptions of R-A theory
used for factor selection (Lambe, Spekman and Hunt, 2002; Hunt, Lambe and Wittmann, 2002).
This finding might indicate that in case multiple SAS components are taken into consideration, the
effect of Resource complementarity on Internal stability prevails. At the same time, regarding
73
External and Internal stability components in separate models is not logical as it is required that
both components are present for an alliance to be overall stable (Zenkevich, Koroleva, Mamedova,
2014a,b).
The effect of Long-term orientation has proven to be positive and significant in relation to
External stability, which supports theoretical assumptions put forward in sub-chapter 2.1. At the
same time, the effect of Long-term orientation on Internal stability has been found insignificant
in the examined model. Contrary to this result, López‐Navarro, Callarisa‐Fiol & Moliner‐Tena,
(2013) find a significant and positive relationship between Long-term orientation and partner
commitment in export joint ventures. The discrepancy in finding might result, firstly, from
difference in sampling. In particular, the current study addressed all alliance types, while the
abovementioned research focuses exclusively on export JVs. Secondly, the discrepancy in findings
might stem from differences in conceptualization of the outcome variable. As it has been
mentioned for (Deitz et al, 2010), the term “commitment” is most closely related to “motivational
stability”, which constitutes one part of Internal stability. Therefore, there is an implication for
further research that Long-term orientation can be regarded as a factor of one of the Internal
stability components, e.g., motivational stability. Thirdly, it can be claimed, that the effect of Longterm orientation on External stability prevails in the model, and makes the effect of Long-term
orientation on Internal stability statistically insignificant. Although, as it was already mentioned,
considering External and Internal stability as outcome variables in separate models does not make
sense.
Speaking of the relationships among SAS factors hypothesized in sub-chapter 2.1, the study
has found support for both. López‐Navarro, Callarisa‐Fiol & Moliner‐Tena, (2013) have found
that Resource complementarity is positively and significantly associated with Trust. This result
corresponds to the findings on the association between Resource complementarity and Trust
demonstrated in the current paper (see Table 3.17). Moreover, Deitz et al (2010) have found that
there is a partial mediation by Trust between Resource complementarity and intent to remain in an
alliance. The same result has been obtained for Trust, Resource complementarity and Internal
stability examined in the current paper (see Table 3.18). Moreover, López‐Navarro, Callarisa‐Fiol
& Moliner‐Tena, (2013) find a significant and positive relationship between Trust and Long-term
orientation, which corresponds to the findings in this paper (see Table 3.17).
The positive and significant effect of External stability on Internal stability has been
identified, as predicted by theory. This finding also corresponds to results provided in the paper
by Fu, Lin, Sun (2013) who have found a positive and significant effect of the increase in economic
74
results of alliance activities, namely, the income increase, on SAS. However, in the current study,
the effect of External stability on Internal stability is not as strong as the influence of other
determinants on particular components of stability.
For research hypotheses testing summary, refer to the Table 3.17.
Table 3.17. Hypotheses test results
Hyp.
H1
H2
H3
H4
H5
H6
H7
H8
Hypothesis formulation
Long-term orientation is positively associated with
external stability of a strategic alliance
Long-term orientation is positively associated with
internal stability of a strategic alliance
Trust is positively associated with internal stability of a
strategic alliance
Trust is positively associated with long-term orientation
in a strategic alliance
Resource complementarity is positively associated with
external stability of a strategic alliance
Resource complementarity is positively associated with
internal stability of a strategic alliance
Resource complementarity is positively associated with
partners’ trust
External stability is positively associated with internal
stability
St.est.
Result
0.433***
Supported
0.025 (ns)
N/A
0.355**
Supported
0.609***
Supported
0.056 (ns)
N/A
0.377***
Supported
0.450***
Supported
0.171*
Supported
ns – not significant
*significantly different from zero at the 0,05 level (two-tailed)
**significantly different from zero at the 0,01 level (two-tailed)
***significantly different from zero at the 0,001 level (two-tailed)
Structural equation modeling results for the model are depicted in the Figure 3.3 to make
a set of interrelations between constructs more comprehensive.
75
External stability
Long-term
orientation
0.609***
0.171*
Trust
Internal stability
0.448***
Resource
complementarity
Figure 3.2. SEM final results. Dependence paths
Considering the fact that direct and indirect effect of each SAS determinant can be
identified, direct and indirect effect for each construct have been calculated in relation to ES and
IS based on the data used for analysis. To differentiate among different effects, 4 models have
been tested (each following model includes all the paths of the previous model plus one new path,
see sub-chapter 2.1): SM with direct effects between SAS factors and SAS components; SM with
an additional path External stability Internal stability; SM with an additional path (Resource
complementarityTrust); SM with an additional path (TrustLong-term orientation). Next, the
analysis of direct and indirect effects has been made based on significant paths. See Table 3.18 for
the reference.
By comparing direct effects in all 4 models in Table 3.18, it can be argued that all the path
coefficients estimates remain approximately the same compared in models with different numbers
of causal relationships. This implies consistency in results for all the models.
76
Table 3.18. Direct and indirect effects. SMs comparison
Model
SEM
(direct
effects)
SEM (ES
IS)
SM (ES
IS, RC
T)
SM (ES
IS, RC
T, T
LTO)
Path(s)
Direct/Indirect
(significant)
significant effect
LTO ES 0.432***
LTO has a positive direct effect on ES
T IS
0. 377***
T has a positive direct effect on IS
RC IS
0.379***
RC has a positive direct effect on IS
LTO ES
T IS
RC IS
ES IS
LTO ES
IS
LTO ES
T IS
RC IS
ES IS
RC T
RC T
IS
LTO ES
IS
LTO ES
T IS
RC IS
ES IS
RC T
RC T
IS
LTO ES
IS
T LTO
T LTO
ES
T LTO
ES IS
RC T
LTO ES
RC T
LTO ES
IS
RC T
LTO
0.435***
0.358***
0.379***
0.173*
0.435 × 0.173 = 0.075
LTO has a positive direct effect on ES
T has a positive direct effect on IS
RC has a positive direct effect on IS
ES has a positive direct effect on IS
LTO has a minor indirect positive effect
on IS through ES
LTO has a positive direct effect on ES
T has a positive direct effect on IS
RC has a positive direct effect on IS
ES has a positive direct effect on IS
RC has a positive direct effect on T
RC has a positive indirect effect on T
T is a mediator for RC and IS
LTO has a minor indirect positive effect
on IS through ES
LTO has a positive direct effect on ES
T has a positive direct effect on IS
RC has a positive direct effect on IS
ES has a positive direct effect on IS
RC has a positive direct effect on T
RC has a positive indirect effect on T
T is a partial mediator for RC and IS
LTO has a minor indirect positive effect
on IS through ES
T has a positive direct effect on LTO
T has an positive indirect effect on ES
through LTO
T has a minor indirect effect on IS
through the path LTO ES
RC has an positive indirect effect on ES
through the path T LTO
RC has a minor positive indirect effect on
IS through the chain T LTO ES
0.420***
0.339***
0.390***
0.173*
0.478***
0.478 × 0.339 = 0.162
0.420 × 0.173 = 0.073
0.433***
0.355***
0.377***
0.171*
0.450***
0.450 × 0.355 = 0.160
0.433 × 0.171 = 0.074
0.609***
0.609 × 0.433 = 0.264
0.609 × 0.433 × 0.171
= 0.045
0.450 × 0.609 × 0.433
= 0.119
0.450 × 0.609 × 0.433
× 0.171 = 0.020
0.450 × 0.609 = 0.274
Conclusion
RC has a positive indirect effect on LTO
through T
77
In the final model with the maximum possible number of effects (hypotheses 1-8), derived
from theory and tested with SEM, there are a total of 6 direct and 7 indirect effects that can be
identified on a basis of significant path coefficients. Out of indirect effects, one is the indirect
relationship between SAS factors (RC T LTO), one is a mediation effect (T is a partial
mediator for RC and IS). Full mediation implies that in a presence of a mediator a direct
relationship between dependent and independent constructs becomes insignificant, while partial
mediation implies that a direct effect between the independent and dependent variables remains
significant after the mediator is introduced into the relationship (Hair, 2010). 6 indirect effects out
of 7 are on ES or IS.
Other effects enable relationships between SAS factors and SAS components, however, 3
out of these effects are minor. The common trait of these 3 effects is that they are indirect and
involve the path ES IS, namely, LTO ES IS, T LTO ES IS, RC T LTO
ES IS. Therefore, it can be concluded that the main effects on IS are created through shorter
path chains (both, direct and indirect), namely, directly through RC IS, T IS, ES IS, and
indirectly through RC T IS. In turn, ES is most significantly influenced by LTO, and through
LTO by both T and RC.
3.5 Chapter 3 concluding remarks
In Chapter 3, the empirical part of the study has been described in details. The two-step
structural equation modeling procedure has been performed based on the conceptual model derived
from theory using the data collected, as described in Chapter 2 (for reference, see sub-chapter 2.1
and subchapter 2.2 respectively).
The first step of SEM involves measurement model assessment (overall model fit, construct
validity and reliability) via confirmatory factor analysis and AVE, CR, DV calculation. The
measurement model itself was developed from other studies by adapting the measures for SA
context. After the initial model assessment, model respecifications were required. After the
respecification, the measurement model has shown an adequate fit to empirical data. Factor
reliability and validity have also proven to be adequate.
The second step of SEM requires structural equation model test, which was conducted in
Chapter 3. The test for SM follows the same steps done for MM assessment during the CFA. In
order to differentiate indirect and direct effects and conduct a deeper analysis of path coefficients
and effects within the model, 4 SM tests have been conducted: for SM only with direct
relationships between SAS factors and SAS components; with an added path ES IS, with an
78
added path RC T; with an added path T LTO. On a basis of this analysis, 6 direct and 7
indirect effects on External and Internal stability have been identified.
In case SAS was a homogenous phenomena, it would be expected that all the relationships
between SAS factors and SAS different components would be the same. However, empirical
results have shown that this is not so. Speaking of direct relationships, the significant determinant
of External SAS appeared to be partners’ Long-term orientation, while Trust, Resource
complementarity and External stability directly influence Internal stability. Moreover, External
stability influences Internal stability to a lesser extent than Trust and Resource complementarity.
Speaking of indirect effects, Trust appeared to play a partial mediator role between Resource
complementarity and Internal stability channeling a part of the effect of the former on the latter.
Otherwise, effects of Resource complementarity, Trust and Long-term orientation through
External stability are negligible. At the same time, the indirect effects of Trust and Resource
complementarity on External stability through the sequence of relationships RC T LTO
ES are worth considering.
79
CONCLUSION AND IMPLICATIONS
Research goal and objectives. During this study, the research goal, formulated in the
Introduction, was reached through covering research objectives, also stated in the Introduction
part. During the research, a set of relationships between strategic alliance stability and interorganizational SAS factors were identified. This goal was attained by (1) defining the term of
strategic alliance stability as a result of extensive literature analysis, then (2) developing a
conceptual model of strategic alliance stability factors and (3) conducting an empirical study to
test relationships between SAS factors and stability components in order to make conclusions
about these relationships.
Answers to research questions. The study was aimed at answering two research questions
in order to fulfill research objectives. The empirical study has provided answers to them.
RQ1: What are the relationships between strategic alliance stability inter-organizational
factors and different components of strategic alliance stability?
The research has provided insights for the issue of SAS factors. The study has differentiated
between two stability components: external and internal stability, which corresponds to the game
theory strategic alliance conceptualization and seems to be an all-inclusive approach. A number
of SAS determinants were chosen from previous academic studies, both conceptual and empirical,
for further analysis. The approach chosen to define SAS factors was R-A theory as the most
integrative approach to SAS factors. Factors included trust and long-term orientation and resource
complementarity.
Speaking of direct relationships, results for external stability have shown that long-term
orientation of partners plays an important role in defining external SAS, while resource
complementarity is an insignificant factor. However, results for internal SAS differ: long-term
orientation becomes an insignificant factor, while trust and resource complementarity appear have
a significant and positive effect on internal stability. This difference in results contradicts to
theoretically hypothesized relationships and is likely to occur in the presence of both internal and
external stability present the model, as each additional construct might change the set of significant
relationships (Hair, 2010). At the same time, the presence if both, internal and external stability,
has to be considered in the SAS factors model as all the stability components have to be present in
order for an alliance to be overall stable (Zenkevich, Koroleva, Mamedova, 2014a).
Moreover, there is a significant causal relationship between external stability and internal
stability. Hence, it was empirically proven that SAS components are not independent from each
80
other. This finding is especially important in the light of SAS concept understanding and further
practical applications, which are hindered by the concept current misspecifications, as described
in Chapter 1.
RQ2: What are potential indirect effects of strategic alliance stability factors on different
components of strategic alliance stability?
The study has shown that some of the SAS factors predicted by other scholars do not always
have a direct effect on all the components of SAS. For example, resource complementarity does
not directly influence external stability of SAs as well as long-term orientation does not directly
influence internal SAS.
However, the set of relationships between different stability factors appeared to correspond
to the expected ones and were found to be statistically significant. Therefore, a set of indirect
effects was differentiated. As for external stability, resource complementarity and trust influence
it indirectly through a set of causal relationships (resource complementarity trust long-term
orientation external stability). As for internal stability, resource complementarity being the most
influential factor, influences it not only directly, but also indirectly through trust, where partial
trust is a mediator for resource complementarity. However, the sequence of relationships resource
complementarity trust long-term orientation external stability internal stability
generates minor effects on internal stability for resource complementarity, trust and long-term
orientation. At the same time, the set of relationships among constructs in the chain resource
complementarity trust long-term orientation has proven its significance.
Theoretical contributions. As it has been mentioned previously and thoroughly discussed
in Chapter 1, the concept of SAS remains largely unidentified and fragmented. Therefore, after a
careful examination of academic literature, a dynamic and all-inclusive game theoretic approach
to SAS definition and conceptualization was adopted from (Zenkevich, Koroleva, Mamedova,
2014a). Hence, SAS was regarded as a multi-dimensional phenomena (see Figure 1.1, Chapter 1).
In particular, 2 stability components were considered in the focus of this study: internal stability
and external stability of strategic alliances (see sub-chapter 1.4 for explanation).
Game theory approach to SAS conceptualization was merged with resource-advantage
theory to examine the influence of particular inter-organizational factors on SAS. Based on recent
studies by López‐Navarro, Callarisa‐Fiol and Moliner‐Tena (2013) and Deitz et.al. (2010), the
following SAS factors were chosen for further examination: resource complementarity, trust, long-
81
term orientation. As a result, the following conceptual model was put forward (see Figure C.11
below).
External stability
Long-term
orientation
+
+
Trust
Internal stability
+
Resource
complementarity
Note: the sign (+) denotes a positive causal relationship among constructs
Figure C.1. Conceptual model: strategic alliance stability factors
The model was tested by firstly considering direct relationships between SAS factors and
SAS components, gradually introducing new relationships into the model and testing 3 interim and
one final model (see sub-chapter 2.1). Figure C.22 represents results of a final empirical test.
Results provided in this paper suggest that in case SAS stability is viewed as a multi-dimensional
construct, different components of it are significantly determined by different factors (see Figure
C.2). Therefore, an important theoretical contribution is that SAS is not a homogenous concept as
it has been empirically proven that different parts of it are influenced by different factors.
Moreover, given a significant and positive relationship between external and internal
stability, it can be claimed that strategic alliance stability components are not independent.
1
2
Corresponds to the Figure 2.1.
Corresponds to the Figure 3.2
82
External stability
Long-term
orientation
0.171*
0.609***
Trust
Internal stability
0.448***
Resource
complementarity
Note:
*significantly different from zero at the 0,05 level
**significantly different from zero at the 0,01 level
***significantly different from zero at the 0,001 level
Figure C.2. SEM final results. Dependence paths
Managerial implications. Given the fact that a more stable SA is likely to survive external
turbulences and experience greater economic success, reaching its strategic goals, it is important
to understand the mechanics behind SAS dynamics and use it for SAS management (Jiang, Li and
Gao, 2008).
While SAS can be assessed using game theory approach by interpreting financial data
along with inside expert estimations (Zenkevich, Koroleva, Mamedova, 2014b), SAS assessment
would be incomplete without SAS management. Theoretical results provided in the paper suggest
which inter-organizational factors could be altered in order to enhance external and internal
stability of strategic alliances, given the importance of either component for the overall alliance
stability.
Results provided in Figure C.2, suggest that direct determinants of internal SAS are trust
and resource complementarity, considering that the latter has a greater effect on internal stability.
Moreover, trust plays a mediating role in a relationship between resource complementarity and
internal stability by interference. The only factor in the model affects external stability directly,
which is long-term orientation. Contrary to expectations that scholars and management
83
practitioners might have, long-term orientation of partners does not directly and significantly affect
internal stability of strategic alliances as well as resource complementarity does not directly affect
external strategic alliance stability. It means that in practice managers who are willing to enhance
the overall stability should manage different SAS factors simultaneously in order to reach a higher
stability level.
While resources are often immobile and it might not be feasible to enhance resource
complementarity during the implementation stage of an alliance, it seems reasonable to enhance
trust among partners and pay closer attention to relationship management. This could include
building communication channels and facilitating communication overall, managing cultural
distance in terms of national, professional and organizational cultures, etc. (Elmuti, Kathawala,
2001). Then, given that resource complementarity is one of the criterion for partner selection in
many alliances, partners should pay close attention to resource complementarity as it does not only
play role at a formation stage of an alliance, but also affects SAS on the implementation stage.
Moreover, it has been found that the effect of the trend of economic results (external
stability) on internal stability is not as strong as the effect of such determinants as trust and resource
complementarity. Therefore, relational factors, often disregarded in strategic alliances (Agarwal,
Croson, Mahoney, 2010) should be subject to constant monitoring during the implementation
phase of an alliance.
LIMITATIONS AND FURTHER RESEARCH
The study is subject to some limitations that can be addressed further. The primary reason
for most of limitations in this study is scarcity of data and difficulties connected with data
collection. First, the research does not differentiate between different alliance types (e.g., equity,
non-equity) because strategic alliances are not easily accessible for the outsider from the point of
information collection, e.g., most alliances do not publish financial data and are restricted to
provide sensitive information (Jiang, Li and Gao, 2008).
Second, given sample characteristics, study results can be best generalized for micro and
small size European alliances, mainly in business service industry. However, some peculiarities
can be found for larger alliances and alliances that operate in different fields. Therefore, results
provided in the current study, should be applied in practice with a careful consideration of
organizational and industrial conditions that an alliance operates with. The same issue can also be
seen as a focus for further examination.
84
Third, given the fact that internal SAS consists of 3 components (dynamic, strategic,
motivational stability; see Figure 1.1), an additional study on interrelationships among them and
on their determinants can be considered further. Based on the mismatch between obtained results,
expected findings and results provided in other empirical papers, there is a rationale to assume
that, e.g., long-term orientation that did not exhibit a significant effect on internal stability overall,
might have an effect on one of its components, most likely, on motivational stability. Similar
conclusions can be made on the effects of resource complementarity on strategic and motivational
stability, which might be different in each case.
Fourth, given current tools for SAS assessment (Zenkevich, Koroleva, Mamedova, 2014b),
it is now possible to make conclusions on the presence of strategic alliance stability, however,
stability level is still hardly quantifiable. Therefore, there is a vast potential for researchers to
address the issue of a quantitative stability level assessment (e.g., developing stability indices).
85
LIST OF REFERENCES
Achrol, R. S. (1991). Evolution of the marketing organization: new forms for turbulent
environments. The Journal of Marketing, 77-93.
Adobor, H. (2011). Alliances as collaborative regimes: An institutional based explanation of
interfirm collaboration. Competitiveness Review: An International Business Journal, 21(1), 6688.
Agarwal, R., Croson, R., and Mahoney, J. T. (2010). The role of incentives and communication in
strategic alliances: an experimental investigation. Strategic Management Journal, 31(4), 413437.
Alderson, W. (1957). Marketing behavior and executive action.
Anderson, E., and Weitz, B. (1989). Determinants of continuity in conventional industrial channel
dyads. Marketing science, 8(4), 310-323.
Anderson, J. C., and Gerbing, D. W. (1988). Structural equation modeling in practice: A review
and recommended two-step approach. Psychological bulletin, 103(3), 411.
Arnett, D. B., and Hunt, S. D. (2002). Competitive irrationality: The influence of moral
philosophy. Business Ethics Quarterly, 12(03), 279-303.
Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.
Bamford, J., and Ernst, D. (2005). Governing joint ventures. McKinsey Quarterly, 15, 12-16.
Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of management,
17(1), 99-120.
Barney, J. B. (1992). Integrating organizational behavior and strategy formulation research: A
resource based analysis. Advances in strategic management, 8(1), 39-61.
Barringer, B. R., and Harrison, J. S. (2000). Walking a tightrope: Creating value through
interorganizational relationships. Journal of management, 26(3), 367-403.
Beamish, P. W., and Inkpen, A. C. (1995). Keeping international joint ventures stable and
profitable. Long Range Planning, 28(3), 2-36.
Berg, S., and Friedman, P. (1978). Joint ventures in American industry: an overview. Mergers and
Acquisitions, 13, 28-41.
86
Bidault, F., and Salgado, M. (2001). Stability and Complexity of Inter-Firm Co-operation:: The
case of Multi-Point Alliances. European Management Journal, 19(6), 619-628.
Bowman, E. H., and Hurry, D. (1993). Strategy through the option lens: An integrated view of
resource investments and the incremental-choice process. Academy of management review,
18(4), 760-782.
Boyle, S.E. (1968) An estimate of the number and size distribution of domestic joint subsidiaries.
Antitrust Law and Economics Review, 1, 81-92.
Brown, J. R., Dev, C. S., and Lee, D. J. (2000). Managing marketing channel opportunism: the
efficacy of alternative governance mechanisms. Journal of Marketing, 64(2), 51-65.
Buffenoir, E., and Bourdon, I. (2013). Managing Extended Organizations and Data Governance.
In Digital Enterprise Design and Management 2013 (pp. 135-145). Springer Berlin Heidelberg.
Caves, R. E., Crookell, H., and Killing, J. P. (1983). THE IMPERFECT MARKET FOR
TECHNOLOGY LICENSES*. Oxford Bulletin of Economics and statistics, 45(3), 249-267.
Christoffersen, J. (2013). A review of antecedents of international strategic alliance performance:
synthesized evidence and new directions for core constructs. International Journal of
Management Reviews, 15(1), 66-85.
Christoffersen, J., Plenborg, T., and Robson, M. J. (2014). Measures of strategic alliance
performance, classified and assessed. International Business Review, 23(3), 479-489.
Conner, K. R. (1991). A historical comparison of resource-based theory and five schools of
thought within industrial organization economics: do we have a new theory of the firm?.
Journal of management, 17(1), 121-154.
Cooper, D. R., and Schindler, P. S. (2006). Marketing research. New York: McGraw-Hill/Irwin.
Dalton, L., D'Netto, B., and Bhanugopan, R. (2015). Cultural diversity competencies of managers
in the Australian energy industry. The Journal of Developing Areas, 49(6), 387-394.
Das, T. K., and Teng, B. S. (1998). Between trust and control: Developing confidence in partner
cooperation in alliances. Academy of management review, 23(3), 491-512.
Das, T. K., and Teng, B. S. (2000). Instabilities of strategic alliances: An internal tensions
perspective. Organization science, 11(1), 77-101.
Das, T. K., and Rahman, N. (2001). Partner misbehaviour in strategic alliances: Guidelines for
effective deterrence. Journal of General Management, 27(1), 43-70.
87
Das, T. K., and Rahman, N. (2010). Determinants of partner opportunism in strategic alliances: a
conceptual framework. Journal of Business and Psychology, 25(1), 55-74.
Das, T. K., and Teng, B. S. (1998). Between trust and control: Developing confidence in partner
cooperation in alliances. Academy of management review, 23(3), 491-512.
Das, T. K., and Teng, B. S. (2001). Trust, control, and risk in strategic alliances: An integrated
framework. Organization studies, 22(2), 251-283.
Deitz, G. D., Tokman, M., Richey, R. G., and Morgan, R. M. (2010). Joint venture stability and
cooperation: Direct, indirect and contingent effects of resource complementarity and trust.
Industrial Marketing Management, 39(5), 862-873.
Delios, A., and Beamish, P. W. (2004). Joint venture performance revisited: Japanese foreign
subsidiaries worldwide. MIR: Management International Review, 69-91.
Douma, M. U., Bilderbeek, J., Idenburg, P. J., and Looise, J. K. (2000). Strategic alliances:
managing the dynamics of fit. Long Range Planning, 33(4), 579-598.
Doz, Y. L., and Hamel, G. (1998). Alliance advantage: The art of creating value through
partnering. Harvard Business Press.
Dussauge, P., and Garrette, B. (1995). Determinants of success in international strategic alliances:
Evidence from the global aerospace industry. Journal of International Business Studies, 505530.
Dwyer, F. R., Schurr, P. H., and Oh, S. (1987). Developing buyer-seller relationships. The Journal
of marketing, 11-27.
Dyer, J. H. (1996). Specialized supplier networks as a source of competitive advantage: Evidence
from the auto industry. Strategic management journal, 17(4), 271-291.
Dyer, J. H., and Hatch, N. W. (2006). Relation‐specific capabilities and barriers to knowledge
transfers: creating advantage through network relationships. Strategic management journal,
27(8), 701-719.
Dyer, J. H., and Singh, H. (1998). The relational view: Cooperative strategy and sources of
interorganizational competitive advantage. Academy of management review, 23(4), 660-679.
Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management
review, 14(1), 57-74.
88
Elmuti D, Kathawala Y. (2001). An overview of strategic alliances. Management decision. Apr
1;39(3):205-18.
Ernst, D., and Bamford, J. (2005). Your alliances are too stable. Harvard Business Review, 83(6),
133-141.
Ferreira, M. P., Storopoli, J. E., and Serra, F. R. (2014). Two decades of research on strategic
alliances: Analysis of citations, co-citations and themes researched. Revista de Administração
Contemporânea, 18(SPE), 109-133.
Foss, N. J., and Ishikawa, I. (2007). Towards a dynamic resource-based view: Insights from
Austrian capital and entrepreneurship theory. Organization Studies, 28(5), 749-772.
Franko, L. G. (1971). Joint venture divorce in the multinational company. The International
Executive, 13(4), 8-10.
Friedmann, W. G., and Kalmanoff, G. (1961). Joint international business ventures. The
International Executive, 3(3), 1-4.
Fu, S., Lin, J., and Sun, L. (2013). An empirical examination of the stability of the alliance of “a
company+ farmers” From the perspective of farmers. Chinese Management Studies, 7(3), 382402.
Ganesan, S. (1994). Determinants of long-term orientation in buyer-seller relationships. The
Journal of Marketing, 1-19.
Geringer, J. M. (1988). Joint venture partner selection: Strategies for developed countries. Praeger
Pub Text.
Geringer, J. M. (1991). Strategic determinants of partner selection criteria in international joint
ventures. Journal of international business studies, 41-62.
Gibbs, M. R., and Humphries, M. A. (2015). Enterprise Relationship Management: A Paradigm
for Alliance Success. Ashgate Publishing, Ltd.
Gill, J., and Butler, R. J. (2003). Managing instability in cross-cultural alliances. Long range
planning, 36(6), 543-563.
Glaister, K. W., and Buckley, P. J. (1998). Measures of performance in UK international alliances.
Organization Studies, 19(1), 89-118.
Gomes-Casseres, B. (1987). Joint venture instability: Is it a problem. Division of Research,
Harvard University.
89
Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness.
American journal of sociology, 481-510.
Griffith, D. A., Harvey, M. G., and Lusch, R. F. (2006). Social exchange in supply chain
relationships: The resulting benefits of procedural and distributive justice. Journal of
operations management, 24(2), 85-98.
Grover, V., and Malhotra, M. K. (2003). Transaction cost framework in operations and supply
chain management research: theory and measurement. Journal of Operations management,
21(4), 457-473.
Gulati, R., and Singh, H. (1998). The architecture of cooperation: Managing coordination costs
and appropriation concerns in strategic alliances. Administrative science quarterly, 781-814.
Gulati, R., Khanna, T., and Nohria, N. (1994). Unilateral commitments and the importance of
process in alliances. Sloan Management Review, 35(3), 61.
Hagedoorn, J. (1993). Understanding the rationale of strategic technology partnering:
Nterorganizational modes of cooperation and sectoral differences. Strategic management
journal, 14(5), 371-385.
Hair, J. F. (2010). Multivariate data analysis.
Hair, J. F., Ringle, C. M., and Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of
Marketing theory and Practice, 19(2), 139-152.
Hamel, G. (1991). Competition for competence and interpartner learning within international
strategic alliances. Strategic management journal, 12(S1), 83-103.
Heide, J. B., and Miner, A. S. (1992). The shadow of the future: Effects of anticipated interaction
and frequency of contact on buyer-seller cooperation. Academy of management journal, 35(2),
265-291.
Hennart, J. F. (1988). A transaction costs theory of equity joint ventures. Strategic management
journal, 9(4), 361-374.
Hennart, J. F., and Larimo, J. (1998). The impact of culture on the strategy of multinational
enterprises: does national origin affect ownership decisions?. Journal of International Business
Studies, 515-538.
Hennart, J. F., Kim, D. J., and Zeng, M. (1998). The impact of joint venture status on the longevity
of Japanese stakes in US manufacturing affiliates. Organization Science, 9(3), 382-395.
90
Holmstrom, B. (1982). Moral hazard in teams. The Bell Journal of Economics, 324-340.
Hong, J. I., Yu, H. U. A. N. G., and Zhichao, C. A. O. (2011). Research About Measures of
Enhancing Stability of Competitive Strategic Alliance. Chinese Business Review, 10(12).
Hoxie, R. F., Hoxie, L. B., and Fine, N. (1923). Trade unionism in the United States. D. Appleton
and Company.
Huang, Y. (2003). Selling China: Foreign direct investment during the reform era. Cambridge
University Press.
Humphries, A., and Gibbs, R. (2016). Enterprise Relationship Management: A Paradigm For
Alliance Success. Routledge.
Hunt, S. D. (1997). Competing through relationships: Grounding relationship marketing in
resource‐advantage theory. Journal of Marketing Management, 13(5), 431-445.
Hunt, S. D., and Arnett, D. B. (2001). Competition as an evolutionary process and antitrust policy.
Journal of Public Policy and Marketing, 20(1), 15-26.
Hunt, S. D., and Arnett, D. B. (2003). Resource-advantage theory and embeddedness: Explaining
RA theory’s explanatory success. Journal of Marketing Theory and Practice, 11(1), 1-17.
Hunt, S. D., and Lambe, C. J. (2000). Marketing’s contribution to business strategy: market
orientation, relationship marketing and resource‐advantage theory. International Journal of
Management Reviews, 2(1), 17-43.
Hunt, S. D., and Morgan, R. M. (1995). The resource-advantage theory of competition. Journal of
Management Inquiry, 4(4), 317-32.
Hunt, S. D., and Morgan, R. M. (1996). The Resource-Advantage Theory of Competition:
Dynamics, Path Dependencies, and Evolutionary Dimensions. Journal of Marketing, 60, 107114.
Hunt, S. D., and Morgan, R. M. (1997). Resource-advantage theory: a snake swallowing its tail or
a general theory of competition?. The Journal of Marketing, 74-82.
Hunt, S. D., and Arnett, D. B. (2003). Resource-advantage theory and embeddedness: Explaining
RA theory’s explanatory success. Journal of Marketing Theory and Practice, 11(1), 1-17.
Hunt, S. D., Lambe, C. J., and Wittmann, C. M. (2002). A theory and model of business alliance
success. Journal of Relationship Marketing, 1(1), 17-35.
91
Huo, B., Ye, Y., and Zhao, X. (2015). The impacts of trust and contracts on opportunism in the
3PL industry: The moderating role of demand uncertainty. International Journal of Production
Economics, 170, 160-170.
Hussey, J., and Hussey, R. (1997). Business research. Hampshire: Palgrave.
Inkpen, A. C., and Beamish, P. W. (1997). Knowledge, bargaining power, and the instability of
international joint ventures. Academy of management review, 22(1), 177-202.
Isidor, R., Schwens, C., Hornung, F., and Kabst, R. (2015). The impact of structural and attitudinal
antecedents on the instability of international joint ventures: The mediating role of
asymmetrical changes in commitment. International Business Review, 24(2), 298-310.
Jap, S. D. (1999). Pie-expansion efforts: collaboration processes in buyer-supplier relationships.
Journal of marketing Research, 461-475.
Jensen, M. C., and Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs
and ownership structure. Journal of financial economics, 3(4), 305-360.
Jiang, X., Li, Y., and Gao, S. (2008). The stability of strategic alliances: Characteristics, factors
and stages. Journal of International Management, 14(2), 173-189.
Geringer, J.M., Hebert, L. (1991). Measuring joint venture performance. Journal of International
Business Studies, 22 (2), pp. 249–263.
Kale, P., Dyer, J. H., and Singh, H. (2002). Alliance capability, stock market response, and long‐
term alliance success: the role of the alliance function. Strategic Management Journal, 23(8),
747-767.
Kang, I., Han, S., Lee, J., and Olfman, L. (2016). An evolutionary perspective of opportunism in
high-technology alliance: the evidence from South Korean companies. Asia Pacific Business
Review, 22(2), 238-261.
Kelley, H. H., and Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence (p.
341). New York: Wiley.
Khanna, T., Gulati, R., and Nohria, N. (1998). The dynamics of learning alliances: Competition,
cooperation, and relative scope. Strategic management journal, 19(3), 193-210.
Killing, J. P. (1982). How to make a global joint venture work. Harvard business review, 60(3),
120-127.
92
Kim, J. S., Kaye, J., and Wright, L. K. (2001). Moderating and mediating effects in causal models.
Issues in Mental Health Nursing, 22(1), 63-75.
Klein, S., Frazier, G. L., and Roth, V. J. (1990). A transaction cost analysis model of channel
integration in international markets. Journal of Marketing research, 196-208.
Kogut, B. (1988). Joint ventures: Theoretical and empirical perspectives. Strategic management
journal, 9(4), 319-332.
Kogut, B. (1989). The stability of joint ventures: Reciprocity and competitive rivalry. The Journal
of Industrial Economics, 183-198.
Kogut, B. (1991). Joint ventures and the option to expand and acquire. Management science, 37(1),
19-33.
Kumar, M. V. (2011). Are joint ventures positive sum games? The relative effects of cooperative
and noncooperative behavior. Strategic Management Journal, 32(1), 32-54.
Kumar, N., Scheer, L. K., and Steenkamp, J. B. E. (1995). The effects of perceived
interdependence on dealer attitudes. Journal of marketing research, 348-356.
Lambe, C. J., Spekman, R. E., and Hunt, S. D. (2000). Interimistic relational exchange:
Conceptualization and propositional development. Journal of the Academy of Marketing
Science, 28(2), 212-225.
Lambe, C. J., Spekman, R. E., and Hunt, S. D. (2002). Alliance competence, resources, and
alliance success: conceptualization, measurement, and initial test. Journal of the academy of
Marketing Science, 30(2), 141-158.
Lee, D. Y., and Dawes, P. L. (2005). Guanxi, trust, and long-term orientation in Chinese business
markets. Journal of international marketing, 13(2), 28-56.
Likert, R. (1932). A technique for the measurement of attitudes. Archives of psychology.
López‐Navarro, M. Á., Callarisa‐Fiol, L., and Moliner‐Tena, M. Á. (2013). Long‐Term
Orientation and Commitment in Export Joint Ventures among Small and Medium‐Sized Firms.
Journal of Small Business Management, 51(1), 100-113.
Lu, H., Feng, S., Trienekens, J. H., and Omta, S. W. F. (2012). Network strength, transactionspecific investments, inter-personal trust, and relationship satisfaction in Chinese agri-food
SMEs. China Agricultural Economic Review, 4(3), 363-378.
93
Lu, J. W., and Beamish, P. W. (2006). Partnering strategies and performance of SMEs'
international joint ventures. Journal of Business Venturing, 21(4), 461-486.
Luo, Y. (1999). The structure-performance relationship in a transitional economy: An
empiricalstudy of multinational alliances in China. Journal of Business Research, 46(1), 15-30.
Lusch, R. F., and Brown, J. R. (1996). Interdependency, contracting, and relational behavior in
marketing channels. The Journal of Marketing, 19-38.
Madhok, A. (1995). Revisiting multinational firms' tolerance for joint ventures: A trust-based
approach. Journal of international Business studies, 117-137.
Malhotra, N. K., and Birks, D. F. (2007). Marketing research: An applied approach. Pearson
Education.
Malhotra, N., and Birks, D. (2006). Marketing Research: An Applied Perspective.
Mayer, R. C., Davis, J. H., and Schoorman, F. D. (1995). An integrative model of organizational
trust. Academy of management review, 20(3), 709-734.
Mellat-Parast, M., and Digman, L. A. (2007). A framework for quality management practices in
strategic alliances. Management Decision, 45(4), 802-818.
Moor, R. E. (1971). The Use of Economics in Investment Analysis. Financial Analysts Journal,
27(6), 63-69.
Moorman, C., Zaltman, G., and Deshpande, R. (1992). Relationships between providers and users
of market research: the dynamics of trust within and between organizations. Journal of
marketing research, 29(3), 314.
Morgan, R. M., and Hunt, S. D. (1994). The commitment-trust theory of relationship marketing.
The journal of marketing, 20-38.
Mulaik, S. A., James, L. R., Van Alstine, J., Bennett, N., Lind, S., and Stilwell, C. D. (1989).
Evaluation of goodness-of-fit indices for structural equation models. Psychological bulletin,
105(3), 430.
Nielsen, B. B. (2007). Determining international strategic alliance performance: A
multidimensional approach. International Business Review, 16(3), 337-361.
Nielsen, B. B., and Gudergan, S. (2012). Exploration and exploitation fit and performance in
international strategic alliances. International Business Review, 21(4), 558-574.
94
Ozorhon, B., Arditi, D., Dikmen, I., and Birgonul, M. T. (2008). Effect of partner fit in
international construction joint ventures. Journal of Management in Engineering, 24(1), 12-20.
Park, S. H., and Ungson, G. R. (1997). The effect of national culture, organizational
complementarity, and economic motivation on joint venture dissolution. Academy of
Management journal, 40(2), 279-307.
Park, S. H., and Ungson, G. R. (2001). Interfirm rivalry and managerial complexity: A conceptual
framework of alliance failure. Organization science, 12(1), 37-53.
Peteraf, M. A. (1993). The cornerstones of competitive advantage: A resource‐based view.
Strategic management journal, 14(3), 179-191.
Pfeffer, J., and Nowak, P. (1976). Joint ventures and interorganizational interdependence.
Administrative science quarterly, 398-418.
Pfeffer, J., and Salancik, G. R. (1978). The external control of organizations: A resource
dependence approach. NY: Harper and Row Publishers.
Pisano, G., and Teece, D. J. (1989). COLLABORATIVE ARRANGEMENTS AND GLOBAL
TECHNOLOGY STRATEGY: SOME EVIDENCE FROM THE.
Porter, M. E. (2008). Competitive strategy: Techniques for analyzing industries and competitors.
Simon and Schuster.
Qing, X., and Zhang, W. (2015). Co-opetition and the Stability of Competitive Contractual
Strategic Alliance: Thinking Based on the Modified Lotka-Voterra Model. International
Journal of u-and e-Service, Science and Technology, 8(1), 67-78.
Ring, P. S., and Van de Ven, A. H. (1994). Developmental processes of cooperative
interorganizational relationships. Academy of management review, 19(1), 90-118.
Riordan, M. H., and Williamson, O. E. (1985). Asset specificity and economic organization.
International Journal of Industrial Organization, 3(4), 365-378.
Rumelt, R. P. (1991). How much does industry matter?. Strategic management journal, 12(3),
167-185.
Ryu, S., Park, J. E., and Min, S. (2007). Factors of determining long-term orientation in interfirm
relationships. Journal of Business Research, 60(12), 1225-1233.
Sanchez, R. (2003). Knowledge management and organizational competence. Oxford University
Press.
95
Sarkar, M. B., Echambadi, R., Cavusgil, S. T., and Aulakh, P. S. (2001). The influence of
complementarity, compatibility, and relationship capital on alliance performance. Journal of
the academy of marketing science, 29(4), 358-373.
Schelling, T. C. (1980). The strategy of conflict. Harvard university press.
Seabright, M. A., Levinthal, D. A., and Fichman, M. (1992). Role of individual attachments in the
dissolution of interorganizational relationships. Academy of Management Journal, 35(1), 122160.
Shan, W., Walker, G., and Kogut, B. (1994). Interfirm cooperation and startup innovation in the
biotechnology industry. Strategic management journal, 15(5), 387-394.
Sheth, J. N., and Parvatiyar, A. (1992). Towards a theory of business alliance formation.
Scandinavian International Business Review, 1(3), 71-87.
Sim, A. B., and Ali, M. Y. (2000). Determinants of stability in international joint ventures:
Evidence from a developing country context. Asia Pacific Journal of Management, 17(3), 373397.
Smith, K. G., Carroll, S. J., and Ashford, S. J. (1995). Intra-and interorganizational cooperation:
Toward a research agenda. Academy of Management journal, 38(1), 7-23.
Spender, J. C. (1996). Making knowledge the basis of a dynamic theory of the firm. Strategic
management journal, 17(S2), 45-62.
Steensma, H. K., and Lyles, M. A. (2000). Explaining IJV survival in a transitional economy
through social exchange and knowledge-based perspectives. Strategic Management Journal,
21(8), 831-851.
Suhr, D. (2006). The basics of structural equation modeling. Presented: Irvine, CA, SAS User
Group of the Western Region of the United States (WUSS).
Umukoroa, F. G., Sulaimonb, A. H. A., and Kuyeb, O. L. (2009). Strategic alliance: an insight into
cost of structuring. Serbian Journal of Management, 4(2), 259-272.
Van Dijk, M. L. (2014). Cross-border Collaboration in European-Russian Supply Chains:
Integrative Approach of Provisions on Design, Performance and Impediments. Master thesis.
St. Petersburg: Vysshaya Shkola Menedzhmenta.
Varadarajan, P. R., and Cunningham, M. H. (1995). Strategic alliances: a synthesis of conceptual
foundations. Journal of the Academy of Marketing Science, 23(4), 282-296.
96
Varadarajan, P. R., and Jayachandran, S. (1999). Marketing strategy: an assessment of the state of
the field and outlook. Journal of the Academy of Marketing Science, 27(2), 120-143.
Vyas, N. M., Shelburn, W. L., and Rogers, D. C. (1995). An analysis of strategic alliances: forms,
functions and framework. Journal of business and industrial marketing, 10(3), 47-60.
Wernerfelt, B. (1984). A resource‐based view of the firm. Strategic management journal, 5(2),
171-180.
Whipple, J. M., and Frankel, R. (2000). Strategic alliance success factors. Journal of Supply Chain
Management, 36(2), 21-28.
Wilkinson, I., and Young, L. (2002). On cooperating: firms, relations and networks. Journal of
Business Research, 55(2), 123-132.
Williamson, O. E. (1975). Markets and hierarchies. New York, 26-30.
Williamson, O. E. (1979). Transaction-cost economics: the governance of contractual relations.
The journal of law and economics, 22(2), 233-261.
Williamson, O. E. (1991). Comparative economic organization: The analysis of discrete structural
alternatives. Administrative science quarterly, 269-296.
Wong, A., Tjosvold, D., and Zhang, P. (2005). Developing relationships in strategic alliances:
Commitment to quality and cooperative interdependence. Industrial Marketing Management,
34(7), 722-731.
Wright, L. L., Lane, H. W., and Beamish, P. W. (1988). International management research:
Lessons from the field. International Studies of Management and Organization, 18(3), 55-71.
Wu, W. P., and Choi, W. L. (2004). Transaction cost, social capital and firms' synergy creation in
Chinese business networks: an integrative approach. Asia Pacific Journal of Management,
21(3), 325-343.
Yan, A. (1998). Structural stability and reconfiguration of international joint ventures. Journal of
international business studies, 773-795.
Yan, A., and Luo, Y. (2001). International joint ventures: Theory and practice. ME Sharpe.
Yan, A., and Zeng, M. (1999). International joint venture instability: A critique of previous
research, a reconceptualization, and directions for future research. Journal of international
Business studies, 397-414.
97
Yeung, D. W., and Petrosyan, L. (2006). Dynamically stable corporate joint ventures. Automatica,
42(3), 365-370.
Young Baek, H., Min, S., and Ryu, S. (2006). The effects of agency problems on the stability of
the international joint venture. Multinational Business Review, 14(3), 53-70.
Yu, J. P., and Pysarchik, D. T. (2002). Economic and non-economic factors of Korean
manufacturer-retailer relations. The International Review of Retail, Distribution and Consumer
Research, 12(3), 297-318.
Zaheer, A., and Venkatraman, N. (1995). Relational governance as an interorganizational strategy:
An empirical test of the role of trust in economic exchange. Strategic management journal,
16(5), 373-392.
Zaheer, A., and Venkatraman, N. (1995). Relational governance as an interorganizational strategy:
An empirical test of the role of trust in economic exchange. Strategic management journal,
16(5), 373-392.
Zenkevich N.A., Koroleva A.F., Mamedova Zh.A. (2014). Concept of Joint Venture Stability.
Vestnik of St. Petersburg University. Ser. Management. Iss. 1. P. 28-56.
Zenkevich N.A., Koroleva A.F., Mamedova Zh.A. (2014). Joint Venture Stability Assessment
methodology. Vestnik of St. Petersburg University. Ser. Management. Iss. 3. P. 41-74.
Zenkevich, N. A. (2009). Modelirovanie ustojchivogo sovmestnogo predprijatija. Nauchnie
doklady, 1(R), St. Petersburg: Vysshaya Shkola Menedzhmenta.
Zenkevich, N. A., and Petrosjan, L. A. (2006). Time-consistency of cooperative solutions.
Zenkevich, N. A., Petrosyan, L. A., and Yang, D. V. K. (2009). Dinamicheskie igry i ikh
prilozheniya v menedzhmente (Dynamic Games and Their Applications in Management), St.
Petersburg: Vysshaya Shkola Menedzhmenta.
Zenkevich, N. A., Petrosyan, L. A., and Yang, D. V. K. (2009). Dinamicheskie igry i ikh
prilozheniya v menedzhmente (Dynamic Games and Their Applications in Management), St.
Petersburg: Vysshaya Shkola Menedzhmenta.
Zhao, Y., and Cavusgil, S. T. (2006). The effect of supplier's market orientation on manufacturer's
trust. Industrial Marketing Management, 35(4), 405-414.
98
APPENDICES
Appendix 1: Survey Cover Letter
Dear Participant,
I am a graduate student at Graduate School of Management, Saint-Petersburg State University,
Russia. The school is ranked as a number one business school in Eastern Europe by EdUniversal,
and is listed among 60 best European business schools by Financial Times.
For my Master thesis, I am examining the issue of strategic alliances stability. I am inviting you
to participate in my research by completing the survey. This study will contribute to the theory of
strategic alliance management and will provide practical implications for both management of
companies that are involved in strategic alliances and for strategic alliance managers. I am
conducting the study under the supervision of Zenkevich Nikolay A., Candidate of Mathematical
Science, Associate Professor, Deputy Chair of annual International Conference “Game Theory and
Management”.
The main assumption of the survey: you or your company have experience of dealing with strategic
alliances, e.g. the company you are (were) working in is (was) involved in a strategic alliance as a
partner who forms an alliance, or you personally work(ed) in a strategic alliance itself. At the same
time, there is no specific requirement about the type of a strategic alliance.
The questionnaire will require from 15 to a maximum of 30 minutes to complete. Please, note that
all the responses will be treated confidentially and reported only in form of entities. Nevertheless,
in case you are interested in receiving results of this study, please, feel free to contact me on my
email: st027633@student.spbu.ru.
http://www.surveygizmo.com/s3/2604128/Strategic-Alliances
Thank you for taking the time to assist me in my educational endeavors!
Kind regards,
Anastasiia Reusova
99
Appendix 2: Survey Questionnaire Questions
Section A: Alliance Profile
1. Select the country of your alliance operations
2. Which industry best describes the core activities of your strategic alliance
Aerospace and Aircraft
Oil and Gas Extraction
Business Services
Paper and Allied Products
Chemicals and Allied Products
Petroleum
Coal Mining
Computer and Office Equipment
Prepackaged Software
Computer Integrated Systems Design
Public Administration
Computers, Peripheral Equipment and
Security Systems
Software
Telecommunications
Textile Mill Products
Preparation and Processing
Transportation Equipment
Drugs
Transportation by Air
Electronic and Electrical Equipment
Transportation and Shipping (except
Food and Kindred Products
Health Services
Wholesale Trade-Durable Goods
Investment and Commodity Firms,
Wholesale Trade-Nondurable Goods
Dealers, Exchanges
Other
Computer
Refining
and
Related
Industries
Processing
and
Data
Machinery
Measuring, Medical, Photo Equipment;
air)
Clocks
Metal and Metal Products
100
3. What is the type of alliance you work in/have worked in
Minority equity alliance (a member holds equity in the partner, or partners cross-hold equity
in each other)
Joint venture
Non-equity alliance (does not involve any equity or the transfer of ownership)
4. Select how many full-time employees are involved in work with a strategic alliance in the
organization (in case the alliance is not a Joint Venture)
1-9
10-49
50-249
250 or more
5. Select how many full-time employees work in the strategic alliance (if the alliance is a Joint
Venture)3
1-9
10-49
50-249
250 or more
6. Estimate the period of alliance existence
<1 year
1-3 years
3-5 years
More than 5 years
7. Please, select the type of your alliance involvement
3
Strategic alliance management team
Partner (participant) company management team
In case the alliance is not a Joint Venture, please, select the same option as for the previous question
101
Employed by a strategic alliance (not management team)
Employed by a company that participates in a strategic alliance (not management team)
8. Estimate for how long you have been involved into the alliance activities
<3 years
3-5 years
More than 5 years
Section B: Strategic Alliances Stability Indicators
Please evaluate the following statements about economic performance of your alliance (1Completely Disagree, 2-Disagree, 3-Somewhat Disagree, 4-Neutral, 5-Somewhat Agree, 6-Agree,
7-Completely Agree):
9. Generally, there is a constant improvement in alliance's economic results (code: ES-1)
10. Most often, alliance meets its economic objectives (e.g., revenue, net profit, additional
benefits generated for partner companies, etc.) (code: ES-2)
11. Overall, revenue trend of the alliance can be characterized as rising (code: ES-3)
12. Overall, net profit trend of the alliance can be characterized as rising (code: ES-4)
Please evaluate the following statements about participants' motivation, benefits sharing and
cooperation in your alliance (1-Completely Disagree, 2-Disagree, 3-Somewhat Disagree, 4Neutral, 5-Somewhat Agree, 6-Agree, 7-Completely Agree):
13. There is a well-established procedure of how benefits from the strategic alliance are shared
among participants (code: IS-1)
14. There is a mutual understanding on how the benefits from the strategic alliance should be
shared among alliance participants (code: IS-2)
15. Participants are absolutely satisfied with this form of cooperation compared to other
possibilities (e.g., different forms of cooperation with other companies, such as other alliances)
(code: IS-3)
16. Participants will continue cooperation in this alliance form until the termination date (code:
IS-4)
102
17. Participants are involved in solving alliance issues (code: IS-5)
Section C: Stability Factors Indicators
Please evaluate the following statements about trust in your alliance (1-Completely Disagree, 2Disagree, 3-Somewhat Disagree, 4-Neutral, 5-Somewhat Agree, 6-Agree, 7-Completely Agree):
18. Participants believe that another participant (other participants) is (are) honest (code: T-1)
19. Participants consider each others perspective (code: T-2)
20. Participants are always faithful (code: T-3)
21. Partners found it necessary to be cautious in dealing among themselves (code: T-4, reversed
score)
22. Participant(s) are honest and truthful among themselves (T-5)
23. Participants interact with each other fairly and justly (T-6)
Please evaluate the following statements about participants’ long-term orientation in your alliance
(1-Completely Disagree, 2-Disagree, 3-Somewhat Disagree, 4-Neutral, 5-Somewhat Agree, 6Agree, 7-Completely Agree):
24. Participants believe that over the long-run the alliance will be profitable (code: LTO-1)
25. Maintaining a long-term relationship among the participants is important for them (code:
LTO-2)
26. Participants focus on long-term goals in this alliance (code: LTO-3)
27. Participants believe that any concessions they make to help out among them will even out in
the long run (code: LTO-4)
28. Participants expect working together for a long time (code: LTO-5)
29. Participants are willing to make sacrifices to help out among them from time to time (code:
LTO-6)
103
Please evaluate the following statements about resource complementarity4 in your alliance5 (1Completely Disagree, 2-Disagree, 3-Somewhat Disagree, 4-Neutral, 5-Somewhat Agree, 6Agree, 7-Completely Agree):
30. Together, participants have been adding a substantial value to the alliance (code: RC-1)
31. Alliance participants bring to the table resources and competencies that complement those of
other participants (code: RC-2)
32. Strategic fit between participants could not be better (code: RC-3)
33. All participants contribute different resources that help achieve their mutual goal (code: RC4)
34. Participants have complementary strengths that are useful to their relationship (code: RC-5)
35. Each participant has separate abilities that, when combined together, enable them to achieve
goals beyond their individual reach (code: RC-6)
4
Resource complementarity - the degree to which the joint use of distinct sets of resources produces a higher total
return than the sum of returns that could be achieved if each set of resources were utilized independently.
5
Example: Nestle and Coca-Cola have entered a strategic alliance for cold tea distribution. Nestle owns an attractive
product as a resource, Coca-Cola owns a well-developed distribution network as a complementary resource. In case
they collaborate, they are likely to experience synergetic effects of this collaboration.
104
Appendix 3: Cronbach’s Alpha Test Results
Latent construct: External stability
Item-Total Statistics
Corrected Item-
Squared
Cronbach's
Scale Mean if
Scale Variance
Total
Multiple
Alpha if Item
Item Deleted
if Item Deleted
Correlation
Correlation
Deleted
ES-1
15,10
10,565
,706
,498
,800
ES-2
15,01
12,515
,551
,319
,861
ES-3
15,13
10,728
,734
,614
,788
ES-4
15,39
10,254
,766
,636
,773
Latent construct: Internal stability
Item-Total Statistics
Corrected Item-
Squared
Cronbach's
Scale Mean if
Scale Variance
Total
Multiple
Alpha if Item
Item Deleted
if Item Deleted
Correlation
Correlation
Deleted
IS-1
20,49
17,849
,396
,188
,791
IS-2
21,24
14,604
,509
,278
,765
IS-3
20,66
13,435
,684
,506
,701
IS-4
21,25
13,653
,620
,401
,725
IS-5
20,50
15,267
,622
,425
,728
105
Latent construct: Trust
Item-Total Statistics
Corrected Item-
Squared
Cronbach's
Scale Mean if
Scale Variance
Total
Multiple
Alpha if Item
Item Deleted
if Item Deleted
Correlation
Correlation
Deleted
T-1
23,41
22,810
,652
,503
,753
T-2
23,45
23,891
,636
,473
,760
T-3
24,08
21,434
,691
,646
,742
T-4
24,99
31,268
-,020
,044
,894
T-5
23,72
21,726
,772
,777
,726
T-6
23,58
21,320
,791
,707
,720
Latent construct: Long-term orientation
Item-Total Statistics
Corrected Item-
Squared
Cronbach's
Scale Mean if
Scale Variance
Total
Multiple
Alpha if Item
Item Deleted
if Item Deleted
Correlation
Correlation
Deleted
LTO-1
27,06
23,220
,760
,664
,887
LTO-2
27,11
22,413
,825
,758
,877
LTO-3
27,45
22,309
,782
,637
,883
LTO-4
27,68
23,786
,688
,516
,897
LTO-5
27,22
22,950
,767
,602
,886
LTO-6
27,77
23,954
,632
,420
,906
Latent construct: Resource complementarity
Item-Total Statistics
Corrected Item-
Squared
Cronbach's
Scale Mean if
Scale Variance
Total
Multiple
Alpha if Item
Item Deleted
if Item Deleted
Correlation
Correlation
Deleted
RC-1
26,83
21,131
,488
,345
,844
RC-2
26,84
18,204
,761
,613
,796
RC-3
27,95
17,400
,552
,386
,846
RC-4
27,22
17,829
,703
,536
,805
RC-5
26,90
18,904
,692
,570
,809
RC-6
26,85
19,185
,630
,468
,820
106
Appendix 4: Initial Confirmatory Factor Analysis Output
Regression Weights: (Group number 1 - Default model)
Estimate
S.E.
C.R.
P
ES1 <---
ExternalStability
1,000
ES2 <---
ExternalStability
,710
,099
7,200
***
ES3 <---
ExternalStability
1,102
,102
10,853
***
ES4 <---
ExternalStability
1,169
,106
10,988
***
IS1
<---
InternalStability
1,000
IS2
<---
InternalStability
1,216
,125
9,712
***
IS3
<---
InternalStability
,997
,121
8,248
***
IS4
<---
InternalStability
,839
,104
8,068
***
IS5
<---
InternalStability
,567
,092
6,177
***
LTO1 <---
LongtermOrientation
,898
,070
12,917
***
LTO2 <---
LongtermOrientation
1,000
LTO3 <---
LongtermOrientation
,939
,070
13,441
***
LTO4 <---
LongtermOrientation
,786
,073
10,789
***
LTO5 <---
LongtermOrientation
,899
,067
13,394
***
LTO6 <---
LongtermOrientation
,799
,081
9,830
***
RC3 <---
ResourceComplementarity
1,000
RC4 <---
ResourceComplementarity
1,141
,144
7,950
***
RC5 <---
ResourceComplementarity
,882
,118
7,494
***
RC6 <---
ResourceComplementarity
,819
,117
6,990
***
T1
<---
Trust
1,000
T3
<---
Trust
,737
,086
8,609
***
T5
<---
Trust
1,116
,093
12,032
***
T6
<---
Trust
1,183
,094
12,526
***
107
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
ES1
<---
ExternalStability
,742
ES2
<---
ExternalStability
,569
ES3
<---
ExternalStability
,854
ES4
<---
ExternalStability
,872
IS1
<---
InternalStability
,703
IS2
<---
InternalStability
,849
IS3
<---
InternalStability
,685
IS4
<---
InternalStability
,679
IS5
<---
InternalStability
,523
LTO1 <---
LongtermOrientation
,792
LTO2 <---
LongtermOrientation
,875
LTO3 <---
LongtermOrientation
,811
LTO4 <---
LongtermOrientation
,703
LTO5 <---
LongtermOrientation
,808
LTO6 <---
LongtermOrientation
,662
RC1
<---
ResourceComplementarity
,566
RC2
<---
ResourceComplementarity
,812
RC3
<---
ResourceComplementarity
,601
RC4
<---
ResourceComplementarity
,771
RC5
<---
ResourceComplementarity
,779
RC6
<---
ResourceComplementarity
,705
T1
<---
Trust
,761
T2
<---
Trust
,761
T3
<---
Trust
,672
T4
<---
Trust
,742
T5
<---
Trust
,880
T6
<---
Trust
,893
108
Appendix 5: Modification Indices
Regression Weights: (Group number 1 - Default model)
M.I.
Par Change
IS1
<---
T4
5,994
-,131
IS1
<---
T5
4,462
-,128
IS1
<---
T3
5,520
-,160
IS2
<---
LongtermOrientation
7,267
-,195
IS2
<---
T2
4,446
-,116
IS2
<---
LTO4
4,180
-,127
IS2
<---
LTO3
6,090
-,147
IS2
<---
LTO5
7,284
-,168
IS2
<---
LTO2
5,976
-,148
ES1
<---
InternalStability
4,875
,174
ES1
<---
Trust
6,010
,178
ES1
<---
IS4
9,740
,185
ES1
<---
IS3
4,868
,113
ES1
<---
LTO1
6,121
,156
ES1
<---
T6
4,184
,109
ES1
<---
T5
8,518
,158
ES1
<---
T1
6,349
,134
ES3
<---
IS2
4,715
-,092
ES3
<---
T1
4,674
-,095
ES4
<---
IS4
5,940
-,122
ES4
<---
LTO5
4,317
-,113
LTO1 <---
ResourceComplementarity
5,993
,168
LTO1 <---
ExternalStability
4,610
,133
LTO1 <---
RC2
7,296
,143
LTO1 <---
RC5
5,188
,124
LTO1 <---
ES3
4,786
,099
LTO1 <---
ES1
6,827
,113
LTO2 <---
Trust
4,151
-,101
LTO2 <---
T4
11,014
-,108
LTO2 <---
IS3
6,172
-,087
109
M.I.
Par Change
LTO2 <---
T5
8,109
-,105
LTO5 <---
RC3
5,893
-,090
LTO5 <---
ES4
4,246
-,085
LTO6 <---
T4
8,155
,135
LTO6 <---
T2
4,321
,118
LTO6 <---
RC1
4,518
,160
LTO6 <---
IS4
5,322
-,136
LTO6 <---
T3
9,609
,187
IS5
<---
ResourceComplementarity
9,052
,261
IS5
<---
RC2
16,288
,271
IS5
<---
RC1
16,263
,307
IS5
<---
RC3
4,822
,108
IS5
<---
T3
4,374
,128
RC1
<---
LongtermOrientation
11,255
,213
RC1
<---
Trust
5,507
,144
RC1
<---
ExternalStability
9,365
,203
RC1
<---
RC2
4,368
,119
RC1
<---
IS5
8,583
,164
RC1
<---
T4
5,681
,096
RC1
<---
LTO4
4,081
,110
RC1
<---
LTO3
14,057
,196
RC1
<---
LTO1
4,578
,114
RC1
<---
LTO6
11,203
,168
RC1
<---
LTO5
11,697
,187
RC1
<---
LTO2
4,964
,118
RC1
<---
T3
6,646
,133
RC1
<---
T1
7,828
,126
RC1
<---
ES4
6,518
,119
RC1
<---
ES3
8,699
,143
RC1
<---
ES1
5,575
,109
RC2
<---
IS5
6,720
,129
RC2
<---
T4
8,996
-,107
110
M.I.
Par Change
RC2
<---
RC1
10,690
,186
RC2
<---
IS3
8,171
-,109
RC2
<---
T6
4,017
-,080
RC3
<---
T4
8,936
,182
RC4
<---
Trust
5,786
,153
RC4
<---
T4
5,447
,098
RC4
<---
T2
5,257
,115
RC4
<---
RC1
5,589
-,158
RC4
<---
IS3
8,756
,133
RC4
<---
RC3
6,264
,108
RC4
<---
T6
5,727
,112
RC4
<---
T5
4,021
,095
RC4
<---
T3
5,024
,120
RC5
<---
RC3
4,266
-,077
RC5
<---
RC6
4,606
,109
RC5
<---
ES1
4,790
-,092
RC6
<---
LongtermOrientation
4,110
-,130
RC6
<---
T4
7,693
-,113
RC6
<---
LTO3
4,588
-,114
RC6
<---
T5
7,039
-,123
RC6
<---
T1
4,322
-,095
RC6
<---
ES1
5,010
-,105
IS3
<---
Trust
9,566
,258
IS3
<---
ExternalStability
4,605
,194
IS3
<---
T4
15,275
,214
IS3
<---
LTO5
5,992
,182
IS3
<---
T6
7,589
,169
IS3
<---
T5
11,003
,206
IS3
<---
T3
6,285
,176
IS3
<---
ES1
4,955
,140
IS4
<---
LongtermOrientation
4,454
,158
IS4
<---
LTO4
4,385
,134
111
M.I.
Par Change
IS4
<---
LTO3
5,167
,140
IS4
<---
RC6
5,348
-,154
IS4
<---
LTO5
8,535
,188
IS4
<---
LTO2
4,187
,128
T1
<---
IS4
9,833
,181
T1
<---
RC3
9,425
-,147
T2
<---
LongtermOrientation
9,157
,205
T2
<---
LTO3
8,000
,158
T2
<---
LTO1
16,024
,229
T2
<---
LTO6
9,024
,161
T2
<---
LTO2
11,496
,193
T2
<---
T3
6,370
,139
T3
<---
T2
4,687
,118
T3
<---
LTO6
8,136
,160
T4
<---
RC2
7,747
-,209
T4
<---
IS1
5,192
-,131
T4
<---
RC6
7,375
-,202
T4
<---
LTO1
8,614
-,207
T4
<---
T5
5,106
,136
T4
<---
LTO2
8,299
-,202
T5
<---
T4
12,266
,128
T5
<---
T2
4,567
-,094
T5
<---
RC6
4,296
-,106
T5
<---
LTO2
6,092
-,119
T6
<---
RC6
6,411
,128
112
Appendix 6: Final Confirmatory Factor Analysis Output
Regression Weights: (Group number 1 - Default model)
Estimate
S.E.
C.R.
P
ES1
<---
ExternalStability
1,000
ES2
<---
ExternalStability
,709
,099
7,175
***
ES3
<---
ExternalStability
1,106
,102
10,843
***
ES4
<---
ExternalStability
1,171
,107
10,966
***
IS1
<---
InternalStability
1,000
IS2
<---
InternalStability
1,135
,115
9,852
***
IS3
<---
InternalStability
1,011
,127
7,968
***
IS4
<---
InternalStability
,795
,097
8,221
***
IS5
<---
InternalStability
,521
,086
6,058
***
LTO1 <---
LongtermOrientation
,873
,064
13,589
***
LTO2 <---
LongtermOrientation
1,000
LTO3 <---
LongtermOrientation
1,001
,078
12,807
***
LTO4 <---
LongtermOrientation
,834
,080
10,481
***
LTO5 <---
LongtermOrientation
,955
,075
12,714
***
LTO6 <---
LongtermOrientation
,842
,088
9,534
***
RC3
<---
ResourceComplementarity
1,000
RC4
<---
ResourceComplementarity
1,234
,155
7,963
***
RC5
<---
ResourceComplementarity
,770
,107
7,204
***
RC6
<---
ResourceComplementarity
,686
,107
6,419
***
T1
<---
Trust
1,000
T3
<---
Trust
,738
,087
8,461
***
T5
<---
Trust
1,116
,093
11,963
***
T6
<---
Trust
1,181
,094
12,501
***
113
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
ES1
<---
ExternalStability
,741
ES2
<---
ExternalStability
,568
ES3
<---
ExternalStability
,855
ES4
<---
ExternalStability
,871
IS1
<---
InternalStability
,736
IS2
<---
InternalStability
,829
IS3
<---
InternalStability
,736
IS4
<---
InternalStability
,672
IS5
<---
InternalStability
,491
LTO1 <---
LongtermOrientation
,738
LTO2 <---
LongtermOrientation
,840
LTO3 <---
LongtermOrientation
,828
LTO4 <---
LongtermOrientation
,717
LTO5 <---
LongtermOrientation
,824
LTO6 <---
LongtermOrientation
,667
RC3
<---
ResourceComplementarity
,613
RC4
<---
ResourceComplementarity
,932
RC5
<---
ResourceComplementarity
,662
RC6
<---
ResourceComplementarity
,573
T1
<---
Trust
,765
T3
<---
Trust
,647
T5
<---
Trust
,867
T6
<---
Trust
,905
114
Appendix 7: Structural Model Output
Regression Weights: (Group number 1 - Default model)
Estimate
S.E.
C.R.
P
ResourceComplementarity
,570
,113
5,071
***
LongtermOrientation <---
Trust
,563
,078
7,184
***
ExternalStability
<---
LongtermOrientation
,437
,090
4,828
***
ExternalStability
<---
ResourceComplementarity
,066
,099
,670
,503
InternalStability
<---
Trust
,184
,058
3,193
,001
InternalStability
<---
LongtermOrientation
,014
,054
,260
,795
InternalStability
<---
ResourceComplementarity
,248
,064
3,858
***
InternalStability
<---
ExternalStability
,096
,046
2,093
,036
ES1
<---
ExternalStability
1,000
ES2
<---
ExternalStability
,710
,100
7,129
***
ES3
<---
ExternalStability
1,114
,103
10,792
***
ES4
<---
ExternalStability
1,176
,108
10,886
***
IS1
<---
InternalStability
1,882
,308
6,110
***
IS2
<---
InternalStability
2,159
,330
6,532
***
IS3
<---
InternalStability
1,875
,309
6,065
***
IS5
<---
InternalStability
1,000
LTO1
<---
LongtermOrientation
,874
,064
13,572
***
LTO2
<---
LongtermOrientation
1,000
LTO3
<---
LongtermOrientation
1,005
,079
12,781
***
LTO4
<---
LongtermOrientation
,837
,080
10,465
***
LTO5
<---
LongtermOrientation
,958
,076
12,667
***
LTO6
<---
LongtermOrientation
,844
,089
9,505
***
RC2
<---
ResourceComplementarity
1,000
RC4
<---
ResourceComplementarity
1,114
,113
9,867
***
RC5
<---
ResourceComplementarity
1,048
,100
10,521
***
RC6
<---
ResourceComplementarity
,968
,102
9,465
***
SS2
<---
InternalStability
1,506
,253
5,944
***
T1
<---
Trust
1,000
T3
<---
Trust
,736
,087
8,498
***
T5
<---
Trust
1,106
,092
11,967
***
T6
<---
Trust
1,175
,094
12,560
***
Trust
<---
115
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
Trust
<---
ResourceComplementarity
,448
LongtermOrientation <---
Trust
,603
ExternalStability
<---
Trust
,206
ExternalStability
<---
LongtermOrientation
,319
ExternalStability
<---
ResourceComplementarity
InternalStability
<---
ExternalStability
,168
InternalStability
<---
Trust
,350
InternalStability
<---
LongtermOrientation
,029
InternalStability
<---
ResourceComplementarity
,376
ES1
<---
ExternalStability
,741
ES2
<---
ExternalStability
,567
ES3
<---
ExternalStability
,855
ES4
<---
ExternalStability
,872
IS1
<---
InternalStability
,733
IS2
<---
InternalStability
,835
IS3
<---
InternalStability
,723
IS4
<---
InternalStability
,673
IS5
<---
InternalStability
,499
LTO1
<---
LongtermOrientation
,736
LTO2
<---
LongtermOrientation
,839
LTO3
<---
LongtermOrientation
,828
LTO4
<---
LongtermOrientation
,718
LTO5
<---
LongtermOrientation
,825
LTO6
<---
LongtermOrientation
,667
RC2
<---
ResourceComplementarity
,765
RC4
<---
ResourceComplementarity
,766
RC5
<---
ResourceComplementarity
,822
RC6
<---
ResourceComplementarity
,736
T1
<---
Trust
,768
T3
<---
Trust
,646
T5
<---
Trust
,865
T6
<---
Trust
,903
-,006
116
Отзывы:
Авторизуйтесь, чтобы оставить отзыв